report
stringlengths 320
1.32M
| summary
stringlengths 127
13.7k
|
|---|---|
Over the last decade, the Army has begun to transform its warfighting capabilities to more effectively counter a broad and complex set of potential threats. According to Army officials, the transformation is the most comprehensive change in the Army in over a century, and will affect all aspects of its organizations, training, doctrine, leadership, and strategic plans as well as its acquisitions. Through this transformation, the Army expects to establish a force that provides both the lethality and survivability of today’s heavily armored units and the deployability and responsiveness of today’s lighter combat units. As envisioned, the future force will operate very differently than forces have in the past. It will function in smaller, more agile and deployable modular brigade combat teams (composed of roughly 3,000 to 4,000 personnel) that can react quickly to changing missions and circumstances. To be effective, force components—soldiers, platforms, weapons, and sensors—must be “net- centric,” that is, closely linked and able to operate seamlessly together. The transformation involves two major, interrelated acquisitions: (1) development of new advanced communications and networking systems— computers, software, and a wireless tactical internet—to acquire, exchange, and employ timely information throughout the battlespace and (2) development of a new generation of battlefield vehicles, weapons, and sensors. The Army has taken initial steps toward transformation through its Digitization and Stryker programs. Under the Digitization program, the Army installed computers, software, and interfaces to communications systems on Abrams tanks, Bradley fighting vehicles, and other vehicles in selected units that enable both in-theater and higher commands to share battlefield data with lower-level units. The Stryker program introduced a new family of vehicles expected to make units more lethal, mobile, and survivable than today’s light forces. In addition, the Army has initiated a major restructuring of its force into modular brigade combat teams— brigade-sized units that will have a common organizational design. FCS is the culminating stage in the Army’s ongoing transformation to a lighter, more agile and capable force. It is a large and complex development effort to provide a networked family of weapons and other systems for the future force. Establishing reliable, robust communications and networking capabilities is essential to FCS. Without these capabilities, the lighter, more decentralized units would be vulnerable to enemy attack. Currently, the armed forces have limited communications and networking capabilities on the battlefield, making it necessary to patch together or reroute information through multiple radio, data terminal, and network systems to get critical information to the warfighter and commanders. Current “dial-up speed” data rates further delay forces’ ability to identify, assess, and respond to time-critical targets. FCS’s networked on-the-move communications for voice, data, video, and imagery are expected to be a revolutionary improvement over current communications capabilities (see table 1). The FCS communications and networking capabilities are being designed around five components: Platforms and sensors: Under FCS, the Army is developing new warfighting systems, including manned and unmanned aerial and ground vehicles that will provide and use intelligence, surveillance, and reconnaissance information. Applications: Software applications will support battlefield command functions, including command and control, logistics support, training, and modeling and simulation. Network services: SOSCOE will be the network-centric operating system, or middleware, that enables the integration of separate FCS communications software packages, independent of their location and the technology used to develop them. The Army likens the SOSCOE architecture to Microsoft Windows, but many times larger. SOSCOE represents about 10 percent of the more than 30 million lines of FCS software code. Transport systems: Transport systems—primarily JTRS and WIN-T— will provide wireless communication capabilities to transport information within the FCS network and the broader DOD-wide network. Standards: Standards implement DOD-wide policies and doctrine developed by offices such as the Office of the Secretary of Defense, Networks and Information Integration, and the Joint Chiefs of Staff. Two critical objectives of these standards are net-centric operations and inter-service interoperability. Figure 1 shows a representation of the five FCS network components. JTRS is a software-reprogrammable radio that is intended to operate with many different legacy radio systems and provide the warfighter with additional communications and networking capabilities—including seamless interoperability and increased data throughput—to simultaneously access maps and other visual data, communicate via voice and video with other units and levels of command, and obtain information directly from battlefield sensors. A key component of JTRS is developing waveforms to operate with legacy radios as well as new waveforms to provide advanced networking capabilities, such as the Wideband Networking Waveform. The Wideband Networking Waveform represents a new, critical capability for DOD. The development of the Wideband Networking Waveform is intended to address many of the current limitations associated with DOD tactical wireless networking, including line-of-sight limitations that cause many network partitions, unique network monitoring systems, and predefined security enclaves that require hardware for each security level. The waveform is expected to provide data rates of 5 megabits per second or more-–hundreds of times faster than existing communications systems-–and facilitate the routing of large amounts of information among users anywhere in the battlespace. DOD has structured the JTRS development effort into several programs clustered by requirements. The JTRS Cluster 1 program is developing radios for ground vehicles and helicopters to equip the current force as well as FCS. The program is expected to cost $15.6 billion to develop and acquire over 100,000 Cluster 1 radios. The JTRS Cluster 5 program is developing handheld and manpack radios for soldiers as well as several smaller varieties of radios for use in weight- and power-constrained platforms—such as Unattended Ground Sensors and Intelligent Munitions Systems. The program is expected to cost $8.5 billion to develop and acquire over 300,000 Cluster 5 radios. The WIN-T program is developing communications equipment that supports an expanded area of battlefield operations and interfaces with JTRS radios to connect warfighters and command centers, including joint, allied, and coalition forces, providing commanders with access to on-the- move communications—that is, continuously updated, real-time multimedia information from dispersed locations throughout the theater. It will replace existing communications networks that have limited capacity to support on-the-move communications. Leveraging advanced commercial technologies that enable mobile communications, the WIN-T system includes data routing and switching hardware, computers, video and teleconferencing equipment, high-capacity line-of-sight radios and satellite terminals—all of which make up a tactical operation center’s communications element. WIN-T is being developed in three blocks, with each block adding capabilities. Based on current plans, Block 1 is projected to cost approximately $10 billion; Blocks 2 and 3 have yet to be funded. The SOSCOE software will reside within each FCS platform’s integrated computer system and provide a number of services for the users of the integrated computer system. These services include interoperability services, information assurance services, and communications services. SOSCOE will enable integrated management of the network and will allow systems within the network to access sources of information. The Army estimates that SOSCOE software development will be completed in 2011. The Army plans to field the SOSCOE software in increments to align with the overall FCS software builds and planned FCS spirals. When FCS began system development in May 2003, the JTRS and WIN-T programs were under way with schedules that aligned with FCS planned fielding. However, the Army restructured the FCS program in July 2004 to address development risks. The restructuring added 4 years to develop the platform systems and established an evaluation unit to demonstrate FCS capabilities. Even though the restructuring provided additional time to the program, it also emphasized developing FCS capabilities in spirals and accelerating the development of the network into the current force. The Army now plans to test and field its FCS capabilities incrementally between 2008 and 2014 through four spirals. A 2-year period of testing will precede the actual fielding of capabilities in each spiral. The Army has defined the initial spiral of FCS around the capabilities needed by the current force, to include the main components of the communications network--JTRS Cluster 1 and 5 radios and the wideband waveforms, some form of WIN-T communications capability, and SOSCOE. The capabilities for the other FCS spirals will be defined over time. Figure 2 shows the FCS spirals’ timeline. The JTRS Cluster 1 program began system development and demonstration in 2002 with an aggressive schedule, immature technologies, and a lack of clearly defined and stable requirements. These factors have contributed to significant cost and schedule problems that the program has not recovered from. The program has not been able to mature the technologies needed to produce radios that generate sufficient power as well as meet platform size and weight constraints and has been forced to make design changes to accommodate evolving security requirements. Because of cost, schedule, and performance problems, in December 2004, the Army proposed restructuring the program by adding $458 million and 24 months to the development schedule. However, recently DOD directed that work on the Cluster 1 radios be stopped while an assessment is completed to determine the future of the program. In addition, because of increased concern about the contractor’s ability to develop the radios, the Army notified the contractor that it was considering contract termination. At this point it is not clear what the outcome will be and what impact this will have on the future of the program. As a result, it is unlikely JTRS Cluster 1 radios will be available for the first FCS network spiral, slated to begin in fiscal year 2008. FCS and other users dependent on Cluster 1 radios, such as Army helicopters, will have to rely on legacy radios to fill the gap. Prior to the start of system development in 2002, the JTRS Cluster 1 schedule was accelerated 27 months to meet the Army’s plan to modernize its helicopters with various technological upgrades including advanced communications. Cluster 1 proceeded into the system development and demonstration phase with none of the program’s 20 critical technologies sufficiently matured and with requirements not clearly defined—contrary to best practices and DOD guidance. Although many of the technologies had been used in other radio applications, significant technical advances were nonetheless required for developing key components of the radio. The program’s acquisition strategy, for example, highlighted technology risks associated with the following requirements: Wideband Networking Waveform: As the core of the JTRS networking capability, the Wideband Networking Waveform is to operate across a wide range of radio frequency spectrum, 2 megahertz (MHz) to 2 gigahertz (GHz), and provide increased routing and networking capabilities. The Wideband Networking Waveform must also be compliant with the Software Communications Architecture, which demands a modular approach to waveform design, imposing much greater processing and memory requirements. This is especially critical for FCS, as the waveform is to provide what has been called the backbone or main conduit of the FCS network. Security: The JTRS radio set is intended to operate applications at multiple levels of security. For it to do so, developers not only have to be concerned with traditional radio security issues but also must be prepared to implement the features required for network and computer security. This will require development of new technologies, obtaining certification through a rigorous process by the National Security Agency, and accommodating an expected growth in security requirements. Interference mitigation: Prior to JTRS, tactical radios were largely designed for single channel and single band operations. Because JTRS radio sets will operate multiple channels—-as many as eight channels—-simultaneously within the same radio set, developers must ensure that communications over one channel do not interfere with communications over another, because such interference would degrade the quality of service and limit the radio’s high data rate capability. The accelerated acquisition strategy compressed the development cycle and allowed little time for testing prior to key development decisions. For example, the schedule called for making the initial production decision for selected platforms immediately following an early operational assessment of a partially functioning prototype of the JTRS radio in surrogate vehicles (see fig. 3). This is in contrast to the knowledge-based approach captured in best practices, which advocates making production decisions based on an assessment of production-representative prototypes in a realistic environment. Historically, programs that must define requirements, develop technology, and design products concurrently have experienced cost increases and schedule delays. While the Army recognized the risk of moving forward with immature technologies, it expected that emerging technologies in radio software technology would enable it to develop the critical technologies and integrate them into the product quickly. Despite the Army’s expectations to leverage current and emerging radio technologies, the critical technologies for the JTRS Cluster 1 radio have generally not matured. The program is also struggling to derive detailed specifications for Cluster 1 requirements. Despite the lack of mature technologies and detailed specifications, the Army held the program’s critical design review—the point at which design stability is to be achieved and demonstrated—in December 2003. However, with the requirements still evolving, the program expects to make several costly hardware and software design modifications. For example: The current processing and memory capacity of the Network INFOSEC Unit, which contains the operating software, is insufficient to support full systems operation, including waveform processing, enhanced security, and power management. The program plans to double the Network INFOSEC Unit’s capacity from 256 megabytes of memory to 512 megabytes, which will require changes to the hardware design. The National Security Agency has recently determined that the current design is not sufficient to meet security requirements to operate in an open networked environment. Specifically, particular versions of JTRS radios will be used by allied and coalition forces, requiring the Army to release specific source code of the software architecture to these forces. To address the release, the National Security Agency has required changes to the security architecture. While the program has not finalized or funded the changes, the current plan is to separate the networking and radio functions into two separate processors. A key technical challenge in developing the Cluster 1 radio is meeting the size, weight, and power requirements for ground vehicles and helicopters. To realize the full capabilities of the Wideband Networking Waveform, including transmission range, the Cluster 1 radio requires significant amounts of memory and processing power, which add to the size, weight, and power consumption of the radio. The added size and weight are the result of efforts to ensure electronic parts in the radio are not overheated by the electricity needed to power the additional memory and processing. Thus far, the program has not been able to develop radios that meet size, weight, and power requirements, and the current projected transmission range is only 3 kilometers—well short of the 10-kilometer range required for the Wideband Networking Waveform. As a consequence, more unmanned aerial vehicles may be needed to relay information. Intended ground vehicle users have accepted a deviation in the design—to have some of the radio’s hardware mounted separately outside the vehicle— with the expectation that the contractor will develop a better solution later on. However, deviations were not accepted for the helicopters because it would necessitate major design changes to the aircraft and adversely affect the aircraft modernization schedules. Unlike ground vehicles, aviation platforms are limited in their ability to compromise on size, weight, and power issues because of the difficulty in maintaining equilibrium while airborne. The Cluster 1 radio’s size, weight, and peak power consumption exceeds helicopter platform requirements by as much as 80 percent (see fig. 4). To meet the JTRS size, weight, and power requirements and realize the full capabilities of the Wideband Networking Waveform, significant technology advances in power amplification and cooling are essential. The Army has initiated science and technology development efforts to address these issues, but it will take time to evolve the technologies to an acceptable level of maturity. In addition to conducting other research, the Army is evaluating technologies associated with a communications and navigation system that was being developed as part of the Comanche helicopter program. The Army approved further development of this system and plans to integrate it into the JTRS system and conduct a demonstration of its capabilities later this year. However, the Army will not be able to deliver Cluster 1 radios to support the helicopter fielding schedules and will have to purchase legacy radios instead. The FCS program is exploring solutions to meet a key transportability requirement that FCS vehicles must be limited to 19 tons in order to be airlifted by a C-130 transport aircraft. To meet this transportability requirement, the program recently proposed significant size and weight reductions for vehicle components, including communications equipment. While Cluster 1 currently has no size, weight, and power requirements for the systems to be fielded in FCS, the JTRS radios may require further redesign to meet FCS’s aggressive weight requirements. Such a reduction would likely have a significant impact on the design of JTRS radios for the FCS vehicles. Since the program entered systems development, in 2002, the contractor has overrun cost estimates by $93 million-–nearly 28 percent above what was planned (see app. II). Although the program attempted to stabilize costs by adding approximately $200 million to the contract in January 2004, costs continued to grow steadily thereafter. In addition, the contractor has increasingly fallen behind schedule and has had to devote more resources than originally planned. In January 2005, the prime contractor estimated that the total costs for the Cluster 1 radio and waveform development would be $531 million more than what was originally budgeted, reaching about $898 million at completion. However, according to program officials, since contract award, the prime contractor has not demonstrated strong cost estimating and cost management techniques, and it is difficult to estimate with any confidence what the overall program is likely to cost. Key issues driving the cost growth are unanticipated complexity associated with developing the hardware, Wideband Networking Waveform, and other software. As a result, the unit costs for early prototypes have increased from the prime contractor’s original proposal. According to one DOD official, until the requirements’ specifications are stabilized, cost and schedule problems are likely to continue. For example, according to the Defense Contract Management Agency, meeting the design changes for security requirements is expected to cost an estimated $80 million. In light of the technical problems and cost growth, the Army in December 2004 delayed the initial production decision, which was scheduled for the third quarter of fiscal year 2005, and proposed to add $458 million and 24 months to the program. Before carrying out this restructure, the Office of the Secretary of Defense directed the Army in January 2005 to stop work on portions of the Cluster 1 development and focus on preparing for an early operational assessment of the radio, which was intended to test the basic functionality of pre-engineering development models of the radio. In April 2005, however, the Army suspended the operational assessment and notified the contractor that it was considering contract termination. This action was taken based on initial findings of an assessment of the Cluster 1 program conducted by a newly established JTRS Joint Program Executive Office, which concluded that the current program structure is not executable and the contractor’s ability to develop the radio is questionable. At this point it is not clear whether the contract will be terminated and what impact a termination would have on the future of the program. The Joint Program Executive Office is expected to complete its assessment of the program, and a Defense Acquisition Board review will be held at the end of fiscal year 2005 to determine the future of the program. Program officials anticipate a new program acquisition strategy will evolve, with greater emphasis on developing the radio in blocks. If development resumes, it is anticipated that there will be start-up delays—3 to 12 months, according to agency officials—associated with restaffing the contractor’s development team and bringing the team up the learning curve. Adding to the program’s uncertainty is the impact of pending requirements on program cost and schedule. According to agency officials, the program will likely be tasked with new requirements from key stakeholders. For example: To meet FCS requirements for accessing intelligence, surveillance, and reconnaissance data on the battlefield, FCS will need a new network data link operating in the radio frequency range above 2 GHz. According to the Army, developing the new network data link is expected to cost approximately $170 million. Furthermore, additional costs are likely because the new network data link may require changes to the already challenging JTRS Cluster 1 radio design—which operates over a large 2 MHz to 2 GHz range—to operate at an even higher frequency. An analysis of alternatives is currently under way to determine how best to meet this requirement. According to FCS officials, a decision on the new network data link is needed by the end of the year to keep the FCS program on track. To comply with the standards of the Global Information Grid, DOD has directed all systems to transition to the use of Internet Protocol Version 6 in the future. Cluster 1, which has been designed with Version 4, not only will need to upgrade but will need additional hardware and software to ensure Version 4 and Version 6 systems can interoperate. Reconciling security requirements for Version 6 is also expected to be a challenge. Given the many program uncertainties, it is unlikely that JTRS radios will be available to support intended users: the first increment of the FCS network slated for fiscal year 2008, Stryker Brigade Combat Team ground vehicles, and helicopters. The Army plans to purchase legacy radios, which have limited capabilities, for the Stryker Brigade Combat Teams and helicopters. According to Army officials, FCS is planning to experiment with early prototypes of JTRS radios and the Wideband Networking Waveform, but they will not know when the fully capable Cluster 1 radios would be available until after the program is restructured at the end of fiscal year 2005. In addition, because of ongoing military operations in Afghanistan and Iraq, the Army has purchased a large number of legacy radios over the past few years. The fielding of so many new radios to the current force may call into question the affordability of replacing them prematurely with JTRS sets. The Army is assessing JTRS fielding plans in light of the additional investments in legacy radios and JTRS Cluster 1 cost, schedule, and technical problems. As with the Cluster 1 program, radio size, weight, power, and data- processing requirements have presented significant technical challenges for the JTRS Cluster 5 program, which is developing a series of radios much smaller than those for the Cluster 1 program. Several programmatic changes and a contract award bid protest have contributed to disruptions in the progress of the Cluster 5 program. As a result, the Cluster 5 program is no longer synchronized with the FCS program. The Army is currently assessing the feasibility of accelerating the development of selected small form Cluster 5 radios. However, in light of the unresolved technical issues with the Cluster 1 program, the JTRS Joint Program Executive Office has initiated an assessment to restructure the Cluster 5 program into increments. In the event that Cluster 5 radios are not available, the Army plans to use surrogate radios for the initial spiral of FCS. In addition, users depending on the Cluster 5 radios, such as the Army’s Land Warrior program, have decided to move forward with surrogate radios. Meeting requirements for Cluster 5 radios is even more challenging than for Cluster 1 because of their smaller size, weight, power, and large data- processing requirements. For example, a one-channel handheld version of the Cluster 5 radios has a maximum weight specification of 2 pounds and a volume of 40 cubic inches (see table 2). A two-channel manpack radio has weight and volume of 9 pounds and 400 cubic inches, respectively. A one- channel small form radio weighs about 1 pound and occupies 40 cubic inches. In comparison, a Cluster 1 two-channel radio weighs 84 pounds and occupies 1,732 cubic inches. Despite their extreme size and weight limitations, Cluster 5 radios are still required to store multiple waveforms. For instance, manpack radios will be required to store at least 10 waveforms, handheld sets 6 waveforms, and the small form sets 2 waveforms. The Cluster 5 program began system development and demonstration with immature technologies, especially those related to the handheld and smaller variants because of the limited size, weight, and power allowances (see fig. 5). According to the Army, the requirements for two-channel small form radios—wideband radio frequency capabilities up to 2500 MHz, thermal management and packaging, and complex security architecture— all introduce unique technological challenges. Cluster 5 program officials had expected to leverage technology from the Cluster 1 program. However, the Cluster 1 technologies have not matured as anticipated. Program officials stated that backup technology will be identified as a part of a risk mitigation plan. The JTRS Cluster 5 program has identified six critical technologies as follows: Microelectronics: Microelectronics addresses the processes for producing and packaging the electronic circuits and systems that make up the Cluster 5 radios. Miniaturization technology and microelectronics components are critical to the feasibility of Cluster 5 radios because of their extremely small size. Environmental protection: Environmental protection describes the technologies, tools, or design considerations necessary to protect the radios from potentially harsh effects of the operational environment, including, for example, lightning, short-duration force impacts, or radioactive contaminants. Power management: One of the greatest challenges in designing and implementing the Cluster 5 radios is the management and conservation of the limited amount of available battery power. Power management refers to the set of technologies that facilitate a reduction in energy consumption or an increase in battery capacity with the goal of obtaining longer operating time and a reduced battery size and weight. Multichannel architecture: Multichannel JTRS radios are required to provide multiple, independent channels to simultaneously transmit and receive information using different waveforms. The compact size of the Cluster 5 radios and requirement for simultaneous multichannel operation present a co-site interference mitigation challenge. Antennas: Cluster 5 JTRS radios are required to transmit and receive multiple waveforms over the large frequency range 2 MHz to 2.5 GHz and are further required to transmit and receive two separate waveforms simultaneously with a maximum of three antennas. The requirements impose unique technical challenges for both antenna and radio designs. Security: Cluster 5 security framework must support Multiple Single Levels of Security to allow the processing of information with different classifications and categories. It also must support an over-the-air download capability of waveforms, which will entail large software files. It has yet to be demonstrated in a relevant environment. The Cluster 5 radios are required to store and operate the Wideband Networking Waveform. This will provide high data rates and networking capabilities for mobile forces. The full Wideband Networking Waveform requires significant amounts of memory and processing power, which may not be available for the Cluster 5 radios. According to the program office, the principal challenge in operating the Wideband Networking Waveform on Cluster 5 radios stems from the significantly smaller size, weight, and power requirements when compared with those for Cluster 1, as well as safety and heat considerations for the soldier. Because of the difficulties in overcoming these challenges, the Cluster 5 program is seeking to ease the waveform’s requirements and reduce the power demands of the software. The Cluster 5 program is also developing another new, wideband waveform called the Soldier Radio Waveform. Although less powerful than the Wideband Networking Waveform, it is expected to provide the needed network services for battery-powered radios with limited power and antenna size such as the handheld and the small form varieties. Cluster 5 radios with the Soldier Radio Waveform will enable squad-level communications and interoperability with other radios and work on a network based on the Wideband Networking Waveform. The Soldier Radio Waveform is expected to be available in 2008. However, the development of this waveform is being managed as a science and technology effort by the Army’s Communications-Electronics Research Development and Engineering Center until it is matured and can be transitioned into the JTRS program. To support the first FCS spiral in the 2008-2010 timeframe, the Army has acknowledged that it may have to use an early version of the Soldier Radio Waveform and a surrogate radio to operate the waveform. Compounding the challenges in developing the waveform is the Army’s assessment that developing the Soldier Radio Waveform’s network manager is high risk and has yet to be funded. Without the network manager functionality, the Soldier Radio Waveform will not be able to interface with the Wideband Networking Waveform. A number of JTRS Cluster 5 technologies are interdependent (see fig. 6) that, in our opinion, can exacerbate the technical and program risks of moving forward with immature technologies. For example, power management is dependent upon microelectronics, multichannel architecture, antennas, and security. A lag in the development of any of these technologies could result in a lag in the development of power management. Because of the criticality of the size, weight, and power challenge faced by all variants of the JTRS radios, the program office is pursuing various solutions to the problem. The program, for example, hopes to benefit from the Army’s science and technology research on developing wideband power amplifiers and advanced passive cooling technology. Several programmatic changes have significantly affected the Cluster 5 schedule, and the program has focused on delivering manpack radios for the near term and handheld and small form radios later. However, the availability of small form JTRS radios is of greater importance to FCS because they are needed for the planned fielding of three core systems in FCS spiral 1. The Army has concluded that the small form radios may not be able to meet the FCS schedule and may need to use surrogate radios to support the first FCS spiral. In May 2003, the responsibility for developing the JTRS handheld and manpack radios was shifted from the Special Operations Command to the Army because of difficulties in resolving differences over requirements and funding among the services. At the same time, the Acting Under Secretary of Defense for Acquisition, Technology, and Logistics noted that the Cluster 5 capabilities would have to be delivered in at least two spirals and set an expectation that the Army would deliver prototype handheld and manpack radios in the third quarter of fiscal year 2005 and low rate initial production would begin by the fourth quarter of fiscal year 2006. In May 2004, the Army Acquisition Executive approved the Cluster 5 program for the system development and demonstration phase of acquisition. The Army Acquisition Executive moved the Cluster 5 handheld radios to spiral 2, and it delayed the delivery of the spiral 1 prototype manpack radios to the fourth quarter of fiscal year 2005 and the low-rate initial production manpack radios to the first quarter of fiscal year 2007. The Army awarded the Cluster 5 contract in the middle of July 2004, but had to issue a stop-work order to the contractor by the end of July because of the filing of a bid protest by the losing contractor. The bid protest was not upheld, but the program was delayed another 3 months while the protest was decided. In authorizing the May 2004 Cluster 5 program’s entry into the system development and demonstration phase, the Army Acquisition Executive noted the criticality of the JTRS Cluster 5 radio and directed that a review be conducted to assess the plans for the spiral 2 portion of the program. At a minimum, the review was to assess development schedule synchronization, technical performance expectations, integration and performance risks, waveform development, maturity, baseline, and program affordability. The review was scheduled for the spring of 2005. However, because of the ongoing cost, schedule, and technical problems with the Cluster 1 program, the JTRS Joint Program Executive Office has begun a broader assessment of the Cluster 5 program. On the basis of the initial findings of the assessment, development work on the Cluster 5 spiral 1 radios has been suspended because the office determined that key waveforms being developed as part of the Cluster 1 program would not be delivered to Cluster 5 when needed. According to the JTRS Joint Program Executive Office, a restructuring of Cluster 5 spiral 1 and 2 is being developed, and it will identify more well defined and executable increments. While the Cluster 5 manpack and handheld radios are important deliverables, of greater urgency for the first spiral of FCS is the availability of the small form Cluster 5 radios. These radios will be embedded in a variety of sensors and weapons systems. In fact, three FCS core systems— Unattended Ground Sensors, Intelligent Munitions Systems, and the Non Line of Sight Launch System—need Cluster 5 small form radios to support their planned inclusion in the first FCS spiral scheduled for the 2008-2010 timeframe. The Army has concluded that the schedule for the small form radios is not synchronized with the FCS schedule and has asked the contractor for a plan to accelerate deliveries. The Army has acknowledged that it may have to use surrogate radios, which have limited capabilities, if the Cluster 5 small form radios are not available to support the initial fielding of the three FCS core systems. In addition, other users depending on the Cluster 5 radios, such as the Army’s Land Warrior program, have decided to move forward with surrogate radios. The WIN-T program entered the system development and demonstration phase with only 3 of its 12 critical technologies close to full maturity. None of the critical technologies will be fully mature at the time production begins in March 2006. Because there are significant interdependencies among critical technologies, any delay in maturing an individual technology further increases overall program risk. WIN-T has gone through a number of program changes, including shifts in the program’s focus. In the fall of 2004, the Office of the Secretary of Defense approved the Army’s proposal to combine the work of two contractors to facilitate early delivery of WIN-T capabilities to the warfighter while continuing to focus on the restructured FCS program. A decision has recently been made not to accelerate the program or develop capabilities sooner. It remains unclear what WIN-T capabilities will be provided to the first FCS spiral. The changes, along with existing technical challenges, put the program at risk of cost and schedule overruns and failure to achieve performance objectives. During WIN-T’s 32-month systems development and demonstration schedule, the program must mature 9 of its 12 critical technologies. Although risk mitigation plans were developed in mid-2003 for the 9 immature technologies, a program review sponsored by the Army in July 2004 concluded that the plans lacked sufficient detail. Eight backup technologies have been identified, but they are less robust and only 3 are close to full maturity. Relying on these substitutes may degrade network performance resulting in reduced operational capability. Contrary to best practices under knowledge-based development, the program will continue technology development concurrently with the product development and demonstration phase (see fig.7). The tightly compressed schedule also assumes nearly flawless execution and may not allow sufficient time for correcting problems. For example, the combined testing to demonstrate system performance and operational functionality is slated to occur just 1 month after critical design review. With immature technologies, it will be difficult, at best, to demonstrate the system’s design stability and determine whether the system can be produced affordably and work reliably. In fact, WIN-T program officials may be unable to conclude a reliable operational capability of on-the-move communications until the system is demonstrated in an operational environment early in fiscal year 2009—long after production begins. The significant interdependencies among WIN-T’s critical technologies exacerbate the technical and program risks of moving forward with immature technologies. For example, the on-the-move satellite communications technologies rely on wideband waveforms, antennas, and other technologies to achieve their performance objectives. Therefore, a lag in the development of any of these technologies may result in a lag in the overall development of mobile communications technologies—a critical component of the operational concept for WIN-T. Not only is the program faced with technical challenges, but its dependence on other programs puts the WIN-T program at risk. WIN-T’s ability to significantly improve upon current communications capabilities relies on demonstrating integrated network operations and the ability to work on the move. The WIN-T system depends on other programs to provide needed capabilities. Although separate from the WIN-T program, changes or delays in these external programs may impair WIN-T’s ability to perform. For WIN-T, unmanned aerial vehicles are fundamental to the program as they route information and extend transmission range that ground systems are constrained by—preserving network reliability, connectivity, and mobile throughput. Citing their capacity to fly at high altitudes, program officials have identified two platforms to support WIN-T, the Extended Range Multi-Purpose Unmanned Aerial Vehicle or the High Altitude Airship. However, one is not adequately funded for a dedicated communications capability, and the other is still in the concept development phase. Therefore, a study is under way to assess the consequence of not having unmanned aerial vehicles and its resulting effect on the network. It is unclear whether the issue will be resolved in time for the upcoming development test/operational test event. The program plans to use a surrogate plane, but it is unknown whether this will adequately assess network reliability and critical on-the-move communications. Central to the WIN-T operational effectiveness is the development of a software-programmable radio and wideband waveforms. Together, the radio and waveforms are expected to allow warfighters to receive large volumes of data while moving around the battlefield at increasing speeds. However, given the uncertainty of whether a JTRS radio would be available to support WIN-T, the program plans to develop its own high- capacity radio, operating above the 2 GHz radio frequency range. To meet FCS requirements, the WIN-T radio is expected to run above 2 GHz with two new waveforms—a net-centric waveform and a high-capacity waveform--and the existing Global Broadcast Service waveform. In particular, these waveforms enable distribution of intelligence, surveillance, and reconnaissance data to provide a more detailed picture of the battlefield. To address the need for waveforms operating above 2 GHz, the Office of the Secretary of Defense is conducting an assessment to identify solutions. However, the results of the study may not be available by the critical design review. Since the WIN-T program was conceived nearly 5 years ago, the program strategy has shifted several times. Originally, the program focused on designing a network that would meet current force needs. In 2002, the program was realigned to focus on a network that would support future force needs. Two contractors were to work independently on designing the future force network architecture, and the program office would select the better of the two. The contractors were given significant flexibility in designing the network architecture and developing system performance specifications. Two years later, with the global war on terrorism and military operations in Afghanistan and Iraq, WIN-T was directed to focus on developing and fielding network capabilities to meet both current and future force needs. To expedite completion of the architecture’s design, the Army eliminated competition between the two contractors in September 2004. Army officials believe that the combined team provides a stronger technical solution by taking the best elements of each contractor’s proposed architecture and maintains some competition because over 50 percent of the work will still be competed among sub-contractors. In fact, the contractors working together completed the network architecture by January 2005—a year earlier than previously planned. According to Army officials, the early completion of the network architecture allows other Army programs, particularly FCS, to stabilize their network designs earlier than planned. In conjunction with the WIN-T program’s shift in focus to address both current and future force needs, the Army fielded a separate program, in 2004, a beyond-line-of-sight communications network to units deployed in Iraq: the Joint Network Transport Capability-Spiral (JNTC-S). Although an improvement over past capabilities, JNTC-S is stationary —units must come to a standstill and set up their satellite equipment to communicate. In contrast, WIN-T is expected to maintain satellite connection— regardless of distance, weather conditions, or terrain—-while units are in motion. Currently, the Army is assessing how best to transition JNTC-S to WIN-T. In addition, the Army is assessing whether the WIN-T program can be modified to address the restructured FCS plan to field communications and networking capabilities in spirals. Army officials concede that, based on available technologies and resources, WIN-T block 1 performance requirements may need to be scaled back to meet the FCS spiral 1 time frame. For example, the data rate requirements for block 1 WIN-T—which calls for an unprecedented data throughput rate of 256 kilobits per second while units are moving at 25 miles per hour—may need to be reduced. Although the Army has decided not to accelerate development of WIN-T, it is unclear when plans to migrate from the JNTC-S program and address FCS needs will be completed. The Army assesses SOSCOE as high-risk. SOSCOE software may not reach the necessary technical maturity level required to meet FCS milestones. In addition, FCS system-level requirements are still being defined, which could affect the SOSCOE design. Consequently, it is unclear whether SOSCOE will be sufficiently developed to support the initial fielding of FCS beginning in fiscal year 2008. Because SOSCOE software will tie together FCS systems, support battle command applications, and enable interoperability with current and future forces, it is the fundamental building block upon which a substantial portion of FCS will be built. Thus, delays in SOSCOE software development could affect FCS’ ability to meet production and fielding milestones. Since the start of system development, the Army has assessed SOSCOE software availability and maturity as high-risk. According to program officials, SOSCOE development does not require “cutting edge” software technology. However, there are some aspects of particular service families that are more challenging than others and result in an overall SOSCOE development effort that varies in complexity. The key to SOSCOE development is the “threading model,” which is intended to allow an interface between different subsystem operating systems. The high risk is derived from the fact that SOSCOE may not reach the necessary technical maturity level required to meet program milestones. The SOSCOE risk mitigation strategy is to develop and deliver the software in increments to provide the functionality required by SOSCOE users when they need it. Specifically, the SOSCOE software is scheduled for delivery in a series of seven software builds between the end of 2005 and 2011. FCS functionality will increase with each successive software build. The Army will need about one-half of the SOSCOE software in time for the fielding of the initial FCS capability in fiscal year 2008. If the software risks materialize, the SOSCOE build plan may have to be modified, deferring some functionality to later software builds. Higher-level FCS specifications are still evolving nearly 2 years after the program started development. As in most engineering efforts, FCS requirements are first defined at a general or high level. Once these are defined, more detailed specifications that flow down to the subsystem level are derived. It is the specifications that provide the details necessary to design subsystems like SOSCOE. In the case of FCS, very few specifications have flowed to SOSCOE, as higher-level specifications are still being defined. The lack of specific requirements flow-down could affect the SOSCOE software build needed to support the first FCS spiral. In addition, program officials are concerned that SOSCOE will have difficulty meeting emerging requirements without significant cost and schedule impacts. Costs are likely to grow as SOSCOE is reworked to meet new requirements, or applications software is reworked to accommodate the limitations of SOSCOE. Further, if design assumptions underlying SOCOE during the spiral 1 and 2 builds are wrong, because of incomplete technical information, requirements for future software builds might not be met or the software could require extensive rework, resulting in cost and schedule problems. As part of the original FCS schedule, a DOD-level Network Maturity Milestone Decision was scheduled for 2008 to assess demonstrated communications and networked functions. The demonstration was to verify the performance of FCS software, including SOSCOE. The purpose of the demonstration would have been to provide confidence that all networked operations software would meet initial operational capability objectives and to use the results of the milestone decision to initiate long- lead production for the network equipment. However, the restructuring of the overall FCS program allowed the reduction of the high concurrency in the SOSCOE development and fielding schedule. The development schedule has now been extended to 2011. The DOD-level assessment of demonstrated network capabilities will be deferred until the formal FCS production milestone decision in 2012. Although DOD and the military services have produced the best armed forces in the world, their effectiveness in carrying out military operations has been hampered by communications and networking systems that lack interoperability and have limited capacity to transfer information where and when it is needed. The Army’s efforts to develop JTRS, WIN-T, and the SOSCOE as components of the network are essential to overcoming these limitations. However, to achieve the desired capabilities, not only must each program be successfully executed, but because the programs are interdependent, they must be closely synchronized. In particular, the successful fielding of FCS capabilities is critically dependent on the outcome of the JTRS and WIN-T programs. If they do not work as intended, there will not be sufficient battlefield information for the future force to operate effectively. As currently structured, the JTRS, WIN-T, and SOSCOE programs are at risk of not delivering intended capabilities when needed, particularly for the first spiral of FCS. They continue to struggle to meet an ambitious set of user requirements, steep technical challenges, and stringent timeframes. While the Army’s restructuring of the FCS program last year into spiral increments was a positive step, the first spiral may not demonstrate key networking capabilities. The first spiral of FCS should provide a meaningful demonstration of the networking capabilities that can then serve as a basis to support further development of the future force. In particular, demonstrating the capability of the Wideband Networking Waveform is important, given that the design of FCS vehicles and systems in later spirals is predicated on this capability. It is reasonable that such a demonstration should include JTRS with the Wideband Networking Waveform, WIN-T, and basic capability from SOSCOE. Since (1) an enhanced Army communications network is critical for a successful transformation to FCS and (2) JTRS, including the advanced wideband waveforms, WIN-T, and SOSCOE are the key pillars of the communications network, the timing of the first FCS spiral should be based on when the pacing capabilities to be provided by JTRS and WIN-T will be demonstrated. Therefore, we recommend that the Secretary of Defense: establish low-risk schedules for demonstrating JTRS, WIN-T, and synchronize the FCS spiral schedule with such schedules for JTRS, WIN-T, and SOSCOE; and develop an operational test and evaluation strategy that supports an evaluation of network maturity as part of FCS spiral production decisions. In addition, in light of the delays in JTRS Cluster 1 and the criticality of the Wideband Networking Waveform for FCS, we recommend the Secretary of Defense assess whether a greater priority should be placed on demonstrating the Wideband Networking Waveform on a JTRS radio prototype over other Cluster 1 capabilities in the remainder of the Cluster 1 development program. In its letter commenting on a draft of our report, DOD concurred with our findings and three of our recommendations and partially concurred with a fourth recommendation. (DOD’s letter is reprinted in app III.) As part of its comments, DOD provided some information on actions it has begun to take to address each of our recommendations. While these actions should help strengthen the management of JTRS, WIN-T, and SOSCOE, we remain concerned that a demonstration of FCS’s communications and networking capabilities will not be known for some time. Until these capabilities are demonstrated, investment in FCS platforms and systems carries substantial risk. DOD also provided technical comments, which we incorporated where appropriate. Regarding our first recommendation—that the Secretary of Defense establish low-risk schedules for demonstrating JTRS, WIN-T, and SOSCOE capabilities—DOD concurred, noting (1) that its newly established JTRS Joint Program Executive Office is evaluating the condition of each JTRS product line and will make recommendations to ensure effective control of cost, schedule, and performance and (2) that the Army is managing risks associated with WIN-T and SOSCOE and the Office of the Secretary of Defense is applying the appropriate level of oversight. While the actions being taken by DOD and the Army will help, it remains unclear whether they will be sufficient to ensure JTRS, WIN-T, and SOSCOE—the critical components of the enhanced communications network—are successfully executed. We remain concerned that the requisite knowledge needed to effectively manage program development risks has not been sufficiently developed. A low-risk fielding schedule for each of the components should set the pace for the Army’s transformation to FCS. Regarding our second recommendation—that the Secretary of Defense synchronize the FCS spiral schedule with the fielding schedules for JTRS, WIN-T, and SOSCOE—DOD partially concurred, but stated that “the Army’s strategy for spiraling out FCS technology is not constrained to any one particular element of the program. The strategy aims to make available mature and military useful system capability in increments, leveraging opportunities to integrate new and mature technology with current force capability.” DOD further stated that “the FCS spirals will make use of technologies as they become available or leverage the use of surrogate applications where they apply.” DOD also noted that the Army did not define the first FCS spiral around the main components of the communications network, but around the capabilities needed by the current force. While we agree with DOD that mature and military useful capabilities should be fielded as expeditiously as possible, we believe that the first spiral should demonstrate meaningful capabilities for FCS. In particular, we believe that the first spiral of FCS should demonstrate critical networking capabilities, and that its schedule be predicated on demonstrating core capabilities, such as the JTRS Wideband Networking Waveform. Progress made on these capabilities should guide the future investments, such as on ground vehicles that depend on network performance. In addition, reliance on surrogate applications has the potential to result in costly replacement of the surrogate applications once the target applications are fully mature. Regarding our third recommendation—that the Secretary of Defense develop an operational test and evaluation strategy that supports an evaluation of network maturity as part of FCS spiral production decisions—DOD concurred, stating that FCS will initially field a mix of both new and legacy communications and network capabilities, and that iterative operational test and evaluation will be stressed to ensure strong capability verification and validation. DOD also noted that network maturity will be assessed at each spiral’s production decision. While it is appropriate to assess network maturity at each spiral’s production decision, to measure progress in developing the FCS communications network, these assessments will need to culminate in a full demonstration that the network will perform as intended before committing to produce equipment for FCS units of action. Finally, regarding our fourth recommendation—that the Secretary of Defense assess whether a greater priority should be placed on demonstrating the Wideband Networking Waveform on a JTRS radio prototype over other Cluster 1 capabilities in the remainder of the Cluster 1 development program—DOD concurred, noting that the newly established JTRS Joint Program Executive Office is assessing the JTRS Cluster 1 development path and that the development of the Wideband Networking Waveform will be included in the assessment. As agreed with your office, unless you announce its contents, we will not distribute this report further until 30 days after the date of this letter. At that time, we will send copies to the Chairmen and Ranking Minority Members of other Senate and House committees and subcommittees that have jurisdiction and oversight responsibilities for DOD. We will also send copies to the Secretary of Defense, the Secretary of the Army, and the Director, Office of Management and Budget. Copies will also be available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-2811, or Assistant Director John Oppenheim at (202) 512-3111. Major contributors to this report were Ridge Bowman, Subrata Ghoshroy, Karen Sloan, Hai Tran, Paul Williams, and Candice Wright. To determine the development risks associated with the Joint Tactical Radio System-Tactical (JTRS) Cluster 1, JTRS Cluster 5, and WIN-T programs, we obtained briefings on acquisition plans, analyzed documents describing the maturity of critical technologies, and interviewed project and product officials from the Warfighter Information Network-Tactical (WIN-T) Program Management Office, Fort Monmouth, New Jersey. To determine the status of JTRS waveforms, we obtained briefings on wideband waveform development efforts and interviewed officials from the JTRS Joint Program Office, Arlington, Virginia. We also reviewed selected acquisition reports, technology readiness assessments, test and evaluation plans, defense acquisition executive summaries, and acquisition decision memorandums for individual programs. To obtain information related to the planned use of JTRS Cluster 1 radios in rotary wing platforms, we interviewed officials from the Program Executive Office, Aviation, Arlington, Virginia. To obtain information related to JTRS Cluster 1 contract performance data, we interviewed Defense Contract Management Agency officials in Anaheim, California, and obtained cost performance reports and other cost analysis documentation. To assess cost and schedule performance for JTRS Cluster 1 and waveform development for the period between August 2003 and January 2005, we used cost and schedule variances reported in contractor cost performance reports. Results were presented in graphical form to determine the period’s trends. We also obtained likely cost at the completion of the prime contract from the reports. We confirmed that the prime contractor’s earned value management system had been validated by the Defense Contract Management Agency. The cost and schedule results include both prime and subcontractors. The development of the waveforms was included in our analysis of Cluster 1 because, although the effort is managed separately under the Joint Program Office, it is being executed under the same contract. To determine the development risks associated with the System of Systems Common Operating Environment (SOSCOE), we obtained briefings on fielding plans, analyzed documents describing SOSCOE software availability and maturity, and interviewed project officials from the Project Manager for FCS Network Systems Integration, Fort Monmouth, New Jersey. We also attended FCS in-process reviews and a board of directors meeting in St. Louis, Missouri, organized by the Program Manager, Unit of Action. To obtain the perspective of organizations that provide policy guidance, oversight, and technology support for the JTRS, WIN-T, and Future Combat Systems (FCS) programs, we interviewed officials from the Office of the Secretary of Defense, Networks and Information Integration, Arlington, Virginia; Assistant Secretary of the Army for Acquisition, Logistics, and Technology, Arlington, Virginia; and, the Army’s Communications-Electronics Research, Development and Engineering Center, Fort Monmouth, New Jersey. Our review was conducted from January 2004 through May 2005 in accordance with generally accepted government auditing standards. Since Cluster 1 entered systems development, in 2002, the contractor has overrun cost estimates by almost $93 million-–nearly 28 percent above what was planned. We used contractor cost performance reports to assess the prime contractor’s progress toward meeting the Army’s cost and schedule goals during the period August 2003-January 2005. The government routinely uses such reports to independently evaluate the prime contractor’s performance. Generally, the reports detail deviations in cost and schedule relative to expectations established under the contract. Deviations are referred to as variances. Positive variances—activities costing less or completed ahead of schedule—are considered as good news, and negative variances—activities costing more or falling behind schedule—as bad news. Although the program attempted to stabilize cost growth by adding approximately $200 million to the contract in January 2004, the cost variance continued to decline steadily thereafter. Key issues driving the cost growth are unanticipated complexity associated with developing the hardware, Wideband Networking Waveform, and other software. As a result, the unit costs for early prototypes have increased from the prime contractor’s original proposal. In January 2005, the prime contractor estimated that the total costs for the Cluster 1 radio and waveform development would be $531 million more than was originally budgeted, reaching about $898 million at completion (see fig. 8). However, the program office noted that, since contract award, the prime contractor has not demonstrated strong cost estimating and cost management techniques. Cluster 1 has also experienced unfavorable schedule variance. Figure 9 indicates that the contractor increasingly fell behind schedule during the period August 2003-January 2005. If a program is not only overrun in costs, but is also behind schedule, additional costs can be expected because of potential schedule slippage or from acceleration of the effort to finish on time. The schedule variance stabilized briefly after the program rebaselined in January 2004, but then it continued to increase again. By January 2005, the value of planned work that the contractor was behind schedule was about $25 million. Delays in software build completions, software/hardware integration, and the delivery of key technologies to the waveform developers have contributed to schedule problems. | The Army has embarked on a major transformation of its force. Central to this transformation is the Future Combat Systems (FCS), a $108 billion effort to provide warfighters with the vehicles, weapons, and communications needed to identify and respond to threats with speed, precision, and lethality. Establishing reliable, robust communications and networking capabilities is key to FCS's success. Each of the systems integral to the FCS communications network--the Joint Tactical Radio System (JTRS), the Warfighter Information Network-Tactical (WIN-T), and the System of Systems Common Operating Environment (SOSCOE)--rely on significant advances in current technologies and must be fully integrated to realize FCS. Given the complexity and costs of this undertaking, GAO was asked to review each of these key development efforts to identify any risks that may jeopardize the successful fielding of FCS. Each of the programs for developing FCS's communications network is struggling to meet ambitious sets of user requirements and steep technical challenges within highly compressed schedules. As currently structured, the programs are at risk of not delivering intended capabilities for the first spiral of FCS, slated to start in fiscal year 2008. The JTRS Cluster 1 program--a program to develop radios for ground vehicles and helicopters--began development with an aggressive schedule, immature technologies, and a lack of clearly defined and stable requirements. As currently designed, the radio will only have a transmission range of only 3 kilometers--well short of the required 10 kilometers--and will not meet security requirements for operating in an open networked environment. The program's struggle to mature and integrate key technologies has contributed to significant cost and schedule growth. A recent review of the program concluded that the current program structure is not executable, and in April 2005, DOD directed the Army to stop work and notify the contractor that it was considering terminating the contract. Meeting requirements for JTRS Cluster 5 radios--miniaturized radios, including those that soldiers carry--is even more technically challenging given their smaller size, weight, and power needs. The smallest of these radios weighs only about 1 pound, compared with 84 pounds for Cluster 1 radios. Several programmatic changes and a contract award bid protest have further slowed program progress. The Army is considering options for restructuring the program to meet the needs of FCS and address the technical issues encountered in the Cluster 1 program. The Army does not expect to fully mature the technologies for WIN-T--communications equipment that supports an expanded area of battlefield operations and interfaces with JTRS radios--when production begins in March 2006. Moreover, the compressed schedule assumes nearly flawless execution and does not allow sufficient time for correcting problems. Significant interdependencies among the critical technologies further increase overall program risk. The program was directed to deliver networking and communications capabilities sooner to meet near-term warfighting needs and synchronize with the restructured FCS program. A plan for how to develop and field WIN-T capabilities sooner to address FCS needs remains undetermined. According to Army network system integration officials, SOSCOE--the operating software to integrate the communications network--may not reach the necessary technical maturity level required to meet program milestones. In addition, top-level FCS requirements are still evolving and have not been translated into more detailed specifications necessary for writing SOSCOE software. |
Flaws in software code that could cause a program to malfunction generally result from programming errors that occur during software development. The increasing complexity and size of software programs contribute to the growth in software flaws. For example, Microsoft Windows 2000 reportedly contains about 35 million lines of code, compared with about 15 million lines for Windows 95. As reported by the National Institute of Standards and Technology (NIST), based on various studies of code inspections, most estimates suggest that there are as many as 20 flaws per thousand lines of software code. While most flaws do not create security vulnerabilities, the potential for these errors reflects the difficulty and complexity involved in delivering trustworthy code. By exploiting software vulnerabilities, hackers and others who spread malicious code can cause significant damage, ranging from Web site defacement to taking control of entire systems, and thereby being able to read, modify, or delete sensitive information, destroy systems, disrupt operations, or launch attacks against other organizations’ systems. Between 1995 and the first half of 2003, the CERT Coordination Center (CERT/CC) reported 11,155 security vulnerabilities that resulted from software flaws. Figure 1 illustrates the dramatic growth in security vulnerabilities over these years. The growing number of known vulnerabilities increases the number of potential attacks created by the hacker community. As vulnerabilities are discovered, attackers may attempt to exploit them. Attacks can be launched against specific targets or widely distributed through viruses and worms. Worms and viruses are commonly used to launch denial-of-service attacks, which generally flood targeted networks and systems with so much transmission of data that regular traffic is either slowed or completely interrupted. Such attacks have been utilized ever since the groundbreaking Morris worm, which brought 10 percent of the systems connected to Internet systems to a halt in November 1988. In 2001, the Code Red worm used a denial-of-service attack to affect millions of computer users by shutting down Web sites, slowing Internet service, and disrupting business and government operations. This type of attack continues to be used by recent worms, including Blaster, which I will discuss further later in my testimony. The sophistication and effectiveness of cyber attack have steadily advanced. Because automated tools now exist, CERT/CC has noted, attacks that once took weeks or months to propagate over the Internet now take just hours, or even minutes. Code Red achieved an infection rate of over 20,000 systems within 10 minutes, foreshadowing more damaging and devastating attacks. Indeed, earlier this year, the Slammer worm, which successfully attacked at least 75,000 systems, became the fastest computer worm in history, infecting more than 90 percent of vulnerable systems within 10 minutes. Frequently, skilled hackers develop exploitation tools and post them on Internet hacking sites. These tools are then readily available for others to download, allowing even inexperienced programmers to create a computer virus or to literally point and click to launch an attack. According to a NIST publication, 30 to 40 new attack tools are posted to the Internet every month. The threat to systems connected to the Internet is illustrated by the increasing number of computer security incidents reported to CERT/CC. This number rose from just under 10,000 in 1999 to over 52,000 in 2001, to about 82,000 in 2002, and to over 76,000 for the first and second quarters of 2003. And these are only the incidents that are reported. According to the Director of CERT/CC, as much as 80 percent of actual incidents go unreported, in most cases because the organization was either unable to recognize that its systems had been penetrated (because there were no indications of penetration or attack) or because it was reluctant to report an incident. Figure 2 illustrates the number of incidents reported to CERT/CC from 1995 through the second quarter of 2003. According to CERT/CC, about 95 percent of all network intrusions could be avoided by keeping systems up to date with appropriate patches; however, such patches are often not quickly or correctly applied. Maintaining current patches is becoming more difficult, as the length of time between the awareness of a vulnerability and the introduction of an exploit is shrinking. For example, the Blaster worm was released almost simultaneously with the announcement of the vulnerability it exploited. Successful attacks on unpatched software vulnerabilities have caused billions of dollars in damage. Following are examples of significant damage caused by worms that could have been prevented had available patches been effectively installed: In September 2001 the Nimda worm appeared, reportedly infecting hundreds of thousands of computers around the world, using some of the most significant attack methods of Code Red II and 1999’s Melissa virus that allowed it to spread widely in a short amount of time. A patch had been made publicly available the previous month. On January 25, 2003, Slammer triggered a global Internet slowdown and caused considerable harm through network outages and other unforeseen consequences. As we discussed in our April testimony, the worm reportedly shut down a 911 emergency call center, canceled airline flights, and caused automated teller machine (ATM) failures. According to media reports, First USA Inc., an Internet service provider, experienced network performance problems after an attack by the Slammer worm, due to a failure to patch three of its systems. Additionally, the Nuclear Regulatory Commission reported that Slammer also infected a nuclear power plant’s network, resulting in the inability of the computers to communicate with each other, disrupting two important systems at the facility. In July 2002, Microsoft had released a patch for its software vulnerability that was exploited by Slammer. Nevertheless, according to media reports, some of Microsoft’s own systems were infected by Slammer. In addition to understanding the threat posed by security vulnerabilities, it is useful to understand the process of vulnerability identification and response. In general, when security vulnerabilities are discovered, a process is initiated to effectively address the situation through appropriate reporting and response. Typically, this process begins when security vulnerabilities are discovered by software vendors, security research groups, users, or other interested parties, including the hacker community. When a software vendor is made aware of a vulnerability in its product, the vendor typically first validates that the vulnerability indeed exists. If the vulnerability is deemed critical, the vendor may convene a group of experts, including major clients and key incident-response groups such the Federal Computer Incident Response Center (FedCIRC) and CERT/CC, to discuss and plan remediation and response efforts. After a vulnerability is validated, the software vendor develops and tests a patch and/or workaround. A workaround may entail blocking access to or disabling vulnerable programs. The incident response groups and the vendor typically prepare a detailed public advisory to be released at a set time. The advisory often contains a description of the vulnerability, including its level of criticality; systems that are affected; potential impact if exploited; recommendations for workarounds, and Web site links where a patch (if publicly available) can be downloaded. Incident-response groups as well as software vendors may continue to issue updates as new information about the vulnerability is discovered. When a worm or virus is reported that exploits a vulnerability, virus detection software vendors also participate in the process. Such vendors develop and make available to their subscribers downloadable “signature files” that are used, in conjunction with their software, to identify and stop the virus or worm from infecting systems protected by their software. The Organization for Internet Safety (OIS), which consists of leading security researchers and vendors, recently issued a voluntary framework for vulnerability reporting and response. Recently, two critical vulnerabilities were discovered in widely used commercial software products. The federal government and the private- sector security community took steps, described below in chronological order, to collaboratively respond to the threat of potential attacks against these vulnerabilities. Last Stage of Delirium Research Group discovered a security vulnerability in Microsoft’s Windows Distributed Component Object Model (DCOM) Remote Procedure Call (RPC) interface. This vulnerability would allow an attacker to gain complete control over a remote computer. On June 28, 2003, the group notified Microsoft about the RPC vulnerability. Within hours of being notified, Microsoft verified the vulnerability. On July 16, Microsoft issued a security bulletin publicly announcing the critical vulnerability and providing workaround instructions and a patch. The following day, CERT/CC issued its first advisory. Nine days after Microsoft’s announcement, on July 25, Xfocus, an organization that researches and demonstrates security vulnerabilities, released code that could be used to exploit the vulnerability. On July 31, CERT/CC issued a second advisory reporting that multiple exploits had been publicly released, and encouraged all users to apply the patches. On August 11, 2003, the Blaster worm (also known as Lovsan) was launched to exploit this vulnerability. When the worm is successfully executed, it can cause the operating system to crash. Experts consider Blaster, which affected a range of systems, to be one of the worst exploits of 2003. Although the security community had received advisories from CERT/CC and other organizations to patch this critical vulnerability, Blaster reportedly infected more than 120,000 unpatched computers in the first 36 hours. By the following day, reports began to state that many users were experiencing slowness and disruptions to their Internet service, such as the need to frequently reboot. The Maryland Motor Vehicle Administration was forced to shut down, and systems in both national and international arenas have also been affected. The worm was programmed to launch a denial-of-service attack on Microsoft’s Windows Update Web site www.windowsupdate.com (where users can download security patches) on August 16. Microsoft preempted the worm’s attack by disabling the Windows Update Web site. On August 14, two variants to the original Blaster worm were released. Federal agencies reported problems associated with these worms to FedCIRC. On August 18, Welchia, a worm that also exploits this vulnerability, was reported. Among other things, it attempts to apply the patch for the RPC vulnerability to vulnerable systems, but reportedly creates such high volumes of network traffic that it effectively denies services in infected networks. Media reports indicate that Welchia affected several federal agencies, including components of the Departments of Defense and Veterans Affairs. The federal government’s response to this vulnerability included coordination with the private sector to mitigate the effects of the worm. On July 17, FedCIRC issued an advisory to encourage federal agencies to patch the vulnerability, followed by several advisories from the Department of Homeland Security (DHS). The following week, on July 24, DHS issued its first advisory to heighten public awareness of the potential impact of an exploit of this vulnerability. On July 28, on behalf of the Office of Management and Budget (OMB), FedCIRC requested that federal agencies report on the status of their actions to patch the vulnerability. From August 12 to August 18, DHS’s National Cyber Security Division hosted several teleconferences with federal agencies, CERT/CC, and Microsoft. Figure 3 is a timeline of selected responses to the Blaster Internet worm. Based on an analysis of the agencies reported actions, as requested on July 28, FedCIRC indicated that many respondents had completed patch installation on all systems at the time of their report and that only a minimal number of infections by the Blaster worm were reported. Cisco Systems, Inc., which controls approximately 82 percent of the worldwide share of the Internet router market, discovered a critical vulnerability in its Internet operating system (IOS) software. This vulnerability could allow an intruder to effectively shut down unpatched routers, blocking network traffic. Cisco had informed the federal government of the vulnerability prior to public disclosure, and worked with different security organizations and government organizations to encourage prompt patching. On July 16, 2003, Cisco issued a security bulletin to publicly announce the critical vulnerability in its IOS software, and provide workaround instructions and a patch. Cisco had planned to officially notify the public of the vulnerability on July 17, but early media disclosure led them to announce the vulnerability a day earlier. In addition, FedCIRC issued advisories to federal agencies and DHS advised private-sector entities of the vulnerability. In the week that the vulnerability was disclosed, FedCIRC, OMB, and DHS’s National Cyber Security Division held a number of teleconferences with representatives from the executive branch. On July 17, OMB requested that federal agencies report to CERT/CC on the status of their actions to patch the vulnerability by July 24. On July 18, DHS issued an advisory update in response to an exploit that was posted online, and OMB moved up the agencies’ reporting deadline to July 22. CERT/CC has received reports of attempts to exploit this vulnerability, but as of September 5, no incidents have yet been reported. Patch management is a process used to help alleviate many of the challenges involved with securing computing systems from attack. It is a component of configuration management that includes acquiring, testing, and applying patches to a computer system. I will now discuss common patch management practices, federal efforts to address software vulnerabilities in agencies, and services and tools to assist in carrying out the patch management process. Effective patch management practices have been identified in security- related literature from several groups, including NIST, Microsoft, patch management software vendors, and other computer-security experts. Common elements identified include the following: Senior executive support. Management recognition of information security risk and interest in taking steps to manage and understand risks, including ensuring that appropriate patches are deployed, is important to successfully implementing any information security–related process and ensuring that appropriate resources are applied. Standardized patch management policies, procedures, and tools. Without standardized policies and procedures in place, patch management can remain an ad-hoc process—potentially allowing each subgroup within an entity to implement patch management differently or not at all. Policies provide the foundation for ensuring that requirements are communicated across an entity. In addition, selecting and implementing appropriate patch management tools is an important consideration for facilitating effective and efficient patch management. Dedicated resources and clearly assigned responsibilities. It is important that the organization assign clear responsibility for ensuring that the patch management process is effective. NIST recommends creating a designated group whose duties would include supporting administrators in finding and fixing vulnerabilities in the organization’s software. It is also important that the individuals involved in patch management have the skills and knowledge needed to perform their responsibilities, and that systems administrators be trained regarding how to identify new patches and vulnerabilities. Current technology inventory. Creating and maintaining a current inventory of all hardware equipment, software packages, services, and other technologies installed and used by the organization is an essential element of successful patch management. This systems inventory assists in determining the number of systems that are vulnerable and require remediation, as well as in locating the systems and identifying their owners. Identification of relevant vulnerabilities and patches. It is important to proactively monitor for vulnerabilities and patches for all software identified in the systems inventory. Various tools and services are available to assist in identifying vulnerabilities and their respective patches. Using multiple sources can help to provide a more comprehensive view of vulnerabilities. Risk assessment. When a vulnerability is discovered and a related patch and/or alternative workaround is released, the entity should consider the importance of the system to operations, the criticality of the vulnerability, and the risk of applying the patch. Since some patches can cause unexpected disruption to entities’ systems, organizations may choose not to apply every patch, at least not immediately, even though it may be deemed critical by the software vendor that created it. The likelihood that the patch will disrupt the system is a key factor to consider, as is the criticality of the system or process that the patch affects. Testing. Another critical step is to test each individual patch against various systems configurations in a test environment before installing it enterprisewide to determine any impact on the network. Such testing will help determine whether the patch functions as intended and its potential for adversely affecting the entity’s systems. In addition, while patches are being tested, organizations should also be aware of workarounds, which can provide temporary relief until a patch is applied. Testing has been identified as a challenge by government and private-sector officials, since the urgency in remediating a security vulnerability can limit or delay comprehensive testing. Time pressures can also result in software vendors’ issuing poorly written patches that can degrade system performance and require yet another patch to remediate the problem. For instance, Microsoft has admittedly issued security patches that have been recalled because they have caused systems to crash or are too large for a computer’s capacity. Further, a complex, heterogeneous systems environment can lengthen this already time-consuming and time-sensitive process because it takes longer to test the patch in various systems configurations. Distributing patches. Organizations can deploy patches to systems manually or by using an automated tool. One challenge to deploying patches appropriately is that remote users may not be connected at the time of deployment, leaving the entity’s networks vulnerable from the remote user’s system because they have not yet been patched. One private- sector entity stated that its network first became affected by the Microsoft RPC vulnerability when remote users plugged their laptops into the network after being exposed to the vulnerability from other sources. Monitoring through network and host vulnerability scanning. Networks can be scanned on a regular basis to assess the network environment, and whether patches have been effectively applied. Systems administrators can take proactive steps to preempt computer security incidents within their entities by regularly monitoring the status of patches once they are deployed. This will help to ensure patch compliance with the network’s configuration. The federal government has taken several steps to address security vulnerabilities that affect federal agency systems, including efforts to improve patch management. NIST has taken a number of steps, including, as I previously mentioned, providing a handbook for patch management. In addition, NIST offers a source of vulnerability data, which I will discuss later in this testimony. Further, in accordance with OMB’s reporting instructions for the first year implementation of the Federal Information Security Management Act (FISMA), maintaining up-to-date patches is a part of FISMA’s system configuration requirements. As such, OMB requires that agencies report how they confirm that patches have been tested and installed in a timely manner. In addition, certain governmentwide services are offered to federal agencies to assist them in ensuring that software vulnerabilities are patched. For example, FedCIRC was established to provide a central focal point for incident reporting, handling, prevention, and recognition for the federal government. Its purpose is to ensure that the government has critical services available in order to withstand or quickly recover from attacks against its information resources. In addition, for the two recent vulnerabilities just discussed in my testimony, OMB and FedCIRC held teleconferences with agency Chief information officers to discuss vulnerabilities and request that agencies report on the status of their actions to patch them. An OMB official indicated that they planned to hold meetings with agencies to discuss ways to improve communication of and followup on critical vulnerabilities, including addressing some of the challenges identified in the two recent exercises, such as delays in reaching key security personnel in certain instances. FedCIRC also initiated PADC to provide users with a method of obtaining information on security patches relevant to their enterprise and access to patches that have been tested in a laboratory environment. The federal government offers PADC to federal civilian agencies at no cost. According to FedCIRC, as of last month, 41 agencies were using PADC. Table 2 lists its features and benefits, as reported by FedCIRC. OMB reported that while many agencies have established PADC accounts, actual usage of those accounts is extremely low. To participate in PADC, subscribers (who could be one or more individuals within an agency) receive an account license that allows them to receive notifications and log into the secure Web site to download the patch. To establish an account, each subscriber must set up a profile defining the technologies that they use. The profiles act much like a filtering service and allow PADC to notify agencies of only the patches that pertain to their systems. The profiles do not include system–specific information because of the sensitivity of that information. Subscribers using PADC receive notification of threats, vulnerabilities, and the availability of patches on the basis of the submitted profiles. They are notified by E-mail or pager message that a vulnerability or patch has been posted to a secure Web site that affects one or all of their systems. When a patch is identified, FedCIRC, through contractor support, ensures that it originates from a reliable source. The patch is then tested on a system to which it applies. The installation of the patch and the operation of the system are monitored to ensure that the patch causes no problem. Next, if an exploit had been developed, exploit testing is performed to ensure that the patch fixes the vulnerability. Any issues identified with a patch are summarized and provided to the users. The validated patch is then uploaded to PADC servers and made available to users. A patch is considered validated when it has been downloaded from a trusted source, authenticated, loaded onto an appropriate system, tested, exploit–tested, verified, and posted to the PADC server. This type of testing and validation is performed for over 60 technologies that, according to FedCIRC officials, account for approximately 80 percent of the technologies used by federal agencies. Also available is notification of patches that are not validated for over 25,000 additional technologies. According to FedCIRC officials, high-priority patches are to be tested and posted on the PADC server within the same business day of availability. Medium- and low-priority patches are to be completed by the following business day, but are generally available sooner. However, because PADC has several early warning mechanisms in place and arrangements with software vendors, some patches may be available as soon as a vulnerability is made public. FedCIRC officials emphasize that although the contractor tests the security patches, these tests do not ensure that the patch can be successfully deployed in another environment; therefore, agencies still need to test the patch for compatibility with their own business processes and technology. PADC offers a reporting capability that is hierarchical. Senior managers can look at their complete system and see which subsystems have been patched. These enterprisewide reports and statistics can be generated for a “reporting user” subscriber who has read-only capability within the system. According to agency officials, there are limitations to the PADC service. Although it is free to agencies, only about 2,000 licenses or accounts are available because of monetary constraints. According to FedCIRC officials, this requires them to work closely with participating agencies to balance the number of licenses that a single agency requires with the need to allow multiple agencies to participate. For example, the National Aeronautics and Space Administration initially requested over 3,000 licenses—one for each system administrator. Another agency, NIST, thought that each of its users should have his or her own PADC account. Another limitation is the level of services currently provided by PADC. At present, the government is considering broadening the scope of these services and capabilities, along with the number of users. Several services and automated tools are available to assist entities in performing the patch management function, including tools designed to be stand-alone patch management systems. In addition, systems management tools can be used to deploy patches across an entity’s network. Some of the features in services and tools typically include methods to inventory computers and the software applications and patches installed; identify relevant patches and workarounds and gather them in one location; group systems by departments, machine types, or other logical divisions to easily manage patch deployment; scan a network to determine the status of the patches and other corrections made to network machines (hosts and/or clients); assess the machines against set criteria; access a database of patches; deploy effective patches; and report information to various levels of management about the status of the network. Patch management vendors also offer central databases of the latest patches, incidents, and methods for mitigating risks before a patch can be deployed or a patch has been released. Some vendors provide support for multiple software platforms, such as Microsoft, Solaris, Linux, and others, while others focus on certain platforms exclusively, such as Microsoft. Patch management tools can be either scanner–based (non agent) or agent–based. While scanner–based tools can scan a network, check for missing patches, and allow an administrator to patch multiple computers, these tools are best suited for smaller organizations due to their inability to serve a large number of users without breaking down or requiring major changes in procedure. Another difficulty with scanner-based tools is that part-time users and turned-off systems will not be scanned. Agent-based products place small programs, or agents, on each computer, to periodically poll a patch database—a server on the network—for new updates, giving the administrator the option of applying the patch. Agent- based products require up-front work to integrate agents into the workstations and in the server deployment process, but are better suited to large organizations due to their ability to generate less network traffic and provide a real-time network view. The agents maintain information that can be reported when needed. Finally, some patch management tools are hybrids—allowing the user to utilize agents or not. Instead of an automated stand-alone system, entities can also use other methods and tools to perform patch management. For example, they can maintain a database of the versions and latest patches for each server and each client in their network and track the security alerts and patches manually. While labor-intensive, this can be done. In addition, entities can employ systems management tools with patch-updating capabilities to deploy the patches. This method requires monitoring for the latest security alerts and patches. Entities may also need to develop better relationships with their vendors to be alerted to vulnerabilities and patches prior to public release. In addition, software vendors may provide automated tools with customized features to alert system administrators and users of the need to patch, and if desired, automatically apply patches. A variety of resources are also available to provide information related to vulnerabilities and their exploits. As I mentioned earlier, one resource is CERT/CC, a major center for analyzing and reporting vulnerabilities as well as providing information on possible solutions. Another useful resource is NIST’s ICAT, which offers a searchable index leading users to vulnerability resources and patch information. ICAT links users to publicly available vulnerability databases and patch sites, thus enabling them to find and fix vulnerabilities existing on their systems. It is based on common vulnerability and exposures (commonly referred to as CVE) naming standards. These are standardized names for vulnerabilities and other information security exposures, compiled in an effort to make it easier to share data across separate vulnerability databases and tools. Many other organizations exist, including the Last Stage of Delirium Research Group, that research security vulnerabilities and maintain databases of such vulnerabilities. In addition, mailing lists, such as BugTraq, provide forums for announcing and discussing vulnerabilities, including information on how to fix them. In addition, Security Focus monitors thousands of products to maintain a vulnerability database and provide security alerts. Finally, vendors such as Microsoft and Cisco provide software updates on their products, including notices of known vulnerabilities and their corresponding patches. In addition to implementing effective patch management practices, several additional steps can be considered when addressing software vulnerabilities, including: deploying other technologies, such as antivirus software, firewalls, and other network security tools to provide additional defenses against attacks; employing more rigorous engineering practices in designing, implementing, and testing software products to reduce the number of potential vulnerabilities; improving tools to more effectively and efficiently manage patching; researching and developing technologies to prevent, detect, and recover from attacks as well as identify their perpetrators, such as more sophisticated firewalls to keep serious attackers out, better intrusion- detection systems that can distinguish serious attacks from nuisance probes and scans, systems that can isolate compromised areas and reconfigure while continuing to operate, and techniques to identify individuals responsible for specific incidents; and ensuring effective, tested contingency planning processes and procedures. Actions are already underway in many, if not all, of these areas. For example, CERT/CC has a research program, one goal of which is to try to find ways to improve technical approaches for identifying and preventing security flaws, for limiting the damage from attacks, and for ensuring that systems continue to provide essential services in spite of compromises and failures. Also, Microsoft recently initiated its Trustworthy Computing strategy to incorporate security-focused software engineering practices throughout the design and deployment of its software, and is reportedly considering the use of automated patching in future products. In summary, it is clear from the increasing number of reported attacks on information systems that both federal and private-sector operations and assets are at considerable—and growing—risk. Patch management can be an important element in mitigating the risks associated with software vulnerabilities, part of overall network configuration management and information security programs. The challenge will be ensuring that a patch management process has adequate resources and appropriate policies, procedures, and tools to effectively identify vulnerabilities and patches that place an entity’s systems at risk. Also critical is the capability to adequately test and deploy the patches, and then monitor progress to ensure that they work. Although this can currently be performed, the eventual solution will likely come from research and development to better build security into software and tools from the beginning. Mr. Chairman, this concludes my statement. I would be pleased to answer any questions that you or other members of the Subcommittee may have at this time. Should you have any further questions about this testimony, please contact me at (202) 512-3317 or at daceyr@gao.gov. Individuals making key contributions to this testimony included Shannin G. Addison, Michael P. Fruitman, Michael W. Gilmore, Sophia Harrison, Elizabeth L. Johnston, Min S. Lee, and Tracy C. Pierson. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Attacks on computer systems--in government and the private sector--are increasing at an alarming rate, placing both federal and private-sector operations and assets at considerable risk. By exploiting software vulnerabilities, hackers can cause significant damage. While patches, or software fixes, for these vulnerabilities are often well publicized and available, they are frequently not quickly or correctly applied. The federal government recently awarded a contract for a government-wide patch notification service designed to provide agencies with information to support effective patching. Forty-one agencies now subscribe to this service. At the request of the Chairman of the Subcommittee on Technology, Information Policy, Intergovernmental Relations, and the Census, GAO reviewed (1) two recent software vulnerabilities and related responses; (2) effective patch management practices, related federal efforts, and other available tools; and (3) additional steps that can be taken to better protect sensitive information systems from software vulnerabilities. The increase in reported information systems vulnerabilities has been staggering, especially in the past 3 years. Automated attacks are successfully exploiting such software vulnerabilities, as increasingly sophisticated hacking tools become more readily available and easier to use. The response to two recent critical vulnerabilities in Microsoft Corporation and Cisco Systems, Inc., products illustrates the collaborative efforts between federal entities and the information security community to combat potential attacks. Patch management is one means of dealing with these increasing vulnerabilities to cybersecurity. Critical elements to the patch management process include management support, standardized policies, dedicated resources, risk assessment, and testing. In addition to working with software vendors and security research groups to develop patches or temporary solutions, the federal government has taken a number of other steps to address software vulnerabilities. For example, offered without charge to federal agencies, the federal patch notification service provides subscribers with information on trusted, authenticated patches available for their technologies. At present, the government is considering broadening the scope of these services and capabilities, along with the number of users. Other specific tools also exist that can assist in performing patch management. In addition to implementing effective patch management practices, several additional steps can be taken when addressing software vulnerabilities. Such steps include stronger software engineering practices and continuing research and development into new approaches toward computer security. |
In March 2003, DHS began operations and set about the daunting task of merging 22 separate and autonomous federal agencies with homeland security-related missions under the centralized leadership of a single department. In doing so, DHS assumed operational control of about 209,000 civilian and military positions from these agencies. As we have previously reported, the creation and transformation of DHS is critically important and poses significant management and leadership challenges, and failure to address these challenges could have serious consequences for our national security. Consequently, in 2003, we first designated the department’s implementation and transformation as high risk, and we continue to do so today. IT is a critical tool in DHS’s quest to transform itself and carry out the department’s critical missions on a day-to-day basis. For fiscal year 2008 alone, the department is requesting almost $4 billion in IT budgetary authority. The department’s ability to effectively and efficiently invest these funds and deliver IT systems and infrastructure that perform as intended depends in large part on the capabilities of its IT human capital. As we have reported, DHS and the other federal agencies historically have been challenged in their ability to strategically manage human capital. For this reason, we first designated strategic human capital management as a governmentwide high-risk area in 2001, and we continue to do so today. To accomplish its mission, the department is organized into various agencies and directorates, each of which is responsible for specific homeland security missions and for coordinating related efforts with other DHS organizations, as well as external entities. Table 1 shows DHS’s principal organizations and their respective missions. Within DHS, responsibility for IT human capital management resides with the Management Directorate—specifically, the Offices of the CIO and the CHCO—and with component agency CIO and human capital offices. More specifically, the management directive of the DHS Office of the CIO’s states that the office is responsible for leveraging the best available technologies and applying proven IT management and human capital practices to provide shared services, coordinate acquisition strategies, maintain an enterprise architecture, and advocate and enable business transformation, among other things. To assist in managing these matters, DHS established the DHS CIO Council made up of the CIOs from each of DHS’s component organizations. The council identified eight priorities, including IT human capital, and for each priority, it assigned an executive sponsor that is responsible for overseeing the department’s efforts in that area. The council also established the IT Human Capital Resource Center (formerly called the IT Human Capital Center of Excellence) to support the council and the executive sponsor responsible for IT human capital. In short, the center is responsible for setting a DHS-wide vision and strategy for IT human capital and the functions that IT staff perform. The center is staffed by the component CIO organizations and, among other things, is responsible for coordinating the implementation of the department’s IT human capital initiatives. Figure 1 shows a simplified and partial DHS organizational structure, including the CIO IT human capital-related entities. According to the DHS overall strategic human capital plan, which covers IT and non-IT personnel, the Office of the CHCO is responsible for implementing initiatives to achieve strategic human capital goals in support of the department’s mission. With regard to IT, this includes planning and managing human capital to meet current and future mission needs, recruiting a high-quality workforce, developing a strong and capable workforce, motivating and retaining high performers, and fostering a culture of continuous learning and improvement. It also includes applying human capital best practices in carrying out these responsibilities. Each of the department’s component agencies has its own CIO and human capital director to, among other things, manage the implementation of their respective IT human capital initiatives. According to DHS, this includes recruiting staff to close competency and skill gaps, coordinating and delivering mission-essential training, analyzing workforce data, and aligning component human capital plans with the department human capital plans to achieve agency and department missions. To accomplish its mission, DHS relies extensively on IT. For example, in fiscal year 2007, about $4.16 billion dollars in funding was requested to support 278 major IT programs. Table 2 shows the fiscal year 2007 IT funding for key DHS components. To manage the use of these funds and carry out these programs, the department reports that it employs about 2,600 IT personnel. While these personnel represent about 1 percent of the department’s total workforce, they nonetheless perform critical mission functions. Specifically, IT personnel develop, manage, and operate mission-critical systems that are intended to unify the department under a common IT infrastructure and to facilitate agencies’ ability to analyze intelligence to identify threats, guard U.S. borders and airports, protect critical infrastructure, coordinate national responses to emergencies, and implement other security measures. Moreover, IT staff track and oversee the efforts of a sizable workforce of support contractors. According to DHS, the need to successfully manage its IT human capital is essential to effectively and efficiently leveraging technology in achieving the department’s mission. This need is compounded by the fact that the department faces major near-term IT human capital challenges. For example, DHS estimates that between 2005 and 2010, approximately 35 percent of its IT workforce will be eligible for retirement. Moreover, it reports that in light of the continued growth in demand of experienced IT professionals and the high rate of turnover experienced thus far, the department faces significant risk of critical skill shortages, which could hamper its mission imperatives. During the last 3 years, we have reported on the importance of DHS adopting a strategic approach to addressing its IT human capital challenges. For example, in August 2004, we reported that DHS had begun strategic planning for IT human capital at the headquarters level, but it had not yet systematically gathered baseline data about its existing workforce. We also reported on CIO staffing concerns and slow progress in this area. Accordingly, we recommended that the department analyze whether it had appropriately allocated and deployed IT staff with the relevant skills to obtain its institutional and program-related goals. In response, the DHS CIO approved funding for the IT Human Capital Resource Center in July 2004. Among other things, the center subsequently began work to complete an IT human capital plan. Consistent with our recommendation, the center was to ensure that the completed plan provided for an analysis of IT workforce skill sets. In May 2005, the DHS CIO issued a draft version of the IT human capital plan. This draft version was sent to the Senate and House Appropriations Committees on August 30, 2006, as part of the CIO’s report pursuant to requirements in DHS’s fiscal year 2006 appropriations act. According to the CIO Council senior executive leading the effort to develop this plan, it was developed in partnership with the DHS CHCO’s office and intended to direct the department’s IT human capital efforts. In March 2006, we testified on a number of IT human capital and other management challenges at DHS. We noted that DHS had undertaken a departmentwide human capital initiative, MAXHR, which was to provide greater flexibility and accountability in the way employees are paid, developed, evaluated, afforded due process, and represented by labor organizations. Part of this initiative involved the development of departmentwide workforce competencies. We testified that the department had intended to implement MAXHR in the summer of 2005 but had encountered delays. More recently, DHS officials stated that MAXHR had been canceled and is to be replaced by another initiative called the Human Capital Operational Plan. In May 2007, we reported that while DHS continues work to develop and implement departmentwide human capital initiatives, its overall progress in managing its IT and non-IT human capital had been limited. Since 2002, we have also reported on human capital management weaknesses associated with key DHS IT programs. For example: In September 2005, we reported that the program office for the Atlas program was not adequately staffed. Accordingly, we recommended that the Atlas program conduct a staffing needs assessment to determine the positions and the level of staffing needed for all Atlas projects, and that it develop a human capital strategy for meeting its staffing needs. DHS agreed with our recommendations and has since completed a needs assessment, developed a human capital strategy, and used it to staff the program office and projects. In February 2006, we reported that the US-VISIT program had developed a human capital strategy, as we had recommended 2 years earlier, and had begun implementing it. However, we also reported that several activities in the plan had not been implemented, such as assessing the extent of current employees’ competency gaps and developing a listing of competency-based training courses. To address this shortfall, among other things, the program recently developed a new human capital plan. We have not yet reviewed the new plan. In May 2006, we reported that the Automated Commercial Environment (ACE) program had yet to develop and implement a human capital management strategy, as we had recommended several years earlier. Instead, program officials told us that they were following a less formal approach to bolstering ACE’s workforce. Accordingly, we recommended that the department report to its appropriations committees on its strategy for managing ACE human capital needs. DHS agreed with our recommendation and has since been working to develop a strategy. A strategic approach to human capital management includes viewing people as assets whose value to an organization can be enhanced by investing in them. Such an approach enables organizations to effectively use their people and determine how well they integrate human capital considerations into daily decision making and planning for mission results. It also helps organizations remain aware of and be prepared for current and future needs as an organization, ensuring that they have the knowledge, skills, and abilities needed to pursue their missions. On the basis of our experience with leading organizations, we issued a model in 2002 for strategic human capital management. The model is built around four cornerstones: (1) leadership; (2) strategic human capital planning; (3) acquiring, developing, and retaining talent; and (4) results- oriented organizational cultures. We also issued a set of key practices in 2003 for effective strategic human capital management. These practices are generic, applying to any organization or component, such as an agency’s IT organization. Since then, OPM, in conjunction with OMB and us, issued a strategic human capital framework—called the Human Capital Assessment and Accountability Framework—to provide a consistent, comprehensive representation of human capital management to guide federal agencies. Consistent with our 2002 model, OPM’s framework provides six standards, along with associated indicators (practices) for achieving success. The six standards for success and related definitions are as follows: Strategic alignment. The organization’s human capital strategy is aligned with mission, goals, and organizational objectives and integrated into its strategic plans, performance plans, and budgets. Workforce planning and deployment. Among other things, the organization strategically uses staff in order to achieve mission goals in the most efficient ways. Leadership and knowledge management. The organization’s leaders and managers effectively manage people, ensure continuity of leadership, and sustain a learning environment that drives continuous improvement in performance. Results-oriented performance culture. The organization has a diverse, results-oriented, high-performance workforce and a performance management system that effectively differentiates between high and low performance and links individual, team, or unit performance to organizational goals and desired results. Talent management. The organization makes progress toward closing gaps or making up deficiencies in most mission-critical skills, knowledge, and competencies. Accountability. The organization’s human capital decisions are guided by a data-driven, results-oriented planning and accountability system. Our recent work has shown that DHS and other federal agencies, such as the Securities and Exchange Commission, have begun to use OPM’s framework as the basis for preparing strategic IT and other human capital plans. According to DHS CIO officials, they used the OPM framework in developing the IT human capital plan that they included in the August 2006 report to Congress. DHS has developed an IT human capital plan that is largely consistent with OPM guidance. Specifically, of 27 key practices in OPM’s framework, the department’s plan and related documentation fully address 15 practices and partially address the other 12, meaning that these 12 are missing elements that are essential to having a well-defined and executable plan. DHS officials responsible for developing the plan attributed the missing elements to, among other things, the department’s decision to focus its resources on other IT priorities. These officials also stated that until the missing elements are fully addressed, it is unlikely that the plan will be effectively and efficiently implemented, which in turn will continue to put DHS at risk of not having sufficient people with the right knowledge, skills, and abilities to manage and deliver its mission-critical IT systems. Examples of the key practices that DHS has fully and partially addressed in its IT human capital plan and related documentation, organized according to OPM’s six standards for success, are given in the following text. Also, table 3 is a summary of the DHS plan’s satisfaction of all 27 key practices. Appendix II contains our full analysis of the plan’s satisfaction of these 27 practices. Both the summary and full analysis contain examples to demonstrate full or partial satisfaction of the practices. They do not contain all examples of DHS’s accomplishments or limitations to a given key practice. Strategic alignment. DHS’s plan and related documentation satisfy a number of strategic alignment practices. For example, they specify human capital goals for the IT organization and provide for linking them to departmental human capital goals. More specifically, the plan identifies such IT human capital goals as meeting current and future mission needs, recruiting a high-quality IT workforce, and motivating and retaining high performers. The plan further states that IT human capital programs and initiatives should produce performance outcomes that support the overall DHS strategic goal of operational excellence. In addition, the plan calls for involving key stakeholders—such as the CIO, the CHCO, and their component agency counterparts—in carrying out a range of workforce planning activities, such as conducting a workforce analysis, developing an inventory of current staff skills, and identifying the future skills that are needed for mission-critical positions. By addressing these key practices, the plan helps set the overall direction and tone for strategic management of IT human capital and lays a foundation for demonstrating management commitment and promoting buy-in across the organization. However, the plan and related documentation do not fully satisfy other key practices. For example, they do not include specific milestones for when most defined activities and steps are to be completed. This is a serious limitation because milestones help to ensure that resources needed to execute plans are allocated, and they provide a basis for measuring progress. In addition, although the plan provides for involving key stakeholders, it does not assign stakeholders responsibility and accountability for specific activities. Without fully addressing these practices, the plan does not provide an adequate basis for promoting accountability for results, and thus ensuring that the plan will be effectively implemented. Workforce planning and deployment. The plan and related documentation satisfy a number of key practices in this standards area, including provision for incentives for new recruits, training for existing staff, and an exchange program to draw on private sector personnel with necessary skills. This is important because such practices are essential ingredients to acquiring, training, and deploying an effective workforce. However, the plan does not provide for regular collection and analysis of data on promotions, conversions, separations, and retirements to show an understanding of trends and related indicators of performance. Without this information, DHS will be limited in its ability to know whether the techniques being employed are effective, and thus performance results and accountability goals are being met. Leadership and knowledge management. DHS’s plan and supporting documentation provide for a number of leadership and knowledge management practices. For example, DHS planning documents (e.g., DHS Succession Management Plan FY 2006–2009) supporting the IT human capital plan describe and encourage leadership development across all DHS components through application of the department’s Leadership Competency Framework and succession approach to workforce planning efforts. The plan also identifies succession planning goals and objectives, implementation strategies, and program evaluation critical success factors to achieve expected leadership outcomes. These efforts are important because they show how the department and components plan for and minimize the impact of changes to its leadership team arising from retirements and separations. However, the plan does not address how these activities are to be linked to and reflected in department annual performance plans and budgets. Having performance plans and budgets that address the IT human capital goals is vital to ensuring that the plan is properly funded to ensure implementation. Results-oriented performance culture. DHS’s plan and supporting documentation satisfy key practice elements under this standards area, such as identifying outcome-based human capital goals for its IT workforce and linking these goals to departmental strategic plans. However, the plan does not address linking each work unit’s efforts and performance to these goals. Linking the work units to goals is important because it provides a framework for setting performance expectations, determining whether expectations are met, and establishing accountability, each of which is critical to effective and efficient plan implementation. Talent management. DHS’s plan addresses important practices related to talent management, including documenting mission-critical occupations, strategizing how to reduce competency gaps between the workforce’s current skills and those needed to achieve mission goals, and tracking efforts to implement strategies. In particular, it provides for a monthly forum hosted by the IT Human Capital Resource Center for DHS components to share ideas and strategies for recruitment, retention, and training of their workforces. These initiatives are important because they provide a disciplined and systematic approach to identifying and reducing organizational skill shortfalls, and thus contribute to better ensuring that DHS has the right people with the right skills. However, neither the plan nor supporting documents fully provide for measuring whether its recruitment and training efforts are closing competency gaps. Such performance measurement is vital to effective plan implementation because it provides feedback on the effectiveness of efforts and the need for corrective action. Accountability. The plan addresses the key practice for establishing and using applicable merit principles and standards in appraising IT staff performance, and for establishing a process for employee grievances to be considered and addressed. However, the plan does not fully address other accountability-related practices. For example, it does not provide for proactively identifying where the department is at risk with regard to attaining its IT human capital goals and developing initiatives to mitigate any high risks. This is a significant omission because proactively managing risks is a proven means for avoiding problems before they can occur. According to DHS officials responsible for developing the plan, the 12 key practices were not fully addressed for several reasons. Specifically, they stated that uncertainty surrounding the source of resources for implementing the plan led to a lack of a clear definition of stakeholder roles and responsibilities, which in turn made setting realistic milestones impractical. They added that a number of other IT priorities that were competing for resources, such as consolidation of data centers, also contributed to the 12 practices not being addressed, while other omissions were purely an unintended oversight, such as not addressing central management of risks. According to the officials, the next version of the plan, which is tentatively scheduled to be released in the second quarter of fiscal year 2008 based on the assumption that resources are made available, is to address all of these omissions. Without a comprehensive IT human capital plan, DHS does not have an effective means for ensuring that it has the right people in the right place at the right time to achieve the department’s mission-related IT goals. The department has acknowledged this risk and estimates there is currently a medium-to-high level of risk of not meeting DHS’s mission due to personnel and competency and skill shortages. The DHS departmental offices and component agencies that share responsibility for implementing the IT human capital plan have collectively made little progress in doing so. In general, the DHS Offices of the CHCO and the CIO have done more to implement the plan than have the DHS component agencies, as described in the following text. The plan’s state of implementation is due to both a lack of clarity around the respective implementation-related roles and responsibilities of the various DHS organizations involved, as well as the lower funding priority that these organizations have given to the plan’s implementation relative to other competing IT efforts. Until a complete and well-defined IT human capital plan is effectively and efficiently implemented, the department will continue to run the risk of not having the people it needs to leverage technology in achieving organizational transformation and mission goals. At the department level, the CIO and the CHCO organizations, working with the CIO Council’s Human Capital Resource Center, have together performed some of the tasks in the plan. For example, they have performed a gap analysis between existing and future skill needs and have begun examining strategies for reducing the identified gaps. They have also identified mission-critical occupations and skills necessary to achieve departmental goals. However, it is unclear which organization has primary responsibility for the plan. According to officials from both the Offices of the CHCO and the CIO, primary responsibility for the IT human capital plan and its implementation has recently moved from the CIO to the CHCO. However, these officials have yet to provide us with documentation of this transfer in responsibility. Despite the previously noted positive steps toward implementing the plan, officials from the CIO and the CHCO offices told us that the plan is largely not implemented. For example, while DHS is collecting information on the number of increasing, decreasing, and new mission-critical occupations, it is not identifying and analyzing year-to-year changes and trends to determine whether recruitment and retention strategies need to be updated to meet current organizational needs. In addition, although the department has documented performance goals and objectives for some work units (e.g., managers in Customs and Border Protection) and linked them to department-level organizational goals, it had not done so for much of the department. At the component level, none of the three agencies that we reviewed had begun implementing the plan, as described in the following text. The Coast Guard had not implemented the plan. According to Coast Guard officials, including the Director, Future Force, and the Chief of Human Resource Information Services, they were aware of the plan’s existence, but were unaware of any requirement to implement it. However, they stated that their own human capital efforts satisfy everything in the plan. For example, these officials said that they had performed workforce analyses to determine skill and competency gaps and have employed a range of strategies, such as strategic recruitment through direct hiring authority and internal training, to fill the gaps. The Coast Guard has yet to provide us with documentation to substantiate these statements. Customs and Border Protection had not implemented the plan, although officials from its Office of Information Technology and the Office of Human Resources Management told us that they were aware of the plan and the need to implement it. According to these officials, the agency is in the process of developing a strategy to implement the plan. They also stated that the strategy was to be completed in June 2007, but it is still under development. On August 30, 2007, the officials reported that the strategy had been completed. We have not yet received the strategy and had an opportunity to analyze it. The Federal Emergency Management Agency had not implemented the plan. Agency officials, including the Deputy CIO and the Chief of the Human Capital Branch, stated that they were aware of the plan but were unaware of a requirement to implement it. They also stated that their agency human capital efforts nevertheless were fully consistent with the plan. However, the officials have yet to provide analysis and related documentation to support these statements. In addition, the officials added that they are in the process of developing an agencywide human capital plan—addressing both IT and non-IT personnel—that is to be consistent with the plan and is to be issued on October 1, 2007. The lack of implementation progress can be attributed in part to ambiguity surrounding implementation roles and responsibilities. In particular, the plan itself is in large part silent on implementation roles and responsibilities as well as implementation accountability mechanisms. Moreover, as we have previously noted, the plan does not address important aspects of OPM’s key practices that are implementation related. To help clarify the plan, including implementation roles and responsibilities, the DHS CIO Council’s Human Capital Resource Center developed an implementation briefing and provided it to the CIO Council members in November 2005. However, the briefing does not assign specific implementation activities to specific organizations. Rather, it groups implementation activities into solution sets and then broadly assigns these sets to department and component agency CIOs, CHCOs, and human capital directors. As a result, department and component agency officials told us that they were not clear on who was responsible for what, particularly with regard to the sources of funding and staff. Moreover, as we have previously noted, officials for at least one component agency were not even aware that they were required to implement it, or what their roles and responsibilities were relative to implementation. The lack of implementation progress can also be attributed to resources being assigned to competing IT initiatives that were judged to be higher priorities. According to DHS CIO officials, including the CIO Council senior executive leading the effort, when it came time to fund implementation of the plan, the department and components decided to fund other priorities, such as DHS’s effort to consolidate multiple component data centers and create a unified departmental network. Furthermore, the IT Human Capital Resource Center program manager responsible for implementing the plan resigned in January 2006, and his replacement left in November 2006. According to DHS CIO and CHCO officials, the department has not provided funding to fill the position, which still remains vacant. Department and component officials agreed that the IT human capital plan is largely not implemented. However, they stated that they are nonetheless following many of the OPM framework practices in the plan as a by- product of fulfilling their periodic reporting requirements to OMB on the President’s Management Agenda human capital initiatives. Specifically, the department and its components are required to report quarterly to OMB on progress in meeting certain human capital goals, such as filling mission-critical positions and delivering training to strengthen key IT knowledge, skills, and abilities. For example, the actions reported to OMB require the department and components to identify mission-critical occupations and competencies, develop recruitment strategies to maintain mission-critical competencies at desired levels, and report on progress toward achieving human capital goals, which are also called for by the plan and OPM’s framework. Our analysis showed that efforts related to this reporting requirement align with about 12 of the 27 practices that we examined. DHS officials did not disagree with this analysis. This means that despite a number of IT human capital-related activities, the department and its component agencies are not implementing the full range of practices needed for effective management of IT human capital. An effective DHS IT workforce is essential to the department’s efforts to leverage technology in transforming itself and achieving mission goals and outcomes. Central to creating and sustaining such a workforce is developing a comprehensive IT human capital plan that reflects relevant guidance and best practices, and ensuring that the plan is effectively implemented. While much of such a plan has been developed, and thus a planning foundation exists upon which to build, this plan is nevertheless lacking with respect to relevant guidance and best practices aimed at, among other things, ensuring that the plan is effectively implemented. Moreover, actual implementation of the plan to date has been limited, with much remaining to be accomplished by the department CIO and CHCO organizations as well as their DHS component agency counterparts. The status of the plan and its implementation is largely attributable to the lack of clarity surrounding implementation roles and responsibilities, and the lack of priority being given to the plan’s implementation relative to competing IT priorities at the department and component agency levels. Until DHS has a comprehensive plan and follows through to ensure that it is effectively implemented departmentwide, it will remain challenged in its ability to have sufficient people with the right knowledge, skills, and abilities to effectively leverage technology in support of transformation and mission goals. To strengthen DHS’s management of IT human capital, we recommend that the Secretary of Homeland Security direct the Under Secretary for Management and the head of each DHS component agency to instruct their respective CIOs and human capital directors to make development and implementation of a comprehensive IT human capital plan an imperative within each organization. In this regard, we recommend that the Secretary direct the Under Secretary and the component agency heads to ensure that (1) IT human capital planning efforts fully satisfy relevant federal guidance and related best practices, (2) roles and responsibilities for implementing the resulting IT human capital plan and all supporting plans are clearly defined and understood, (3) resources needed to effectively and efficiently implement the plans are made available, and (4) progress in implementing the plans is regularly measured and periodically reported to DHS leadership and Congress. In written comments on a draft of this report, signed by the Director, Departmental GAO/Office of Inspector General Liaison and reprinted in appendix III, the department stated that it agreed with our recommendations. Consistent with our report, it also stated that the state of IT human capital management varies widely across DHS component organizations, and it acknowledged that a lower priority has been assigned to IT human capital relative to other IT-related matters. In addition, DHS stated that it understands the importance of IT human capital planning and that it will dedicate the resources needed to ensure that it has a highly skilled and effective IT workforce. DHS also provided what it termed additional information about ongoing and planned activities to update and clarify the status of its IT human capital efforts, particularly with regard to the key practices that we determined to be “partially satisfied.” Among other things, DHS stated that some of our determinations were based on the DHS IT Human Capital Strategic Plan (2005), which was not intended to include certain details relative to achieving results, such as milestones, time frames, and roles and responsibilities. According to DHS, this plan is a high-level strategy and not a “blueprint for execution.” Rather, it said that the IT Gap Analysis Report and Improvement Plan (2007) is the department’s “operative diagram” for achieving its human capital goals and results. We agree that the IT Gap Analysis Report and Improvement Plan (2007) is relevant to our determinations. However, we disagree that our determinations were based solely on the strategic plan. As described in our report’s scope and methodology, our determinations were based on examining all relevant documentation that the department provided for each key practice, including the IT Gap Analysis Report and Improvement Plan (2007), as well as on interviews with key officials from DHS’s Offices of the CIO and CHCO, the CIO Council executive sponsor for Human Capital issues, and officials from the department’s IT Human Capital Resource Center. Accordingly, the determinations in our draft report already recognized most of the additional information that DHS provided. In cases where new information was provided, we have incorporated, or otherwise recognized, this information in our report as appropriate. We are sending copies of this report to the Chairmen and Ranking Members of the Senate and House committees that have authorization and oversight responsibilities for homeland security and other interested congressional committees. We are also sending copies to the Directors of OMB and OPM; the DHS Secretary, Undersecretary for Management, CHCO, and CIO; the component agency heads; and other interested parties. In addition, the report will also be available without charge on GAO’s Web site at http://www.gao.gov. Should you have any questions about matters discussed in this report, please contact me at (202) 512-3439 or by e-mail at hiter@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix IV. The objectives of our review were to determine (1) whether the Department of Homeland Security’s (DHS) information technology (IT) human capital plan is consistent with federal guidance and associated best practices and (2) the status of the plan’s implementation. To address our first objective, we reviewed the department’s May 20, 2005, IT human capital plan, which DHS labeled as “Draft Final for Discussion Purposes” and submitted on August 30, 2006, to the Senate and House Appropriations Committees pursuant to requirements in DHS’s fiscal year 2006 appropriations act. We evaluated this plan and supporting documentation against selected practices in the Office of Personnel Management’s (OPM) Human Capital Assessment and Accountability Framework. We used this framework because it is the federal guidance that DHS used in developing its plan, and because the framework reflects the human capital best practices in GAO’s strategic human capital model. In addition, this framework provides a method for assessing the adequacy of a human capital plan. In applying this method, we focused on 27 practices in the framework that are essential to a well-defined and useful plan and that span the six standards areas in the framework. We also validated our use of the 27 practices with OPM. Using the framework’s method, we compared the DHS IT human capital plan and supporting documentation with each of the elements comprising the 27 practices. We also interviewed (1) officials from DHS’s Offices of the Chief Information Officer (CIO) and the Chief Human Capital Officer (CHCO); (2) the CIO Council executive sponsor for Human Capital issues; and (3) officials from the department’s IT Human Capital Resource Center, which helped develop the IT human capital plan and supporting documentation. In performing our comparative analysis, we determined if the practice was fully satisfied, partially satisfied, or not satisfied. For purposes of this review, we defined “fully satisfied” to mean that the agency demonstrated, through verifiable evidence, that it had addressed all aspects of the key practice; “partially satisfied” to mean that such evidence showed that some, but not all, aspects of the key practice had been addressed; and “not satisfied” to mean that such evidence showed that none of the aspects of the key practice had been addressed. In addition, we shared all of our preliminary determinations with officials from the DHS CIO Council and the DHS Office of the CHCO and provided them with an opportunity to comment on these determinations. These officials agreed with many of our determinations but also provided additional evidence to support revising others, which we have done and incorporated in this report. For our second objective, we reviewed plan implementation activities within the DHS Offices of the CIO and the CHCO and three DHS component agencies: the Coast Guard, Customs and Border Protection, and the Federal Emergency Management Administration. We selected these components because based on DHS’s fiscal year 2006 budget, they were among the largest with respect to total budget, IT budget, and IT staff positions. Thus, the scope of our component agency coverage extended to about $20 billion of DHS’s $40 billion total budget; $720 million of the department’s $2.2 billion IT budget; and 60 percent of its IT personnel. In each of these organizations, we requested and reviewed available documentation on its respective efforts to implement the plan, including development of supporting implementation plans, completion of tasks, and the status of ongoing efforts related to IT human capital. We also interviewed responsible officials from DHS’s Offices of the CIO and the CHCO; the Coast Guard’s Human Resources Directorate; Customs and Border Protection’s Office of Human Resources Management; and the Federal Emergency Management Administration’s Office of the CIO and its Office of Human Resources Management. We performed our work at DHS headquarters in Washington, D.C., from October 2006 through July 2007, in accordance with generally accepted government auditing standards. Key stakeholders—identified by DHS as including the CIO, the CHCO, component agency CIO and human capital directors, and the IT Human Capital Resource Center— participated in the development of the department’s IT human capital plan and workforce planning and analysis efforts. For example, in March 2005, DHS held an off-site meeting with these stakeholders to facilitate collaboration and to gather stakeholder input as part of plan development efforts. DHS’s IT human capital plan also states that the department intends to involve these stakeholders in efforts to periodically revise the plan to reflect current priorities and conditions. Furthermore, the plan and supporting documentation identify these stakeholders as participating in analyzing and identifying the department’s workforce needs and in developing a departmentwide workforce plan to fill identified gaps. The organization defines successful achievement of its mission in terms of valid and reliable data, including both long- and short-term human capital performance goals. In its IT human capital plan and supporting documentation, DHS defines accomplishing its near-term and long-term IT human capital goals and objectives in terms of qualitative and quantitative measures that are to be based on valid and reliable data, and links them to accomplishing DHS’s mission. Specifically, the plan identifies departmental human capital goals, such as recruiting a high-quality IT workforce, training its IT workforce to be capable, and retaining high performers. It also describes how these goals support the strategic goal of empowering the IT workforce and how this helps to achieve DHS’s mission. Trends in mission-critical occupations are analyzed in terms of suggested factors in order to continually adjust the agency’s recruitment and retention strategy to its current state of need. In the IT human capital plan and supporting documentation, the department provides updates for fiscal years 2004 and 2005 on, for example, the number of mission-critical occupations that are increasing, decreasing, or new. However, the plan and supporting documentation generally do not identify and analyze the year-to-year trends. For example, supporting documentation (e.g., the DHS Workforce Plan FY 2005–2008) has data for 2003 and 2004, but the year-to-year changes and trends in occupations are not identified and analyzed to determine whether the recruitment and retention strategy needs to be updated to meet the current state of organizational need. In addition, while DHS officials noted an example of one component agency (Transportation Security Administration) adjusting its recruitment and retention strategy to meet the current state of need, they stated that most components are not adjusting recruitment strategies on the basis of available occupation data. An integrated human capital planning process is in use, including representatives from the agency/unit human capital team, the primary IT human capital officer, and senior leaders and managers from mission- specific program areas. DHS’s IT human capital plan and supporting documentation identify use of a human capital planning process that includes stakeholders from across the department and component agencies. For example, in developing the IT human capital plan, the department used a process involving representatives from the department’s CHCO, CIO, and component offices, among others. This was also the case with regard to other supporting documentation. For example, in developing the DHS Workforce Plan FY 2005–2008, the department brought together stakeholders from across the department to collaborate on and produce this product. This workforce plan also defines a human capital planning process whose stated purpose is to help identify, in an integrated and cost- effective manner, the human capital resources needed to meet mission goals and develop strategies for developing or acquiring those resources. Mission-critical occupations and competencies are identified in the agency’s strategic plan and/or performance plan, and its strategic human capital plan. Although documentation supporting DHS’s IT human capital plan identifies mission-critical occupations (e.g., IT project managers and IT security specialists), the IT human capital plan and other DHS strategic and human capital plans do not. Specifically, DHS’s IT Human Capital Plan to Mitigate IT Competency and Skill Gaps and the DHS Workforce Plan FY 2005–2008 identify technical competencies and skills needed for IT occupations. However, the DHS IT human capital plan and the departmentwide strategic and human capital plans do not identify mission-critical IT occupations and competencies. While documents supporting DHS’s IT human capital plan (e.g., the November 2005 implementation briefing and the May 2007 IT Gap Analysis Report and Improvement Plan) include milestones and assign roles and responsibilities, neither these documents nor the IT human capital plan include specific time frames or milestones for when most defined activities and steps are to be completed. In addition, although the supporting documents and the plan provide for involving key stakeholders, they do not assign stakeholders responsibility and accountability for specific activities. Key human capital leaders and agency stakeholders utilize collaborative mechanisms/forums that provide a venue for consistent dialogue in the planning process (e.g., team members of review boards, working groups, or executive off-sites). DHS’s IT human capital plan and supporting documentation describe the department’s collaborative mechanisms and forums for planning strategic human capital activities. They include, for example, the DHS CIO Council, which is made up of component agency CIOs and which has monthly meetings to discuss, among other things, human capital matters. The council used this forum and off-site meetings to collaborate with the DHS CHCO office, the IT Human Capital Resource Center, and component human capital directors, among others, in developing the IT human capital plan. In addition, the department tasks the Human Capital Resource Center to bring together representatives from DHS and the components on a monthly basis to share ideas and strategies on emerging IT human capital issues. Furthermore, DHS established a Workforce Planning Council, comprising department and component agency officials, to develop a workforce plan and provide for analysis across DHS. In the IT human capital plan and supporting documentation, DHS documents a change management process that identifies human capital practices needed to achieve the department’s human capital objectives. For example, in the DHS Workforce Plan FY 2005–2008, the department describes its change management process that includes steps such as identifying departmental goals, identifying workforce requirements, developing a workforce strategy, and evaluating the effectiveness of the planning process. In addition, the IT human capital plan identifies certain practices—such as analyzing workforce needs and capabilities, developing an IT training strategy, implementing an IT leadership development program, and developing performance measures for accountability—as being critical to achieving DHS human capital objectives. Moreover, supporting documentation (e.g., DHS’s November 2005 implementation briefing) identifies traceable linkages between the practices it is intended to implement and IT human capital goals and objectives. Studies indicate which occupations and competencies are essential to achieving the agency’s strategic goals. Documentation supporting the IT human capital plan (e.g., DHS’s IT Human Capital Plan to Mitigate IT Competency and Skill Gaps) identifies occupations and competencies to achieve the agency’s strategic goals. For example, the department identified competencies within IT project management, information security, and enterprise architecture as being critical to achieving the department’s mission goals. Line managers and key staff, including human resources, consider and prepare for possible workforce changes in areas such as mission/goals, technology, program additions or deletions, functions, and outsourcing initiatives. DHS’s IT human capital plan and supporting documentation include guidance for managers and key staff to consider, plan, and prepare for changes in the department’s mission, programs, and workforce composition. Specifically, DHS’s IT human capital plan states that DHS managers should consider and prepare for changes in organizational goals, personnel, and technology. In addition, supporting documentation (e.g., the DHS Workforce Plan FY 2005–2008) acknowledges the possibility of workforce changes due to retirements and attrition. The workforce plan also states that it will serve as an integrated approach for addressing future business needs, and identifies steps that department managers should go through in planning for changes, including considering how changes will impact mission goals, programs, functions, and workforce composition. The workforce plan also states that managers should consider using alternative strategies, such as outsourcing. Turnover indicators are monitored regularly. Documentation supporting the IT human capital plan (specifically, the DHS Workforce Plan FY 2005–2008) identifies several factors to be monitored, including appointments, separations, and retirements, and assigns the responsibility for monitoring the factors to the department’s Office of the CHCO. This documentation also reports on the department’s appointments, separations, and retirements during fiscal years 2004 and 2005. A workforce analysis process is used on a regular basis for assessment and planning, and to drive human capital decisions. The IT human capital plan and supporting documentation show that DHS uses a workforce analysis process for human capital assessment, planning, and decisions. For example, supporting documentation (e.g., the DHS Workforce Plan FY 2005–2008) identifies workforce trends analyzed among cross-cutting and high-profile mission- critical occupations and the process established and followed to develop such trend data. Furthermore, the documentation also shows that DHS established a Workforce Planning Council that is responsible for ensuring that workforce planning and human capital initiatives are integrated consistently and cost-effectively across DHS. According to DHS CHCO officials, the department intends to conduct workforce analysis efforts every 2 years. However, these officials also report that not all components are using the workforce data on a regular basis to drive human capital decisions. The agency has a clearly defined strategy and plan to facilitate human capital changes. The IT human capital plan and supporting documents clearly identify human capital strategies and goals, but do not fully provide for how and when human capital changes will be made. For example, the plan defines strategic goals and objectives and states that an implementation plan is to be developed and executed with performance measures, such as milestones, deadlines, and assignment of personnel responsible for achieving them. However, as we have previously discussed, DHS developed such a plan in November 2005 (i.e., the November 2005 implementation briefing) and later updated it in the May 2007 IT Gap Analysis Report and Improvement Plan, but these documents do not include specific time frames or milestones for when most defined activities and steps are to be completed. In addition, although the document provides for involving key stakeholders, it does not assign stakeholders responsibility and accountability for specific activities. Staffing data showing trends in appointments, promotions, conversions, separations, and retirements are analyzed regularly, and management decisions regarding workforce deployment are based on documented data. The IT human capital plan and supporting documentation include analyses of staffing data for appointments, separations, and retirements that are reported to the Office of Management and Budget (OMB) on a quarterly basis. In addition, these documents (e.g., the DHS Workforce Plan FY 2005–2008) include workforce trends analyses among cross-cutting and mission-critical occupations. However, trends in these data are not fully analyzed, and, according to DHS CHCO officials, not all components are using the data on a regular basis to drive human capital decisions. The agency uses multifaceted techniques to close competency gaps within the organization (e.g., strategic recruitment, midcareer hiring, and training). The IT human capital plan and supporting documentation provide for a variety of recruitment and training techniques to be used in closing competency gaps. For example, supporting documents (e.g., DHS’s IT Human Capital Plan to Mitigate IT Competency and Skill Gaps) describe efforts planned and under way to mitigate gaps using strategic recruitment through outsourcing, private/public cross training, internal training, and e-training. Leadership development and succession needs are considered, reflected in human capital plans and strategies, and addressed through related human capital management efforts/programs. The IT human capital plan and supporting documentation, in particular DHS’s Succession Management Plan FY 2006–2009, describe practices to be followed in developing the leadership skills of DHS personnel. These documents also identify succession planning goals and objectives, implementation strategies, and program evaluation critical success factors to measure whether expected outcomes are being achieved. The agency has a strategy and plan for communication of human capital changes and progress, and to capture employee feedback related to human capital practices and needs. In its IT human capital plan and supporting documentation, DHS identifies strategies and plans for communicating changes and progress to employees. For example, the IT human capital plan includes initiatives to improve communication on human capital changes and progress, such as developing training materials and courses to educate supervisors on how to (1) take advantage of hiring flexibilities; (2) promote the use and accessibility of departmentwide training opportunities, including e-learning; and (3) provide Web-based information on training and human capital policies and procedures. In addition, supporting documentation, such as the DHS Workforce Plan FY 2005–2008, includes a communications plan on how to keep DHS personnel informed on workforce changes, including the department’s progress in implementing them. Furthermore, according to DHS CHCO and CIO officials, the department captures employee feedback on its practices through representatives to the IT Human Capital Resource Center and also through annual IT staff surveys. Annual performance plans, budgets, and performance reports document plans for and progress toward human capital goals. As directed by OMB, DHS reports quarterly on its progress on human capital goals. However, DHS’s IT human capital plan and supporting documentation do not provide for developing annual performance plans, budget documents, or performance reports that discuss plans for and progress against human capital goals. In addition, the information reported to OMB is primarily on DHS efforts to close IT competencies and skills gaps, which is just one of the multiple goals and objectives in DHS’s plan and supporting documentation. Work units have documented performance goals and objectives linked to the agency strategic plan and performance plan. Although DHS’s IT human capital plan and related documentation support having measurable performance goals for work units, such performance plans and measures have not been fully developed. For example, DHS CIO and CHCO officials stated that although the department has documented performance goals and objectives for some work units (e.g., managers in Customs and Border Protection) and linked them to department-level organizational goals, it had not done so for much of the department. Specifically, only managers in the DHS CHCO’s office and also at Customs and Border Protection have performance objectives that are linked to strategic plans. The agency’s strategic planning process documents and tracks mission-critical occupations and competency gap-reduction efforts. DHS’s IT human capital plan and supporting documentation provide details on the department’s strategic planning process, including the reporting and tracking of mission-critical occupations and efforts to reduce competency gaps. For example, in supporting documentation (e.g., the DHS Workforce Plan FY 2005– 2008), the department describes a workforce planning process that is to help identify the human capital resources needed to meet mission goals and develop strategies for developing or acquiring those resources. In addition, other supporting documentation (e.g., the IT Human Capital Plan to Mitigate IT Competency and Skill Gaps) identifies mission-critical IT occupations and high-level efforts needed to close its competency gaps. The department uses its OMB quarterly reports to document and track the status of efforts to close those competency gaps. Strategies are developed and implemented for reducing competency gaps through training, development, or alternative sources. As described in its IT human capital plan and supporting documentation, DHS’s strategies to close its competency gaps consist of a number of human capital initiatives, including training, staff development, and an outside executive exchange program. Specifically, DHS’s IT Human Capital Plan to Mitigate IT Competency and Skill Gaps details planned and ongoing efforts to mitigate gaps using, for example, strategic recruitment through outsourcing, private/public cross training, internal training, and e-training. The IT human capital plan and supporting documentation state that data on its progress toward meeting human capital goals will be reported to OMB and DHS management as required; they do not, however, specify what data are to be reported. The department reports quarterly to OMB on the status of efforts to close competency gaps. In addition, a recently completed (May 2007) DHS workforce survey and gap analysis identify existing IT competency gaps, but do not indicate any progress in closing them. According to DHS CHCO and CIO officials, the department to date has had limited resources and data available to assess the effectiveness of ongoing efforts to close competency gaps. They further stated that they intend to use the newly completed analysis as a baseline for measuring the success of future efforts. Recruitment strategies are created to maintain mission-critical competencies at the desired level using business forecasting and workforce analysis results. Documentation supporting DHS’s IT human capital plan provides for developing recruiting strategies based on workforce forecasting and analysis results. For example, the DHS Workforce Plan FY 2005–2008 states that the department is to use a strategic approach to recruitment and workforce planning. In addition, it identifies a DHS corporate recruitment workgroup, which includes senior human resources and civil rights staff throughout DHS, who are to assess departmentwide recruitment activities and tools; coordinate participation in recruitment fairs; and develop recruitment strategies and activities for crosscutting occupations, primarily entry-level positions. In May 2007, the department developed an improvement plan that provides updated strategies for addressing competency gaps and maintaining mission-critical competencies. This plan is based on the analysis of a recently completed workforce survey. Human capital risks are tracked, documented, and reported to a central advisory or management board, and action is taken to mitigate high-risk areas. Documents supporting DHS’s IT human capital plan (e.g., the IT Human Capital Plan to Mitigate IT Competency and Skill Gaps and the IT Gap Analysis Report and Improvement Plan) identify and document some but not all key human capital risks and do not provide for reporting risks to management or a management board. More specifically, these plans document that failure to fill critical competency and skill sets (e.g., IT project management and IT security) poses a medium-to-high human capital risk to DHS’s ability to achieve mission goals. However, DHS CHCO and CIO officials acknowledged that the department does not track these risks through any formal mechanism. In addition, they stated that DHS has not established a comprehensive effort to identify and track the full range of human capital risks facing the department, as well as reporting those risks to management or a central advisory or management board. Applicable merit principles and standards are upheld, and employee grievances are considered and addressed. DHS’s IT human capital plan and supporting documentation provide for the application and enforcement of merit principles and standards and for considering and addressing employee grievances. For example, the plan recognizes that the department has legislative and regulatory requirements to implement performance-based management practices, including merit principles and standards, for its IT workforce. In addition, an OPM analysis of DHS’s human resources management operations reports that the department’s human resources management operates in a consistent manner with merit principles. Moreover, DHS has a policy directive that defines the process for administering its employee grievance system. The department also regularly reports to management and employees on the number of grievances filed and resolved as well as the number of cases outstanding. Program and initiative implementation efforts include published plans that clearly outline periodic review of performance and desired outcomes. DHS’s IT human capital plan and supporting documentation provide for performance reviews of desired outcomes. For example, supporting documentation (specifically, DHS’s IT Human Capital Plan to Mitigate IT Competency and Skill Gaps and IT Gap Analysis Report and Improvement Plan) provides analyses and snapshots of the department’s performance in trying to close gaps in mission-critical competencies. However, these competency gap snapshots do not constitute a comprehensive review and evaluation of progress against all of the objectives established in the human capital plan. In addition, DHS’s IT human capital plan and supporting documentation do not clearly outline or identify time frames for periodic review. DHS CHCO and CIO officials stated that they intend to address this in future revisions to the plan. DHS’s IT human capital plan does not clearly assign accountability for human capital improvements or provide for regular assessments of that accountability. However, documents supporting the plan (specifically, the department’s November 2005 implementation briefing and the May 2007 IT Gap Analysis Report and Improvement Plan) do assign accountability and responsibility for human capital improvements. For example, the documents assign accountability to the DHS CHCO, DHS CIO, and component agency heads to make improvements related to closing selected competency gaps. However, the documents do not provide for assessing accountability on a regular basis and using the results as an input into future planning and resource allocation decisions. DHS CHCO and CIO officials stated that while data related to competency gaps are used as an input, data regarding accountability are not. In addition to the individual named above, Gerard Aflague, Mathew Bader, Justin Booth, Barbara Collier, S. Mike Davis, Bill Doherty (Assistant Director), and Gary Mountjoy (Assistant Director) made key contributions to this report. | In performing its missions, the Department of Homeland Security (DHS) relies extensively on information technology (IT). Recognizing this, DHS's fiscal year 2006 appropriations act required its Chief Information Officer (CIO) to submit a report to congressional appropriations committees that includes, among other things, an IT human capital plan, and the act directs GAO to review the report. GAO's review addressed (1) whether the IT human capital plan is consistent with federal guidance and associated best practices and (2) the status of the plan's implementation. In performing its review, GAO compared DHS's plan and supporting documentation with 27 practices in the Human Capital Assessment and Accountability Framework of the Office of Personnel Management, and examined plan implementation activities at three DHS component agencies. DHS's IT human capital plan is largely consistent with federal guidance and associated best practices; however, it does not fully address a number of important practices that GAO examined. Specifically, the plan and supporting documentation fully address 15 practices; for example, they provide for developing a complete inventory of existing staff skills, identifying IT skills that will be needed to achieve agency goals, determining skill gaps, and developing plans to address such gaps. They also provides for involving key stakeholders--such as the CIO, the Chief Human Capital Officer (CHCO), and component agency CIOs and human capital directors--in carrying out the skill gap analyses and other workforce planning activities. Nevertheless, elements of 12 of the 27 practices are not included in the plan or related documentation. For example, although the plan and supporting documents describe the department's IT human capital goals and steps necessary to implement them, most steps do not include associated milestones. In addition, although the plan and supporting documents provide for involving key stakeholders, they do not specifically assign these stakeholders responsibility and accountability for carrying out planned activities. These and other missing elements of the practices are important because they help ensure that the plan is implemented efficiently and effectively. DHS officials provided various reasons why the missing practices were omitted, including uncertainty surrounding the source of resources for implementing the plan and the demands of other IT priorities, such as consolidating component agency data centers. To date, DHS has made limited progress in implementing the plan, according to officials from the offices of the department's CIO and CHCO and three DHS agencies (the Coast Guard, Customs and Border Protection, and the Federal Emergency Management Agency). These officials said that they are nonetheless following several of the practices because they are required to report quarterly to the Office of Management and Budget on progress in meeting such human capital goals as filling mission-critical positions and delivering key IT training. DHS officials stated that the department's limited progress in implementing the plan was due to its focus on other priorities, and ambiguity surrounding plan implementation roles and responsibilities. Until DHS has a complete plan that fully addresses all practices and the department and components implement the plan, DHS will continue to be at risk of not having sufficient people with the right knowledge, skills, and abilities to manage and deliver the IT systems that are essential to executing the department's mission and achieving its transformation goals. |
Initial concerns about the threat posed by nuclear smuggling were focused on nuclear materials originating in the former Soviet Union. As a result, the first major initiatives concentrated on deploying radiation detection equipment at borders in countries of the former Soviet Union and in Central and Eastern Europe. In particular, in 1998, DOE established the Second Line of Defense program, which, through the end of fiscal year 2005, had installed equipment at 83 sites mostly in Russia. In 2003, DOE implemented a second program, the Megaports Initiative, to focus on the threat posed by nuclear smuggling overseas by installing radiation detection equipment at major seaports around the world.In the United States, the U.S. Customs Service began providing its inspectors with portable radiation detection devices in 1998. After September 11, 2001, the agency expanded its efforts to include the deployment of portal monitors— large-scale radiation detectors that can be used to screen vehicles and cargo. In March 2003, the U.S. Customs Service was transferred to DHS, and the border inspection functions of the Customs Service, including radiation detection, became the responsibility of CBP. Deploying radiation detection equipment at U.S. borders is part of DHS’s strategy for addressing the threat of nuclear and radiological terrorism. DHS’s strategy includes: (1) countering proliferation at the source by assisting foreign governments in their efforts to detect and interdict nuclear and radiological smuggling; (2) controlling the illegal export of technology and equipment from the United States that terrorists could use to develop a nuclear or radiological weapon; (3) detecting and interdicting potential smuggling attempts before they reach the United States; and (4) securing U.S. ports-of-entry through multiple technologies that include radiation detection and nonintrusive inspections to view images of cargo in sea containers. CBP plans to deploy radiation portal monitors in five phases, or “categories of entry”: (1) international mail and express courier facilities; (2) major northern border crossings; (3) major seaports; (4) southwestern border crossings; and (5) all other categories, including international airports, remaining northern border crossings and seaports, and all rail crossings. In this final phase, CBP also plans to replace the currently-fielded portal monitors with newer, more advanced technology. Generally, CBP prioritized these categories according to their perceived vulnerability to the threat of nuclear smuggling. CBP did not, however, conduct a formal threat assessment. International mail and express courier facilities present a potential vulnerability because mail and packages arrive with no advance notice or screening. Northern border crossings are also vulnerable, according to CBP, because of the possible presence of terrorist cells operating in Canada. The third category, major seaports, is considered vulnerable because sea cargo containers are suitable for smuggling and because of the large volume of such cargo. Seaports account for over 95 percent of the cargo entering the United States. Southwestern borders are vulnerable because of the high volume of traffic and because of the smuggling that already occurs there. Although airlines can quickly ship and deliver air cargo, CBP considers air cargo to be a slightly lesser risk because the industry is highly regulated. In deploying radiation detection equipment at U.S. borders, CBP identified the types of nuclear materials that might be smuggled, and the equipment needed to detect its presence. The radiological materials of concern include assembled nuclear weapons; nuclear material that could be used in a nuclear weapon but that is not actually assembled into a weapon (“weapons-grade nuclear material”); radiological disbursal devices, commonly called “dirty bombs;” and other illicit radioactive material, such as contaminated steel or inappropriately marked or manifested material. Detecting actual cases of attempted nuclear smuggling is difficult because there are many sources of radiation that are legal and not harmful when used as intended. These materials can trigger alarms (known as “nuisance alarms”) that are indistinguishable from those alarms that could sound in the event of a true case of nuclear smuggling. Nuisance alarms are caused by patients who have recently had radiological treatment; a wide range of cargo with naturally occurring radiation, such as fertilizer, ceramics, and food products; and legitimate shipments of radiological sources for use in medicine and industry. In addition, detecting highly-enriched uranium, in particular, is difficult because of its relatively low level of radioactivity. Furthermore, a potential terrorist would likely attempt to shield the material to reduce the amount of radiation reaching the detector and thereby decrease the probability of detection. The process of deploying portal monitors begins with a site survey to identify the best location at an entry point for installing the equipment. While in some cases the choice may be obvious, operational considerations at many entry points require analysis to find a location where all or most of the cargo and vehicles can pass through the portal monitor without interfering with the flow of commerce. After identifying the best option, CBP works with local government and private entities to get their support. At many U.S. entry points, the federal government does not own the property and therefore collaborates with these entities to deploy the equipment. It is CBP’s policy to depend exclusively on such negotiations, rather than to use any kind of eminent domain or condemnation proceeding. The actual installation of the portal monitors involves a number of tasks such as pouring concrete, laying electrical groundwork, and hooking up the portal monitors to alarm systems that alert officers when radiation is detected. Finally, PNNL tests the equipment and trains CBP officers on its operation, including how to respond to alarms. To coordinate the national effort to protect the United States from nuclear and radiological threats, in April 2005, the president directed the establishment of DNDO within DHS. The new office’s mission covers a broad spectrum of responsibilities and activities, but is focused primarily on providing a single accountable organization to develop a layered defense system. This system is intended to integrate the federal government’s nuclear detection, notification, and response systems. In addition, under the directive, DNDO is to acquire, develop, and support the deployment of detection equipment in the United States, as well as to coordinate the nation’s nuclear detection research and development efforts. For fiscal year 2006, DNDO’s total budget is approximately $318 million, which includes at least $81 million for research and development of advanced nuclear detection technologies and $125 million for portal monitor purchase and deployment. The Homeland Security Act of 2002 gave DHS responsibility for managing the research, development, and testing of technologies to improve the U.S. capability to detect illicit nuclear material. Prior to the creation of DNDO, DHS’s Science and Technology (S&T) directorate had this responsibility. DNDO has assumed these responsibilities and works with S&T’s Counter Measures Test Beds (CMTB) to test radiation detection equipment in New York and New Jersey. As of January 2006, DNDO has provided $605,000 to DOE national laboratories that support this effort. Additional funding for fiscal year 2006 from S&T and DNDO to support test and evaluation activities at the CMTB is yet to be determined. The Homeland Security Act also provided DHS the authority to use DOE national laboratories for research, development, and testing of new technologies to detect nuclear material. As of December 2005, DHS had completed deployment of portal monitors at two categories of entry—a total of 61 ports-of-entry—and has begun work on two other categories; overall, however, progress has been slower than planned. According to DHS officials, the slow progress has resulted from a late disbursal of funds, and delays in negotiating deployment agreements with seaport operators. Further, we believe the expected cost of the program is uncertain because DHS’s plans to purchase newer, more advanced equipment are not yet finalized; also we project that the program’s final cost will be much higher than CBP currently anticipates. Between October 2000 and October 2005, DHS, mainly through its prime contractor PNNL, has spent about $286 million to deploy radiation detection equipment at U.S. ports-of-entry. As of December 2005, DHS had deployed 670 of 3,034 radiation portal monitors—about 22 percent of the portal monitors DHS plans to deploy. The agency has completed portal monitor deployments at international mail and express courier facilities and the first phase of northern border sites—57 and 217 portal monitors, respectively. In addition, by December 2005, DHS had deployed 143 of 495 portal monitors at seaports and 244 of 360 at southern borders. In addition, three portal monitors had been installed at the Nevada Test Site to analyze their detection capabilities and four had been retrofitted at express mail facilities. As of February 2006, CBP estimated that with these deployments CBP has the ability to screen about 62 percent of all containerized shipments entering the United States, and roughly 77 percent of all private vehicles (POVs). Within these total percentages, CBP can screen 32 percent of all containerized seaborne shipments; 90 percent of commercial trucks and 80 percent of private vehicles entering from Canada; and approximately 88 percent of all commercial trucks and 74 percent of all private vehicles entering from Mexico. CBP does not maintain a firm schedule for deploying handheld radiation detectors, such as pagers and radiation isotope identification devices. This is equipment used mainly to help pinpoint and identify sources of radiation found during inspections. Instead, according to CBP officials, the agency acquires and deploys such equipment each fiscal year as needed. The handheld radiation detectors are procured to coincide with portal monitor deployments to ensure mission support. Since fiscal year 2001, CBP has spent about $24.5 million on pagers, and about $6.6 million on radiation isotope identification devices. At present, CBP can field roughly 12,450 pagers—enough to ensure that all officers conducting primary or secondary inspections at a given time have one. The agency intends to deploy about 6,500 additional pagers. Similarly, CBP’s 549 radiation isotope identification devices are deployed at domestic ports-of-entry. CBP intends to acquire another 900 to ensure that all needs are met. Overall, CBP and PNNL have experienced difficulty meeting the portal monitor deployment schedule. None of the planned portal monitor deployments has progressed according to schedule, and monthly deployments would have to increase by almost 230 percent to meet a September 2009 program completion date. For example, in November 2005, deployments at land crossings were about 20 months and $1.9 million behind schedule, while deployments at the first 22 seaports were about 2 years and $24 million behind schedule. Despite these delays, PNNL reported in November 2005 that the overall project schedule should not extend beyond its current completion date of September 2009. However, our analysis indicates that CBP’s deployment schedule is too optimistic. In fact, for CBP and PNNL to meet the current deployment schedule, they would have to install about 52 portal monitors per month from November 2005 to September 2009. In our view, this is unlikely because it requires a rate of deployment that far exceeds recent experience. For example, during calendar year 2005, PNNL deployed portal monitors at the rate of about 22 per month, and deployments have fallen further and further behind schedule. Between February and December 2005, for example, PNNL did not meet any of its scheduled monthly deployments, never deploying more than 38 portal monitors during any single month. If CBP continues to deploy portal monitors at its 2005 pace, the last monitor would not be deployed until about December 2014. Table 1 details the status of portal monitor deployments, as of December 2005. Further, we analyzed CBP’s earned value management data as of November 2005 and determined that, although CBP planned for the deployment program to be 20.5 percent complete by that date, the program is only about 16 percent complete. In addition, our analysis indicates that since the program’s inception, work valued at $48.6 million has fallen behind schedule. Moreover, the trend over the past 14 months shows CBP and PNNL falling further behind schedule, as seen in figure 1. There have been at least three major sources of delay that have affected the portal monitor deployment program: funding issues, negotiations with seaport terminal operators, and problems in screening rail cars— particularly in a seaport environment. According to CBP and PNNL officials, recurrent difficulties with the project’s funding are the most important explanations of the schedule delays. Specifically, according to DHS and PNNL officials, CBP has been chronically late in providing appropriated funds to PNNL, thereby hindering its ability to meet program deployment goals. For example, PNNL did not receive its fiscal year 2005 funding until September 2005. According to PNNL officials, because of this delay, some contracting activities in all deployment phases had to be delayed or halted, but the adverse effects on seaports were especially severe. For example, PNNL reported in August 2005 that site preparation work at 13 seaports had to cease because the Laboratory had not yet received its fiscal year 2005 funding allocation. According to senior CBP officials, their agency’s inability to provide a timely spending plan to the Congress for the portal monitor deployment program is the main reason for these funding delays. According to the House Appropriations Committee report on the CBP portion of DHS’s fiscal year 2005 budget, CBP should provide the Congress an acquisition and deployment plan for the portal monitor program prior to funding PNNL. However, these plans typically take many months for CBP to finalize—in part because CBP requires that the plans undergo several levels of review—but also because these plans are reviewed by DHS and OMB before being submitted to the Congress. In fiscal year 2005, this process was further delayed by the creation of DNDO, uncertainty regarding DNDO’s responsibilities, and negotiations regarding the expenditure of the fiscal year 2005 appropriations. CBP has tried to address this problem by reprogramming funds when money from other programs is available. In some cases, the amount of reprogrammed funds has been fairly large. For example, about 15 percent of fiscal year 2005’s funding included money reprogrammed from other CBP sources, or almost $14 million. In fiscal year 2004, about $16 million was reprogrammed—or about a third of the fiscal year’s total. And in fiscal year 2003, the total of reprogrammed money was about $18 million—about 20 percent. Negotiations with seaport operators have been slow and have also delayed the portal monitor deployment program. According to CBP and PNNL officials, one of the primary reasons behind the seaport phase’s substantial delay in deployments is the difficulty in obtaining contractual agreements with port and terminal operators at seaports. DHS has not attempted to impose agreements on seaport operators because, according to officials, cooperative arrangements with the port operators are more efficient and, in the long term, probably more timely. According to CBP and PNNL officials, many operators believe screening for radiation will adversely affect the flow of commerce through their ports. In addition, deploying portal monitors in major seaports presents several unique challenges. For example, seaports are much larger than land border crossings, consist of multiple terminals, and may have multiple exits. Because of these multiple exits, seaports require a greater number of portal monitors, which may entail more negotiations with port and terminal operators. In addition, port operators at times have insisted on late-stage design changes, requested various studies prior to proceeding with final designs, insisted on inefficient construction schedules, and delayed their final review and approval of project designs. According to CBP and PNNL, these efforts often reflect the port and terminal operators’ uneasiness with portal monitor deployments, and their resolve to ensure that the outcome of the deployment process maintains their businesses’ competitiveness. For example, port officials at one seaport requested several changes late in the process, including performing an unscheduled survey for laying cable, revising portal monitor locations at two gates, and adding a CBP control booth at a third terminal. According to CBP and PNNL officials, the agency prefers to accommodate these types of changes, even late in the process and even if they slow deployment, because in the long term they believe it is more efficient and effective. The difficulty of devising an effective and efficient way to conduct secondary inspections of rail traffic departing seaports without disrupting commerce has created operational issues that could further delay deployments. Four of the five seaports we visited employ rail cars to ship significant amounts of cargo. In one seaport, the port director estimated that about 80-85 percent of the cargo shipped through his port departs via rail. For the other three seaports, the percentages for rail traffic were 5 percent, 13 percent, and 40 percent respectively. According to port officials, these seaports would like to accommodate CBP’s efforts to install radiation detection equipment designed to screen rail traffic, but they are concerned that the logistics of conducting secondary inspections on trains as they prepare to depart the seaport could back up rail traffic within the port and disrupt rail schedules throughout the region—potentially costing the port tens of thousands of dollars in lost revenue. For example, one senior port authority official told us that his port lacked ample space to park trains for secondary inspections, or to maneuver trains to decouple the rail car(s) that may have caused a primary inspection alarm. As a result, trains that cause a primary alarm would have to wait, in place, for CBP to conduct a secondary inspection, blocking any other trains from leaving the port. According to this port official, any delay whatsoever with a train leaving the port could cause rail problems down the line because track switches are geared to train schedules. To avoid these kinds of problems, CBP has delayed deploying portal monitors in this seaport until technical and operational issues can be overcome. As of December 2005, no portal monitors had been deployed at this seaport, although according to PNNL’s schedule, 5 of its 11 terminals—a total of 19 portal monitors—should have been deployed by October 2005. According to the port director at another seaport we visited, a port that actually has a rail portal monitor installed, similar operational issues exist. However, in addition to backing up rail traffic within the port, trains awaiting secondary inspections at this port could block the entrance/exit to a nearby military base. The director of the state’s port authority told us that his solution has been to simply turn off the portal monitor. According to CBP officials, this was entirely a state decision, since this portal monitor is the state’s responsibility and not part of CBP’s deployment. However, these officials also noted that they agreed with the states and noted that they would not attempt to impose a solution or deadline on either port. CBP officials noted that most seaport operators seem willing to accommodate portal monitors, but until a better portal monitor technology evolves that can help ensure a smooth flow of rail traffic out of the port, negotiations with seaport operators will continue to be slow. According to CBP and port officials, they have considered several potential solutions. For example, there is widespread agreement that screening sea cargo containers before they are placed on rail cars offers the best solution, but this option is operationally difficult in many seaports. Mobile portal monitors, when commercially available, may also offer a partial solution. In addition, CBP is optimistic that advanced portal monitors, when they become commercially available, may help solve some of the problems in the rail environment by limiting the number of nuisance alarms. However, according to the CBP and port officials we contacted, screening rail traffic continues to pose a vexing operational problem for seaports. The concerns that seaport operators and CBP expressed regarding screening rail commerce in seaports may increase and intensify in the future because rail traffic, in general, is expected to increase substantially by 2020. DOT has forecast that by 2020, rail will transport roughly 699 million tons of international freight—up from 358 million tons carried in 1998. Officials at 3 of the 5 seaports we visited expect rail traffic through their facilities to increase dramatically during the next 10 to 15 years. As the volume of trade increases, so too will the economic stakes for the port and terminal operators, while the regulatory burden for CBP is likely to increase as well. Delays—for any reason, including radiation detection— are likely to become more costly, and CBP will likely have ever-increasing numbers of rail cars to screen. In addition, although CBP is not scheduled to begin deploying portal monitors to screen rail shipments at land border crossings until 2007, the agency will likely experience operational challenges at land border crossing similar to those it is now experiencing at seaports. For example, at both land border crossings and seaports, if a rail car alarms as it passes through a portal monitor, that car will possibly have to be separated from the remaining train—sometimes a mile in length—to undergo a secondary inspection. Furthermore, because trains transport numerous types of cargo containing large quantities of naturally occurring radioactive material, CBP faces the challenge of maintaining a nuisance alarm rate that does not adversely affect commerce. CBP and PNNL are currently conducting testing of a prototype rail portal monitor to determine the potential impact of naturally occurring radioactive material on rail operations at land border crossings. Unforeseen design and construction problems have also played a role in delaying portal monitor deployments. For example, deployments at six southern border sites have been delayed to coincide with the sites’ expansion activities. According to CBP officials, there are two approaches to accommodating a port-of-entry’s alterations, both of which may delay portal monitor deployments. First, CBP and PNNL may decide to delay the start of portal monitor projects until the port-of-entry completes its alterations, to make certain that portal monitor placements are properly located. Second, port-of-entry expansion activities may alter existing traffic flows and require that PNNL redesign its portal monitor deployments. The portal monitor deployments at three southern border ports-of-entry has taken much longer than planned because of the port’s expansion activities. According to PNNL, there is now considerable schedule uncertainty associated with these deployments, which may ultimately impact the completion of the southern land border deployments. Portal monitor deployments have also been hampered by poor weather. For example, cold weather at several northern sites caused some unexpected work stoppages and equipment failures that resulted in construction delays of 2 to 3 months. Finally, one southern border site has been delayed because of major flooding problems. The flooding issue must be resolved before the deployment can be completed. DHS’s current estimate to complete the program is $1.3 billion, but this estimate is highly uncertain and overly optimistic. First, DHS’s cost estimate is based on a plan to deploy advanced-technology portal monitors that have so far shown mixed results for detecting radiation compared to currently-fielded portal monitors. Since the efficacy of the advanced portal monitors has not yet been proven conclusively, there is at least some uncertainty over whether—and, if so, how many—of the new portal monitors may be deployed. In addition, the final cost of the new portal monitors has not been established. Second, our analysis of CBP’s earned value data also suggests that the program will likely cost much more than planned. The current deployment plan calls for installing advanced portal monitors at all cargo primary and secondary inspection locations, at all secondary inspection locations for private vehicles, and also retrofitting many sites with the advanced equipment, when it becomes available. However, according to senior officials at DNDO, the advanced technology must meet all of DNDO’s performance criteria, and must be proven superior to the portal monitors already in use, before DNDO will procure it for use in the United States. Recent tests of the new portal monitors indicate that DNDO’s criteria have not yet been met. For example, S&T sponsored research in 2004 that compared the detection capabilities of currently- fielded portal monitors with the advanced portal monitors. The results of that research suggested that, in some scenarios, the detection abilities of the two portal monitor types were nearly equivalent. In other scenarios, the new equipment’s detection capability was significantly better. S&T concluded that more work remains to be done in optimizing and comparing portal monitors so as to understand how they can be used to the greatest effect at U.S. ports-of-entry. In 2005, DNDO sponsored additional research designed to compare the two types of portal monitor, and determined that the advanced portal monitors’ detection capabilities were somewhat better than those of the currently-fielded equipment. In addition, in October 2005, DNDO completed the first comprehensive tests for these advanced portal monitors at the Nevada Test Site. This advanced technology combines the ability to detect radiation and identify its source. According to an official who helped supervise these tests, the new portal monitors’ performance did not meet all of DNDO’s expectations with regard to providing significant detection improvements over currently-fielded equipment in all scenarios. CBP and DNDO officials also expressed concerns regarding the advanced portal monitors’ detection capabilities in light of the Nevada test results. In particular, senior CBP officials questioned whether the advanced portal monitors would be worth their considerable extra costs, and emphasized finding the right mix of current and advanced-technology equipment based on the needs at individual ports-of-entry. According to DNDO officials, the potential improvement over currently fielded portal monitors in capability to identify radioactive sources, and hence to detect actual threats as opposed to simply detecting radiation, has not yet been quantified. However, these officials believe that the results to date have been promising, and DNDO intends to continue supporting the advanced portal monitor’s development and believe the new technology may be ready for deployment early in calendar year 2007. There is also considerable uncertainty regarding the eventual cost of the advanced portal monitors—if they become commercially available, and if DNDO opts to use them. Experts we contacted estimated that the new portal monitors could cost between $330,000 and $460,000 each. These estimates are highly uncertain because advanced portal monitors are not yet commercially available. As a point of reference, the portal monitors currently in use typically cost between $49,000 and $60,000. These costs include only the purchase price of the equipment, not its installation. According to CBP and PNNL officials, installation costs vary, but average about $200,000 per portal monitor. Even if future test results indicate that the new technology exhibits much better detection and identification capabilities, it would not be clear that the dramatically higher cost for this new equipment would be worth the considerable investment, without the agency having first rigorously compared the portal monitors’ capabilities taking their costs into account. Currently, DNDO and CBP are working together to determine the most appropriate technologies and concepts of operation for each port-of-entry site. The two agencies are also trying to determine the highest priority sites for advanced-technology portal monitors based on the extent to which the new portal monitors show improved performance. In November 2005, PNNL reported that the portal monitor deployment program could experience an overall cost overrun of $36 million. In contrast, our analysis of CBP’s earned value data indicates that the agency should expect a cost overrun of between $88 million and $596 million. We based our cost overrun projections on the rates at which CBP and PNNL deployed portal monitors, through November 2005. The more efficient the agency and its contractor are in deploying portal monitors, the smaller the cost overruns; conversely, when efficiency declines, cost overruns increase. In fact, as shown in figure 2, recent cumulative program cost trends have been negative, indicating that CBP’s cost overruns are deepening over time. PNNL noted that its management reserve of $62 million should cover the anticipated overrun. However, we do not agree. First, we believe the cumulative cost overrun will far exceed PNNL’s estimate of $36 million. We believe an overrun of about $342 million, the midpoint of our projected overrun range, is more likely. Since 1977, we have analyzed over 700 acquisition projects on which EVM techniques have been applied. These analyses consistently show that once a program is 15 percent complete (as is the case with this program), cost performance almost never improves and, in most cases, declines. PNNL’s recent cost trend follows this pattern. Second, based on these 700-plus studies, our estimate takes a more realistic view that the portal monitor deployment program’s cost performance most likely will continue to decline; hence the management reserve will be consumed over time as the program incurs unexpected expenses. Finally, to meet the deployment program’s planned costs, PNNL would have to greatly improve its work efficiency. However, our analysis of prior EVM- based projects indicates that productivity rates nearly always decline over the course of a project. We determined that PNNL’s efficiency rate for the most recent 8 months has averaged about 86 percent—PNNL has been delivering about $.86 worth of work for every dollar spent. In order to complete the remaining work with available funding, PNNL’s efficiency rate would have to climb to around 98 percent, a rate of improvement unprecedented in the 700-plus studies we have analyzed. Federal agencies are required by OMB to track the progress of major systems acquisitions using a validated EVM system and to conduct an integrated baseline review. We found that PNNL has an EVM system but has not certified it to show that it complies with guidance developed by the American National Standards Institute/Electronic Industries Alliance. This guidance identifies 32 criteria that reliable EVM systems should meet. In addition, we found that PNNL has not conducted an integrated baseline review—a necessary step to ensure that the EVM baseline for the portal monitor program represents all work to be completed, and adequate resources are available. However, although the EVM data have not been independently validated, we examined the EVM data and found that they did not show any anomalies and were very detailed. Therefore, we used them to analyze the portal monitor program status and to make independent projections of the program’s final costs at completion. CBP officers we observed conducting primary and secondary inspections appeared to use radiation detection equipment correctly and to follow the agency’s inspection procedures. In fact, in some cases, CBP officers exceeded standard inspection procedure requirements by opening and entering containers to better identify radiation sources. In contrast, in 2003, when we issued our last report on domestic radiation detection, CBP officers sometimes deviated from standard inspection procedures and, at times, used detection equipment incorrectly. However, the agency’s inspection procedures could be strengthened. During this review, at the 10 ports-of-entry that we visited, the CBP officers we observed conducting primary and secondary inspections appeared to follow inspection procedures and to use radiation detection equipment correctly. The officers’ current proficiency in these areas follows increases in training and in CBP’s experience using the detection equipment. In contrast, in 2003 we reported that CBP officers sometimes used radiation detection equipment in ways that reduced its effectiveness. CBP has increased the number of its officers trained to use radiation detection equipment; in fact, the agency now requires that officers receive training before they operate radiation detection equipment. As of February 2006, CBP had trained 6,410 officers to use radiation isotope identification devices, 8,461 to use portal monitors, and 22,180 to use pagers. Many CBP officers received training on more than one piece of equipment and about 900 have since left the agency. Generally, today CBP officers receive radiation detection training from 4 sources: the CBP Academy in Glynco, Georgia; the Border Patrol Academy in Artesia, New Mexico; a DOE- sponsored 3-day training course for interdicting weapons of mass destruction, in Washington state; and on-the-job training at ports-of-entry. Training at the Academies in Georgia and New Mexico includes formal classroom instruction, as well as hands-on exercises on how to use portal monitors, isotope identifiers, and pagers. This training includes simulated scenarios in which officers use radiation detection equipment to conduct searches for nuclear and radiological materials. On-the-job instruction continues at field locations as senior CBP officers, as well as PNNL and other DHS contractor staff, work closely with inexperienced officers to provide them with practical training on how the radiation detection equipment works and how to respond to alarms. According to senior CBP officials, all of the instructors that offer training on using radiation detection equipment are certified in its use. Trainees must demonstrate proficiency in the use of each system prior to assuming full responsibility for radiation detection inspections. About 1,600 CBP officers have participated in DOE’s 3-day training course designed to acquaint CBP officers with detection equipment. CBP is currently developing refresher training courses on the use of radiation detection equipment. To further enhance officers’ ability to effectively respond to real or potential threats, several of the field locations that we visited conduct “table-top exercises” that simulate scenarios in which the equipment detects an illicit radiological source. According to several of the CBP field supervisors we contacted, many officers have gained proficiency in following procedures and using radiation detection equipment through substantial field experience responding to alarms. The number of alarms officers typically handle varies according to the size of the site, its location, and type. For example, an isolated land border site would probably experience fewer alarms than a major seaport because of the differences in the volume of traffic. However, it was common for several of the locations we visited to experience 15 to 60 alarms per day. One seaport we visited had 9 terminals, usually with 2 primary and 1 secondary portal monitors. According to CBP officials, each terminal recorded about 8 to 12 alarms per day. The director of port security for a major eastern seaport we visited estimated that her facility records roughly 150 portal monitor alarms each day. Virtually all have been nuisance alarms, but CBP officials still believe they gained valuable experience in using the equipment and following procedures. All of the primary and secondary inspections we witnessed were nuisance alarms. In all of these cases except one, officers followed CBP’s guidance— as well as local variations meant to address issues unique to the area—and correctly used detection equipment. The lone exception occurred at a site whose primary inspection station was staffed by a state port police officer. After the station’s portal monitor registered an alarm for a truck departing the site, the police officer did not follow CBP’s procedures. For example, he did not collect any documentation from the driver. At all other sites we visited, when a primary portal monitor sounded, CBP officers gathered the cargo’s manifest, the vehicle registration, and the driver’s license prior to sending the vehicle through secondary inspection. Officers use these documents to check the driver and vehicle cargo. The port police officer told us that he recognized the driver in this case, and so the officer did not believe it was necessary to collect such information. A CBP officer performed the secondary inspection in line with agency guidance. In fact, after using a radiation isotope identification device to conduct an external inspection and determine the source of the alarm—potassium hydroxide— the officer required that the driver open the back of the truck so she could make a visual check of the cargo. From the time of the initial alarm, until the truck departed the site boundary, about 35 minutes elapsed. According to port and CBP officials, this particular alarm, its resolution, and the amount of time it took to resolve are typical of the site. We also discussed the site’s radiation detection efforts with the truck driver, in particular the delay associated with this alarm. He noted that he considers the delays experienced at this site to be relatively minor, and that the delays have not had any adverse effects on his business. We also visited a seaport that experienced a legitimate alarm in which CBP officers used the detection equipment correctly and responded according to procedures. Uranium hexafluoride, a potentially hazardous chemical containing low levels of radioactivity, caused this alarm. A primary portal monitor at the seaport sounded as a truck carrying one container attempted to exit a terminal. Following standard operating procedures, the truck was diverted to a secondary inspection station, where a secondary portal monitor also alarmed. A CBP officer then scanned the container and cab of the truck with an isotope identifier, which indicated that the radiation source was located in the cab within several metal pails. The isotope identifier identified two radiation sources, one of which was uranium-235—potentially a weapons-usable material. The other source was uranium-238. Again following procedures, CBP officers isolated the sources of radiation and provided LSS scientists with information collected by the isotope identifier. Officers also reviewed the driver’s delivery papers; used various CBP databases to check the driver, importer, and consignee’s history of transporting goods; and contacted the driver’s dispatcher and the U.S. consignee to gather information on and assess the legitimacy of the shipment. The consignee explained that the pails contained trace amounts of uranium hexafluoride that had been sent to the company’s laboratory for testing. Following additional investigation, which included an X-ray of the pails and a review of DOT requirements regarding radiation-warning placard requirements, CBP determined that the event was not a security threat and released the driver and conveyance. Senior officials at this seaport told us that CBP’s radiation detection guidance served as an effective and successful guide to resolving this alarm. We identified two potential issues in CBP’s national inspection procedures that could increase the nation’s vulnerability to nuclear smuggling. The first potential issue involves NRC documentation. Generally, NRC requires that importers obtain an NRC license for their legitimate shipments of radiological materials into the United States. However, NRC regulations do not require that the license accompany the shipment, although in some cases importers choose to voluntarily include the license. According to CBP officials, CBP lacks access to NRC license data that could be used to verify that importers actually acquired the necessary licenses or to authenticate a license at the border. At present, CBP officers employ a variety of investigative techniques to try to determine if individuals or organizations are authorized to transport a radiological shipment. For example, CBP officers review their entry paperwork, such as shipping papers. Officers also often interview drivers about the details of the delivery and observe their behavior for any suspicious or unusual signs. At one land border crossing we visited, officers told us that frequent and legitimate shippers of radiological material provide advance notice that a radiological shipment will be transported. This can lead to law enforcement personnel being called in to escort the shipment through the port-of-entry. The second potential issue pertains to CBP’s secondary inspection guidelines. Generally, CBP’s guidelines require that CBP officers locate, isolate, and identify the radiation source(s) identified during primary inspections. Customarily, officers use a radiation isotope identification device to perform an external examination of cargo containers in these situations. (See fig. 3.) However, the effectiveness of a radiation isotope identification device is diminished as its distance from the radioactive source increases, and by the thickness of the metal container housing the radioactive source. As a result, secondary inspections that rely solely on external examinations may not always be able to locate, isolate, and identify an illicit shipment of nuclear material. The local procedures at some ports-of-entry we visited go beyond the requirements established by CBP’s guidelines by having CBP officers open and, if necessary, enter containers when conducting secondary inspections. (See fig. 4.) For example, at one high-volume seaport we visited, the local inspection procedures require officers to open and, if necessary, enter a container to locate and identify a radiological source if an external examination with an isotope identifier is unable to do so. Under such circumstances, the port’s procedures require the officer to open the container doors, locate the source, and obtain another reading as close to the source as possible. By entering the container, an officer may be able to reduce the isotope identifier’s distance from the radioactive source, and thus obtain a more accurate reading. If the isotope identifier is unable to detect and identify the source after two readings within the container, officers must contact LSS for further guidance. Officers at this seaport have opened containers in the past when the isotope identifier had been unable to detect naturally occurring radioactive material, such as granite or ceramic tile, which is low in radioactive emissions. CBP supervisors at this seaport said that this occurs infrequently and that it adds a very minimal amount of time to the inspection process. In addition, at a land border crossing we visited, the local standard operating procedures instruct CBP officers to conduct a physical examination on vehicles that alarm for the presence of radiation. Officials at this particular port-of-entry said that they have entered vehicles with an isotope identifier when the device has been unable to detect or identify the radioactive source from vehicles’ exterior. During a physical examination, officers are supposed to open the vehicle and look for high-density materials, such as lead or steel, which can be used to shield gamma radiation and solid objects with large quantities of liquid that could be used to shield neutron radiation. Because the majority of alarms at this land border crossing are caused by medical isotopes in people, CBP officers physically inspect vehicles on an infrequent basis. Finally, we also visited a land border crossing where CBP officers routinely open and enter commercial trucks to conduct secondary inspections, even though the site’s local procedures do not require this additional examination. Officials at this crossing said that they open up containers to verify that the container’s manifest and reading from the isotope identifier are consistent with the container’s load. If they are not consistent, CBP officers are supposed to contact LSS for further guidance. During our visit, we observed a truck that alarmed at primary and secondary portal monitors. CBP officers then required the driver to park at a loading dock, where officers first used an isotope identifier to screen the truck from the outside; the reading from the isotope identifier was inconclusive, however. Officers then opened and entered the container with an isotope identifier, conducted a second reading of the radioactive source, and determined that the material inside the container was a non-threatening radioactive source that matched the manifest. A CBP supervisor released the truck. This inspection, from the time of the original alarm to the truck’s release took about 25 minutes—slightly greater than the 20-minute average for this site. According to CBP supervisors, officers at this port-of-entry follow this practice routinely, even during the site’s peak hours. This approach enables the officers to get closer to the source and obtain a more accurate reading. Furthermore, since this practice enables officers to conduct a more thorough examination of the containers’ contents, it may increase the likelihood that CBP officers will find any illicit radioactive material. According to senior CBP officials at this port-of-entry, despite being implemented at one of the busiest commercial ports-of-entry in the nation, this additional procedure has had little negative impact on the flow of commerce and has not increased the cost of CBP inspections. DHS has managed research, development, and testing activities that attempt to address the inherent limitations of currently-fielded radiation detection equipment and to produce new, advanced technologies with even greater detection capabilities. DHS is enhancing its ability to test detection equipment by building a new test facility at DOE’s Nevada Test Site. In addition, DHS tests radiation detection equipment under real-life conditions at S&T’s CMTB in New York and New Jersey. However, much work remains for the agency to achieve consistently better detection capabilities, as the efforts undertaken so far have achieved only mixed results. Currently-fielded radiation portal monitors have two main limitations. First, they are limited by the physical properties of the radiation they are designed to detect, specifically with regard to the range of detection (some radioactive material emits more radiation than others). Further, this limitation can be exacerbated because sufficient amounts of high-density materials, such as lead or steel, can shield radiation emissions to prevent their detection. Second, currently-fielded portal monitors cannot distinguish between different types of radioactive materials, i.e., they cannot differentiate naturally occurring radioactive material from radiological threat materials. CBP officers are required to conduct secondary inspections on all portal monitor alarms, including nuisance alarms. According to the CBP field supervisors with whom we spoke, nuisance alarms comprise almost all of the radiation alerts at their ports-of- entry. Port operators noted a concern that nuisance alarms might become so numerous that commerce could be impeded, but thus far these alarms have not greatly slowed the flow of commerce through their ports-of-entry. CBP’s currently-fielded radiation isotope identification devices also have inherent limitations. For example, during some secondary inspections, radiation isotope identification devices are unable to identify radiological material. In these cases, CBP standard procedures require that officers consult LSS to conclusively identify the source. According to CBP officers at two of the ports we visited, this usually lengthens secondary inspections by 20 to 30 minutes, although in some cases an hour or more was needed to resolve the alarm. Furthermore, a 2003 Los Alamos National Laboratory evaluation of seven isotope identifiers, including the one deployed by CBP, concluded that all devices had difficulty recognizing radioactive material and correctly identifying the material they did recognize. The Los Alamos finding is consistent with our field observations, as CBP officers at several of the ports-of-entry we visited reported similar trouble with their radiation isotope identification devices. Laboratory testing of currently-fielded radiation detection equipment has further demonstrated their limitations in effectively detecting and identifying nuclear material. For example, in February 2005, DHS sponsored testing of commercially available portal monitors, isotope identifiers, and pagers against criteria set out in American National Standards Institute (ANSI) standards. The ANSI standards provide performance specifications and test methods for testing radiation detection equipment, including portal monitors and handheld devices. The actual testing was performed by four DOE laboratories, with coordination, technical management, and data evaluation provided by the Department of Commerce’s National Institute for Standards and Technology (NIST). The laboratories tested a total of 14 portal monitors from 8 manufacturers against 29 performance requirements in the ANSI standards. Overall, none of the radiation detection equipment, including the portal monitors and handheld devices deployed by CBP, met all of the performance requirements in this first round of testing. However, according to S&T officials, many of the limitations noted in CBP’s equipment were related to withstanding environmental conditions—not radiation detection or isotope identification. However, in some tests, the portal monitors that CBP employs, along with many others, exhibited poor results. For example, in tests conducted to evaluate the portal monitors’ response to neutron radiation, of which plutonium is a primary source, almost all monitors, including a portal monitor fielded by CBP, failed to meet the ANSI requirement. However, according to S&T officials, the test was conducted using the manufacturer’s standard configuration, rather than the configuration CBP uses in its field operations. In another test, one that used CBP’s typical field parameters rather than the manufacturer’s, the portal monitor passed all the radiation detection performance requirements. S&T believes that the portals used by CBP would meet all the radiation performance requirements if set up with the parameters and configuration as used in the field. In addition, isotope identifiers displayed weaknesses. For example, the isotope identifier currently in use by CBP was not able to simultaneously identify two different isotopes, as required by the ANSI standards. When tested with barium-133 and plutonium-239, the isotope identifier was able to recognize the barium but failed to recognize the plutonium—a weapons-grade nuclear material. As this was a first round of testing and modifications were made to both the standards and testing protocols after the procedures were completed, NIST plans to manage testing of the equipment again in early 2006. The results from both rounds of testing are intended to provide guidance for federal, state, and local officials in evaluating and purchasing radiation detection equipment, and to enable manufacturers to improve their equipment’s performance. DHS has sponsored research efforts designed to improve the detection capabilities of the currently-fielded portal monitors and to provide them with the ability to distinguish radiological sources. For example, PNNL researched, developed, and tested a new software—known as “energy windowing”—to address the currently-fielded portal monitors’ inability to distinguish between radiological materials. Energy-windowing is supposed to identify and screen out material, such as fertilizer or kitty litter, that cause nuisance alarms and thereby reduce the number of such alarms at cargo screening facilities, while also improving the portal monitor’s sensitivity to identify nuclear material of concern. PNNL has activated energy-windowing on the 556 portal monitors it has deployed at land border crossings and seaports. At a few ports-of-entry that we visited, CBP officials said that the software has been effective in significantly reducing the number of nuisance alarms. However, tests of the software have shown that its effectiveness in reducing nuisance alarms largely depends on the types of radiation sources it has been programmed to detect and differentiate. In tests involving some common, unshielded radiation sources, such as cobalt-57 and barium-153, the new software has shown improved detection and discrimination capabilities. However, during scenarios that target other common, shielded threat sources—such as those that might be used in a shielded radiological dispersal device or nuclear weapon—the software has been less able to detect and discriminate. Experts have recommended further testing to fully explore the software’s capabilities. DHS is also sponsoring the development of three new technologies that are designed to address the main inherent limitations of currently-fielded portal monitors. CBP’s deployment plan currently calls for the widespread installation of the first of these technologies, “advanced spectroscopic portal monitors.” According to DNDO, the advanced spectroscopic technology uses different detection materials that are capable of both detecting the presence of radiation and identifying the isotope causing the alarm. It is hoped that the spectroscopic portal monitor can more quickly identify the sources of alarms, thereby reducing the number of nuisance alarms. This increased operational effectiveness may allow the portal monitors to be set at a lower detection threshold, thus allowing for greater sensitivity to materials of concern. DHS commissioned PNNL to determine whether spectroscopic portal monitors provide improved performance capabilities over the currently-fielded monitors. In July 2004 and July 2005, PNNL conducted two small-scale preliminary studies to compare the two types of portal monitors in side-by-side tests using shielded and unshielded radioactive materials. In the first test, PNNL concluded that the relative performance of spectroscopic and currently-fielded portal monitors is highly dependent on variables such as the radioactive sources being targeted and the analytic methods being used. The results of these tests were mixed. In some situations, spectroscopic portal monitors outperformed the current technology; in other cases, they performed equally well. In the second test, PNNL concluded that the spectroscopic monitor’s ability to detect the shielded threat sources was equal to, but no better than, those of the currently-fielded portal monitors. However, because spectroscopic portal monitors have the ability to identify isotopes, they produced fewer nuisance alarms than the current portal monitors. PNNL noted that because the studies were limited in scope, more testing is needed. In October 2005, DNDO completed the first round of comprehensive testing of spectroscopic portal monitors at its testbed at the Nevada Test Site. DNDO tested 10 spectroscopic portal monitors against 3 currently-fielded monitors in 7,000 test runs involving the portal monitors’ ability to detect a variety of radiological materials under many different cargo configurations. According to senior DNDO officials who supervised these tests, preliminary analysis of test data indicates that the spectroscopic portal monitors’ performance demonstrated somewhat mixed results. Spectroscopic portal monitors outperformed currently-fielded equipment in detecting numerous small, medium-sized, and threat-like radioactive objects, and were able to identify and dismiss most naturally occurring radioactive material. However, as the amount of source material declined in size, the detection capabilities of both types of portal monitors converged. Because the data produced by the test runs is voluminous and complex, NIST and another contractor are still in the process of analyzing the test data and plan to produce a report summarizing the results of the testing in 2006. DNDO received responses to the Advanced Spectroscopic Portal Request for Proposal in February 2006, and intends to use the data from the Nevada Test Site to help evaluate these responses. In fiscal year 2006, DNDO also intends to award contracts to two or three manufacturers for further engineering development and production. The second new technology is “high-Z detection,” which is designed to better detect high atomic number (high-Z) materials—such as Special Nuclear Material (SNM)—and shielding materials—such as lead—that could be used to shield gamma radiation from portal monitors. The Cargo Advanced Automated Radiography System (CAARS) program within DNDO is intended to develop the technologies necessary for automated detection of high-Z material. DNDO envisions using the advanced portal monitor technology for the detection of lightly shielded nuclear threats and radiological dispersal devices, and using CAARS technology for the detection of high-Z materials. The third new technology is “active interrogation,” which is designed to better detect nuclear material, especially shielded sources, and DNDO expects it to play a role further in the future than advanced portal monitors and CAARS. DHS and DOE are supporting research at DOE national laboratories, such as Los Alamos and Lawrence Livermore, to develop these systems. Active interrogation systems probe or “interrogate” containers with neutron or gamma rays to induce additional radiation emissions from radioactive material within the container. According to DNDO, these systems are too large and costly to consider for current use. In addition, because these systems emit radiation, care will have to be taken to ensure personnel safety before any deployments are made. In addition to these relatively near-term research and development efforts, DNDO intends to solicit proposals from private, public, academic, and federally funded research centers to pursue radiation detection projects with a more long-term orientation. The solicitation identifies five areas of research: mobile detection systems that can be used to detect potential radiological threats that are in transit, at fixed locations, and at special events; detection systems that can be integrated into ships, trucks, planes, or active detection technologies, including portal monitors and handheld devices that can detect and verify the presence of shielded nuclear materials; innovative detector materials that provide improved detection and isotope identification capabilities over existing materials, in addition to technologies that lead to reductions in the costs to manufacture detector materials, increasing the size and choice of the shapes of detector materials without a loss in performance; and alternate means to detect and identify nuclear material other than through radiation detection such as mass, density, or temperature. DHS is testing commercially available portal monitors, advanced portal monitors, and handheld devices at its new Radiological and Nuclear Countermeasures Test and Evaluation Complex at the Nevada Test Site (NTS). DNDO, with assistance from DOE’s National Nuclear Security Administration, began construction of the complex in 2005. While construction work is under way, an Interim Test Track was built nearby. The complex is to support the DNDO’s development, testing, acquisition, and support of the deployment of radiation detection technologies. When completed, the complex will be comprised of several operating areas where testing and evaluation of detection systems will be conducted, such as a testing facility to evaluate active interrogation technologies; and a large, instrumented outdoor testing area to test mobile detection systems. The complex will also have a vehicle choke point where detection systems for land border crossings, toll plazas, and entrances to tunnels and bridges can be evaluated. According to DNDO officials, an important advantage of using NTS is that it provides the necessary facilities to test detection system capabilities with special nuclear materials in threat-representative configurations. The complex will be open to other organizations within DHS, including CBP, S&T, the Transportation Security Administration, and the U.S. Coast Guard. It will also be open to DOE national laboratories, universities, and private companies conducting radiation detection development and production for DHS. The facility is expected to become fully operational in January 2007. In addition to the Nevada complex, DHS manages CMTB to test radiation detection equipment in an operational environment. The CMTB originated as a DOE funded demonstration project in fiscal year 2003, but transferred to DHS in August 2003. The scientific, engineering, and technical staff of the CMTB are drawn predominantly from the national laboratories. The test bed encompasses various operational settings, such as major seaports, airports, roadways, and railways. The CMTB deploys commercially available and advanced radiation detection equipment at these venues to test and evaluate their performance in real-world situations, to develop better standard operating procedures, and to assess the impact the equipment has on the flow of commerce. At present, CMTB is testing portal monitors at toll crossings of two tunnels and one bridge, two seaport terminals, and two air cargo facilities. In addition, CMTB is developing several advanced secondary inspection mobile technologies. (See fig. 5.) The advanced spectroscopic portal monitors that DNDO is developing will likely be evaluated at the CMTB, once testing is completed at the Nevada Test Site. DHS works with DOE, DOD, and other federal, state, and local agencies, as well as the private sector to carry out radiation detection programs. The newly established DNDO was set up to serve as DHS’s main instrument for coordinating these efforts. Since its creation in April 2005, DNDO has entered into working relationships with other agencies and is taking the lead in developing what it calls a “global architecture,” an integrated approach to detecting and stopping nuclear smuggling. However, because DNDO was created so recently, these efforts are in their early stages of development and implementation. Historically, cooperation among agencies engaged in domestic radiation detection has been limited. In April 2005, however, the president signed a joint presidential directive that directed the establishment of DNDO to, among other things, improve such cooperation by creating a single accountable organization with the responsibility for establishing strong linkages across the federal government and with other entities. As currently envisioned under the directive, DNDO’s mission covers a broad spectrum of radiological and nuclear protective measures, but focuses mainly on nuclear detection. The directive includes several provisions directing DNDO to coordinate its activities with other entities. For example, DNDO is to work with DOE, DOD, the Departments of State and Justice, state and local agencies, and the private sector to develop programs to thwart illicit movements of nuclear materials. In addition, provisions of the directive require consultation between DNDO, law enforcement and nonproliferation centers, as well as other related federal and state agencies. Table 2 provides a summary of the cooperation brought about by the presidential directive. According to senior DNDO officials, although the close cooperation called for in DNDO’s mandate has been difficult to achieve, there are two factors that may help DNDO succeed in this effort. First, the presidential directive is explicit in directing other federal agencies to support DNDO’s efforts. The directive transfers primary responsibility for radiation and nuclear detection activities in the United States to DNDO, and requires DNDO to include personnel from other agencies in its organization. For example, under the directive, DOE will provide DNDO with information received from overseas programs, including the Megaports Initiative and others, as well as information from DOE’s international partners involved with radiological and nuclear detection systems. Second, all of the radiological and nuclear detection programs and staff of S&T became part of DNDO. DOE’s Second Line of Defense program supports DNDO efforts by working with the agency to exchange information, data, and lessons learned from overseas deployments. According to senior officials at DNDO, the data from overseas deployments are needed to help DNDO efforts to develop profiles of potential risks to the United States. In addition, the performance of these systems, as evidenced by these data, can help improve domestic portal monitors’ ability to detect radiation. In addition, DOE provides equipment training opportunities for DHS personnel. In April 2005, DOE and DHS formalized certain aspects of this cooperation in a memorandum of understanding. Specifically, the areas of cooperation include, among other things: discussing procedures for the rapid analysis of cargo and for operational/emergency responses, training CBP officers, exchanging technical and lessons learned information, and providing updates on their respective programs’ implementation. DHS has also entered into formal agreements with state and local governments to coordinate their radiation detection efforts. For example, in April 2005, just prior to DNDO’s creation, DHS and the Port Authority of New York and New Jersey finalized a memorandum of understanding to provide services, personnel, and equipment to run the CMTB program. Specifically, the program is designed to evaluate and assess the role of threat detection technologies, develop and exercise various concepts of operation and response tools, integrate lessons learned from field experiences, and provide detection and monitoring capabilities for testing and evaluation purposes. The agreement spells out each partner’s responsibilities, including coordination with other agencies. According to a senior DNDO official, DNDO now has responsibility for this and other similar agreements under its authority to develop and evaluate new radiation detection equipment. Finally, DNDO officials also believe that the way the agency has been staffed and organized will aid its cooperation efforts. For example, staff from DHS, DOD, DOE, the Departments of State and Justice, and other agencies, have been detailed to DNDO. All of DNDO’s major organizational units are staffed with personnel from multiple agencies. For example, the strategic planning staff within the Office of the Director has employees from DOE, DOD, CBP, Federal Bureau of Investigation (FBI), and DHS’s Office of State and Local Government Coordination and Preparedness. Significantly, DNDO’s Office of Operations Support, which is designed to provide real-time situational data as well as technical support to field units, is headed by an FBI executive with senior staff from CBP, DOE, and DHS’s Transportation Security Administration providing direct management support. According to a senior DNDO official, having this broad range of agencies represented in DNDO decision making helps ensure that agencies’ views are heard and fully considered, thereby helping to achieve the greatest possible consensus even for difficult decisions. Further, agency personnel detailed to DNDO have the authority to “bind” their respective agencies, i.e., whatever decisions or agreements are reached under the auspices of DNDO will bind their agency to comply to the extent permitted by law. Finally, according to senior officials in DOE and CBP, the current organizational arrangement appears to be working. Officials noted that early in DNDO’s history, communication was difficult, but has recently improved. For example, CBP and DOE officials told us they had hoped to have greater input into DNDO’s early efforts to develop integrated radiation detection systems. However, these officials noted that by October 2005, DNDO seemed to have heard and acted upon their recommendations. However, although these officials were optimistic about future collaborations with DNDO, they also noted that DNDO has not yet completed a large enough body of work to conclude firmly that its coordination efforts will always be similarly successful. Among the main purposes in creating the DNDO, according to its Director, is to develop a global nuclear detection system that he characterized as a “global architecture.” DNDO’s intention in developing such an approach is to coordinate other agencies’ efforts, such as the Second Line of Defense and Container Security Initiative, with the domestic deployment program to create an integrated, worldwide system. The resulting “global architecture” would be a multi-layered defense strategy that includes programs that attempt to secure nuclear materials and detect their movements overseas; to develop intelligence information on nuclear materials’ trans-shipments and possible movement to the United States; and to integrate these elements with domestic efforts undertaken by governments—federal, state, local, and tribal—and the private sector. Much of DNDO’s work in terms of acquiring and supporting the deployment of radiation detection equipment, as well as in supporting research, development, and testing of new detection equipment supports the office’s mission to develop the U.S. domestic portion this global architecture. In addition, DHS, in conjunction with selected state and local organizations, as well as other federal agencies and the private sector, began two pilot projects in fiscal year 2003 to demonstrate a layered defense system designed to protect the United States against radiological and nuclear threats. DHS’s Radiological Pilot Programs Office coordinated the projects’ initial efforts, and DNDO assumed responsibility in October 2005. Field work began in fiscal year 2004 and will be completed in fiscal year 2007. The project leaders expect the final report and lessons learned to be issued in fiscal year 2007. Both pilot projects featured a broad selection of federal, state, and local agencies, including state law enforcement, counter-terrorism, emergency management, transportation, and port authorities. DHS has made progress deploying radiation detection equipment at U.S. ports-of-entry; notably, the department achieved these gains without greatly impeding the flow of commerce (i.e., the movement of cargo containers out of ports-of-entry). However, we believe that DHS will find it difficult under current plans and assumptions to meet its current portal monitor deployment schedule at U.S. borders because it would have to increase its current rate of deployment by 230 percent to meet its September 2009 deadline. Our analysis of CBP’s and PNNL’s earned value data suggests that millions of dollars worth of work is being deferred each month and that the work that is completed is costing millions more than planned. Currently, we estimate that CBP is facing a likely cost overrun of about $340 million, and that the last portal monitor may not be installed until late 2014. Unless CBP and PNNL make immediate improvements in the schedule performance, then additional slippage in the deployment schedule is likely. A key overriding cause for these delays is the late disbursal of funds to DHS contractors. This late dispersal disrupts and delays some ongoing installation projects. In this regard, DHS approval processes for documentation requested by the House Appropriations Committee are lengthy and cumbersome. In one case, for example, funds for fiscal year 2005 were not made available to the DHS contractor until September 2005, the last month of the fiscal year. This process is taking too long and needs to be shortened. Further, the unsure efficacy and uncertain cost associated with the advanced portal monitor technology means that DHS cannot determine, with confidence, how much the program will eventually cost. In particular, even if the advanced portal monitor technology can be shown superior to current technology—which currently does not seem certain—DHS does not yet know whether the new technology will be worth its considerable additional cost. Only after testing of the advanced portal monitors has been completed and DHS has rigorously compared currently-fielded and advanced portal monitors, taking into account their differences in cost, will DHS be able to answer this question. CBP has experienced difficulty deploying portal monitors at seaports, at least in part because it has been unable to reach agreements with many seaport operators, who are concerned that radiation detection efforts may delay the flow of commerce through their ports. As a result, the agency has fallen 2 years behind its seaport deployment schedule—and seaports continue to be vulnerable to nuclear smuggling. Significantly, there is no clear solution and no reason to be optimistic that progress can be made soon. CBP’s policy of negotiating deployment agreements with seaport terminal operators has not yet yielded agreements at many seaports and this has caused significant delays in the deployment of portal monitors at some seaports. CBP has chosen not to attempt to force terminal operators to cooperate. A subset of this issue concerns screening rail traffic leaving seaports, which is a particularly difficult problem. The operational concerns of performing secondary rail inspections in seaports are daunting. Some port operators as well as a national study strongly suggest that rail transport will increase over the next 10 years. However, unless an effective and efficient means to screen rail traffic is developed and deployed, seaports will likely continue to either avoid installing detection equipment altogether, or simply turn it off when its operation might prove to be inconvenient. Without more progress on this front, we risk rail cargo becoming a burgeoning gap in our defenses against nuclear terrorism. CBP appears to have made progress in using radiation detection equipment correctly and adhering to inspection procedures. At several ports-of-entry we visited, CBP officers physically opened and inspected cargo containers to confirm the nature of the radiological source under certain circumstances. They did this when they were unable to confirm the type of radiological material through current approved procedures. Since the currently deployed handheld equipment is limited in its ability to accurately identify sources of radiation, opening the container allows CBP officers to get closer to the source of the alarm and thereby improve their chances of accurately identifying the source. It also enables officers to verify that the container’s contents are consistent with the isotope identifier’s initial reading and the container’s manifest. Furthermore, since DHS and DOE officials have expressed concerns that illicit radiological material could be shielded, this practice enables officers to conduct a more thorough examination of the containers’ contents—thereby increasing the likelihood that CBP officers will find any illicit radioactive material. Importantly, this process, according to border security officials, did not impede the progress of commerce through any port-of-entry. On the other hand, because CBP officers do not have access to NRC licensing data, it is difficult for them to verify that shippers have obtained necessary NRC licenses and to verify the authenticity of any NRC licenses that may accompany shipments of radioactive materials. As a result, unless nuclear smugglers in possession of faked license documents raised suspicions in some other way, CBP officers could follow agency guidelines yet unwittingly allow them to enter the country with their illegal nuclear cargo. As we see it, this is a significant gap in CBP’s national procedures that should be closed. Since DHS provides the Congress with information concerning the acquisition and deployment of portal monitors, and since DHS’s procedures to obtain internal agreement on this information are lengthy and cumbersome—often resulting in delays—we recommend that the Secretary of Homeland Security, working with the Director of DNDO and the Commissioner of CBP, review these approval procedures and take actions necessary to ensure that DHS submits information to the Congress early in the fiscal year. In order to complete the radiation portal monitor deployment program, as planned, we recommend that the Secretary of Homeland Security, working with the Director of DNDO, and in concert with CBP and PNNL, devise a plan to close the gap between the current deployment rate and the rate needed to complete deployments by September 2009. To ensure that DHS’s substantial investment in radiation detection technology yields the greatest possible level of detection capability at the lowest possible cost, we recommend that once the costs and capabilities of advanced technology portal monitors are well understood, and before any of the new equipment is purchased, the Secretary of Homeland Security work with the Director of DNDO to analyze the benefits and costs of deploying advanced portal monitors. This analysis should focus on determining whether any additional detection capability provided by the advanced equipment is worth its additional cost. After completing this cost- benefit analysis, the Secretary of Homeland Security, working with the Director of DNDO, should revise its total program cost estimates to reflect current decisions. To help speed seaport deployments and to help ensure that future rail deployments proceed on time, we recommend that the Secretary of Homeland Security, in cooperation with the Commissioner of CBP, develop procedures for effectively screening rail containers and develop new technologies to facilitate inspections. To increase the chances that CBP officers find illicit radiological material, we recommend that the Secretary of Homeland Security, working with the Commissioner of CBP, consider modifying the agency’s standard operating procedures for secondary inspections to include physically opening cargo containers during secondary inspections at all ports-of-entry when the external inspection does not conclusively identify the radiological material inside. To further increase the chances that CBP officers identify illicit radiological material, we recommend that the Secretary of Homeland Security, working with the Chairman of NRC, develop a way for CBP border officers to determine whether radiological shipments have the necessary NRC licenses and to verify the authenticity of NRC licenses that accompany such shipments. To ensure that CBP is receiving reliable cost and schedule data, we recommend that the Secretary of Homeland Security direct PNNL to have its earned value management system validated so that it complies with guidance developed by the American National Standards Institute/ Electronic Industries Alliance. In addition, we recommend the Secretary of Homeland Security direct CBP and PNNL to conduct an Integrated Baseline Review to ensure its earned value management data is reliable for assessing risk and developing alternatives. We provided a draft of this report to DHS for comment. In response, we received written comments from DHS officials. DHS noted that the report is factually correct. Further, the Department agreed with our recommendations and committed to implementing them. DHS officials also commented that our review did not completely capture the enormity or complexity of the Radiation Portal Monitor program. We agree that this program is a massive undertaking, and our original draft reflected this perspective in several places. In commenting on our recommendation to develop a better means for CBP border officers to verify NRC license information, DHS stated that “NRC licenses are required to accompany certain legitimate shipments of radiological materials…” However, according to senior NRC officials, no requirement that the license accompany the shipment exists. Finally, DHS provided some clarifying comments that we incorporated into this report, as appropriate. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies to the congressional committees with jurisdiction over DHS and its activities; the Secretary of Homeland Security; the Director of OMB; and interested congressional committees. We will also make copies of the report available to others upon request. This report will also be available at no charge on GAO’s home page at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-3841. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix IV. To assess the Department of Homeland Security’s (DHS) progress in deploying radiation detection equipment, including radiation portal monitors, radiation isotope identification devices, and pagers at U.S. ports- of-entry and any problems associated with that deployment, we reviewed documents and interviewed officials from the U.S. Customs and Border Protection (CBP), Domestic Nuclear Detection Office (DNDO), and Pacific Northwest National Laboratory (PNNL). We focused primarily on the issues surrounding radiation portal monitors because they are a major tool in the federal government’s efforts to thwart nuclear smuggling, and because the budget and other resources devoted to these machines far exceeds the handheld equipment also used at U.S. ports-of-entry. Further, we focused on the use of radiation detection equipment in primary and secondary inspections, but we did not examine their use as a part of CBP’s targeted inspections. To assess CBP’s current progress in deploying portal monitors, we compared PNNL’s December 2004 project execution plan for deploying radiation portal monitors—including the project’s schedule and estimated cost. We analyzed budget, cost, and deployment data on portal monitors to determine differences between PNNL’s plan and its current progress. We also assessed PNNL’s cost and schedule performance using earned value analysis techniques based on data captured in PNNL’s contract performance reports. We also developed a forecast of future cost growth. We based the lower end of our forecast range on the sum of costs spent to date and the forecast cost of work remaining. The remaining work was forecast using an average of the current cost performance index efficiency factor. For the upper end of our cost range, we relied on the actual costs spent to date added to the forecast of remaining work with an average monthly cost and schedule performance index. We also visited a nonprobability sample of CBP ports-of-entry, including two international mail and express courier facilities, five seaports, and three land border crossings. We selected these ports-of-entry by using criteria such as the types of ports-of-entry where CBP plans to deploy equipment; ports-of-entry with wide geographic coverage; and ports-of- entry where portal monitors have been—or are planned to be—installed. During each visit, we spoke with CBP inspectors and local port authority officials on the progress made, and any problems experienced in deploying the equipment at their locations. To assess CBP officers’ use of radiation detection equipment, and how inspection procedures are implemented at U.S. ports-of-entry, and any problems associated with the use of the equipment, we reviewed CBP’s standard operating procedures for radiation detection; documents on its training curriculum; and training materials on how to use the equipment. We participated in a 3-day hands-on training course for CBP officers at PNNL on how to use radiation detection equipment. We also interviewed officials from CBP field and headquarters to discuss problems associated with the use of the equipment. During our site visits, we toured the facilities, observed the equipment in use, and interviewed CBP officers about radiation detection policies and procedures and the deployment of equipment at their locations. We discussed with CBP officers how they determine the validity of Nuclear Regulatory Commission (NRC) licenses when legitimate shipments of radioactive material enter the nation. To assess DHS’s progress in improving and testing radiation detection equipment capabilities, we reviewed documents and interviewed officials from CBP, DNDO, Science and Technology Directorate (S&T), DOE, PNNL, and the National Institute for Standards and Technology (NIST). We reviewed S&T’s April 2005 Program Execution Plan; DHS documentation on the development of advanced radiation detection technologies; and test results and assessments of the performance of both commercially available radiation detection equipment and advanced technologies. We visited four national laboratories—Lawrence Livermore, Los Alamos, Pacific Northwest, and Sandia—that are involved in the research, development, and testing of radiation detection technologies. In addition, we visited the Counter Measures Test Bed (CMTB) in New York and New Jersey, the Nevada Test Site, and the Department of Defense’s (DOD) test site at a U.S. Air Force base to observe the testing of radiation detection equipment and discuss progress in improving and testing radiation detection equipment with onsite experts. To assess the level of cooperation between DHS and other federal agencies in conducting radiation detection programs, we interviewed officials from CBP; S&T; the Transportation Security Administration; DOD’s Defense Threat Reduction Agency; DOE’s National Nuclear Security Administration; and Lawrence Livermore, Los Alamos, Pacific Northwest, and Sandia National Laboratories. We discussed the current extent of coordination and whether more coordination could result in improvements to DHS’s deployment, development, and testing of radiation detection equipment and technologies. We reviewed agency agreements to cooperate, including a memorandum of understanding between DHS and DOE to exchange information on radiation detection technologies and deployments, and a memorandum of understanding between DHS and the Port Authority of New York and New Jersey to integrate lessons learned into domestic radiation detection efforts. In addition, we reviewed an organizational chart from DNDO as well as our past reports on coordination between federal agencies on deployment and testing. We received training data from CBP, cost and budget data from CBP, and deployment data from CBP and PNNL. We obtained responses from key database officials to a number of questions focused on data reliability covering issues such as data entry access, internal control procedures, and the accuracy and completeness of the data. We determined these data were sufficiently reliable for the purposes of this report. We conducted our review from March 2005 to February 2006 in accordance with generally accepted government auditing standards. In addition to the contact named above, Jim Shafer; Nancy Crothers; Emily Gupta; Brandon Haller; Richard Hung; Winston Le; Greg Marchand; Judy Pagano; Karen Richey; Keith Rhodes, GAO’s Chief Technologist; and Eugene Wisnoski made key contributions to this report. Combating Nuclear Smuggling: Corruption, Maintenance, and Coordination Problems Challenge U.S. Efforts to Provide Radiation Detection Equipment to Other Countries. GAO-06-311. Washington, D.C.: March 14, 2006. Combating Nuclear Smuggling: Efforts to Deploy Radiation Detection Equipment in the United States and in Other Countries. GAO-05-840T. Washington, D.C.: June 21, 2005. Homeland Security: Key Cargo Security Programs Can Be Improved. GAO-05-466T. Washington, D.C.: May 25, 2005. Container Security: A Flexible Staffing Model and Minimum Equipment Requirements Would Improve Overseas Targeting and Inspection Efforts. GAO-05-557. Washington, D.C.: April 26, 2005. Preventing Nuclear Smuggling: DOE Has Made Limited Progress in Installing Radiation Detection Equipment at Highest Priority Foreign Seaports. GAO-05-375. Washington, D.C.: March 31, 2005. Customs Service: Acquisition and Deployment of Radiation Detection Equipment. GAO-03-235T. Washington, D.C.: October 17, 2002. Nuclear Nonproliferation: U.S. Efforts to Help Other Countries Combat Nuclear Smuggling Need Strengthened Coordination and Planning. GAO- 02-426. Washington, D.C.: May 16, 2002. | Preventing radioactive material from being smuggled into the United States is a key national security objective. To help address this threat, in October 2002, DHS began deploying radiation detection equipment at U.S. ports-of-entry. This report reviews recent progress DHS has made (1) deploying radiation detection equipment, (2) using radiation detection equipment, (3) improving the capabilities and testing of this equipment, and (4) increasing cooperation between DHS and other federal agencies in conducting radiation detection programs. The Department of Homeland Security (DHS) has made progress in deploying radiation detection equipment at U.S. ports-of-entry, but the agency's program goals are unrealistic and the program cost estimate is uncertain. As of December 2005, DHS had deployed 670 portal monitors and over 19,000 pieces of handheld radiation detection equipment. However, the deployment of portal monitors has fallen behind schedule, making DHS's goal of deploying 3,034 by September 2009 unlikely. In particular, two factors have contributed to the schedule delay. First, DHS provides the Congress with information on portal monitor acquisitions and deployments before releasing any funds. However, DHS's lengthy review process has caused delays in providing such information to the Congress. Second, difficult negotiations with seaport operators about placement of portal monitors and how to most efficiently screen rail cars have delayed deployments at seaports. Regarding the uncertainty of the program's cost estimate, DHS would like to deploy advanced technology portals that will likely cost significantly more than the currently deployed portals, but tests have not yet shown that these portals are demonstrably more effective than the current portals. Consequently, it is not clear that the benefits of the new portals would be worth any increased cost to the program. Also, our analysis of the program's costs indicates that DHS may incur a $342 million cost overrun. DHS has improved in using detection equipment and in following the agency's inspection procedures since 2003, but we identified two potential issues in Customs and Border Protection (CBP) inspection procedures. First, although radiological materials being transported into the United States are generally required to have a Nuclear Regulatory Commission (NRC) license, regulations do not require that the license accompany the shipment. Further, CBP officers do not have access to data that could be used to verify that shippers have acquired the necessary documentation. Second, CBP inspection procedures do not require officers to open containers and inspect them, although under some circumstances, doing so could improve security. In addition, DHS has sponsored research, development, and testing activities to address the inherent limitations of currently fielded equipment. However, much work remains to achieve consistently better detection capabilities. DHS seems to have made progress in coordinating with other agencies to conduct radiation detection programs; however, because the DHS office created to achieve the coordination is less than 1 year old, its working relationships with other agencies are in their early stages of development and implementation. In the future, this office plans to develop a "global architecture" to integrate several agencies' radiation detection efforts, including several international programs. |
Interior’s ongoing reorganization of bureaus with oil and gas functions will require time and resources, and undertaking such an endeavor while continuing to meet ongoing responsibilities may pose new challenges. Historically, BLM managed onshore federal oil and gas activities, while MMS managed offshore activities and collected royalties for all leases. In May 2010, the Secretary of the Interior announced plans to reorganize MMS into three separate bureaus. The Secretary stated that dividing MMS’s responsibilities among separate bureaus would help ensure that each of the three newly established bureaus have a distinct and independent mission. Interior recently began implementing this restructuring effort, transferring offshore oversight responsibilities to the newly created BOEMRE and revenue collection to ONRR. Interior plans to continue restructuring BOEMRE to establish two additional separate bureaus—the Bureau of Ocean and Energy Management, which will focus on leasing and environmental reviews, and the Bureau of Safety and Environmental Enforcement, which will focus on permitting and inspection functions. While this reorganization may eventually lead to more effective operations, we have reported that organizational transformations are not simple endeavors and require the concentrated efforts of both leaders and employees to realize intended synergies and accomplish new organizational goals. In that report, we stated that for effective organizational transformation, top leaders must balance continued delivery of services with transformational activities. Given that, as of December 2010, Interior had not implemented many recommendations we made to address numerous weaknesses and challenges, we are concerned about Interior’s ability to undertake this reorganization while (1) providing reasonable assurance that billions of dollars of revenues owed to the public are being properly assessed and collected and (2) maintaining focus on its oil and gas oversight responsibilities. We have reported that Interior has experienced several challenges in meeting its obligations to make federal oil and gas resources available for leasing and development while simultaneously meeting its responsibilities for managing public lands for other uses, including wildlife habitat, recreation, and wilderness. In January 2010, we reported that while BLM requires oil and gas operators to reclaim the land they disturb and post a bond to help ensure they do so, not all operators perform such reclamation. In general, the goal is to plug the well and reclaim the site so that it matches the surrounding natural environment to the extent possible, allowing the land to be used for purposes other than oil and gas production, such as wildlife habitat. If the bond is not sufficient to cover well plugging and surface reclamation, and there are no responsible or liable parties, the well is considered “orphaned,” and BLM uses federal dollars to fund reclamation. For fiscal years 1988 through 2009, BLM spent about $3.8 million to reclaim 295 orphaned wells, and BLM has identified another 144 wells yet to be reclaimed. In addition, in a July 2010 report on federal oil and gas lease sale decisions in the Mountain West, we found that the extent to which BLM tracked and made available to the public information related to protests filed during the leasing process varied by state and was generally limited in scope. We also found that stakeholders—including environmental and hunting interests, and state and local governments protesting BLM lease offerings—wanted additional time to participate in the leasing process and more information from BLM about its leasing decisions. Moreover, we found that BLM had been unable to manage an increased workload associated with public protests and had missed deadlines for issuing leases. In May 2010, the Secretary of the Interior announced several departmentwide leasing reforms that are to take place at BLM that may address these concerns, such as providing additional public review and comment opportunity during the leasing process. Further, in March 2010, we found that Interior faced challenges in ensuring consistent implementation of environmental requirements, both within and across MMS’s regional offices, leaving it vulnerable with regard to litigation and allegations of scientific misconduct. We recommended that Interior develop comprehensive environmental guidance materials for MMS staff. Interior concurred with this recommendation and is currently developing such guidance. Finally, in September 2009, we reported that BLM’s use of categorical exclusions under Section 390 of the Energy Policy Act of 2005—which authorized BLM, for certain oil and gas activities, to approve projects without preparing new environmental analyses that would normally be required in accordance with the National Environmental Policy Act—was frequently out of compliance with the law and BLM’s internal guidance. As a result, we recommended that BLM take steps to improve the implementation of Section 390 categorical exclusions through clarification of its guidance, standardizing decision documents, and increasing oversight. We have reported that BLM and MMS have encountered persistent problems in hiring, training, and retaining sufficient staff to meet Interior’s oversight and management responsibilities for oil and gas operations on federal lands and waters. For example, in March 2010, we reported that BLM and MMS experienced high turnover rates in key oil and gas inspection and engineering positions responsible for production verification activities. As a result, Interior faces challenges meeting its responsibilities to oversee oil and gas development on federal leases, potentially placing both the environment and royalties at risk. We made a number of recommendations to address these issues. While Interior’s reorganization of MMS includes plans to hire additional staff with expertise in oil and gas inspections and engineering, these plans have not been fully implemented, and it remains unclear whether Interior will be fully successful in hiring, training, and retaining these additional staff. Moreover, the human capital issues we identified with BLM’s management of onshore oil and gas continue, and these issues have not yet been addressed in Interior’s reorganization plans. Federal oil and gas resources generate billions of dollars annually in revenues that are shared among federal, state, and tribal governments; however, we found Interior may not be properly assessing and collecting these revenues. In September 2008, we reported that Interior collected lower levels of revenues for oil and gas production in the deep water of the U.S. Gulf of Mexico than all but 11 of 104 oil and gas resource owners whose revenue collection systems were evaluated in a comprehensive industry study—these resource owners included other countries as well as some states. However, despite significant changes in the oil and gas industry over the past several decades, we found that Interior had not systematically re-examined how the U.S. government is compensated for extraction of oil and gas for over 25 years. GAO recommended Interior conduct a comprehensive review of the federal oil and gas system using an independent panel. After Interior initially disagreed with our recommendations, we recommended that Congress consider directing the Secretary of the Interior to convene an independent panel to perform a comprehensive review of the federal system for collecting oil and gas revenue. More recently, in response to our report, Interior has commissioned a study that will include such a reassessment, which, according to officials, the department expects will be complete in 2011. The results of the study may reveal the potential for greater revenues to the federal government. We also reported in March 2010 that Interior was not taking the steps needed to ensure that oil and gas produced from federal lands was accurately measured. For example, we found that neither BLM nor MMS had consistently met their agency goals for oil and gas production verification inspections. Without such verification, Interior cannot provide reasonable assurance that the public is collecting its share of revenue from oil and gas development on federal lands and waters. As a result of this work, we identified 19 recommendations for specific improvements to oversight of production verification activities. Interior generally agreed with our recommendations and has begun implementing some of them. Additionally, we reported in October 2010 that Interior’s data likely underestimated the amount of natural gas produced on federal leases, because some unquantified amount of gas is released directly to the atmosphere (vented) or is burned (flared). This vented and flared gas contributes to greenhouse gases and represents lost royalties. We recommended that Interior improve its data and address limitations in its regulations and guidance to reduce this lost gas. Interior generally agreed with our recommendations and is taking initial steps to implement these recommendations. Furthermore, we reported in July 2009 on numerous problems with Interior’s efforts to collect data on oil and gas produced on federal lands, including missing data, errors in company-reported data on oil and gas production, and sales data that did not reflect prevailing market prices for oil and gas. As a result of Interior’s lack of consistent and reliable data on the production and sale of oil and gas from federal lands, Interior could not provide reasonable assurance that it was assessing and collecting the appropriate amount of royalties on this production. We made a number of recommendations to Interior to improve controls on the accuracy and reliability of royalty data. Interior generally agreed with our recommendations and is working to implement many of them, but these efforts are not complete, and it is uncertain at this time if the efforts will fully address our concerns. In October 2008, we reported that Interior could do more do encourage the development of existing oil and gas leases. Our review of Interior oil and gas leasing data from 1987 through 2006 found that the number of leases issued had generally increased toward the end of this period, but that offshore and onshore leasing had followed different historical patterns. Offshore leases issued peaked in 1988 and in 1997, and generally rose from 1999 through 2006. Onshore leases issued peaked in 1988, then rapidly declined until about 1992, and remained at a consistently low level until about 2003, when they began to increase moderately. We also analyzed 55,000 offshore and onshore leases issued from 1987 through 1996 to determine how development occurred on leases that had expired or been extended beyond their primary terms. Our analysis identified three key findings. First, a majority of leases expired without being drilled or reaching production. Second, shorter leases were generally developed more quickly than longer leases but not necessarily at comparable rates. Third, a substantial percentage of leases were drilled after the initial primary term following a lease extension or suspension. We also compared Interior’s efforts to encourage development of federal oil and gas leases to states’ and private landowners’ efforts. We found that Interior does less to encourage development of federal leases than some states and private landowners. Federal leases contain one provision–– increasing rental rates over time for offshore 5-year leases and onshore leases—to encourage development. In addition to using increasing rental rates, some states undertake additional efforts to encourage lessees to develop oil and gas leases more quickly, including shorter lease terms and graduated royalty rates—royalty rates that rise over the life of the lease. In addition, compared to limited federal efforts, some states do more to structure leases to reflect the likelihood of oil and gas production, which may also encourage faster development. Based on the limited information available on private leases, private landowners also use tools similar to states to encourage development. In conclusion, as concerns rise over the recent increase in oil prices and as demands are made for additional drilling on federal lands and waters, it is important that Interior meet its current oversight responsibilities. Interior is now in the midst of a major reorganization, which makes balancing delivery of services with transformational activities challenging for an organization. Managing this change in a fiscally constrained environment only exacerbates the challenge. If steps are not taken to improve Interior’s oversight of oil and gas leasing, we are concerned about the department’s ability to manage the nation’s oil and gas resources, ensure the safe operation of onshore and offshore leases, provide adequate environmental protection, and provide reasonable assurance that the U.S. government is collecting the revenue to which it is entitled. Chairman Hastings, Ranking Member Markey, and Members of the Committee, this concludes our prepared statement. We would be pleased to answer any questions that you or other Members of the Committee may have at this time. For further information on this statement, please contact Frank Rusco at (202) 512-3841 or ruscof@gao.gov. Contact points for our Congressional Relations and Public Affairs offices may be found on the last page of this statement. Other staff that made key contributions to this testimony include, Jeffrey Barron, Glenn C. Fischer, Jon Ludwigson, Alison O'Neil, Kiki Theodoropoulos, and Barbara Timmerman. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The Department of the Interior oversees oil and gas activities on leased federal lands and waters. Revenue generated from federal oil and gas production is one of the largest nontax sources of federal government funds, accounting for about $9 billion in fiscal year 2009. For onshore leases, Interior's Bureau of Land Management (BLM) has oversight responsibilities. For offshore leases, the newly created Bureau of Ocean Energy Management, Regulation, and Enforcement (BOEMRE), has oversight responsibilities. Prior to BOEMRE, the Minerals Management Service's (MMS) Offshore Energy and Minerals Management Office oversaw offshore oil and gas activities, while MMS's Minerals Revenue Management Office collected revenues from all oil and gas produced on federal leases. Over the past several years, GAO has issued numerous recommendations to the Secretary of the Interior to improve the agency's management of oil and gas resources. In 2011, GAO identified Interior's management of oil and gas resources as a high risk issue. GAO's work in this area identified challenges in five areas: (1) reorganization, (2) balancing responsibilities, (3) human capital, (4) revenue collection, and (5) development of existing leases. Reorganization: Interior's reorganization of activities previously overseen by MMS will require time and resources and may pose new challenges. Interior began a reorganization in May 2010 that will divide MMS into three separate bureaus--one focusing on revenue collection, another on leasing and environmental reviews, and yet another on permitting and inspections. While this reorganization may eventually lead to more effective operations, GAO has reported that organizational transformations are not simple endeavors. GAO is concerned with Interior's ability to undertake this reorganization while meeting its revenue collection and oil and gas oversight responsibilities. Balancing Responsibilities: GAO has reported that Interior has experienced several challenges with meeting its responsibilities for providing for the development of oil and gas resources while managing public lands for other uses, including wildlife habitat. In January 2010, GAO reported that, while BLM requires oil and gas operators to reclaim the land they disturb and post a bond to help ensure they do so, not all operators perform reclamation. For fiscal years 1988 through 2009, BLM spent about $3.8 million to reclaim 295 so-called "orphaned" wells--because reclamation had not been done, and other resources, including the bond, were insufficient to pay for it. Human Capital: GAO has reported that BLM and MMS have encountered persistent problems in hiring, training, and retaining sufficient staff to meet their oversight and management responsibilities for oil and gas operations. For example, in March 2010, GAO reported that BLM and MMS experienced high turnover rates in key oil and gas inspection and engineering positions responsible for production verification activities. As a result, Interior faces challenges meeting its responsibilities to oversee oil and gas development on federal leases, potentially placing both the environment and royalties at risk. Revenue Collection: While federal oil and gas resources generate billions of dollars in annual revenues, past GAO work has found that Interior may not be properly assessing and collecting these revenues. In September 2008, GAO reported that Interior collected lower levels of revenues for oil and gas production in the deep water of the U.S. Gulf of Mexico than all but 11 of 104 oil and gas resource owners whose revenue collection systems were evaluated in a comprehensive industry study. Nonetheless, Interior has not completed a comprehensive assessment of its revenue collection policies and processes in over 25 years. Additionally, in March 2010, GAO reported that Interior was not consistently completing inspections to verify volumes of oil and gas produced from federal leases. Development of Existing Leases: In October 2008, GAO reported that Interior could do more to encourage the development of existing oil and gas leases. Federal leases contain one provision--increasing rental rates over time for offshore 5-year leases and onshore leases--to encourage development. In addition to escalating rental rates, states undertake additional efforts to encourage lessees to develop oil and gas leases more quickly, including shorter lease terms and graduated royalty rates. |
USAID’s 2011-2015 education strategy, issued in February 2011, includes three goals: improved reading skills for 100 million children in primary grades by 2015 (Goal 1), improved ability of tertiary and workforce development programs to produce a workforce with relevant skills to support country development goals by 2015 (Goal 2), and increased equitable access to education in crisis and conflict environments for 15 million learners by 2015 (Goal 3). Goal 1 represents a shift from an earlier broader focus on achieving universal enrollment of children in primary school as well as improving the quality of education. In August 2011, USAID issued guidance for implementing the education strategy, directing missions with basic education programs in non-crisis and conflict environments to narrow the focus of their interventions to achieving the goal of improving early grade reading by the end of fiscal year 2012. According to the guidance, missions that were not in conflict or crisis-affected environments were to focus resources on primary grade reading and discontinue programming in such areas as early childhood education, secondary education, or adult literacy. In addition, missions’ programming was to be based on evidence of what works to improve reading outcomes. The guidance stated that the timing of missions’ alignment of education programs with the education strategy was critical because programs required at least 3 years of implementation to contribute to the strategy’s numerical target. USAID requires that missions’ primary grade reading programs adhere to the agency’s performance monitoring and evaluation framework requirements in order to track progress annually against the full set of expected results. Performance monitoring periodically tracks changes to determine whether desired results are occurring and whether projects are implemented as designed. USAID missions and implementing partners monitor programs by using performance management plans or equivalent documents that define what is to be measured, the unit of measurement, and the level of data disaggregation, among other information. USAID guidance for implementing the education strategy requires that these plans include, at a minimum, either of two standard outcome indicators related to primary grade reading, if relevant: The proportion of students who demonstrate by the end of two grades of primary schooling that they can read and understand the meaning of grade-level text. The proportion of students who by the end of the primary school years are able to read and demonstrate understanding as defined by a country curriculum or standards or as agreed to by national experts. While performance monitoring is required for all programs, USAID’s evaluation policy requires performance or impact evaluations of all large projects. The policy states that performance evaluations should focus on questions such as what a project or program has achieved, how it is being implemented, and how it is perceived and valued. Impact evaluations measure the change in a development outcome that is attributable to the project or program, based on models of cause and effect. In addition, USAID’s guidance for implementing the education strategy instructs missions to integrate plans for performance or impact evaluation into program design during the planning stages and to derive data for evaluations from reading assessments. The guidance also states that performance and impact evaluations should be provided by an external, separately contracted team. A comparison of baseline assessment data (i.e., data collected before a reading program starts) and endline assessment data (i.e., data collected at or near the end of the program or before the end of fiscal year 2015) can be used by missions to report on the standard outcome indicators for performance monitoring and by the agency’s contractors to estimate progress toward the goal of 100 million students with improved reading skills by December 2015. USAID officials noted that the need to assess learners toward the end of a school year, which often ends in May or June, may result in some missions assessing learning outcomes later than December 2015. See U.S. Agency for International Development, 2011 USAID Education Strategy Technical Notes (Washington, D.C.: revised April 2012). least two phases of assessment are necessary to measure reading improvement.the Philippines. In fiscal year 2013—the first year in which missions’ education programs were expected to be fully aligned with the goals of the education strategy—USAID allocated about $790 million for basic education assistance divided between programs supporting the education strategy’s Goal 1—improving primary grade reading—and programs supporting Goal 3—increasing equitable access to education in crisis and conflict environments. USAID provided basic education funding to 41 countries for primary grade reading programs in fiscal year 2013. Table 1 shows USAID’s allocations for basic education for fiscal years 2010 through 2014 by appropriation account. More than half of the 2013 funding was allocated for basic education activities in countries in Sub-Saharan Africa, where baseline reading scores were among the lowest globally, according to USAID. Appendix II shows the fiscal year 2013 allocations for basic education by country. USAID’s basic education programs are planned and implemented primarily by USAID’s overseas missions. In Washington, D.C., the Office of Education in USAID’s Bureau for Economic Growth, Education and Environment provides technical leadership, research, and field support. The office also assists in the development, implementation, and evaluation of the USAID education strategy and resource allocation priorities. In addition, USAID’s geographic bureaus advise missions on their education programs. In the five countries where we conducted fieldwork, the USAID missions had implemented primary grade reading programs that included intervention elements recommended in USAID’s guidance related to its education strategy. In several of the countries, the need to develop new reading materials in mother-tongue languages and train classroom teachers in the use of these materials (see fig. 3) resulted in long lead times before fully reaching the targeted population of students. (Table 2 shows information about the programs in the five countries.) As a result, portions of targeted student populations in four countries are not expected to benefit from the programs in time to contribute to USAID’s strategic goal of 100 million students with improved reading skills by December 2015. The programs’ monitoring efforts produced data showing very low baseline reading skills among targeted students. In addition, the programs in the five countries all included plans for external performance or impact evaluations. Our fieldwork at the USAID missions in Ethiopia, Malawi, the Philippines, Peru, and Uganda showed that each mission had implemented primary grade reading programs with elements recommended in USAID’s technical notes to the education strategy (see text box for key recommendations from the technical notes).countries, the programs included teacher training and coaching in reading instruction; outreach to encourage community and parental support for reading during and after school (see fig. 4); and emphasis on five key components of reading—phonemic awareness, phonics, fluency, vocabulary, and comprehension. Moreover, in four of the countries— Ethiopia, the Philippines, Malawi, and Uganda—the development of new reading instruction materials in mother-tongue languages as well as training and coaching teachers in the use of these materials were central to the reading programs, according to mission officials. The programs in these countries also included improving reading instruction in English. Instruction in a language that students speak and understand, particularly in the early years of school. Instruction focused on five key component skills of reading alphabetic languages—phonemic awareness, phonics, fluency, vocabulary, and comprehension. Teacher training and coaching in reading instruction. Development of instructional material, including textbooks, decodable readers, teaching manuals, and lesson plans. Regular classroom-based, teacher-led assessment of students’ progress. Periodic formal assessment to measure impact over time. Increasing community and parental support for reading, such as support for parent-teacher associations, teacher’s aides, and after-school tutoring. Improving host-country education policy, such as increasing instruction time and adopting mother-tongue curriculum. Capacity building, such as helping host countries conduct early-grade reading assessments. Performance monitoring efforts in the five countries where we conducted fieldwork reveal very low reading skills among targeted student populations. Consistent with USAID’s guidance for performance monitoring, all five programs included in their performance monitoring plans the following standard outcome indicator: “Proportion of students who, by the end of two grades of primary schooling, demonstrate that they can read and understand the meaning of grade level text.” To develop this outcome indicator, program implementers in each country, after reaching consensus with host governments, set benchmarks for reading grade-level text with fluency and comprehension. Program implementers in each country also conducted standardized reading assessments to establish baselines for the proportion of students meeting the benchmarks. Because they are tied to unique circumstances and linguistic differences, the benchmarks may vary by country or language. For example, Ethiopia’s second-grade benchmark for oral reading fluency ranges from 55 words read correctly per minute for Tigrignia to 40 for Hadiyyisa. In Malawi and Uganda, the benchmark for reading fluently is 40 words read correctly per minute. With the exception of Peru, where the host government has well-established benchmarks for grade-level reading, the benchmarks used are not final and may be adjusted as reading assessment data become available for further analysis by program implementers and local stakeholders, such as ministries of education. A USAID official in one of the missions we visited noted that some countries may decide to set more aspirational targets than others to try to motivate regions and partners. Uganda’s benchmark for reading with comprehension is 80 percent of comprehension questions (i.e., 4 out of 5) answered correctly. shows, the program in Ethiopia set similar targets for improving reading despite students’ very low baseline skills. Reflecting the emphasis on program evaluation in the education strategy and other guidance, all five missions where we conducted fieldwork planned for performance or impact evaluations of their reading programs, to be performed under separate contracts with external teams. The missions planned to conduct impact evaluations, using experimental designs that include both treatment and control groups, in accordance with USAID’s evaluation policy, and integrated the evaluation plans in their program designs. According to USAID officials, host countries are sometimes reluctant to accept program designs that intentionally withhold program interventions to some groups of recipients. For example, in Malawi, according to a senior government official, the government strongly preferred that all schools in the 11 participating school districts receive the program interventions but, to allow for USAID’s planned impact evaluation, ultimately agreed to some schools being part of the control group and not receiving the program’s intervention. USAID officials stated that in Ethiopia, where USAID’s program targets all schools, the program will receive a performance evaluation of certain intervention components, such as mother-tongue curriculum and teacher training, and an impact evaluation of other components, such as community outreach. Final evaluations of the reading programs in these countries are planned for 2017 or later. As of April 2015, USAID lacked the ability to estimate overall progress toward its primary grade reading goal because of four factors. First, because it took some missions longer to implement primary grade reading programs than USAID estimated in its implementation guidance to the education strategy, only about two-thirds of missions are expected to have data that could be used to estimate progress toward USAID’s reading goal by the end of 2015. Second, only a small number of the missions that have collected reading assessment data have provided complete data and supporting information needed to aggregate and analyze missions’ reading assessment results. Agency officials attributed the incomplete data and information to a lack of timely guidance, among other factors. In 2014, USAID updated its guidance to clarify the agency’s reporting requirements. Third, USAID has not selected and documented a methodology for estimating total numbers of children with improved reading skills, owing in part to the lack of complete reading assessment data and supporting information. Documenting methodologies used to derive an estimate and reporting the information used to evaluate its reliability, such as information on sampling designs, sample sizes, and modes of data collection, help ensure that stakeholders understand the appropriate uses of the estimate and interpret the results according to generally accepted survey guidelines. Fourth, the education strategy did not set interim targets for measuring progress toward USAID’s reading goal in comparison to planned performance, as suggested by leading performance management practices. Without a methodology and interim targets, USAID officials cannot determine USAID’s progress toward its goal or identify a realistic goal for a future education strategy. Because it took some missions longer to implement primary grade reading programs than USAID estimated in its implementation guidance to the education strategy, USAID expects only about two-thirds of missions with funding for such programs to provide baseline and midline or endline assessment data in time for the data to be used in estimating As figure 8 the agency’s progress toward its primary grade reading goal.shows, 26 missions (62 percent) of the 42 with funding for such programs had implemented them by the end of fiscal year 2012, the target date established in the education strategy’s implementation guidance to allow at least 3 years of implementation. Thirty-six of the 42 missions (86 percent) had implemented primary grade reading programs by the end of fiscal year 2014. USAID expects that most of the missions that implemented primary grade reading programs by 2012, as well as several that implemented their programs in fiscal years 2013 or 2014, will complete baseline and midline or endline reading assessments in time to contribute to meeting the agency’s goal by the end of 2015. Of the 26 missions that implemented their programs by the end of fiscal year 2012, all but 3 missions are expected to have completed a baseline assessment and at least a midline or an endline assessment by calendar year 2015. Of the 36 missions that implemented primary grade reading programs by the end of fiscal year 2014, 25 missions completed a baseline reading assessment by the end of calendar year 2013, generally consistent with the agency’s fiscal year 2013 target date. USAID expects that 28 of the 36 missions will have completed a baseline assessment as well as a midline or an endline assessment by calendar year 2015. Figure 9 shows the calendar years when USAID missions with funding for primary grade reading programs completed, or are expected to complete, baseline, midline, and endline reading assessments. (See app. III for information about the quality of the assessment data collected at the five missions we selected for fieldwork.) USAID officials said that several factors contributed to the length of time required for missions to implement primary grade reading programs and collect assessment data in accordance with the education strategy. Designing projects in collaboration with host governments and then awarding contracts or cooperative agreements with implementing partners can take a year or more. USAID officials cited the following factors that contributed to the time needed to begin reading programs aligned with the education strategy: Host governments may have had different education priorities than those articulated in USAID’s education strategy and may not fully support a focus on primary grade reading. In such cases, there is a need for discussion with host government officials to balance the education priorities of USAID with those of the host government. For example, the Jordanian government’s education reform efforts were focused on early childhood education and secondary education rather than primary grade reading. In December 2012, the mission in Jordan received USAID’s approval for exemption from full compliance with the education strategy; however, the mission stated that it would add an emphasis on primary grade reading to its education program. Some missions had other, competing development objectives. For example, the mission in Morocco delayed implementing a primary grade reading program because the mission’s education assistance was tied to the competing objective of addressing at-risk youths. However, in 2013 the mission began planning for implementation of a primary grade reading program to support school retention and facilitate reenrollment for primary grade dropouts. Designing new primary grade reading programs in countries where USAID did not have an existing program required additional time. For example, the mission in Nepal had not implemented a primary grade reading program before the education strategy was issued in 2011 and, as a result, needed time to assess the host government’s capacity to operate such a program. Although the USAID contractor responsible for aggregating and analyzing missions’ reading assessment results has begun to collect available data, as of April 2015, only a small number of the missions with completed reading assessments had provided both complete data and the supporting information that the contractor needs. In April 2015, the contractor reported that it had received complete baseline assessment data from 27 missions and midline or endline assessment data from 14 missions. However, the contractor reported that it had received complete supporting information about the programs and assessments from only 4 missions. Examples of the supporting information that the contractor needs to finish aggregating and analyzing the data include a description of the reading interventions, the implementation dates for the interventions, the numbers of children exposed to the interventions, the dates of the assessments and the assessment instrument, and a description of the sampling design for the assessments and population of children assessed. Using this information, the contractor checks the quality of the data; analyzes the effects of the reading interventions on children’s reading skills; compares sampled children’s reading skills at baseline (before the reading interventions begin) and at endline (at or near the end of the interventions); and extrapolates the assessment results for the samples of children to estimate improvements in reading skills for all children exposed to the reading interventions, including children who were not assessed. Senior USAID officials noted that prior to the 2011-2015 education strategy, USAID’s Office of Education had no precedent for such a data collection effort, which required USAID to develop a contractual and legal framework and guidance to facilitate the transfer of reading assessment data and supporting information from missions’ implementing partners. and implementing partners to request and verify the data and to correct errors or account for missing information. Building on its 2012 implementation guidance, in August 2014, USAID issued updated reporting guidance to help ensure that missions have a clear understanding of the requirements and processes for reporting assessment results. The updated guidance clarifies USAID’s expectations for the control and use of the assessment data and supporting information. The guidance also specifies the time frames and procedures for the transfer of assessment data and supporting information to USAID. In addition, the guidance specifies the types of data and supporting information required as well as the formats for the transfer of the data and information. Contractor staff said that the updated guidance has made the transfer of the required assessment data and supporting information more straightforward and consistent among missions and implementing partners. As of April 2015, USAID had not yet selected and documented a methodology for using missions’ assessment data and supporting information to estimate progress toward its goal of improved reading skills for 100 million children by 2015. According to generally accepted survey guidelines, documenting methodologies used to derive an estimate and reporting the information used to evaluate its reliability—for example, sampling designs, sample sizes, and modes of data collection—is important to help ensure that stakeholders are able to understand the appropriate uses of the estimate and interpret the results accurately. U.S. Agency for International Development, 2011 USAID Education Strategy Technical Notes. capacity to test the proposed methodology. For example, according to the contractor, the fact that few missions had provided data from both baseline and midline or endline assessments made it difficult to test for the minimum levels of reading improvement that children must meet to be counted as improved readers in different languages. In addition, the lack of information from multiple countries about the populations of children exposed directly or indirectly to the programs’ reading interventions limited the contractor’s ability to develop possible methods for extrapolating assessment results for samples of children to all children exposed to the interventions. According to contractor staff, as missions and implementing partners transfer data from additional assessments, with complete supporting information, the contractor will be better able to test and implement a methodology for measuring reading improvements in samples of assessed children and to extrapolate the results. Although USAID proposed a methodology in 2012, it recognized that the methodology had limitations that needed to be resolved prior to its application. After unsuccessfully attempting to resolve these limitations, USAID requested that a second contractor examine the methodology and explore alternative methodologies in July 2014. USAID also asked the second contractor to conduct further technical work to implement a methodology, using selected reading programs for which baseline and midline or endline assessment data were available. The second contractor reported key challenges to developing a methodology. These challenges included the difficulty of selecting a methodology that is applicable to different programs and of using a cross- sectional approach, in which different samples of children are assessed at baseline and at endline, as recommended by USAID’s implementation guidance. According to the second contractor, before USAID can produce an estimate of improved readers and monitor progress toward its goal, the agency must define the minimum increase in reading levels that children must demonstrate to be counted as improved readers, select a methodology for calculating the percentages of assessed children demonstrating improved reading skills, and select procedures for extrapolating the results from the assessed sample to the targeted population. In November 2014, after completing its technical work, the second contractor reported to USAID regarding two possible methodologies for calculating the percentages of assessed children demonstrating improved reading skills: the threshold methodology, which USAID proposed in its technical notes to the education strategy, and the gain score methodology, which the contractor proposed as an alternative. According to the second contractor, once USAID selects a methodology, the contractor can calculate the percentages of assessed children who demonstrate improved reading skills. The contractor will then extrapolate these percentages to the populations of children exposed to the programs to estimate total numbers of children with improved reading skills. Threshold methodology. The threshold methodology calculates the percentages of improved readers on the basis of changes in the proportion of children who improve from below a threshold level of reading in a baseline assessment to a level at or above that threshold in an endline assessment. For example, different threshold levels of reading are created using the average numbers of words read correctly per minute and numbers of reading comprehension questions answered correctly for the sample of assessed children at baseline. The percentage of improved readers is then calculated on the basis of the differences between the proportions of children at each of these threshold levels at baseline and at endline. Gain score methodology. The gain score methodology ranks assessment results for a sample of children assessed at baseline and a sample assessed at endline and matches results for individuals in each sample to calculate percentages with improved reading skills. For example, assessment results for the child reading the 10th fewest words correctly per minute at baseline are matched with results for the child reading the 10th fewest words correctly at endline. The percentages of children with improved reading skills are calculated on the basis of the numbers of matched pairs whose reading skills increased by a given minimum—for instance, who read an additional 1, 5, or 10 words correctly per minute. The USAID contractor staff used assessment data from eight USAID programs to compare percentages of improved readers that they calculated using the two methodologies and various minimum reading levels. According to the contractor’s analysis, using the two methodologies to analyze the same assessment results produced different estimates of the percentages of children demonstrating improved reading skills. Specifically, the contractor found that the gain score methodology consistently produced higher estimates of improved readers than did the threshold methodology. According to the contractor, the discrepancy in the estimates from the two methodologies occurs because the threshold methodology only counts children whose reading improves across a threshold level. The threshold methodology does not count children who demonstrate some level of improved reading, but whose reading skills remain below or above a threshold level. The USAID contractor has also reviewed possible procedures for extrapolating the percentages of samples of assessed children demonstrating reading improvements to all children exposed to the reading interventions. However, the contractor noted that it needs to conduct further technical work before making recommendations regarding these procedures to USAID. Lack of information about progress relative to USAID’s primary grade reading goal limits the agency’s ability to assess its current approach to achieving its reading goal and establish a realistic goal for a post-2015 education strategy. Because USAID has not selected a methodology for estimating total numbers of children with improved reading skills resulting from exposure to primary grade reading interventions, it has not produced interim estimates of its progress toward the goal of 100 million children with improved reading skills by December 2015. In addition, USAID’s education strategy and guidance did not establish any interim targets for measuring progress toward USAID’s primary grade reading goal. As a result, USAID officials and other stakeholders currently lack aggregate information needed to monitor and report on the numbers of children whose reading skills have improved as a result of the USAID-supported programs. Establishing interim performance targets and using reliable estimates to assess progress toward an agency goal allow managers and stakeholders to address any performance issues that may affect the outcome of the agency’s goal, according to leading practices for performance management. Frequently reporting progress in achieving interim performance targets allows managers to review the information in time to make improvements. Without interim targets and reliable estimates, USAID cannot assess its actual progress against its planned performance and use the information to make any adjustments needed to better achieve its goal. Furthermore, as the current education strategy’s December 2015 end date approaches, the lack of aggregate information about USAID’s progress in achieving its primary grade reading goal may constrain its efforts to develop a new education strategy. According to agency officials, USAID is presently engaged in developing its post-2015 strategy. We have previously reported that agencies can use performance information to make decisions that affect future strategies. helps program managers decide among competing priorities and reassess their performance goals and strategies. The lack of information about the outcome of USAID’s current primary grade reading goal could limit USAID officials’ capacity to assess USAID’s current strategy and goal and make effective decisions about a realistic goal for USAID’s education strategy after 2015. See GAO-05-927. improved reading skills by December 2015 and to provide data for program monitoring and evaluation. In the five countries where we conducted fieldwork, the USAID missions have implemented primary grade reading programs with elements recommended in USAID guidance and have conducted standardized reading assessments. However, the very low reading skills revealed by data from these assessments suggest that improving skills to the point where students can read with fluency and comprehension will be a long-term effort in these and other developing countries. USAID’s use of standardized reading assessments represents an important step in its efforts to measure the outcomes of its basic education assistance. However, in part, because of the time required for missions to design and implement programs and assess children’s reading skills, USAID’s contractor has limited access to reading assessment data. USAID officials acknowledged that the education strategy’s time frames were unrealistic. Moreover, USAID has not yet selected and documented a methodology—including reporting information on the sampling designs, sample sizes, and modes of data collection—for estimating total numbers of children whose reading skills have improved. Until it selects and documents a methodology for calculating the percentages of assessed children demonstrating improved reading skills, and extrapolating the results to estimate the total numbers of children with improved reading skills, USAID cannot monitor and reliably report estimates of the results of its efforts to improve reading skills of 100 million children. Furthermore, unless it selects and documents a methodology and sets targets for measuring progress toward a reading goal in any future education strategy, USAID will be limited in its ability to assess progress toward that goal in comparison with planned performance. Accurate reporting of USAID’s progress toward its strategic goal would inform stakeholders about the impact of the agency’s efforts to improve children’s reading skills in the final year of the current education strategy and could help inform USAID officials’ efforts to develop a realistic goal for a future education strategy. We recommend that the Acting USAID Administrator take the following three actions. To help ensure that USAID reports a reliable estimate of the results of its efforts to achieve the goal of improving reading skills of 100 million children, we recommend that the Acting Administrator select a methodology for estimating the total numbers of children with improved reading skills as a result of exposure to primary grade reading programs and document a description of the selected methodology as well as the information necessary to evaluate the estimate (i.e., sampling designs, sample sizes, and modes of data collection) when reporting progress toward the reading goal. To improve USAID’s ability to measure progress in achieving a quantitative reading goal in any future education strategy, we recommend that the Acting Administrator ensure that the future strategy includes targets that will allow USAID to monitor interim progress toward its goal in comparison with planned performance. We provided a draft of this report to USAID for comment. In its written comments, reproduced in appendix IV, USAID agreed with our findings and recommendations. USAID also provided technical comments, which we incorporated throughout the report, as appropriate. We are sending copies of this report to interested congressional committees and the Acting USAID Administrator. In addition, the report is available at no charge on the GAO website at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-3149 or gootnickd@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix V. We were asked to review several aspects of the U.S. Agency for International Development’s (USAID) basic education programming. We focused our review on basic education programming related to Goal 1 of USAID’s education strategy. This report (1) examines five USAID missions’ implementation, performance monitoring, and evaluation of primary grade reading programs and (2) assesses USAID’s efforts to estimate progress toward the 2011-2015 education strategy’s primary grade reading goal. For both of these objectives, we analyzed funding for basic education for fiscal years 2010 through 2014 and reviewed USAID’s 2011-2015 education strategy, as well as guidance for implementing the education strategy and USAID’s technical notes to the education strategy. We also interviewed USAID officials in the Bureau for Economic Growth, Education and Environment; the Bureau for Policy, Planning, and Learning; and four of the agency’s geographic bureaus. In addition, we conducted fieldwork in five countries—Ethiopia, Malawi, Peru, the Philippines, and Uganda—which we selected because they are geographically diverse and received relatively large levels of U.S. funding for basic education in fiscal year 2013. We selected the three African countries because this region received more than 50 percent of USAID’s allocation for basic education assistance in fiscal year 2013 and was identified in the education strategy as a priority region for investment of resources. In addition, we selected the five countries because the USAID missions in these countries had primary grade reading programs under way and had conducted at least baseline reading assessments. During our fieldwork in these countries, we met with USAID mission officials; USAID’s implementing partners; host government ministry of education officials; local school district officials; school principals; teachers; and members of parent groups, such as parent-teacher associations. To examine the five missions’ implementation, monitoring, and evaluation of primary grade reading programs—our first objective—we reviewed key planning and reporting documents associated with USAID’s primary grade reading program in each of the five countries. These documents included, among others, contract awards or cooperative agreements, monitoring and evaluation plans, performance management plans, and implementing partners’ annual work plans and quarterly and annual reports. To determine whether the reading programs in the five countries included key program elements that USAID’s education strategy recommends as effective in improving reading outcomes, we compared elements in these programs, as identified in their planning documents, with those recommended in USAID’s technical notes to the education strategy. We also visited schools, where we observed classroom reading lessons that used instructional materials developed through USAID’s programs and met with local teachers, education officials, and community organizations to discuss program interventions. In Ethiopia, we observed local teams of teachers, story writers, editors, and typists in the process of developing reading curriculums in seven mother-tongue languages for grades 5 through 8. To identify the programs’ timelines for delivering improved reading instruction to targeted students, we reviewed the programs’ performance management plans and annual work plans and met with USAID officials at the country missions as well as implementing partner officials. To examine the five USAID missions’ performance monitoring of the current primary grade reading programs, we reviewed outcome and output measures identified in the missions’ and reading programs’ monitoring and evaluation plans and relevant standardized reading assessment reports. We identified each country’s benchmarks for reading at grade level—number of correct words read per minute and percentage of comprehension questions answered correctly—and summarized each program’s baseline data showing the percentage of students meeting these benchmarks. To examine each mission’s plans for conducting performance or impact evaluations of current primary grade reading and interviewed programs, we reviewed USAID’s evaluation policyUSAID officials at missions and implementing partners responsible for conducting external evaluations of these programs. We also reviewed the missions’ related requests for proposals and award agreements, as well as work plans and interim or quarterly reports submitted to the missions by these implementing partners. To assess USAID’s efforts to estimate progress toward the education strategy’s primary grade reading goal—our second objective—we reviewed the USAID education strategy, guidance for implementing the strategy, technical notes to the strategy, and the strategy’s reporting guidance to identify the agency’s policies and guidelines for completing reading assessments at USAID missions and for analyzing the data and supporting information. We analyzed USAID reports on missions’ collection of reading assessment data. We also analyzed reports prepared by the USAID contractor responsible for aggregating and analyzing the reading assessment data, to determine the missions’ progress in transferring reading assessment data and supporting information to the contractor. In addition, we interviewed USAID officials and USAID contractor staff in Washington, D.C., regarding the guidance and procedures for collecting and analyzing the reading assessment data and supporting information as well as missions’ progress in transferring the data to USAID for aggregation and analysis. To examine USAID’s potential methodologies for estimating total numbers of children with improved reading skills, we reviewed USAID contractors’ documentation of proposed methodologies. We also reviewed the USAID contractors’ reports on the development of potential methodologies.contractor staff in Washington, D.C., about the development and implementation of potential methodologies to measure improved reading skills, and the challenges and limitations to collecting and analyzing missions’ reading assessment data. We used GAO standards and In addition, we interviewed USAID officials and USAID generally accepted survey guidelines to assess USAID’s efforts to estimate progress toward its primary grade reading goal. To assess the reliability of the reading assessment data, we compared the design, data collection, and analysis of results for the assessments at the five missions against generally accepted survey guidelines as outlined in Office of Management and Budget (OMB) guidance and USAID guidance.accepted professional survey practices, including the collection of interview-based survey data. We also found that the majority of OMB guidelines were also reflected in USAID’s own implementation guidelines. GAO methodologists compared the information collected from agency documents and during interviews with agency officials and implementing partners and generally found that the reading assessments met these guidelines. We also determined that the data we obtained for both objectives were sufficiently reliable for the purposes of our review. Many OMB guidelines represent long-established, generally We conducted this performance audit from May 2014 to May 2015 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. In fiscal year 2013—the first year in which missions’ education programs were expected to be fully aligned with the goals of the education strategy—the U.S. Agency for International Development (USAID) allocated funding to 41 countries for programs to improve the reading skills of 100 million children in primary grades by December 2015—Goal 1 of its education strategy. Table 4 shows USAID funding in support of Goals 1 and 3 of the strategy. As figure 10 shows, more than half of USAID’s funding for Goal 1 of the strategy was directed to countries in Sub-Saharan Africa, where baseline reading scores were among the lowest globally, according to USAID. Afghanistan and Pakistan received the next largest share, a combined total of 18 percent of USAID’s 2013 basic education allocations. We found that the design and implementation of the reading assessments in the five countries where we conducted fieldwork—Ethiopia, Malawi, Peru, the Philippines, and Uganda—met generally accepted survey guidelines and U.S. Agency for International Development (USAID) guidance on implementing Early Grade Reading Assessments (EGRA). In four of the five countries we visited, a U.S.-based contractor assisted in the design and data collection using the EGRA. Contractor staff in headquarters provided input into both instrument and sample design and conducted in-country training for assessors and supervisors. For example, in each country, assessor and supervisor reliability scores were calculated during training and those assessors and supervisors with low scores were either retrained or not allowed to collect data. Also, sampling designs appropriately considered analytical goals, such as expected precision of subpopulation estimates, and the geographical distribution of schools and the population of students. In some countries, the sampling design also included information about languages. We also generally found that the contractor appropriately calculated country-level estimates based on the complex sample design after assessment data were collected. In three of four countries that used the EGRA, data were collected electronically on tablets and regularly transmitted to contractor headquarters for review (see fig. 11). Data were reviewed on a timely basis so that any potential problems could be identified and corrected in the field. For example, if an assessor was consistently taking too little time administering a reading assessment, that assessor could be identified and retrained if necessary. One country using an EGRA, Ethiopia, collected data using paper questionnaires—a method with the potential for more errors than electronic data collection. However, certain data quality procedures were in place, such as requiring that a consultant verify all of the data entered. One country where we conducted fieldwork, Peru, did not use an EGRA to collect reading assessment data but instead relied on reading comprehension scores from a national assessment administered by the Ministry of Education to all second graders. Ministry officials told us that certain safeguards are in place to protect the integrity of the data collected. For example, assessors attend training and must pass a test, and ministry officials supervise testing sites where the validity of previous results has been questioned. Ministry officials also told us that the current methodology had been in place since 2007 and asserted that year-to-year comparisons of test scores since 2007 were valid. Assessment scores are provided to each region and school, including those receiving basic education assistance, to track children’s reading comprehension progress. In addition to the individual named above, Michael Rohrback (Assistant Director), Howard Cott, Justin Fisher, Bradley Hunt, Jill Lacey, Reid Lowe, Grace Lui, and Michelle Serfass made significant contributions to this report. | While many developing countries have achieved important gains in primary school enrollment, students' reading skills remain very low. In response, USAID's 2011-2015 Education Strategy set a goal of improved reading skills for 100 million primary grade children by the end of 2015 (Goal 1). The 42 missions with funding for primary grade reading programs were to start programs aligned with the strategy by the end of fiscal year 2012 to allow time to assess results by 2015. GAO was asked to review USAID's efforts to implement primary grade reading programs—Goal 1 of its education strategy. This report (1) examines five USAID missions' implementation, monitoring, and evaluation of reading programs and (2) assesses USAID's efforts to estimate progress toward its reading goal. GAO analyzed USAID's education strategy and other guidance, reading assessment data and methodologies, performance monitoring plans and reports, and program evaluations, and conducted fieldwork in five countries selected based on geographic diversity, funding for basic education, and availability of reading assessment data. U.S. Agency for International Development (USAID) missions in all five countries where GAO conducted fieldwork were implementing primary grade reading interventions recommended in USAID's education strategy guidance. A key intervention in four of these countries is developing new reading instruction materials in mother-tongue languages—in some instances, multiple languages. In Ethiopia, for example, USAID's reading program developed new student textbooks and teacher's guides for each of eight grades and seven languages and involved teams of local teachers, language experts, and story writers, among others. The complexity of these interventions in Ethiopia, Malawi, the Philippines, and Uganda increased the time before they could be introduced in classrooms. As a result, some targeted student populations are not expected to benefit from improved reading instruction until late 2015 or 2016, too late to be measured to contribute to USAID's strategic goal of 100 million students with improved reading skills by the end of 2015. In addition, the missions were monitoring their reading programs and planned to conduct program evaluations consistent with USAID guidance. USAID is currently unable to estimate progress toward the education strategy's reading goal because of four factors. First, because it took some missions longer to implement reading programs than USAID estimated, only about two-thirds of missions are expected to have data to estimate progress toward the goal by the end of 2015. Second, only a small number of missions provided USAID the complete data and supporting information that it needs to aggregate and analyze reading assessment results. Agency officials attributed the lack of complete data and information to an absence of timely guidance. In 2014, USAID issued updated reporting guidance. Third, USAID has not selected a methodology for calculating percentages of assessed children demonstrating improved reading skills, and extrapolating the results to estimate the total numbers of children with improved reading skills. Along with incomplete data and information, other factors contributed to a delay in identifying a methodology—for example, USAID did not ask its contractor to examine the methodology proposed in 2012 and explore alternatives until 2014. Fourth, the education strategy did not set interim targets for assessing progress toward the reading goal as suggested by leading performance management practices. Without a methodology and interim targets, USAID officials and others lack aggregate information about progress toward the goal to assess the current strategy and plan a realistic goal for a future strategy. USAID should select a methodology for estimating total numbers of children with improved reading skills, document a description of the methodology when reporting results, and set interim targets to assess progress toward a reading goal in any future education strategy. USAID concurred with GAO's recommendations. |
The Davis-Bacon Act requires contractors and subcontractors working on federally funded contracts in excess of $2,000 to pay at least locally prevailing wages to laborers and mechanics. The act covers both new construction and the alteration or repair of existing public buildings and works. The Department of Labor sets prevailing wage rates for various job categories in a local area on the basis of periodic surveys it conducts of contractors, unions, public officials, and other interested parties. The Davis-Bacon Act stems from a Depression-era practice of transporting workers from lower-paying areas to bypass local workers who would demand a higher wage. The prevailing wage requirement was meant to prevent this practice by ensuring that workers on federal projects were paid at least the locally prevailing wage. Congress has extended this requirement beyond projects funded directly by the federal government by including Davis-Bacon Act prevailing wage provisions in numerous related laws under which federal agencies assist construction projects through grants, loans, guarantees, insurance, and other methods. Examples of related laws include the Federal-Aid Highway Acts, the Housing and Community Development Act of 1974, and the Federal Water Pollution Control Act. Contractors on projects subject to Davis-Bacon requirements may also be subject to additional prevailing wage requirements under state and local laws. In addition to paying no less than locally prevailing wages, contractors for construction projects that are subject to the Davis-Bacon Act must pay their workers on a weekly basis and submit weekly certified payroll records. The federal contracting or administering agency has primary responsibility for enforcing these requirements, while the Department of Labor has coordination and oversight responsibilities, including the authority to establish regulations and investigate compliance with labor standards as warranted. The prevailing wage provision in section 1606 of the Recovery Act broadly applies Davis-Bacon requirements to all construction projects funded directly or assisted by the federal government under Division A of the act. It reinforces Davis-Bacon Act coverage of construction projects where the federal government is a party to the contract, extends it to projects assisted in whole or in part by Division A of the act, and overri any limitation to Davis-Bacon coverage in related laws under which federal agencies provide financial assistance, such as grants and loan guarantees, to recipients to use for construction projects. It also extends the prevailing wage requirement to some federally assisted projects th would not otherwise be subject to that req uirement. According to federal agency officials, 40 programs are newly subject to Davis-Bacon requirements as a result of the Recovery Act’s prevailing wage provision, as shown in table 1. These programs are spread across 12 of 27 federal agencies that received funding under Division A of the ac Most of the programs existed prior to the Recovery Act and are subject to Davis-Bacon requirements for the first time under the act, while some are newly created programs. t. Federal, state, and local officials involved with the 40 Recovery Act programs that are newly subject to Davis-Bacon requirements differed on whether those requirements would increase program costs. However, these officials generally did not expect Davis-Bacon requirements to inhibit their ability to achieve Recovery Act and program goals. Federal officials responsible for most of the programs did not expect Davis-Bacon requirements to affect the timing of their program’s Recovery Act efforts, though officials for some programs expected an impact. Federal officials responsible for programs that are newly subject to Davis- Bacon requirements—the prevailing wage rate requirement and the administrative requirements associated with weekly payroll processes— had mixed views on the extent to which they expected these requirements to affect program costs, as table 2 shows. Officials for the 40 programs provided a range of explanations for the extent to which the prevailing wage rate requirement might impact program costs. Specifically: Little to no impact. Department of Energy officials responsible for two programs said construction firms generally pay prevailing wage rates as a standard practice and therefore the prevailing wage rate requirement would have no impact on program costs. According to other program officials, the impact would be small because a relatively small amount of program funds are to be spent on construction activities that are subject to Davis-Bacon requirements. For example, Department of Justice officials responsible for the Transitional Housing Assistance Grants for Victims of Domestic Violence, Dating Violence, Stalking, or Sexual Assault Program said only a small number of grantees requested funds for construction and that less than 5 percent would be allocated for this purpose. Department of Energy officials responsible for the Geothermal Technologies program noted that the prevailing wage rates were in line with what they expected, and Electricity Delivery and Energy Reliability program officials said existing wage rates paid by utility companies were generally high already so any increase in wage expenses due to prevailing wage rates would probably be minimal. Moderate impact. Officials from the Department of Energy’s Weatherization Assistance Program explained that the prevailing wage rate requirement will have at least a moderate impact on program costs. They explained that weatherization projects in buildings taller than four stories will require that workers be paid commercial prevailing wage rates that are higher than the wage rates that would otherwise be used for weatherization projects. Department of Energy officials responsible for the Wind and Hydropower Technologies program explained that the prevailing wage rates for construction workers are sometimes 20 percent higher than what would have been paid for similar work and could increase labor costs by 20 percent. Conversely, these officials said the payment of higher wages could attract a more highly trained laborer and thus possibly result in savings in rework or in adherence to safety guidelines. Large impact. Department of Energy officials from the State Energy Program explained that in some of the areas hardest hit economically, construction workers are paid less than the prevailing wage rate for the county. Therefore, paying the Davis-Bacon prevailing wage rate will increase costs. Likewise, Department of Justice officials responsible for the Correctional Facilities on Tribal Lands Program explained that in some cases the prevailing wage rate may be significantly higher than what the tribe would normally pay for construction. Department of Energy officials responsible for the Energy Efficiency and Conservation Block Grants program anticipated a potentially large impact as a result of the large number of grantees and significant proportion of funds that would be spent on construction labor wages. State and local officials we interviewed also had mixed views about the expected impact of the prevailing wage rate requirement on program costs. District of Columbia weatherization officials explained that the prevailing wage rates were generally in line with what they would expect to pay and some state weatherization officials in California said the prevailing wage rates may be less than what some service providers are currently paying. Data provided by State Energy Program officials in Georgia indicated that contractor wage rates are higher than Davis-Bacon prevailing wage rates in a subset of counties. Conversely, Iowa state weatherization officials explained that the average cost increase would be about 9 percent per home, and one local Lead Hazard Reduction Program official reported that, based on historical experience, the prevailing wage rate requirement could increase program costs by 10 to 13 percent per home. One Mississippi contractor we interviewed said the wage rates would not have an impact on costs because the hourly rates that the company pays its employees are and always have been higher than the Davis-Bacon rates. As table 2 also shows, federal officials reported mixed views on the extent to which they expected Davis-Bacon administrative requirements, such as paying workers weekly, to affect program administrative costs. For example: Little to no impact. Department of Health and Human Services officials responsible for the Capital Improvement Program and Facility Investment Program for Health Centers said the requirements should have no impact on program costs since grantees were asked to include Davis-Bacon compliance in their proposals. Department of Energy officials responsible for the Solar Energy Technologies program expected the administrative requirements to have a small impact because less than 5 percent of funds will be used for construction-related activities subject to Davis-Bacon requirements. Department of Housing and Urban Development officials responsible for the Green Retrofit Program for Multifamily Housing said the administrative requirements would have a small impact because the grant recipients had previous experience with the Davis-Bacon requirements through other federal housing programs and were accustomed to requirements such as paying workers on a weekly basis. Environmental Protection Agency officials responsible for the Diesel Emission Reduction Act Grants program said the impact would be small because the majority of the grant funds are used to purchase equipment and are therefore not subject to the administrative requirements. These program officials added that while the requirements are new to most grantees, the grantees will become familiar with them over time. Moderate impact. State Energy Program officials noted that many construction companies involved with their projects do not maintain payroll records sufficient to meet Davis-Bacon requirements, and as a result, the administrative requirement to pay workers weekly may add to their administrative costs. Large impact. Federal officials responsible for the Weatherization Assistance Program and the Lead Hazard Reduction Program said Davis- Bacon administrative requirements would require a more detailed payroll tracking system, which would be particularly burdensome for small companies. According to Weatherization Assistance Program officials, smaller companies are the ones, generally speaking, that do not usually have experience with the Davis-Bacon requirement for certified weekly payrolls. For these employers, who often employ fewer than five people, it is particularly burdensome to certify payrolls weekly. Lead Hazard Reduction Program officials explained that additional administrative duties necessary for weekly payroll processing will increase administrative costs. To accommodate this increase, the agency is in the process of increasing the cap on how much recipients can spend on administrative costs from 10 to 15 percent of their award. State and local officials we interviewed had mixed views on the impact of Davis-Bacon administrative requirements on program costs. Clean Water State Revolving Fund officials in Georgia and the District of Columbia said they do not anticipate any additional costs as a result of the administrative requirements, whereas program officials in Mississippi and New Jersey noted the requirements would likely increase project costs. Local agencies involved with the Weatherization Assistance Program in California, Michigan, New York, and Ohio reported hiring new staff to process Davis-Bacon paperwork, and local weatherization officials from California noted that the administrative requirements might be particularly burdensome on smaller businesses. Local officials responsible for a Lead Hazard Reduction program in New York said their subcontractor is familiar with the administrative requirements, and the subcontractor has not indicated that these requirements are problematic. Despite these potential cost increases for some Recovery Act programs, most federal officials said the Davis-Bacon requirements will have little, if any, impact on their ability to support their program and Recovery Act goals. Federal officials responsible for 15 of the 40 programs said Davis- Bacon requirements would have no impact on their program’s ability to achieve its goals, and officials from 12 programs reported that the requirements would have little impact (see table 3). However, federal officials from four programs—the Weatherization Assistance Program, State Energy Program, Energy Efficiency and Conservation Block Grants, and Correctional Facilities on Tribal Lands Program—noted that the Davis-Bacon requirements could have a large impact on their ability to support the Recovery Act goal of preserving or creating new jobs. For example, Weatherization Assistance Program officials said that Davis- Bacon requirements will have a large impact in urban areas because they have to pay commercial construction rates to weatherize buildings over four stories tall. These commercial construction wage rates are higher than the wage rates officials were expecting to pay and officials said program goals would be affected because they will have to reduce the number of homes weatherized. State and local officials we interviewed and collected data from also reported that the Davis-Bacon requirements would generally have little impact on their ability to achieve program and Recovery Act goals. California state officials responsible for the State Energy Program stated that even though the requirements may have been an inhibiting factor for some applicants, they do not believe that the requirements will negatively affect the ability to achieve energy policy goals. While Ohio state officials responsible for weatherization said their program goals have not been affected, District of Columbia Lead Hazard Reduction program officials said they had to reduce the number of homes to be treated. Florida officials responsible for the Clean Water State Revolving Fund and Drinking Water State Revolving Fund programs reported that the addition of Davis-Bacon requirements had little impact on their program’s ability to support the underlying Recovery Act goals, such as creating jobs, in part due to the ultra-competitive behavior during the economic downturn and the enormous demand for environmental infrastructure funding. Conversely, Michigan state weatherization officials said the requirements might affect their ability to support the Recovery Act goal of job creation, especially for smaller businesses and contractors. According to these officials, some smaller local contractors who performed weatherization work before the Recovery Act sometimes may not have the capacity to take advantage of the Recovery Act weatherization work because of requirements such as the weekly payroll certifications. According to federal officials responsible for most programs that are newly subject to Davis-Bacon requirements, the requirements are not likely to affect the timing of their program’s Recovery Act efforts. However, officials for some programs expected an impact on their program’s timing. Federal officials for 23 of the 40 programs said they did not expect Davis- Bacon requirements to affect their program’s timing. For example, an official for the National Science Foundation’s Academic Research Infrastructure Program—Recovery and Reinvestment noted that while they had to spend time consulting with the Office of Management and Budget and the Department of Labor on Davis-Bacon requirements, this effort had not significantly altered the program’s timing. Federal officials for 2 programs were uncertain about how Davis-Bacon requirements might affect their program’s timing, while officials for 7 programs did not provide a response in time for our report. In contrast, federal officials for 8 programs expected Davis-Bacon requirements to have an impact on the timing of their program’s Recovery Act efforts. For example, officials with the Department of Energy’s Weatherization Assistance Program stated that Davis-Bacon requirements had significantly affected their program’s timing because the program is newly subject to the requirements so prevailing wage rates for weatherization workers were not immediately available. Some states decided to wait to begin weatherizing homes until the Department of Labor had determined county-by-county prevailing wage rates for their state to avoid having to pay back wages to weatherization workers who started working before the prevailing wage rates were known. For example, state weatherization officials in Arizona said that all but one of their local service providers decided to wait to weatherize homes using Recovery Act funds until the prevailing wage rates were determined because they were concerned about the time required to reconcile differences in wage rates. The timing of the Weatherization Assistance Program’s Recovery Act efforts was also affected by concerns about complying with Davis-Bacon requirements. For example, Pennsylvania weatherization officials stated that delays occurred because some local agencies had initially submitted management plans that had not included language describing how they would comply with those requirements. Officials with the Department of Energy’s Energy Efficiency and Conservation Block Grants and State Energy Program also expected Davis-Bacon requirements to affect the timing of their Recovery Act efforts, while officials with the Department of Housing and Urban Development’s Lead Hazard Reduction Program reported that grantees were provided additional time to complete their work plans to ensure contractors understood the requirements. We provided a draft of this report to all 27 agencies and the Office of Management and Budget for their review and comment. Two agencies provided technical comments that were incorporated into the report as appropriate. The other agencies had no comments. We also provided a copy of relevant sections of the report to GAO teams responsible for reviewing Recovery Act work in the states mentioned in this report. In some cases, those teams forwarded relevant sections to officials within those states. We included these comments as appropriate. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies to all 27 agencies reviewed in this report, and other interested parties. The report will also be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff members have any questions about this report, please contact me at (202) 512-3841 or daltonp@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix I. Other key contributors to this report were Mark Gaffigan, (Director), Ric Cheston (Assistant Director), Kim Gianopoulos (Assistant Director), David Marroni (Analyst-in-Charge), Angela Miles, and Alise Nacson. Important contributions were also made by Anne Rhodes-Kline, Carol Herrnstadt Shulman, and Barbara Timmerman. | The American Recovery and Reinvestment Act of 2009 (Recovery Act) has the broad purpose of stimulating the economy. It includes substantial appropriations for construction projects that, under the act's prevailing wage provision, are subject to Davis-Bacon Act requirements. That is, contractors must pay laborers and mechanics who work on those projects at least the prevailing wage rates set for their local area by the Secretary of Labor. In addition, contractors must submit certified payrolls and pay their workers weekly. Prior to the Recovery Act, some federal programs with construction projects were already subject to Davis-Bacon Act requirements. Others, however, are subject to the requirements for the first time because the Recovery Act extended the requirements to all construction projects supported by the act. GAO was asked to (1) identify the programs that are newly affected by the Recovery Act's prevailing wage provision and (2) examine the extent to which that provision is expected to affect each of those newly affected programs. GAO obtained data from 27 agencies and spoke with federal, state, and local officials as well as contractors involved with the newly affected programs. Although GAO is not making recommendations in this report, these findings may be helpful in considering and designing legislation with similar objectives. Fortyprograms are newly subject to Davis-Bacon requirements as a result of the Recovery Act's prevailing wage provision, according to federal agency officials. Of these, 33 programs existed prior to the Recovery Act and are subject to Davis-Bacon requirements for the first time under the act, while 7 are newly created programs. Together, the 40 programs account for about $102 billion of the $309 billion that was appropriated by the Recovery Act for projects and activities. However, a smaller amount of these funds will be subject to Davis-Bacon requirements because not all of the funds will be used for construction activities and only a portion of those funds will be used to pay labor wages. For those programs that are newly subject to Davis-Bacon requirements, officials had mixed views on the impact of these requirements on program costs and goal achievement. In some cases, officials said Davis-Bacon requirements would have little or no impact on program costs for a few reasons, such as (1) the program having a small amount of construction activities, (2) prevailing wage rates that were in line with expectations, and (3) companies' previous experience with weekly payrolls. In other cases, officials said the requirements would have a moderate to large impact on program costs and/or goals. For example, officials from the Department of Energy's (DOE) Energy Efficiency and Conservation Block Grants program anticipated a potentially large cost impact as a result of the significant amount of funds to be spent on construction labor wages. Officials from DOE's Weatherization Assistance Program reported that weatherization projects in buildings taller than four stories will require workers to be paid a commercial prevailing wage rate under the Davis-Bacon Act that is higher than what would otherwise be used and could potentially reduce the number of homes weatherized. Additionally, weatherization officials said that Davis-Bacon requirements affected the program's timing because prevailing wage rates for weatherization workers were not fully available until September 2009. Further, officials from the Department of Housing and Urban Development's Lead Hazard Reduction Program noted that Davis-Bacon requirements would require a more detailed payroll tracking system that could be particularly burdensome for small companies. Those officials also explained that because administrative costs are likely to increase, the department is in the process of increasing the cap on how much recipients can spend on administrative costs. |
The purchase price of electronic products primarily reflects their technological capabilities; it does not include all of the substantial costs that are incurred throughout the equipment’s life. A study by Gartner Research, for example, shows that computers costing less than $1,000 typically have a total cost of ownership of more than $5,000 per year when all the energy and maintenance costs are included. Furthermore, the purchase price of electronics does not include the often substantial cost of disposal. Lifecycle costs are high, in part, because electronic products are not always designed to facilitate recycling. EPA estimates that across the federal government 10,000 computers are disposed of each week. Once such products reach the end of their original useful lives, federal agencies have several options for disposing of them. Agencies generally can donate their reusable equipment to schools or other nonprofit educational institutions; give them to a recycler; exchange them with other federal, state, or local agencies; sometimes trade them with vendors to offset the costs of new products; or sell them through the General Services Administration’s (GSA) surplus property program, which sells surplus federal government equipment at public auctions. Federal agencies, however, are not required to track the ultimate destination of their donated or recycled e-waste. Instead, agency officials generally consider this to be the recipient organization’s responsibility. Consequently, they often have little assurance that their e-waste is ultimately disposed of in an environmentally responsible manner. In our prior work, we found that some U.S. electronics recyclers—including ones that publicly tout their exemplary environmental practices—were apparently willing to circumvent U.S. hazardous waste export laws and export e-waste to developing countries. Specifically, we posed as foreign buyers of broken cathode-ray tube computer monitors—which are considered hazardous waste and illegal to export without a permit—in Hong Kong, India, Pakistan, and other countries; and 43 U.S. companies expressed willingness to export these items. Some of the companies were willing to export this equipment in apparent violation of U.S. law. As we showed in our August 2008 report, equipment exported to developing countries may be handled in a way that threatens human health and the environment. As we reported in November 2005, existing federal government approaches to ensuring environmentally responsible management of electronic equipment from procurement through disposal rely heavily on two interrelated EPA electronic product stewardship initiatives. The first, the electronic product environmental assessment tool (EPEAT®), assists federal procurement officials in comparing and selecting laptop computers, desktop computers, and monitors with environmentally preferable attributes. The second, the federal electronics challenge (FEC), helps federal agencies fully utilize the benefits of EPEAT-rated electronics by providing resources to help agencies extend these products’ life spans, operate them in an energy efficient way, and expand markets for recycling and recovered materials by recycling them at end of life. EPEAT was developed along the lines of EPA’s and the Department of Energy’s (DOE) Energy Star program in which the federal government rewards manufacturers of energy-efficient products that ultimately save money and protect the environment by providing them with a label for their products that certifies these benefits. EPEAT-registered products are awarded a bronze, silver, or gold certification for increasing levels of energy efficiency and environmental performance. Using EPEAT, an on- line tool, federal procurement officials can evaluate the design of an electronic product for energy conservation, reduced toxicity, extended lifespan, and end of life recycling, among other things. For example, EPEAT can help agency procurement officials choose electronic products with attributes that make the products easier to upgrade. Some computers are now being built with modular features so that hard drives, processors, memory cards, and other components can be upgraded rather than replaced—thus extending their lifecycles. Agency procurement officials can also use EPEAT to choose among products that are designed to make recycling less expensive, such as those without glues or adhesives, with common fasteners and “snap-in” features, and with easily separable plastic and metal components—making their disassembly easier and recycling less costly. Finally, EPEAT can help procurement officials identify electronic products that contain less hazardous materials, which can also lessen their disposal and recycling costs. Products with these attributes can, in many cases, save agencies money over the products’ lifecycles when compared to those with similar technological characteristics but without environmental attributes. For example, according to one computer vendor, a particular desktop computer with energy-saving attributes cost $35 more than a similar model that one federal program office had been buying; however, it will save $15 per year in energy costs. Thus, after slightly more than 2 years of use, the EPEAT-rated desktop computer can save more money in energy savings than the additional increase in purchase price and result in measurable environmental benefits. Currently, in the electronic products industry, purchasers can choose from 170 desktop computers, 637 laptop computers, and 487 monitors that meet one of the three EPEAT levels of environmental performance. The breadth of EPEAT products provides procurement officials with a range of devices to meet their technology and budgetary needs. For example, agencies have the flexibility to choose liquid crystal display monitors that meet all the required EPEAT criteria as well as numerous optional criteria, such as the lower levels of mercury in light switches and a reduced number of different types of plastics—attributes that can make recycling easier and less costly. Agencies can also choose other monitors that meet these and other criteria, including additional reductions in toxic materials, along with end-of-life services such as a take-back and reuse program for packaging material. Of note, these different types of monitors can meet different technology needs, as there are some differences in display characteristics and power consumption. As we said earlier, federal agencies also have the opportunity to participate in FEC—a program that first relies heavily on EPEAT for procurement considerations and then provides guidance to participants on how to extend electronic product life spans, operate them in an energy- efficient way, and reuse or recycle them at end-of-life. FEC differs from EPEAT in that where EPEAT assists officials in procuring environmentally preferable products, FEC provides participating agencies and facilities with resources to help ensure that electronic products are operated and disposed of in a manner that fully utilizes the environmental attributes of the EPEAT product. FEC has two partner levels: agency and facility. To participate, executive branch agencies or their subcomponents must register. According to EPA documents, participation can provide agency officials greater assurance that the e-waste they donate to schools, or send for recycling, is ultimately disposed of in an environmentally responsible manner. For instance, in following FEC guidance, participants are to provide recipients of donated equipment with instructions on how to have the equipment recycled responsibly and how to verify that responsible recycling occurs—procedures known as “downstream auditing.” When donating equipment, FEC instructs agencies and facilities to ensure that recipients contact local or state environmental or solid waste agencies to obtain a database of vendors who recycle e-waste once the equipment is no longer useful to the recipient organization. FEC also recommends that participating agencies and facilities instruct recipients to avoid arrangements with recyclers that are unable or unwilling to share references and cannot explain the final destination of the e-waste they collect. When recycling equipment, participants are to determine how much electronic equipment the recyclers actually recycle, versus the amount they sell to other parties. If the majority of the incoming e-waste is sold, the recycling facility may be sending a significant amount of e-waste into landfills or for export overseas. In addition, FEC instructs participants to physically inspect potential recycler’s facilities. E-waste in trash containers, for example, may indicate that the facility is not recycling it, and the presence of shipping containers may indicate that the facility exports it. As of December 31, 2008, EPA reported that 16 federal agencies and 215 federal facilities—representing slightly more than one-third of all federal employees—participated in the FEC to some extent. In addition, according to the 128 facilities that reported data to EPA, a majority of electronic products purchased during 2008 were EPEAT-registered. This is a sizeable increase from 2005, when we reported that 12 federal agencies and 61 individual federal facilities participated in FEC. Participating agencies include the Departments of Agriculture, Commerce, Defense, Energy, Health and Human Services, Homeland Security, Interior, Justice, Labor, Treasury, Transportation, and Veterans Affairs, as well as the Environmental Protection Agency, Executive Office of the President, General Services Administration, and the United States Postal Service. The benefits of federal agency and facility participation in EPEAT and FEC offer a glimpse of what can be attained through greater federal involvement. For instance, in 2008 FEC participants reported to EPA and the Office of the Federal Environmental Executive that 88 percent of all desktop computers, laptop computers, and monitors they purchased or leased were EPEAT registered. In addition, FEC participants reported that they extended computer life spans so that 63 percent of computers had at least a 4-year useful life. Procurement officials reported purchasing 95 percent of their monitors with energy-efficient power management features enabled and 38 percent of computers with this feature. Finally, participants reported that 50 percent of electronics taken out of service were donated for reuse; 40 percent were recycled; 8 percent were sold; and 2 percent were disposed of. Of those recycled, 95 percent were reportedly done so in an environmentally sound manner. These environmentally preferable choices from “cradle to grave” resulted in $40.3 million in cost savings reported by participating agencies and facilities, energy savings that EPA found to be equivalent to electric power for more than 35,000 U.S. households for 1 year, and emissions savings equivalent to removing nearly 21,000 passenger cars from the road for 1 year. Through participation in the FEC, numerous federal facilities have purchased greener electronic products, reduced the environmental impacts of electronic products during use, and managed obsolete electronics in an environmentally safe way. For example, officials with the Bonneville Power Administration within DOE reported to EPA that they adopted several environmentally responsible practices associated with the procurement and operation of electronic equipment. First, administration officials extended the lifespan of agency computers from 3 to 4 years. With over 500 computers procured each year at an annual cost of more than $500,000, an administration official said that extending computer life spans generated substantial savings. Additionally, Bonneville Power Administration officials procured new flat-screen monitors instead of cathode-ray tube monitors, reducing both hazardous waste tonnage and end of life recycling costs. According to Bonneville Power Administration officials, they expect to save at least $153 per unit over the life of each new monitor. EPA’s region 9 facility in San Francisco, California—a 20-story office building that houses nearly 900 EPA employees—also reported achieving substantial environmental benefits through participation in the FEC. The facility’s energy subcommittee recommended an audit, which found that enabling computer and monitor power management features, such as those configuring computer monitors to the “sleep” mode instead of the screen saver mode, could save about 10 percent in total energy usage at no cost. In addition, with funding eliminated for new electronics purchases, region 9 staff reported that they reused 30 percent to 40 percent of existing electronics and extended the average lifespan of computers to 5 years. Finally, region 9 staff stated that they successfully recycled more than 10 tons of electronics that had been stored in an offsite warehouse. Although the cost of safely recycling the large quantity of electronics was high and regional staff found it difficult to locate a reputable recycler, EPA headquarters provided funds for the recycling costs and helped find a qualified vendor. The EPEAT and FEC accomplishments achieved to date are steps in the right direction, but opportunities exist to significantly increase the breadth and depth of federal agency and facility participation. First, agencies and facilities representing almost two-thirds of the federal workforce are not yet participating in these promising initiatives, despite Executive Order 13423. This executive order, signed by the President on January 24, 2007, generally requires that each agency (1) meets at least 95 percent of its requirements with EPEAT-registered products; (2) enables the energy saving features on agency computers and monitors; (3) establishes and implements policies to extend the useful life of agency electronic equipment; and (4) uses environmentally sound practices with respect to disposition of agency electronic equipment that has reached the end of its useful life. To implement these requirements, the Office of Management and Budget directed each agency and its facilities to either become a partner in the FEC or to implement an equivalent electronics stewardship program that addresses purchase, operation and maintenance, and end-of- life management strategies for electronic assets consistent with FEC’s recommended practices and guidelines. Second, most of agencies and facilities that participate do not fully maximize these programs’ resources or the environmental benefits that can be achieved. While we acknowledge the efforts of FEC participants, the FEC statistics on participation may overstate these participants’ adherence to the goals of the program, and their successes must be taken in context. Participation by 16 agencies and 215 facilities (representing slightly more than one-third of federal employees), for example, does not mean that all electronic products they purchase are procured, operated, and recycled or reused at end of life in an environmentally preferable fashion. Instead, participation simply means these agencies have identified their current practices for managing electronic products and set goals to improve them. Moreover, as the FEC is an initiative aimed to encourage and support participating agencies and facilities, it does not impose consequences on those agencies who do not meet their goals. As a practical matter, only 34 FEC facility participants (16 percent of participants) reported to EPA that they managed electronic products in accordance with FEC goals for at least one of the three lifecycle phases— procurement, operation, or disposal—with only 2 facilities showing they did this for all three phases in 2008. The need for increased federal participation in these initiatives—in both breadth and depth—is further underscored by the federal government e- waste that continues to appear in online auctions and may subsequently end up overseas. As we reported in August 2008, significant demand exists for used electronics from the United States. We observed thousands of requests for such items on e-commerce Web sites—mostly from Asian countries, such as China and India, but also from some African countries. In our prior work, we showed that these countries often lack the capacity to safely handle and dispose of e-waste, as disassembly practices in these countries often involve the open-air burning of wire to recover copper and open acid baths for separating metals. These practices expose people to lead and other hazardous materials. In the several weeks leading up to this hearing, we monitored an e-commerce Web site where surplus federal government equipment is auctioned and found nearly 450,000 pounds of cathode-ray tube monitors for sale—items that, based on our prior work, have a high likelihood of being exported. For perspective, using EPA’s environmental benefits calculator we calculated the benefits that would result under a hypothetical scenario in which federal agencies replaced 500,000 desktop and laptop computers and computer monitors using EPEAT procurement criteria for each tier of environmental performance—bronze, silver, and gold. As part of this calculation, we added the environmental benefits attained if federal agencies operated all EPEAT units in an energy efficient manner (i.e., enabled Energy Star features) and reused and recycled the end-of-life electronics they replaced in accordance with FEC goals. We found that substantial energy savings and environmental benefits would result at all three EPEAT tiers. Specifically, greater participation could lead to environmental benefits 5- to 10-times greater than the accomplishments of FEC participants in 2008 described earlier. Additionally, if federal agencies were to purchase EPEAT-bronze, silver, or gold products, according to the EPA environmental benefits calculator, they would save approximately $207 million at each level of EPEAT performance in energy usage and realize other cost, waste, and emissions reductions over the useful lives of these products. Table 1 shows the net energy savings and reductions in raw material extraction, greenhouse gas emissions, and toxic materials that would result if agencies and facilities recycled electronic products and replaced them with EPEAT-rated units, as compared to non-EPEAT computers and monitors. To help agency officials put in context the environmental and economic benefits that can result from using environmentally preferable electronic products, the EPA environmental benefits calculator also shows the benefits of procurement, operation, and disposal in accordance with FEC goals using common equivalents. Table 2 shows the environmental benefits of these practices when measured as the amount of household energy usage saved annually and the volume of automobile emissions saved annually. Understandably, when procuring electronics in a challenging fiscal environment, agency officials may give greater weight to price than environmental attributes. However, many of the environmental and human health problems associated with e-waste disposal can be averted through environmentally preferable procurement. Using EPEAT to purchase environmentally-friendly products, agency purchasers can often simultaneously meet their technology needs, benefit the environment, and realize dollar savings over the products’ lives. Using the success of the Energy Star program as a precedent, the federal government has taken steps to encourage environmentally preferable choices. We also applaud federal agency and facility donation and recycling practices for providing valuable learning tools to thousands of school children while, at the same time, providing at least some protection against their equipment ending up in landfills or overseas. Such programs have also demonstrated that relatively simple and inexpensive steps can help ensure that donated and recycled e-waste is ultimately managed in a responsible manner. In particular, the FEC provides a framework through which participants can help ensure responsible recycling through downstream auditing of recipient organizations’ disposal practices and by following guidance on how to select responsible recyclers. The federal government has the opportunity to lead by example and to leverage its substantial market power by broadening and deepening agency and facility participation in EPA electronic product stewardship initiatives, but meaningful results will only occur if federal agencies and facilities fully participate and utilize these promising initiatives’ resources. Ms. Chairwoman, this concludes my prepared statement. I would be happy to respond to any questions that you or other Members of the Subcommittee may have at this time. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. For further information about this testimony, please contact John Stephenson, Director, Natural Resources and Environment at (202) 512-3841 or stephensonj@gao.gov. Key contributors to this statement were Steve Elstein (Assistant Director), Nathan Anderson, Elizabeth Beardsley, Alison O’Neill, and Kiki Theodoropoulos. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Advancing technology has led to increasing sales of new electronic devices. With this increase comes the dilemma of managing them at the end of their useful lives. If discarded with common trash, a number of environmental impacts may result, ranging from the loss of valuable resources to the potential release of toxic substances, such as lead. If recycled, they may be exported to countries with waste management systems that are less protective of human health and the environment that those of the United States. The federal government is the world's largest purchaser of electronics, spending nearly $75 billion on electronic products and services in 2009. The Environmental Protection Agency (EPA) has helped implement several product stewardship initiatives to encourage responsible management of electronic products in all three phases of a product's lifecycle--procurement, operation, and end-of-life disposal. In response to a request to provide information on federal procurement and management of electronic products, GAO's testimony describes (1) EPA's electronic product stewardship initiatives, (2) federal agency participation in them, and (3) opportunities for strengthening participation. GAO's testimony is based on its prior work and updated with data from EPA. In our prior report, EPA agreed that increasing federal participation in its initiatives could be encouraged. Agency officials still agree with this finding. Federal government approaches to ensuring environmentally responsible management of electronic equipment from procurement through disposal rely heavily on two interrelated initiatives. The first initiative, the electronic product environmental assessment tool (EPEAT), was developed along the lines of EPA's and the Department of Energy's Energy Star program and assists federal procurement officials in comparing and selecting computers and monitors with environmental attributes that also routinely save money through reduced energy usage over the products' lives. The second initiative--the federal electronics challenge (FEC)--helps federal agencies realize the benefits of EPEAT-rated electronics by providing resources to help agencies extend these products' life spans, operate them in an energy efficient way, and expand markets for recovered materials by recycling them at end of life. The first 5 years of EPA's initiatives have resulted in notable energy savings and environmental benefits reported by participating agencies. According to facilities that reported information to EPA and the Office of the Federal Environmental Executive in 2008, 88 percent of all desktop computers, laptop computers, and monitors the facilities purchased or leased were EPEAT-registered. EPEAT participation reportedly resulted in procurement officials purchasing 95 percent of their monitors with Energy Star power management features enabled and 38 percent of computers with this feature. In addition, 16 federal agencies and 215 federal facilities--representing about one-third of all federal employees--participated in the FEC to some extent in 2008. As a result, participants reported that 50 percent of electronics taken out of service were donated for reuse, 40 percent were recycled, 8 percent were sold, and 2 percent were disposed of. The environmentally responsible choices associated with EPEAT and FEC resulted in a reported $40.3 million in cost savings for participants. The EPEAT and FEC accomplishments are steps in the right direction, but opportunities exist to increase the breadth and depth of federal participation. First, agencies and facilities representing about two-thirds of the federal workforce are not participating in these promising initiatives, despite instructions to do so in implementing Executive Order 13423. Second, few participating agencies and facilities maximize these programs' resources and their potential benefits. For some, participation simply means the agency identified its current practices for managing electronic products and set goals to improve them. Moreover, as the FEC aims to support participating agencies and facilities, it does not impose consequences for those that do not meet their goals. In fact, only 34 FEC facility partners showed they managed electronic products in 2008 in accordance with FEC goals for at least one of the three lifecycle phases, and only 2 facilities showed they did so for all phases. For perspective, GAO calculated that if federal agencies replaced 500,000 desktop and laptop computers and monitors with EPEAT-registered products and operated and disposed of them in accordance with FEC goals, they could achieve substantially greater energy reductions and cost savings. |
FOIA establishes a legal right of access to government records and information, on the basis of the principles of openness and accountability in government. Before the act (originally enacted in 1966), an individual seeking access to federal records had faced the burden of establishing a right to examine them. FOIA established a “right to know” standard for access, instead of a “need to know,” and shifted the burden of proof from the individual to the government agency seeking to deny access. FOIA provides the public with access to government information either through “affirmative agency disclosure”—publishing information in the Federal Register or the Internet, or making it available in reading rooms—or in response to public requests for disclosure. Public requests for disclosure of records are the best known type of FOIA disclosure. Any member of the public may request access to information held by federal agencies, without showing a need or reason for seeking the information. Not all information held by the government is subject to FOIA. The act prescribes nine specific categories of information that are exempt from disclosure: for example, trade secrets and certain privileged commercial or financial information, certain personnel and medical files, and certain law enforcement records or information (attachment 1 provides the complete list). In denying access to material, agencies may cite these exemptions. The act requires agencies to notify requesters of the reasons for any adverse determination (that is, a determination not to provide records) and grants requesters the right to appeal agency decisions to deny access. In addition, agencies are required to meet certain time frames for making key determinations: whether to comply with requests (20 business days from receipt of the request), responses to appeals of adverse determinations (20 business days from filing of the appeal), and whether to provide expedited processing of requests (10 business days from receipt of the request). Congress did not establish a statutory deadline for making releasable records available, but instead required agencies to make them available promptly. Although the specific details of processes for handling FOIA requests vary among agencies, the major steps in handling a request are similar across the government. Agencies receive requests, usually in writing (although they may accept requests by telephone or electronically), which can come from any organization or member of the public. Once received, the request goes through several phases, which include initial processing, searching for and retrieving responsive records, preparing responsive records for release, approving the release of the records, and releasing the records to the requester. Figure 1 is an overview of the process, from the receipt of a request to the release of records. During the initial processing phase, a request is logged into the agency’s FOIA system, and a case file is started. The request is then reviewed to determine its scope, estimate fees, and provide an initial response to the requester (in general, this simply acknowledges receipt of the request). After this point, the FOIA staff begins its search to retrieve responsive records. This step may include searching for records from multiple locations and program offices. After potentially responsive records are located, the documents are reviewed to ensure that they are within the scope of the request. During the next two phases, the agency ensures that appropriate information is to be released under the provisions of the act. First, the agency reviews the responsive records to make any redactions based on the statutory exemptions. Once the exemption review is complete, the final set of responsive records is turned over to the FOIA office, which calculates appropriate fees, if applicable. Before release, the redacted responsive records are then given a final review, possibly by the agency’s general counsel, and then a response letter is generated, summarizing the agency’s actions regarding the request. Finally, the responsive records are released to the requester. Some requests are relatively simple to process, such as requests for specific pieces of information that the requester sends directly to the appropriate office. Other requests may require more extensive processing, depending on their complexity, the volume of information involved, the need for the agency FOIA office to work with offices that have relevant subject-matter expertise to find and obtain information, the need for a FOIA officer to review and redact information in the responsive material, the need to communicate with the requester about the scope of the request, and the need to communicate with the requester about the fees that will be charged for fulfilling the request (or whether fees will be waived). Specific details of agency processes for handling requests vary, depending on the agency’s organizational structure and the complexity of the requests received. While some agencies centralize processing in one main office, other agencies have separate FOIA offices for each agency component and field office. Agencies also vary in how they allow requests to be made. Depending on the agency, requesters can submit requests by telephone, fax, letter, or e-mail or through the Web. In addition, agencies may process requests in two ways, known as “multitrack” and “single track.” Multitrack processing involves dividing requests into two groups: (1) simple requests requiring relatively minimal review, which are placed in one processing track, and (2) more voluminous and complex requests, which are placed in another track. In contrast, single-track processing does not distinguish between simple and complex requests. With single-track processing, agencies process all requests on a first-in/first-out basis. Agencies can also process FOIA requests on an expedited basis when a requester has shown a compelling need or urgency for the information. As agencies process FOIA requests, they generally place them in one of four possible disposition categories: grants, partial grants, denials, and “not disclosed for other reasons.” These categories are defined as follows: ● Grants: Agency decisions to disclose all requested records in full. ● Partial grants: Agency decisions to withhold some records in whole or in part, because such information was determined to fall within one or more exemptions. ● Denials: Agency decisions not to release any part of the requested records because all information in the records is determined to be exempt under one or more statutory exemptions. ● Not disclosed for other reasons: Agency decisions not to release requested information for any of a variety of reasons other than statutory exemptions from disclosing records. The categories and definitions of these “other” reasons for nondisclosure are shown in table 1. When a FOIA request is denied in full or in part, or the requested records are not disclosed for other reasons, the requester is entitled to be told the reason for the denial, to appeal the denial, and to challenge it in court. In addition to FOIA, the Privacy Act of 1974 includes provisions granting individuals the right to gain access to and correct information about themselves held by federal agencies. Thus the Privacy Act serves as a second major legal basis, in addition to FOIA, for the public to use in obtaining government information. The Privacy Act also places limitations on agencies’ collection, disclosure, and use of personal information. Although the two laws differ in scope, procedures in both FOIA and the Privacy Act permit individuals to seek access to records about themselves—known as “first-party” access. Depending on the individual circumstances, one law may allow broader access or more extensive procedural rights than the other, or access may be denied under one act and allowed under the other. Consequently, the Department of Justice’s Office of Information and Privacy issued guidance that it is “good policy for agencies to treat all first-party access requests as FOIA requests (as well as possibly Privacy Act requests), regardless of whether the FOIA is cited in a requester’s letter.” This guidance was intended to help ensure that requesters receive the fullest possible response to their inquiries, regardless of which law they cite. In addition, Justice guidance for the annual FOIA report directs agencies to include Privacy Act requests (that is, first-party requests) in the statistics reported. According to the guidance, “A Privacy Act request is a request for records concerning oneself; such requests are also treated as FOIA requests. (All requests for access to records, regardless of which law is cited by the requester, are included in this report.)” Although FOIA and the Privacy Act can both apply to first-party requests, such requests are not in many cases processed in the same way as described earlier for FOIA requests. For example, most SSA first-party requests are processed by staff other than FOIA staff, specifically, staff in SSA’s field and district offices and teleservice centers. OMB and the Department of Justice both have roles in the implementation of FOIA. The act requires OMB to issue guidelines to “provide for a uniform schedule of fees for all agencies.” OMB issued this guidance in April 1987. The Department of Justice oversees agencies’ compliance with FOIA and is the primary source of policy guidance for agencies. Specifically, Justice’s requirements under the act are to ● make agencies’ annual FOIA reports available through a single electronic access point and notify Congress as to their availability; in consultation with OMB, develop guidelines for the required annual agency reports, so that all reports use common terminology and follow a similar format; and ● submit an annual report on FOIA statistics and the efforts undertaken by Justice to encourage agency compliance. Within the Department of Justice, the Office of Information and Privacy has lead responsibility for providing guidance and support to federal agencies on FOIA issues. This office first issued guidelines for agency preparation and submission of annual reports in the spring of 1997. It also periodically issues additional guidance on annual reports as well as on compliance, provides training, and maintains a counselors service to provide expert, one-on-one assistance to agency FOIA staff. Further, the Office of Information and Privacy also makes a variety of FOIA and Privacy Act resources available to agencies and the public via the Justice Web site and on- line bulletins. In 1996, the Congress amended FOIA to provide for public access to information in an electronic format (among other purposes). These amendments, referred to as e-FOIA, also required that agencies submit a report to the Attorney General on or before February 1 of each year that covers the preceding fiscal year and includes information about agencies’ FOIA operations. The following are examples of information that is to be included in these reports: ● number of requests received, processed, and pending; ● median number of days taken by the agency to process different ● determinations made by the agency not to disclose information and the reasons for not disclosing the information; ● disposition of administrative appeals by requesters; information on the costs associated with handling of FOIA requests; and ● full-time-equivalent staffing information. In addition to providing their annual reports to the Attorney General, agencies are to make them available to the public in electronic form. The Attorney General is required to make all agency reports available on line at a single electronic access point and report to Congress no later than April 1 of each year that these reports are available in electronic form. In 2001, we prepared the first in a series of reports on the implementation of the 1996 amendments to FOIA, starting from fiscal year 1999. In this and subsequent reviews, we examined the contents of these annual reports for 25 major agencies (shown in table 2). They include the 24 major agencies covered by the Chief Financial Officers Act, as well as the Central Intelligence Agency and, until 2003, the Federal Emergency Management Agency (FEMA). In 2003, the creation of the Department of Homeland Security (DHS), which incorporated FEMA, led to a shift in some FOIA requests from agencies affected by the creation of the new department, but the same major component entities are reflected in all the years reviewed. The annual FOIA reports for fiscal year 2005 show that many of the trends of previous years are continuing: Requests received and processed continue to rise; however, excepting one case—SSA—the rate of increase has flattened in recent years. We present SSA’s statistics separately because the agency reported an additional 16 million requests in 2005, dwarfing those for all other agencies combined, which together total about 2.6 million. SSA attributed this rise to an improvement in its method of counting requests. Justice officials have raised questions about the inclusion of these numbers in FOIA statistics. Figure 2 shows total requests reported governmentwide for fiscal years 2002 through 2005, with SSA’s share shown separately. This figure shows the magnitude of SSA’s contribution to the whole FOIA picture, as well as the scale of the jump from 2004 to 2005. Figure 3 presents these statistics on a scale that allows a clearer view of the rate of increase in FOIA requests received and processed in the rest of the government. As this figure shows, when SSA’s numbers are excluded, the rate of increase is modest and has been flattening: from fiscal year 2002 to 2005, requests received increased by about 27 percent, and requests processed increased by about 25 percent. From fiscal year 2004 to 2005, requests received increased about 2.5 percent, and requests processed increased about 2.0 percent. According to SSA, the increases that the agency reported in fiscal year 2005 can be attributed to an improvement in its method of counting a category of requests it calls “simple requests handled by non-FOIA staff.” In the past 4 years, SSA’s FOIA reports have consistently shown significant growth in this category, which has accounted for the major portion of all SSA requests reported (see table 3). In each of these years, SSA has attributed the increases in this category largely to better reporting, as well as actual increases in requests. SSA describes requests in this category as typically being requests by individuals for access to their own records, or else requests in which individuals consent for SSA to supply information about themselves to third parties (such as insurance and mortgage companies) so that they can receive housing assistance, mortgages, disability insurance, and so on. According to SSA’s FOIA report, these requests are handled by personnel in about 1,500 locations in SSA, including field and district offices and teleservice centers. Such requests are almost always granted, according to SSA, and most receive immediate responses. SSA has stated that it does not keep processing statistics (such as median days to process) on these requests, which it reports separately from other FOIA requests (for which processing statistics are kept). According to SSA, in fiscal year 2005, the agency began to use automated systems to capture the numbers of requests processed by non-FOIA staff, generating statistics automatically as requests were processed; the result, according to SSA, is a much more accurate count. However, Justice officials have suggested that SSA consider treating the bulk of these requests as non-FOIA requests and thus not include them in future reports. Besides SSA, agencies reporting large numbers of requests received were the Departments of Agriculture, Defense, Health and Human Services, Homeland Security, Justice, the Treasury, and Veterans Affairs, as shown in table 4. The rest of the agencies combined account for only about 3 percent of the total requests received (if SSA is excluded). Table 4 presents, in descending order of request totals, the numbers of requests received and percentages of the total (calculated with and without SSA’s statistics). Most FOIA requests in 2005 were granted in full, with relatively few being partially granted, denied, or not disclosed for other reasons (statistics are shown in table 5). This generalization holds with or without SSA’s inclusion. However, including SSA’s numbers in the proportion of grants overwhelms the other categories—raising this number from 87 percent of the total to 98 percent. This is to be expected, since SSA reports that it grants the great majority of its simple requests handled by non-FOIA staff, which make up the bulk of SSA’s statistics. Four of the eight agencies that handled the largest numbers of requests (HHS, SSA, USDA, and VA; see table 4) also granted the largest percentages of requests in full, as shown in figure 4. This figure shows, by agency, the disposition of requests processed: that is, whether granted in full, partially granted, denied, or not disclosed for the “other” reasons shown in table 1. As the figure shows, the numbers of fully granted requests varied widely among agencies in fiscal year 2005. Seven agencies made full grants of requested records in over 80 percent of the cases they processed (besides the four already mentioned, these include DOE, OPM, and SBA). In contrast, 13 of 25 made full grants of requested records in less than 40 percent of their cases, including 3 agencies (CIA, NSF, and State) that made full grants in less than 20 percent of cases processed. This variance among agencies in the disposition of requests has been evident in prior years as well. In many cases, the variance can be accounted for by the types of requests that different agencies process. For example, as discussed earlier, SSA grants a very high proportion of requests because they are requests for personal information about individuals that are routinely made available to or for the individuals concerned. Similarly, VA routinely makes medical records available to individual veterans. For 2005, the reported time required to process requests (by track) varied considerably among agencies. Table 6 presents data on median processing times for fiscal year 2005. For agencies that reported processing times by component rather than for the agency as a whole, the table indicates the range of median times reported by the agency’s components. As the table shows, eight agencies had components that reported processing simple requests in less than 10 days (these components are part of the CIA, Energy, the Interior, Justice, Labor, Transportation, the Treasury, and USDA); for each of these agencies, the lower value of the reported ranges is less than 10. On the other hand, median time to process simple requests is relatively long at some organizations (for example, components of HHS, Justice, and USDA, as shown by median ranges whose upper end values are greater than 100 days). For complex requests, the picture is similarly mixed. Components of four agencies (EPA, DHS, the Treasury, and VA) reported processing complex requests quickly—with a median of less than 10 days. In contrast, other components of several agencies (DHS, Energy, EPA, HHS, HUD, Justice, State, Transportation, the Treasury, and USDA) reported relatively long median times to process complex requests, with median days greater than 100. Six agencies (AID, HHS, NSF, OPM, SBA, and SSA) reported using single-track processing. The median processing times for single- track processing varied from 5 days (at an HHS component) to 173 days (at another HHS component). The changes from fiscal year 2004 to 2005 also vary. For agencies that reported agencywide figures, table 7 shows how many showed increased or decreased median processing times. Table 8 shows these numbers for the components that were reported separately. These tables show that no one pattern emerges across tracks and types of reporting, and the numbers of agencies and components involved vary from track to track. The picture that emerges is of great variation according to circumstances. To allow more insight into the variations in median processing times, we provide in attachment 2 tables of median processing times as reported by agencies and components in the annual FOIA reports in fiscal years 2004 and 2005. This attachment also includes information on the number of requests reported by the agencies and components, which provides context for assessing the median times reported. In addition to processing greater numbers of requests, many agencies (11 of 25) also reported that their numbers of pending cases—requests carried over from one year to the next—have increased since 2002. In 2002, pending requests governmentwide were reported to number about 140,000, whereas in 2005, about 200,000—43 percent more—were reported. In addition, the rate of increase grew in fiscal year 2005, rising 24 percent from fiscal year 2004, compared to 11 percent from 2003 to 2004. Figure 5 shows these results, illustrating the accelerating rate at which pending cases have been increasing. These statistics include pending cases reported by SSA, because as the figure shows, these pending cases do not change the governmentwide picture significantly. As previously discussed, SSA’s pending cases do not include simple requests handled by non- FOIA staff, for which SSA does not keep these statistics. Trends for individual agencies show mixed progress in reducing the number of pending requests reported from 2002 to 2005—some agencies have decreased numbers of pending cases, while others’ numbers have increased. Figure 6 shows processing rates at the 25 agencies (that is, the number of requests that an agency processes relative to the number it receives). Eight of the 25 agencies (AID, DHS, the Interior, Education, HHS, HUD, NSF, and OPM) reported processing fewer requests than they received each year for fiscal years 2003, 2004, and 2005; 8 additional agencies processed less than they received in two of these three years. In contrast, two agencies (CIA and DOE) had processing rates above 100 percent in all three years, meaning that each made continued progress in reducing their numbers of pending cases. Fifteen additional agencies were able to make at least a small reduction in their numbers of pending requests in 1 or more years between fiscal years 2003 and 2005. The Executive Order, with its requirement for agencies to develop FOIA improvement plans, serves to focus agency managers’ attention on the important role that FOIA plays in keeping citizens well informed about the operations of their government. By requiring measurable goals and timetables, the Executive Order provides for a results-oriented framework by which agency heads can hold officials accountable for improvements in FOIA processing. Further, the Department of Justice’s guidance on implementing the order provides several tangible suggestions for improving FOIA operations. The Executive Order states that each agency shall develop an improvement plan by June 14, 2006, that includes measurable, outcome-oriented goals to reduce or eliminate backlog, along with timetables that include milestones for these goals. According to this guidance, the goals and milestones in agency plans should focus on outcomes that are measurable and demonstrate whether or not intended results are being achieved. Justice’s implementation guidance directs agencies to include “means of measurement of success (e.g., quantitative assessment of backlog reduction expressed in numbers of pending requests, percentages, or working days)” and provides agencies considerable leeway in choosing measures of timeliness. Most of the 22 agency plans available as of June 30 discussed reducing backlog, but not all consistently followed the Executive Order directions by establishing goals and timetables for reducing or eliminating their backlog. In all, 12 of the 21 agencies that reported a backlog included such goals and timetables, but the remaining 9 did not do so. (The Small Business Administration did not report a backlog.) These 9 agencies accounted for about 29 percent of the almost 200,000 requests pending at the end of fiscal year 2005 that were reported in the annual FOIA reports. Table 9 summarizes the results of our analysis. As table 9 shows, 13 agencies included goals, but 1 of these (SSA) did not include a timetable associated with its goal. The goals chosen by the 13 agencies varied. For example, OPM’s plan set a goal of reducing and eliminating the agency’s backlog by December 31, 2006. EPA’s goal was to reduce its response backlog to less than 10 percent of the number of new FOIA requests received each year. Several agencies set goals to reduce backlog to various percentages of their current backlog (for example, the CIA, Energy, the Interior, Justice, and the Treasury). HUD set an absolute goal of fewer than 400 pending requests. Although the remaining 8 agency plans discussed efforts to improve FOIA processing, they did not contain goals for backlog reduction. In two cases (GSA and NASA), agencies did not include such goals because they did not include backlog reduction among the areas of improvement on which they planned to focus. These agencies did not consider their backlogs significant; nonetheless, the Executive Order specifically instructed agencies to include goals and timetables to address backlog. In other cases, agencies did address backlog reduction but did not define goals. Many of these agencies did define process goals, such as establishing means to monitor and report on backlog, reviewing current processes, and identifying and reviewing tracking systems, but these were not accompanied by goals for backlog reduction: ● For example, the Department of Commerce’s plan stated that, to the extent possible, its components would use current backlog numbers as a ceiling (these generally range from 9 to 13 percent of the workload) and work aggressively to reduce these numbers, focusing particularly on the 10 oldest requests in each component’s backlog. However, although the plan provided milestone dates for FOIA officers to review progress in this area and assess any need to pursue alternatives (such as contract support) for achieving goals, the plan did not provide measurable targets for assessing success, such as percentage of reduction. ● Similarly, the Department of Defense set various process goals (identifying those FOIA Offices with backlogs greater than 50 cases, determining the staffing levels required to significantly reduce the backlog, and seeking the necessary funding to provide this additional staffing). However, it provided no measurable targets for reducing backlog. In the timetables that agencies provided in their plans, 12 agencies provided milestones for goals that they had identified. As mentioned earlier, one agency (SSA) did not include a milestone for its goal of eliminating backlog. SSA provided instead a timetable that addressed process goals: reorganizing its Office of Public Disclosure and developing a new information system. Like SSA, several agencies provided timetables for various activities that they included in their plans to reduce backlog, but these did not include milestones for outcome-oriented goals (for example, Defense provided milestones for the process goals described above). In addition to setting goals and milestones for those goals, in order to demonstrate that goals are achieved, plans should also include baselines against which results can be measured. In the case of backlog, these numbers can differ from day to day, so specifying a baseline is crucial. Baselines can be defined on the basis of a date from which an agency intends to measure, the number it is using as its baseline, or both. Publicly available baselines are important to promote accountability as well as the transparency of government processes. However, most of the agency improvement plans do not clearly define baselines for their existing backlogs. An exception was OPM: in describing its goal to eliminate backlog by December 31, 2006, it specified its present backlog as 107 requests. Similarly, the Department of Education referred to measuring its success in terms of having fewer open cases at the end of each year, based on a backlog that it specified as 480 as of June 2, 2006. In other cases, agencies did not specify whether they planned to measure from the date of their plans, from the end of fiscal year 2005, or from some other baseline. Some agencies did, however, describe plans to perform analyses that would measure their backlogs so that they could then establish the necessary baselines. Our ongoing work suggests that factors contributing to these deficiencies included difficulties in coordinating responses among components in large decentralized agencies and limitations in the way that agency systems track FOIA processing. In addition, neither the Executive Order nor Justice guidance established a baseline date for measuring the backlog or directed agencies to establish such a baseline. Uncertainty regarding defined baselines could hinder the measurement of progress in reducing backlog. Without clearly defined baselines, specific objectives, and timetables for reducing backlog, the risk is that agency heads, Justice, the Congress, and the public could be hampered in determining whether and how well agencies have achieved the Executive Order’s aims of improving FOIA processing and agency disclosure of information. When we complete our ongoing review and analysis, we expect to make recommendations aimed at improving agency implementation of the Executive Order, including efforts to reduce and eliminate backlog. In summary, FOIA continues to be a valuable tool for citizens to obtain information about the operation and decisions of the federal government. Since 2002, agencies have received increasing numbers of requests and have also continued to increase the number of requests that they process. In addition, agencies continue to grant most requests in full. However, the rate of increase in pending requests is accelerating. Given these continuing trends, the President’s Executive Order creates, among other things, a renewed, results-oriented emphasis on improving request processing and reducing the backlog of pending requests. However, our ongoing work suggests that agencies are not yet fully complying with the order’s requirements for measurable, outcome-oriented goals and associated timetables. In addition, agencies have not all established clear baselines for their existing backlogs. Without a baseline measurement and tangible steps for addressing the accumulation of pending cases, the heads of these agencies may be limited in their ability to measure and evaluate success in implementing their plans as the President’s order requires. Accordingly, in moving forward, it will be important for Justice and the agencies to continue to refine these plans so that the goal of reducing backlogs can be fully realized and the federal government can remain responsive to citizen needs. When we complete our ongoing work, we expect to provide recommendations to help move this process forward. Mr. Chairman, this completes my prepared statement. I would be happy to respond to any questions you or other Members of the Subcommittee may have at this time. If you should have questions about this testimony, please contact me at (202) 512-6240 or via e-mail at koontzl@gao.gov. Other major contributors included Barbara Collier, Vernetta Marquis, Alan Stapleton, Shawn Ward, and Elizabeth Zhao. Exemption number Matters that are exempt from FOIA (A) Specifically authorized under criteria established by an Executive Order to be kept secret in the interest of national defense or foreign policy and (B) are in fact properly classified pursuant to such Executive Order. Related solely to the internal personnel rules and practices of an agency. Specifically exempted from disclosure by statute (other than section 552b of this title), provided that such statute (A) requires that matters be withheld from the public in such a manner as to leave no discretion on the issue, or (B) establishes particular criteria for withholding or refers to particular types of matters to be withheld. Trade secrets and commercial or financial information obtained from a person and privileged or confidential. Inter-agency or intra-agency memorandums or letters which would not be available by law to a party other than an agency in litigation with the agency. Personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy. Records or information compiled for law enforcement purposes, but only to the extent that the production of such law enforcement records or information could reasonably be expected to interfere with enforcement proceedings; would deprive a person of a right to a fair trial or impartial adjudication; could reasonably be expected to constitute an unwarranted invasion of personal privacy; could reasonably be expected to disclose the identity of a confidential source, including a State, local, or foreign agency or authority or any private institution which furnished information on a confidential basis, and, in the case of a record or information compiled by a criminal law enforcement authority in the course of a criminal investigation or by an agency conducting a lawful national security intelligence investigation, information furnished by confidential source; would disclose techniques and procedures for law enforcement investigations or prosecutions, or would disclose guidelines for law enforcement investigations or prosecutions if such disclosure could reasonably be expected to risk circumvention of the law; or could reasonably be expected to endanger the life or physical safety of an individual. Contained in or related to examination, operating, or condition reports prepared by, on behalf of, or for the use of an agency responsible for the regulation of supervision of financial institutions. Geological and geophysical information and data, including maps, concerning wells. The attached tables present median processing times as reported by agencies in their annual FOIA reports in fiscal years 2004 and 2005. To provide context, we include numbers of requests processed for each agency or component. We also indicate (in columns headed “±”) whether the median days to process rose (+), fell (–), or remained unchanged (=). (We also use “~” to indicate other types of changes, such as the establishment of a new component.) Agencies report median processing times according to processing tracks: that is, some agencies divide requests into simple and complex categories and process these in separate tracks, whereas others use a single track. Accordingly, the tables show these tracks where applicable. In addition, agencies are required to subject some requests to expedited processing, and these are reported as a separate track. No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2004 2005 2004 2005 ± ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. (a) n/a ~ (a) (a) (a) (a) 14 (a) (a) 48 (a) 28 (a) 61 + 30 ~ 16 +178 + 210 ~ 21 = 149 + (a) 45 + n/a ~ 2 –105,567 85,307 14 (a) 16 (a) 45 + 27,850 19,532 17 ~ (a) 31 (a) (a) (a) 17 – 16 + 35 – n/a n/a ~ 12 – 13 – 242 ~ 3 – ~ other change (change in reporting, new component, etc.) Component did not exist. No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2004 2005 2004 2005 ± 12 – ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2005 16 – 12,972 11,385 2005 2004 2005 ± – ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. (a) (a) (a) (a) (a) No. No. ~ other change (change in reporting, new component, etc.) Component did not exist. No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 6,206 (a) 30 (a) (a) 63 (a) 25 2–64 (a)64 (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) ~ other change (change in reporting, new component, etc.) Statistics not broken down by component. No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 89 + 19 + 16 + 8 + 10 + 8 = (b) (b) (c) (c) No. No. 15 – 22 137 139 + (a) (a) (a) (a) (a) 3.5 + 0 = 26 + ~ other change (change in reporting, new component, etc.) Component did not exist. No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. No. No. ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2005 2004 2005 ± 35 ~ ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2005 2004 2005 ± 19 = ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2004 2005 2004 2005 ± ~ other change (change in reporting, new component, etc.) Two tables are provided for this department, because its components report both multitrack (simple and complex) processing and single-track processing. No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. (a) (a) (a) No. 4 2005 ± 69 + (a) (a) (a) No. 380 Indian Health Services 158,277 151,428 National Institutes of Health 10,583 ~ other change (change in reporting, new component, etc.) Component did not exist. No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. ± 160 – 35 + 2004 2005 ± 22 – 70 + 2004 2005 ± 65 – 21 = ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2004 2005 2004 2005 ± 19 + 2005 2004 2005 ± 15 – ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2004 2005 2004 2005 ± 12 + 2005 2004 2005 ± 20 – ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. 2004 2005 ± 14 – ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. 2005 2004 2005 ± 14 + ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. 2004 2005 2004 2005 ± 14 + 2005 2004 2005 ± 136 – ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. ~ other change (change in reporting, new component, etc.) No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. + increase ~ other change (change in reporting, new component, etc.) The department reports all processing in one track, but it refers to this track as complex, rather than single track. No. = number of requests processed; Days = median days to process; ± = change from 2004 to 2005 No. No. (a) (a) No. No. (a) ~ 60 + ~ other change (change in reporting, new component, etc.) Component did not exist. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The Freedom of Information Act (FOIA) establishes that federal agencies must provide the public with access to government information, thus enabling them to learn about government operations and decisions. To help ensure appropriate implementation, the act requires that agencies report annually to the Attorney General, providing specific information about their FOIA operations. In addition, a recent Executive Order directs agencies to develop plans to improve their FOIA operations, including, among other things, goals to reduce backlogs in FOIA requests. GAO has reported previously on the contents of these annual reports for 25 major agencies. For this hearing, GAO was asked to testify both on the annual reports for fiscal year 2005 and on the recently developed improvement plans for these 25 agencies. GAO based its testimony on its ongoing work on these topics. Upon completion of its ongoing review, GAO expects to make recommendations to improve agency implementation of the Executive Order, including efforts to reduce and eliminate backlog. According to data reported by agencies in their annual reports, the public continues to request and receive increasing amounts of information from the federal government through FOIA; however, excepting one case--the Social Security Administration (SSA)--the rate of increase has flattened in recent years. (SSA reported an additional 16 million requests in 2005, dwarfing those for all other agencies combined, which together total about 2.6 million; SSA attributed this rise to an improvement in its method of counting requests. However, Justice officials have suggested that SSA consider treating the bulk of these requests as non-FOIA requests and thus not include them in future reports.) When SSA's numbers are excluded, data reported by the other 24 major agencies show that the number of requests received increased by 27 percent from fiscal year 2002 to 2005, but by only about 2.5 percent from fiscal year 2004. As more requests come in, agencies also report that they have been processing more of them--25 percent more from 2002 to 2005 (but only about 2.0 percent more than from 2004). Despite processing more requests, agencies have not kept up with the increase in requests being made: the number of pending requests carried over from year to year has been steadily increasing, rising to about 200,000 in fiscal year 2005--43 percent more than in 2002. The rate of increase in requests pending is also growing: the increase from fiscal year 2004 to 2005 is 24 percent, compared to 11 percent from 2003 to 2004. Most of the agency improvement plans discussed reducing backlog, but not all consistently followed the Executive Order or implementing guidance provided by the Justice Department. Of the 25 agencies, 3 had not posted their plans in time to be included in this testimony, and 1 reported no backlog. Of the remaining 21 agencies, 12 followed the Executive Order's instruction to establish measurable, outcome-oriented objectives for reducing or eliminating their backlogs, as well as timetables with milestones for meeting these objectives. Nine agencies did not do this, although they accounted for a substantial fraction--about 29 percent--of the requests reported to be pending at the end of fiscal year 2005. (Most agencies did provide goals and timetables for other kinds of objectives, however, such as performing staffing analyses and reviewing progress.) In addition, agencies generally did not specify the dates or numbers they were using as the baselines for their existing backlogs, which will be important for measuring improvement. GAO's ongoing work suggests that factors contributing to these deficiencies include difficulties in coordinating responses among components in large, decentralized agencies and limitations in the systems that track FOIA processing. In addition, neither the Executive Order nor Justice guidance established a baseline date for measuring the backlog or directed agencies to establish such a date. Without clearly defined baselines, specific objectives, and timetables for reducing backlog, it could be challenging for agency heads, Justice, and the Congress to gauge progress in improving FOIA processes as intended by the Executive Order. |
The 75 percent rule was established in 1983 to distinguish IRFs from other facilities for payment purposes. According to CMS, the conditions on the list in the rule at that time accounted for 75 percent of the admissions to IRFs. In June 2002 CMS suspended the enforcement of the 75 percent rule after its study of the fiscal intermediaries revealed that they were using inconsistent methods to determine whether an IRF was in compliance and that in some cases IRFs were not being reviewed for compliance at all. CMS standardized the verification process that the fiscal intermediaries were to use, and issued a rule—effective July 1, 2004—that increased the number of conditions from 10 to 13 and provided a 3-year transition period, ending in July 2007, to phase in the 75 percent threshold. The current payment and review procedures for IRFs were established in recent years. The inpatient rehabilitation facility prospective payment system (IRF PPS) was implemented in January 2002. Payment is contingent on an IRF’s completing the IRF-PAI after admission and transmitting the resulting data to CMS. Two basic requirements must be met if inpatient hospital stays for rehabilitation services are to be covered: (1) the services must be reasonable and necessary, and (2) it must be reasonable and necessary to furnish the care on an inpatient hospital basis, rather than in a less intensive facility, such as a SNF, or on an outpatient basis. Determinations of whether hospital stays for rehabilitation services are reasonable and necessary must be based on an assessment of each beneficiary’s individual care needs. Beginning in April 2002, the fiscal intermediaries, the entities that conduct compliance reviews, were specifically authorized to conduct reviews for medical necessity to determine whether an individual admission to an IRF was covered under Medicare. As we reported in April 2005, among the 506,662 Medicare patients admitted to an IRF in fiscal year 2003, less than 44 percent were admitted with a primary condition on the list in the 75 percent rule. About another 18 percent of IRF Medicare patients were admitted with a comorbid condition that was on the list in the rule. Among the 194,922 IRF Medicare patients that did not have a primary or comorbid condition on the list in the rule, almost half were admitted for orthopedic conditions, and among those the largest group was joint replacement patients whose condition did not meet the list’s specific criteria. (See figure 1.) Although some joint replacement patients may need admission to an IRF, such as those with comorbidities that affect the patient’s function, our analysis showed that few of these patients had comorbidities that suggested a possible need for the level of services offered by an IRF. Our analysis found that 87 percent of joint replacement patients admitted to IRFs in fiscal year 2003 did not meet the criteria of the rule, and among those, over 84 percent did not have any comorbidities that would have affected the costs of their care based on our analysis of the payment data. Because the data we analyzed were from 2003, when enforcement of the rule was suspended, we also looked at newly released data from July through December 2004, after enforcement had resumed, to determine whether admission patterns had changed. We focused on the largest category of patients admitted to IRFs, joint replacement patients, and found no material change in the admission of joint replacement patients for the same time periods in 2003 and 2004. Across all IRFs, the percentage of Medicare patients admitted for a joint replacement declined by 0.1 percentage point. In conjunction with our finding on the number of patients admitted to IRFs for conditions not on the list in the rule, we determined that only 6 percent of IRFs in fiscal year 2003 were able to meet a 75 percent threshold. Many IRFs were able to meet the lower thresholds that would be in place early in the transition period, but progressively fewer IRFs were able to meet the higher threshold levels. As we stated in our report, the criteria IRFs used to assess patients for admission varied by facility and included patient characteristics in addition to condition. All the IRF officials we interviewed evaluated a patient’s function when assessing whether a patient needed the level of services of an IRF. Whereas some IRF officials reported that they used function to characterize patients who were appropriate for admission (e.g., patients with a potential for functional improvement), others said they used function to characterize patients not appropriate for admission (e.g., patients whose functional level was too high, indicating that they could go home, or too low, indicating that they needed to be in a SNF). Almost half of the IRF officials interviewed stated that function was the main factor that should be considered in assessing the need for IRF services. IRF officials reported to us that they did not admit all the patients they assessed. Typically, the IRF received a request from a physician in the acute care hospital requesting a medical consultation from an IRF physician, or from a hospital discharge planner or social worker indicating that they had a potential patient. An IRF staff member—usually a physician and/or a nurse—conducted an assessment prior to admission to determine whether to admit a patient. CMS, working through its fiscal intermediaries, has not routinely reviewed IRF admission decisions, although it reported that such reviews could be used to target problem areas. Among the 10 fiscal intermediary officials we interviewed, over half were not conducting reviews of patients admitted to IRFs. We concluded that the presence of patients in IRFs who may not need the intense level of services provided by IRFs called for increased scrutiny of IRF admissions, which could target problem areas and vulnerabilities and thereby reduce the number of inappropriate admissions in the future. We recommended that CMS ensure that its fiscal intermediaries routinely conduct targeted reviews for medical necessity for IRF admissions. CMS agreed that targeted reviews are necessary and said that it expected its contractors to direct their resources toward areas of risk. It also reported that it has expanded its efforts to provide greater oversight of IRF admissions through local policies that have been implemented or are being developed by the fiscal intermediaries. As we reported, the experts IOM convened and other experts we interviewed differed on whether conditions should be added to the list in the 75 percent rule but agreed that condition alone does not provide sufficient criteria to identify types of patients appropriate for IRFs. The experts IOM convened generally questioned the strength of the evidence for adding conditions to the list in the rule. They reported that the evidence on the benefits of IRF services is variable, particularly for certain orthopedic conditions, and some of them reported that little information was available on the need for inpatient rehabilitation for cardiac, transplant, pulmonary, or oncology conditions. In general, they reported that, except for a few subpopulations, uncomplicated, unilateral joint replacement patients rarely need to be admitted to an IRF. Most of them called for further research to identify the types of patients that need inpatient rehabilitation and to understand the effectiveness of IRFs in comparison with other settings of care. IRF officials we interviewed did not agree on whether conditions, including a broader category of joint replacements, should be added to the list in the rule. Half of them suggested that joint replacement be more broadly defined to include more patients saying, for example, that the current requirements were too restrictive and arbitrary. Others said that unilateral joint replacement patients were not generally appropriate for IRFs. We recommended that CMS conduct additional activities to encourage research on the effectiveness of intensive inpatient rehabilitation and factors that predict patient need for these services. CMS agreed and said that it has expanded its activities to guide future research efforts by encouraging government research organizations, academic institutions, and the rehabilitation industry to conduct both general and targeted research, and plans to collaborate with the National Institutes of Health to determine how to best promote research. There was general agreement among all the groups of experts we interviewed, including the experts IOM convened, that condition alone is insufficient for identifying appropriate types of patients for inpatient rehabilitation, because not all patients with a condition on the list need to be in an IRF. For example, stroke is on the list, but not all stroke patients need to go to an IRF after their hospitalization. Similarly, cardiac condition is not on the list, but some cardiac patients may need to be admitted to an IRF. Among the experts convened by IOM, functional status was identified most frequently as the information required in addition to condition. Half of them commented on the need to add information about functional status, such as functional need, functional decline, motor and cognitive function, and functional disability. However, some of the experts convened by IOM recognized the challenge of operationalizing a measure of function, and some experts questioned the ability of the current assessment tools to predict which types of patients will improve if treated in an IRF. We concluded that if condition alone is not sufficient for determining which types of patients are most appropriate for IRFs, more conditions should not be added to the list at the present time, and that future efforts should refine the rule to increase its clarity about which types of patients are most appropriate for IRFs. We recommended that CMS use the information obtained from reviews for medical necessity, research activities, and other sources to refine the rule to describe more thoroughly the subgroups of patients within a condition that require IRF services, possibly using functional status or other factors, in addition to condition. CMS stated that while it expected to follow our recommendation, it would need to give this action careful consideration because it could result in a more restrictive policy than the present regulations, and noted that future research could guide the agency’s description of subgroups. As we stated in our report, we believe that action to conduct reviews for medical necessity and to produce more information about the effectiveness of inpatient rehabilitation could support future efforts to refine the rule over time to increase its clarity about which types of patients are most appropriate for IRFs. These actions could help to ensure that Medicare does not pay IRFs for patients who could be treated in a less intensive setting and does not misclassify facilities for payment. Madam Chairman, this concludes my prepared statement. I would be happy to respond to any questions you or other Members of the Subcommittee may have at this time. For further information about this testimony, please contact Marjorie Kanof at (202) 512-7114. Linda Kohn and Roseanne Price also made key contributions to this statement. A facility may be classified as an IRF if it can show that, during a 12-month period at least 75 percent of all its patients, including its Medicare patients, required intensive rehabilitation services for the treatment of one or more of the following conditions:1. Stroke. 2. Spinal cord injury. 3. Congenital deformity. 4. Amputation. 5. Major multiple trauma. 6. Fracture of femur (hip fracture). 7. Brain injury. 8. Neurological disorders (including multiple sclerosis, motor neuron diseases, polyneuropathy, muscular dystrophy, and Parkinson’s disease). 9. Burns. 10. Active, polyarticular rheumatoid arthritis, psoriatic arthritis, and seronegative arthropathies resulting in significant functional impairment of ambulation and other activities of daily living that have not improved after an appropriate, aggressive, and sustained course of outpatient therapy services or services in other less intensive rehabilitation settings immediately preceding the inpatient rehabilitation admission or that result from a systemic disease activation immediately before admission, but have the potential to improve with more intensive rehabilitation. 11. Systemic vasculidities with joint inflammation, resulting in significant functional impairment of ambulation and other activities of daily living that have not improved after an appropriate, aggressive, and sustained course of outpatient therapy services or services in other less intensive rehabilitation settings immediately preceding the inpatient rehabilitation admission or that result from a systemic disease activation immediately before admission, but have the potential to improve with more intensive rehabilitation. 12. Severe or advanced osteoarthritis (osteoarthritis or degenerative joint disease) involving two or more major weight bearing joints (elbow, shoulders, hips, or knees, but not counting a joint with a prosthesis) with joint deformity and substantial loss of range of motion, atrophy of muscles surrounding the joint, significant functional impairment of ambulation and other activities of daily living that have not improved after the patient has participated in an appropriate, aggressive, and sustained course of outpatient therapy services or services in other less intensive rehabilitation settings immediately preceding the inpatient rehabilitation admission but have the potential to improve with more intensive rehabilitation. (A joint replaced by a prosthesis no longer is considered to have osteoarthritis, or other arthritis, even though this condition was the reason for the joint replacement.) 13. Knee or hip joint replacement, or both, during an acute hospitalization immediately preceding the inpatient rehabilitation stay and also meet one or more of the following specific criteria: a. The patient underwent bilateral knee or bilateral hip joint replacement surgery during the acute hospital admission immediately preceding the IRF admission. b. The patient is extremely obese, with a body mass index of at least 50 at the time of admission to the IRF. c. The patient is age 85 or older at the time of admission to the IRF. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Medicare classifies inpatient rehabilitation facilities (IRF) using the "75 percent rule." If a facility can show that during a 12-month period at least 75 percent of its patients required intensive rehabilitation for 1 of 13 listed conditions, it may be classified as an IRF and paid at a higher rate than for less intensive rehabilitation in other settings. Because this difference can be substantial, it is important to classify IRFs correctly. GAO was asked to discuss issues relating to the classification of IRFs, and in April 2005 it issued a report, Medicare: More Specific Criteria Needed to Classify Inpatient Rehabilitation Facilities (GAO-05-366). For that report, GAO analyzed data on all Medicare patients (the majority of patients in IRFs) admitted to IRFs in fiscal year 2003, spoke to IRF medical directors, and had the Institute of Medicine (IOM) convene a meeting of experts to evaluate the use of a list of conditions in the 75 percent rule. This testimony is based on the April 2005 report. As noted in the April 2005 report, GAO found that in fiscal year 2003 fewer than half of all IRF Medicare patients were admitted for having a primary condition on the list in the 75 percent rule. Almost half of all patients with conditions not on the list were admitted for orthopedic conditions, and among those the largest group was joint replacement patients. The experts IOM convened said that uncomplicated unilateral joint replacement patients rarely need to be admitted to an IRF, and GAO analysis suggested that relatively few of the Medicare unilateral joint replacement patients had comorbid conditions that suggested a possible need for the IRF level of services. Additionally, GAO found that only 6 percent of IRFs in fiscal year 2003 were able to meet a 75 percent threshold. GAO also found that IRFs varied in the criteria used to assess patients for admission, using patient characteristics such as functional status, as well as condition. The Centers for Medicare & Medicaid Services (CMS), working through its fiscal intermediaries, had not routinely reviewed IRF admission decisions to determine whether they were medically justified, although it reported that such reviews could be used to target problem areas. The experts IOM convened and other clinical and nonclinical experts GAO interviewed differed on whether conditions should be added to the list in the 75 percent rule. The experts IOM convened questioned the strength of the evidence for adding conditions to the list--finding the evidence for certain orthopedic conditions particularly weak--and some of them reported that little information was available on the need for inpatient rehabilitation for cardiac, transplant, pulmonary, or oncology patients. They called for further research to identify the types of patients that need inpatient rehabilitation and to understand the effectiveness of IRFs. There was general agreement among all the groups of experts interviewed that condition alone is insufficient for identifying appropriate types of patients for inpatient rehabilitation, since within any condition only a subgroup of patients require the level of services of an IRF, and that functional status should also be considered in addition to condition. GAO concluded that if condition alone is not sufficient for determining which types of patients are most appropriate for IRFs, more conditions should not be added to the list at the present time and the rule should be refined to clarify which types of patients should be in IRFs as opposed to another setting. |
OPM collects and maintains personal information on millions of individuals, including sensitive security clearance data. The agency is the central human resources agency for the federal government, overseeing all policy to support federal agencies’ human resources departments— from classification and qualifications systems to pay, leave, and benefit policies. In addition, the agency provides investigative products and services for more than 100 federal agencies to use as the basis for suitability for employment and security clearance determinations. It also provides more than 95 percent of the government’s background investigations, conducting approximately 2.2 million investigations a year. The agency had a fiscal year 2016 discretionary budget authority of about $245 million and around 5,300 full-time equivalent employees. On June 4, 2015, OPM reported that a breach to its information technology systems and data had potentially compromised the personal information of about 4.2 million former and current federal employees. In July 2015, OPM reported that a separate but related cyber incident targeting its databases containing background investigation records was estimated to have compromised security clearance background information of about 21.5 million individuals. In the months that followed the breach, further investigation by the agency revealed that the information compromised as a result of these breaches included employee Social Security numbers, residency and education history, employment history, information about immediate family and other personal and business acquaintances, criminal and financial history, job assignments, performance ratings, training information, and the fingerprints of approximately 5.6 million individuals. An estimated 22.1 million individuals had some form of PII stolen, with 3.6 million being a victim of both breaches. OPM’s director is responsible for ensuring the adequacy of the agency’s information security program, including information security policies, procedures, and practices. The OPM Chief Information Officer (CIO) leads the development, management, operations, and support of the IT infrastructure, with the assistance of the managers and staff in OCIO. The agency’s Chief Information Security Officer (CISO) serves as the CIO’s primary information security adviser and guides the information security activities of the agency’s authorizing officials and information security officers. OPM also has a Management Review Board, which includes the CISO, Deputy CISO, and certain branch chiefs as members. This board is intended to manage information security risks by reviewing and approving the creation and closure actions associated with plans of action and milestones (POA&M), which address identified weaknesses. The Federal Information Security Modernization Act (FISMA) of 2014 is intended to provide a comprehensive framework for ensuring the effectiveness of information security controls over information resources that support federal operations and assets and for ensuring the effective oversight of information security risks, including those throughout civilian, national security, and law enforcement agencies. FISMA assigns responsibility to the head of each agency for providing information security protections commensurate with the risk and magnitude of the harm resulting from unauthorized access, use, disclosure, disruption, modification, or destruction of information systems used or operated by an agency or by a contractor of an agency or other organization on behalf of an agency. The law also delegates to the agency CIO (or comparable official) the authority to ensure compliance with FISMA requirements. FISMA also requires each agency to develop, document, and implement an agency-wide information security program to provide risk-based protections for the information and information systems that support the operations and assets of the agency, including those provided or managed by another agency, contractor, or other source. Such a program includes assessing risk; developing and implementing cost-effective security plans, policies, and procedures; developing plans for providing adequate information security for networks, facilities, and systems; providing security awareness and specialized training; testing and evaluating the effectiveness of controls; planning, implementing, evaluating, and documenting remedial actions to address information security deficiencies; developing and implementing procedures for detecting, reporting, and responding to security incidents; and ensuring continuity of operations. In addition, FISMA requires agencies to comply with NIST standards and the Office of Management and Budget (OMB) requires agencies to comply with NIST guidelines. Further, FISMA requires the operation of a central federal information security incident center, a role now filled by the DHS’s US-CERT. The mission of US-CERT is to strive for a safer, stronger Internet for all Americans by responding to major incidents, analyzing threats, and exchanging critical cybersecurity information with trusted partners around the world. This mission is reflected in the four critical activities it is responsible for carrying out: Providing cybersecurity protection to federal civilian executive branch agencies through intrusion detection and prevention capabilities. Developing timely and actionable information for distribution to federal departments and agencies; state, local, tribal, and territorial governments; critical infrastructure owners and operators; private industry; and international organizations. Some of the information currently distributed includes weekly vulnerability bulletins and technical alerts. Responding to incidents and analyzing data about emerging cyber threats. Collaborating with foreign governments and international entities to enhance the nation’s cybersecurity posture. US-CERT has developed programs and resources to carry out these activities, including the distribution of technical alerts, tips, and weekly vulnerability bulletins. In addition, it occasionally performs reviews at the agency and system levels, an activity that allows it to provide agencies more detailed information on agency- and/or system-specific weaknesses, vulnerabilities, and actions to remediate them. OMB has established various initiatives intended to protect federal systems. These include, but are not limited to, OMB’s 30-day Cybersecurity Sprint, the October 2015 CSIP, and the CAP Goals. Agency participation and compliance with these key initiatives, which are outlined in table 1, are mandatory. In August 2014, we issued a report that examined federal agency oversight of contractor-operated federal IT systems. We reported that OPM, one of six federal agencies reviewed, had generally established security and privacy requirements and had planned for assessments to determine the effectiveness of contractor implementation of controls. However, the system assessments performed were not always effective. We recommended that OPM improve its oversight of contractor testing to ensure that tests are being fully executed for all contractor-operated systems. According to OPM officials, efforts are underway to implement the recommendation. In May 2016, we reported on the implementation of OPM’s information security program and the security of selected high-impact systems. We reported that OPM, one of four agencies reviewed, had implemented numerous controls to protect selected systems, but access controls had not always been implemented effectively. Weaknesses also existed in patching known software vulnerabilities and planning for contingencies. An underlying reason for these weaknesses was that OPM had not fully implemented key elements of their information security program. We recommended that OPM fully implement key elements of its program, including addressing shortcomings related to its security plans, training, and system testing. According to OPM officials, the agency is in the process of taking actions to address these recommendations. In addition, we issued a restricted version of the May 2016 report that identified vulnerabilities specific to each of the two systems we reviewed and made recommendations to resolve access control weaknesses in those systems. In December 2016, OPM indicated its concurrence with the recommendations and provided time frames for implementing them. Since the 2015 data breaches, OPM has made progress in improving its security to prevent, mitigate, and respond to data breaches involving sensitive personal records and background investigations information. After breaches of personnel and background investigation information were reported, US-CERT worked with the agency to resolve issues and develop a comprehensive mitigation strategy. Toward this end, in September 2015, US-CERT made 19 recommendations to OPM to help the agency improve its overall security posture and, thus, improve its ability to protect its systems and information from security breaches. As of May 2017, OPM had fully implemented 11 of the recommendations. For the remaining 8 recommendations, actions for 4 were still in progress. For the other 4 recommendations, OPM indicated that it had completed actions to address them, but we noted further improvements were needed. Further, OPM had not validated actions taken to address the recommendations in a timely manner. Beyond actions associated with the US-CERT recommendations, OPM established a new organization—the National Background Investigation Bureau (NBIB)—to perform background investigation services. It signed a memorandum of agreement with the Department of Defense (DOD) to develop and secure new systems to support the bureau’s mission. OPM has made progress in implementing the 19 US-CERT recommendations, but has opportunities to improve on the actions it has taken. As shown in table 2 and subsequently discussed, the agency completed actions for 11 recommendations; needed to further improve on actions taken for 4 other recommendations it indicated had been completed; and needed to complete actions in progress for the remaining 4. Due to the sensitive nature of the recommendations, we are not providing the specific recommendations or specific examples associated with them. Generally, the recommendations pertained to strengthening activities and controls related to passwords, access permissions, patches, audit and monitoring, among other things. OMB requires agencies to create a POA&M to track efforts to remediate identified weaknesses, such as those leading to the 19 recommendations made by US-CERT. In addition, OPM’s policy requires that scheduled completion dates be included in the plan. The policy also requires a system’s Information System Security Officer (ISSO) to develop a weakness closure package containing evidence of how items in the POA&M have been remediated before the issue (recommendation in this case) can be closed. The policy further requires that the closure package be reviewed by the Management Review Board, which approves closure. Although OPM has a POA&M to address the 19 recommendations, it had not updated completion dates in the plan. The POA&M showed scheduled completion dates of either October or December 2015 for all recommendations. However, a separate recommendation tracking document provided to us indicated a planned completion date in the third quarter of 2017 for at least one recommendation. With such inconsistencies and lack of updates, OPM’s ability to gauge performance is limited. Further, the agency had not validated in a timely manner its actions to implement the recommendations. For example, the plan indicated that all actions to address at least 7 of the 19 recommendations had been completed in September or October 2015. However, as of April 2017, at least 17 months later, OPM had not reviewed closure packages or otherwise validated the effectiveness of the actions taken to implement these recommendations. OPM OCIO officials explained that the Management Review Board, at a December 2016 meeting, had discussed the expected evidence required to demonstrate formal closure. In addition, they noted that the US-CERT recommendations in the POA&M would not be closed out until evidence was collected to validate the actions taken. While the intent to validate evidence of remediation actions is commendable, the length of time OPM has taken to do so is troubling in light of the cybersecurity incidents at the agency and given the heightened risk environment in which it operates. Because the US-CERT recommendations are intended to improve the agency’s security posture, more timely validation of the effectiveness of the actions taken is warranted. Until closure packages are created and the evidence of such actions is validated, OPM has limited assurance that the actions taken have effectively mitigated vulnerabilities that can expose its systems to cybersecurity incidents. In addition to its actions to address the US-CERT recommendations in response to the breaches that affected background investigation data, OPM updated roles and responsibilities for handling background investigations. Following the recommendation of a 90-day suitability and security interagency review that identified changes needed to strengthen the background investigation process, in January 2016, the administration announced its plans to create the NBIB within OPM. As of October 2016, OPM had transferred responsibility for performing personnel background investigations from its old Federal Investigative Services unit to NBIB and entered into a memorandum of agreement with DOD to develop and operate information systems supporting the bureau. According to the bureau’s fact sheet, NBIB is intended to: improve the security of background investigation IT systems through a partnership with DOD, improve access to criminal history records through the creation of a law enforcement liaison unit, improve the automation and management of background investigation records, improve the efficiency of background investigation business processes, and consolidate the management of federal and contract field operations. According to OPM OCIO officials, existing background investigation information systems and data will continue to be maintained at OPM while DOD designs and develops a new system to support NBIB. OPM plans to maintain the legacy data for historical and reporting purposes once the new system is operational. At this time, the legacy background investigation system is expected to operate for another 3 years or until all cases in it can be closed out. To support future NBIB operations, the Defense Information Systems Agency (DISA) is in the initial stages of designing the new system, DOD OCIO officials said. While OPM will remain the owner of the background investigations data and processes, DISA will build, operate, and secure the National Background Investigation System, according to the memorandum of agreement between the agencies. According to the DOD officials, the primary benefit of having DISA host and operate the system is that it will be able to take advantage of DOD’s existing information security controls to help ensure the security of the system. OMB’s CSIP and CAP Goals require federal agencies, including OPM, to take specific cybersecurity actions to bolster their system security. The requirements include identifying high value assets, minimizing the number of privileged users, using multifactor authentication to access resources, limiting the access privileged accounts have, using data encryption at rest and in transit, deploying DHS cyber threat indicators, using anti-phishing and anti-malware technology, and subjecting systems to continuous monitoring. Of the eight required actions we selected for review, OPM had developed and documented specific policies that addressed seven of them, had fully implemented two, and had taken actions to partially implement the remaining six. Until OPM completes implementation of the government-wide requirements, its systems are at greater risk than they need be. High value assets refer to those assets, systems, facilities, data, and datasets that are of particular interest to potential adversaries. These assets, systems, and datasets may contain sensitive controls, instructions, or data used in critical federal operations; or house unique collections of data (by size or content), making them of particular interest to criminal, politically-motivated, or state-sponsored actors. According to CSIP, federal agencies must continue to identify and submit a list of high value assets to DHS. OPM had developed, documented, and implemented an information security policy for identifying its high value assets. For example, OPM’s policy stated that, as a part of its security authorization process, a security categorization of its information systems shall be conducted to include an evaluation to determine if a system will be a high value asset. The ISSO is responsible for documenting those systems identified as being high value assets using prescribed templates. Further, OPM had taken action to identify its high value assets. To do this, according to OCIO officials, the agency considered factors such as 1) Federal Information Processing Standards (FIPS) Publication 199 categorization of high impact; 2) the sensitivity of system information; 3) the amount of PII; and 4) the size, location, and mission of a system. By identifying a list of its high value assets, OPM is better positioned to quickly identify and initiate efforts to protect its high value assets from imminent danger of a cybersecurity attack. According to CSIP, privileged users refer to those users with a network account with elevated privileges that is typically allocated to system administrators, network administrators, database administrators, and others who are responsible for system/application control, monitoring, or administration functions—functions and responsibilities beyond the control of ordinary users. Cyber Sprint and CSIP require federal agencies to tighten policies and practices for privileged users by inventorying and minimizing the number of privileged users. OPM defined the scope of its privileged users and implemented a process for inventorying privileged accounts. The agency’s policy defines users with elevated privileges as being those users who are authorized to perform security-relevant functions that ordinary users are not authorized to perform. OPM demonstrated that it reduced the number of network administrators; however, the agency had not minimized other types of accounts with elevated privileges. To ensure privileged access is required for these other types of accounts, OPM OCIO officials explained that the agency conducts periodic reviews of accounts and that this review consists of ensuring that privileged access for these accounts is still warranted. However, the agency could not provide any artifacts to demonstrate this process, other than periodic reporting of the number of privileged users. Establishing policy and minimizing the number of network administrators is a positive step. However, until OPM demonstrates that privileged access for other relevant accounts is warranted, increased risk exists that individuals may have access privileges not required to perform their job. Multifactor authentication—the use of more than one of the combinations of the following factors: something you know (e.g., a password), something you have (e.g., an identification badge), or something you are (e.g., a fingerprint or other biometric)—is a stronger form of authentication than single-factor authentication. According to OMB M-11-11, existing physical and logical access control systems must be upgraded to use personal identity verification (PIV) credentials, in accordance with NIST guidelines. In 2015, the Cyber Sprint required agencies to “dramatically accelerate” the use of a PIV card or alternative form of multifactor authentication for access to information systems, and CSIP requires agencies to complete implementation. Further, the CAP Goal requires agencies to implement a set of capabilities that ensure the use of multiple factors to authenticate to information technology resources. NIST states that agencies can satisfy certain identification and authentication requirements by complying with the requirements in Homeland Security Presidential Directive 12 and using multifactor authentication, such as PIV cards. OPM had developed and documented information security policies for implementing multifactor authentication, such as enforcing the use of PIV cards. For example, OPM’s policy states that multifactor authentication is required for network and local access for privileged and non-privileged accounts and the use of the cards is required to gain access to information systems, when feasible. OPM had implemented the use of PIV cards on one of three systems we reviewed. However, it did not require or implement the use of PIV cards or other multifactor authentication methods for the two other systems. According to OPM OCIO officials, the agency plans to complete implementation of PIV cards for all systems no later than 2018. In its fiscal year 2016 FISMA audit report, OPM’s Office of the Inspector General (OIG) reported that it had validated the agency’s implementation of network-level multifactor authentication based on a prior recommendation. However, the OIG reported that only 2 of OPM’s 46 major systems (applications) were compliant with OMB requirements related to PIV authentication, and recommended that the OCIO meet the requirements of OMB M-11-11 by upgrading its major information systems to require multifactor authentication using PIV credentials. The office concurred with the recommendation and stated that it would work to continue the implementation of an initial enterprise identity and access management solution for enforcing multifactor authentication, including the use of PIV credentials where feasible and appropriate. OPM officials indicated in the report that they were working towards multifactor authentication at the system level in order to create multiple layers of security, but that implementation efforts are ongoing. Until OPM upgrades all of its information systems to require the use of multifactor authentication mechanisms like PIV cards, its systems and applications may be susceptible to attack if an attacker compromises its network. Because we consider these actions to be in progress, we are not making a recommendation at this time. To further protect systems, agencies should employ the principle of least privilege. This principle involves granting users the most restrictive set of privileges needed to perform authorized tasks. One CAP Goal specifies that agencies should implement a set of capabilities that ensures users have access to only those resources that are required for their job function. Further, the Cyber Sprint initiative and CSIP require agencies to tighten policies and practices for privileged users by, to the greatest extent possible, limiting functions that users can perform when using privileged accounts and ensuring that privileged user activities are logged and that such logs are reviewed regularly. OPM has developed, documented, and partially implemented an information security policy that incorporated the principle of least privilege. For example, OPM’s policy requires that users with privileged accounts only use these accounts for security-related functions and use non-privileged accounts for other system functions. The policy also requires that privileged user activities be logged and reviewed regularly. OPM initiated efforts to implement this policy; however, it has not been fully implemented. Due to the sensitive nature of the recommendation and the actions needed to implement it, these details are provided in a separate report with limited distribution. Encryption of data can be used to help protect the integrity and confidentiality of data and computer programs by rendering data unintelligible to unauthorized users and by protecting the integrity of transmitted (in transit) or stored (at rest) data. According to OMB Circular A-130, when the assessed risk indicates the need, agencies must encrypt federal information at rest and in transit unless otherwise protected by alternative physical and logical safeguards implemented at multiple layers, including networks, systems, applications, and data. It notes that agencies must apply encryption to federal information categorized as either moderate or high impact in accordance with FIPS Publication 199 unless encrypting such information is technically infeasible or would demonstrably affect the ability of agencies to carry out their mission, functions, or operations. OPM has developed, documented, and partially implemented an information security policy relevant to data encryption. For example, OPM policy requires system owners to ensure that moderate- and high-impact information systems and PII are protected using data encryption at rest and in transit in accordance with FIPS 140-2. Nevertheless, OPM’s implementation of data encryption at rest and in transit varied for the three systems we reviewed. Specifically, One system was not configured to encrypt data in transit, but its production database was configured to encrypt data at rest. The second system was configured to encrypt data in transit, but was not configured to encrypt data at rest. For the third system, encryption efforts were nearly complete as of March 2017, and the agency was working to complete this effort. Accordingly, we are not making a recommendation at this time. Until OPM fully deploys encryption for the selected systems, the agency will be at increased risk that the confidentiality and integrity of the data in transit and stored on its systems is not fully protected. Threat indicators play a key role in creating shared situational awareness of malicious cyber activity. According to the Cyber Sprint initiative, agencies must immediately deploy indicators provided by DHS regarding priority threat-actor techniques, tactics, and procedures to scan systems and check logs. In addition, CSIP states that, on an ongoing basis, agencies must scan for indicators of compromise within 24 hours of receipt of the threat indicators from DHS. Further, the Standards for Internal Control in the Federal Government state that management is to document in policies the internal control responsibilities of the organization. The standards further state that controls be documented; documentation of controls is evidence that controls are identified, capable of being communicated to those responsible for their performance, and capable of being monitored and evaluated by the entity. Although OPM had deployed DHS threat indicators and scanned for them, its policy had not been updated for this specific requirement. The agency had developed and documented a policy that describes threat indicators such as signatures. For example, OPM’s information security handbook states that signatures related to malware should be updated. The agency also had a procedure in place that gives the CISO responsibility for ensuring that security alerts and advisories are received on a continuous basis, and notes that scanning will occur periodically. However, neither the policy nor procedure described responsibilities for ensuring the 24-hour scanning requirements are accomplished and monitored. By updating the policy and procedure to specifically document roles and responsibilities associated with DHS threat indicator requirements, such as scanning, OPM could further improve upon its ability to ensure that controls are being communicated to those responsible for their performance and are capable of being monitored and evaluated. Anti-phishing and malware defense is important to protect agencies from phishing attempts, where attackers use social engineering with authentic- looking e-mails, websites, or instant messages to get users to download malware, open malicious attachments, or open links that direct them to a website that requests information or executes malicious code. According to the CAP Goal for anti-phishing and malware defense, agencies must implement technologies, processes, and training that reduce the risk of malware introduced through e-mail and malicious web sites. OPM had deployed several anti-malware technologies, but shortcomings existed with the tools deployed. Due to the sensitive nature of the types of tools deployed and any shortcomings, we are not providing specific examples in this report. This information is provided in a separate report with limited distribution. OPM also implemented an effective training exercise to reduce the risk of malware introduced through e-mail and malicious websites. At the request of OPM, DHS’s National Cybersecurity Assessment and Technical Services conducted e-mail phishing exercises to assess the agency’s potential risk and vulnerabilities to e-mail phishing attacks. Over the course of months, OPM showed improvement with results from the phishing exercises. Specifically, in April 2016, DHS reported that it had sent 3,000 fictitious e-mails with a link that would redirect users to a website that describes the dangers of clicking on untrusted links. Of the 3,000 e-mails, 366 unique users (about 12 percent) clicked on the phishing link. Subsequent testing indicated that users were better informed. In September 2016, for the 882 suspicious e-mails sent over the course of 7 campaigns, DHS reported only one instance (less than 1 percent) of a user clicking on a phishing link. By having personnel better informed to detect phishing attempts, OPM has greater assurance that agency systems will not be compromised by such a threat. Continuous monitoring is an important activity in assessing the security impacts on an information system resulting from planned and unplanned changes to the hardware, software, firmware, or environment of operation. According to NIST, continuous monitoring facilitates ongoing awareness of threats, vulnerabilities, and information security to support organizational risk management decisions. The CAP Goal for Information Security Continuous Monitoring (ISCM) states that agencies should provide ongoing observation, assessment, analysis, and diagnosis of an organization’s cybersecurity posture, hygiene, and operational readiness. Additionally, OMB Memorandum 14-03 requires federal agencies to develop and maintain an ISCM strategy and establish an ISCM program. Further, to facilitate continuous monitoring by agencies, DHS established the Continuous Diagnostics and Mitigation (CDM) program. The program involves the delivery of tools and services intended to provide the ability to enhance and automate existing agency continuous network monitoring capabilities, correlate and analyze critical security-related information, and enhance risk-based decision making at the agency and federal levels. CSIP states that federal agencies are to accelerate the implementation of capabilities and tools to identify risks to their systems and networks, including DHS’s CDM program. Further, NIST SP 800-53 recommends that an organization provide role-based security training to personnel with assigned security roles and responsibilities. OPM had established an ISCM program and took actions to implement continuous monitoring practices. The agency also developed a continuous monitoring strategy that requires system owners to assess security controls for ongoing continuous monitoring of its information systems based on the established security control assessment frequencies specified in the continuous monitoring plan. According to the agency’s fiscal year 2016 continuous monitoring plan, owners are to assess specified controls on each system periodically according to the schedule in the plan. However, OPM did not consistently assess the security controls at defined frequencies for the three selected systems we reviewed. For example, the agency did not assess the controls for the systems for several months at a time. According to OPM OCIO officials, security controls for two systems were not assessed in accordance with defined frequencies due to a shortage of ISSOs at the agency, and the other system was not assessed because the system was undergoing reauthorization. However, until OPM consistently performs periodic assessments of the selected systems, the agency will have less assurance of whether controls for these systems are effective and operating as intended. In its 2016 FISMA report, the OIG identified a weakness in OPM’s information security governance regarding the extremely high employee turnover rate for the ISSO positions and OPM’s lack of filling those positions. Subsequently, the OIG recommended that OPM hire a sufficient number of ISSOs to adequately support all of the agency’s major information systems. The agency concurred with the recommendation and indicated that plans were in place to hire a sufficient number of ISSOs to support all of its major information systems. OPM deployed continuous monitoring tools obtained through DHS’s CDM program, and developed and deployed its own internal continuous monitoring dashboard. Officials stated that the agency completed the first phase of DHS’s CDM program and is in the process of implementing the second phase. However, OPM has not issued role-based training requirements for the individuals configuring and maintaining the deployed CDM tools. OCIO officials told us the agency drafted role-based training requirements for CDM users, but could not demonstrate to us that it has done so. Until OPM develops and implements role-based training requirements for staff using its CDM tools, it will not be able to ensure that staff are using the tools properly. As a result, the usefulness of the tools for monitoring the security of the agency’s information systems may be diminished. OPM has implemented elements of contractor oversight such as recording security assessment findings for contractor-operated systems in remediation plans, but it did not ensure that system security assessments involved comprehensive testing. The agency requires ISSOs to conduct quality assurance reviews that include reviewing security assessments of contractor-operated systems; however, its policy did not include detailed guidance on how the reviews are to be conducted. To determine the effectiveness of contractors’ implementation of federal information security requirements, agencies rely on security control assessments for contractor-operated systems. FISMA requires that agencies periodically test the management, operational, and technical controls for their systems, including for those systems that are operated by a contractor of an agency or other organization on behalf of an agency. In addition, NIST SP 800-53 recommends that agencies assess systems to determine if controls are implemented correctly, operating as intended, and producing the desired results. It also requires that a plan be developed that describes the assessment procedures to be used to determine security control effectiveness, and that the agency produce a report that contains the results of this assessment. Further, NIST SP 800- 53A contains a suggested methodology for control assessments, including personnel to interview, documents to examine, and testing to be performed. It also states that different assessment methodologies may be used as long as they are sufficient to identify weaknesses. OPM ensured that security control assessments had been conducted for the contractor-operated systems we selected for review; however, instances existed where the methodology employed to conduct the assessments was not sufficient to identify potential weaknesses. Specifically, third-party assessors, on behalf of OPM’s contractors, recently conducted security assessments for each of the three contractor- operated systems we reviewed. The methodologies for the assessments included document examinations, interviews, and system testing. We determined that many of the selected controls had been properly assessed, as shown in table 3. The assessors had addressed most controls for at least two systems satisfactorily; nevertheless, shortcomings existed in the completeness of the testing of specific controls. For example, for System 1, 23 of the 120 controls we reviewed had not been properly assessed. For the 23 controls, the assessor did not test the system to determine if the controls were in place. In addition, for System 3, the security control assessment provided few details on the procedures used to evaluate the effectiveness of the system’s security controls. Instead, the assessors recorded the exact same procedures for every control, and few had enough detail in the results to determine if the assessors had employed sufficient procedures. OPM has developed, documented, and implemented a process for overseeing the security of contractor-operated systems. The process involved reviewing the third-party security control assessments, but it did not ensure the comprehensiveness of the assessments. OPM has established a quality assurance process that includes reviewing security control assessments. This process consisted of a standard form letter that the ISSO completes after reviewing the assessment package provided by the third-party assessor. For all three selected contractor- operated systems, security officers completed the letter and noted shortcomings with the security control assessments: System 1: the ISSO noted some problems due to a lack of provided evidence. For example, the assessor concluded that a number of controls were partially satisfied, despite evidence not being provided by the contractor to the assessor. System 2: the security officer identified documentation issues with several controls. For example, the system security plan described the system inconsistently with other documents, such as the contingency plan. System 3: the ISSO stated that most of the evidence provided by the assessor was not referenced to the specific procedures performed. Nevertheless, the security officers did not comment on the comprehensiveness of testing, such as the scope and depth of testing by the third-party assessors. In February 2017, OPM issued an updated authorization guide, which stated that the security officer should review the assessment results and evidence and conduct a quality assurance review on the procedures executed by the independent assessor. However, the guide does not include instructions on how the security officer should perform this review. Until OPM provides more detailed guidance to the ISSOs, the ISSOs may not conduct a rigorous review of the assessment. Therefore, the agency will have less assurance that the security controls for its contractor-operated systems are comprehensively assessed. We have previously made recommendations related to OPM’s procedures for ensuring the sufficiency of security control assessments for its systems, including those operated by contractors on its behalf. In our August 2014 report related to agency oversight of information technology contractors, we recommended that the agency develop, document, and implement oversight procedures for ensuring that a system test is fully executed for each contractor-operated system. The agency concurred with the recommendation, but, as of April 2017, had not provided evidence that the recommendation had been implemented. Further, in our May 2016 report related to the security of high-impact systems, we recommended that OPM reevaluate security control assessments to ensure that they comprehensively test technical controls for two selected high-impact systems, one of which was a contractor- operated system. The agency did not concur with our recommendation. However, without comprehensive security control assessments for these systems, OPM is at increased risk that it may not detect vulnerabilities in the systems. Therefore, we believe the recommendation is warranted. A remedial action plan is a key component of an agency’s information security program, as described in FISMA. Such a plan assists agencies in identifying, assessing, prioritizing, and monitoring progress in correcting security weaknesses that are found in an information system. NIST recommends that agencies develop a POA&M for an information system to document the agency’s planned remedial actions to correct identified weaknesses. OPM developed a POA&M for each of the three reviewed systems and recorded 63 of 64 security weaknesses identified in security assessments for those systems. Tracking the weaknesses that arise from security assessments allows OPM to keep track of the weaknesses as they are remediated or their risk is accepted. Table 4 shows the number of security weaknesses per system reviewed. OPM policy also requires management oversight and review of remediation plans. Contractors are to provide evidence, which is to be reviewed and approved by multiple parties at OPM before a POA&M is considered closed. However, because the assessments we selected were so recent, we were unable to evaluate whether this process had been effectively implemented. OPM collects and maintains highly sensitive personal information on millions of individuals, including sensitive security clearance data. Cyber incidents at OPM demonstrated the impact that increasingly sophisticated cyber threats can cause and underscore the importance of protecting the agency’s systems. OPM has improved its security posture and is in the process of taking numerous actions, such as addressing recommendations from US-CERT and implementing government-wide requirements and initiatives that could decrease the risk of future security breaches if effectively implemented. However, by not validating remedial actions in a timely manner, the agency has limited assurance whether these actions effectively mitigated vulnerabilities that can expose systems to incidents. In addition, OPM had not consistently updated milestones for outstanding US-CERT recommendations or complied with its own plan for conducting periodic control assessments. Also, training needed to ensure proper use of monitoring tools was not being completed because the agency has not documented such requirements. Further, key security controls on selected contractor-operated systems have not always been comprehensively tested. Until OPM further improves controls over its information and information systems, it has limited assurance that sufficient security controls are in place and operating as intended. To further improve security over personnel and other sensitive information at the agency, we are recommending that the Acting Director of OPM implement the following five recommendations: 1. Update the POA&M to reflect expected completion dates for implementing the recommendations made by US-CERT. 2. Improve the timeliness of validating evidence associated with actions taken to address the US-CERT recommendations. 3. Update policy to reflect deployment of DHS threat indicators and the specific 24-hour scanning requirement. 4. Develop and implement role-based training requirements for staff using CDM tools. 5. Provide detailed guidance on the quality assurance process that includes evaluating security control assessments. In a separate report with limited distribution, we are making nine recommendations to the Acting Director to improve upon actions taken to implement the recommendations made by US-CERT and to further implement government-wide requirements. We received written comments on a draft of this report from OPM. In its comments, which are re-printed in appendix II, the agency concurred with four of our five recommendations and partially concurred with one recommendation. OPM partially concurred with our recommendation that it improve the timeliness of validating evidence associated with actions taken to address the US-CERT recommendations. The agency did not specifically address the reason for its partial concurrence, but stated that it will review its management practices to support more timely closure of POA&Ms. Regardless of whether OPM reviews its management practices, performing timely validation of evidence is critical. As noted in this report, at least 17 months had passed without OPM validating evidence that it had effectively implemented the US-CERT recommendations. Until the evidence is validated, OPM will have limited assurance that the actions taken have mitigated vulnerabilities that can expose its systems to cybersecurity incidents. In addition, OPM provided comments concerning the approach of our audit and aspects of our report message. In particular, the agency stated that GAO did not fully acknowledge OPM’s “defense in depth” strategy and that the report does not present a fully accurate picture of the agency’s cybersecurity posture. As we state in this report, our objectives were to evaluate OPM’s actions taken since the 2015 breach, implementation of requirements associated with selected government- wide initiatives, and oversight of contractor-operated systems. We designed and performed audit procedures to collect sufficient evidence to accomplish these objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions. Beyond the results of this audit, we previously reported in 2016 that OPM needed to improve security controls over its high-impact systems that we selected for review. In two prior reports, we made numerous recommendations to enhance the agency’s information security program and to resolve access control weaknesses in those systems. To date, these recommendations remain open. Based on the results of our prior work and this audit, we believe our current report appropriately reflects OPM’s cybersecurity posture, consistent with our audit objectives. OPM also indicated that it has taken steps to enhance its cybersecurity posture in multiple areas through the addition of cybersecurity tools and security updates, staff and agency-wide training, critical personnel hiring, and collaboration with its interagency partners. We agree that these are positive actions. Effective implementation of these actions and our recommendations is essential to ensuring that sufficient security controls are in place and operating as intended. We are sending copies of this report to appropriate congressional committees, the acting Director of the Office of Personnel Management, including its Office of the Inspector General, and other interested congressional parties. In addition, the report is available at no charge on the GAO website at http://www.gao.gov. If you or your staff have any questions about this report, please contact Gregory C. Wilshusen at (202) 512-6244 or wilshuseng@gao.gov, or Dr. Nabajyoti Barkakati at (202) 512-4499 or barkakatin@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix III. Our objectives were to evaluate the Office of Personnel Management’s (OPM) 1) actions since the 2015 data breaches to prevent, mitigate, and respond to data breaches involving sensitive personnel records and information; 2) information security policies and practices as they relate to selected government-wide initiatives and requirements; and 3) procedures for overseeing the security of OPM information maintained by contractors providing information technology services. To address the first objective, we assessed the extent to which the agency had implemented 19 recommendations that the United States Computer Emergency Response Team (US-CERT) made in its 2015 breach investigation report. We chose to focus on the status of US- CERT’s recommendations for this objective in order to avoid duplication of effort. According to US-CERT, its recommendations outlined mitigations and security best practices based on the specific compromise and OPM’s cybersecurity environment, and are tailored to the OPM environment. To determine the extent of implementation and whether further improvements were needed, we reviewed the status (e.g., complete or in progress) of OPM’s actions to address the recommendations made by US-CERT. We focused on one or more actions described in the plans of action and milestones provided to us. We interviewed officials from OPM’s Office of the Chief Information Officer (OCIO) and reviewed the agency’s information security documentation, to include policies, plans, and procedures. We also performed limited testing of the agency’s internal network and software tools to determine the extent to which US- CERT’s recommendations had been implemented. This testing included examining system and device configurations, firewall rules, system status reports, and account listings. We focused on those software tools that OPM indicated it had deployed as actions taken to implement US-CERT recommendations. If, after reviewing the plans provided by OPM and performing our own testing, we could not account for a specific recommendation or disagreed with the implementation status, we followed up with the agency. For this objective, we evaluated the reliability of the data related to the number of workstations at OPM, as well as the number of outstanding patches. We assessed the data reliability by various means, including reviewing related documents, conducting observations of systems generating data, interviewing knowledgeable agency officials, and reviewing internal controls such as agency policies and procedures. We concluded that the data were sufficiently reliable for the purposes of this reporting objective. As part of this objective, because the data breaches involved background investigation information, we also reviewed OPM’s plans to establish the new National Background Investigations Bureau (NBIB), its plans to transfer the information systems supporting NBIB to the Department of Defense (DOD), and DOD’s plans for the National Background Investigations System. To accomplish this, we reviewed the memorandum of agreement between the agencies and implementation plans and interviewed officials from OPM and DOD. To address the second objective, we focused on the security controls agencies are required to implement, as described in the Office of Management and Budget’s October 2015 Cybersecurity Strategy and Implementation Plan (CSIP) and the Cybersecurity Cross-Agency Priority (CAP) Goals. In addition, OMB issued an updated version of its Circular A-130, which we also reviewed. Based on our review of these documents, we determined that government-wide requirements associated with them include: 1) identifying high value assets, 2) minimizing the number of privileged users, 3) using multifactor authentication, 4) limiting the access for privileged accounts, 5) using data encryption, 6) deploying cyber threat indicators issued by the Department of Homeland Security, 7) using anti-phishing and anti- malware technology, and 8) subjecting systems to continuous monitoring. Further, we reviewed and applied the National Institute of Standards and Technology’s (NIST) standards and guidelines that are related to these requirements. To determine the extent to which OPM had implemented the requirements associated with the initiatives, we reviewed agency policies, implementation plans, and other related documents, and interviewed OPM OCIO officials and contractors. We verified this information and collected new information by reviewing relevant controls on three selected systems. We selected these systems based on the agency categorizing them as high-impact systems and them not being recently audited by GAO. Because system weaknesses identified in this report could be exploited, we are not identifying the system names or other specifics related to our selection. In a separate report with limited distribution, we provide more details regarding our scope and methodology. The review we conducted was targeted to acquire information specifically related to this objective and was therefore not comprehensive in nature. Some of the testing we performed included reviewing configuration settings, administrative account privileges and authorizations, and data encryption. To address our third objective, we interviewed OPM officials and reviewed OPM policy and system security assessments. For selected OPM contractor-operated systems, we reviewed the system’s security assessments to determine whether the security control review conducted by the assessor was adequate to evaluate the effectiveness of the controls tested. We selected the two active, contractor-operated systems that were not involved in the breach, but were categorized by the agency as high-impact systems. We also selected one active system that the agency categorized as moderate impact. Because system weaknesses identified in this report could be exploited, we are not identifying the systems’ names or other specifics related to our selection. We focused the scope of our review on assessments of moderate- and high-impact controls in the following NIST Special Publication (SP) 800-53 defined control families: access controls, audit and accountability, security assessment and authorization, configuration management, contingency planning, risk assessments, and system and information integrity. In addition, we excluded controls identified by the assessors as either inherited or not applicable. We determined the sufficiency of the procedures used by comparing the requirements and recommended testing procedures—as defined in NIST SP 800-53 and NIST SP 800-53A—of each control with the testing procedures documented by assessors in the assessment results. For each control, we evaluated whether the type (e.g., interview, observation, technical testing) and scope of the testing performed was such that one could accurately determine whether a given control had been effectively implemented. We also reviewed the plans of action and milestones associated with each assessment to determine whether the assessment results were reflected in the plans. We conducted this performance audit from January 2016 to August 2017 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. In addition to the contacts named above, West Coile and Jeffrey Knott (assistant directors); Justin Palk (analyst-in-charge); Corey Evans, Nancy Glover, David Plocher, Brandon Sanders, Michael Stevens, and Edward Varty made key contributions to this report. | OPM collects and maintains personal data on millions of individuals, including data related to security clearance investigations. In 2015, OPM reported significant breaches of personal information that affected 21.5 million individuals. The Senate report accompanying the Financial Services and General Government Appropriations Act, 2016 included a provision for GAO to review information security at OPM. GAO evaluated OPM's (1) actions since the 2015 reported data breaches to prevent, mitigate, and respond to data breaches involving sensitive personnel records and information; (2) information security policies and practices for implementing selected government-wide initiatives and requirements; and (3) procedures for overseeing the security of OPM information maintained by contractors providing IT services. To do so, GAO examined policies, plans, and procedures and other documents; tested controls for selected systems; and interviewed officials. This is a public version of a sensitive report being issued concurrently. GAO omitted certain specific examples due to the sensitive nature of the information. Since the 2015 data breaches, the Office of Personnel Management (OPM) has taken actions to prevent, mitigate, and respond to data breaches involving sensitive personal and background investigation information, but actions are not complete. OPM implemented or made progress towards implementing 19 recommendations made by the United States Computer Emergency Readiness Team (US-CERT) to bolster OPM's information security practices and controls in the wake of the 2015 breaches. GAO determined that the agency completed actions for 11 of the recommendations and took actions for the remaining 8, with actions for 4 of these 8 requiring further improvement (see table). In addition, OPM did not consistently update completion dates for outstanding recommendations and did not validate corrective actions taken to ensure that the actions effectively addressed the recommendations. OPM also made progress in implementing information security policies and practices associated with selected government-wide initiatives and requirements. However, it did not fully implement all of the requirements. For example, OPM identified its high value assets, such as systems containing sensitive information that might be attractive to potential adversaries, but it did not encrypt stored data on one selected system and did not encrypt transmitted data on another. Until OPM completes implementation of government-wide requirements, its systems are at greater risk than they need be. OPM's procedures for overseeing the security of its contractor-operated systems did not ensure that controls were comprehensively tested. Although the agency has implemented elements of contractor oversight such as recording security assessment findings for contractor-operated systems in remediation plans, it did not ensure that system security assessments involved comprehensive testing. The agency requires information system security officers to conduct quality assurance reviews that include reviewing security assessments of contractor-operated systems; however, its policy did not include detailed guidance on how the reviews are to be conducted. Until such a procedure is clearly defined and documented, OPM will have less assurance that the security controls intended to protect OPM information maintained on contractor-operated systems are sufficiently implemented. GAO is making five recommendations to improve OPM's security. OPM concurred with four of these and partially concurred with the one on validating its corrective actions. GAO continues to believe that implementation of this recommendation is warranted. In GAO's limited distribution report, GAO made nine additional recommendations. |
Today the Social Security program does not face an immediate crisis but rather a long-range and more fundamental financing problem driven largely by known demographic trends. The lack of an immediate solvency crisis affects the nature of the challenge, but it does not eliminate the need for action. Acting soon reduces the likelihood that the Congress will have to choose between imposing severe benefit cuts and unfairly burdening future generations with the program’s rising costs. Acting soon would allow changes to be phased in so the individuals who are most likely to be affected, namely younger and future workers, will have time to adjust their retirement planning while helping to avoid related “expectation gaps.” Mr. Chairman, as you heard earlier this month while hosting the second Annual OECD International Conference of Chairpersons of Parliamentary Budget Committees, we are not alone in facing long-term budget challenges due to an aging population. Our counterparts in many European countries are debating these same issues, and a number of developed and developing countries have already engaged in fundamental reform of their systems to deal with their long-range challenges. Acting soon will also help put the overall federal budget on a more sustainable footing over the long term, thereby promoting both higher economic growth and more fiscal flexibility. The importance of such flexibility was brought dramatically home last September. The budgetary surpluses of recent years put us in a stronger position to respond both to the events of September 11 and to the economic slowdown than would otherwise have been the case. Going forward, the nation’s commitment to surpluses will truly be tested. None of the changes since September 11 have lessened the pressures placed by Social Security, Medicare, and Medicaid on the long-term fiscal outlook. Indeed, the events of September 11 have served to increase our long-range fiscal challenges. Since there is a great deal of confusion about Social Security’s current financing arrangements and the nature of its long-term financing problem, I would like to spend some time describing the nature, timing, and extent of the financing problem. As you all know, Social Security has always been largely a pay-as-you-go system. This means that current workers’ taxes pay current retirees’ benefits. As a result, the relative numbers of workers and beneficiaries has a major impact on the program’s financial condition. This ratio, however, is changing. In the 1960s, the ratio averaged 4.2:1. Today it is 3.4:1 and it is expected to drop to around 2:1 by 2030. The retirement of the baby boom generation is not the only demographic challenge facing the system. People are retiring early and living longer. A falling fertility rate is the other principal factor underlying the growth in the elderly’s share of the population. In the 1960s, the fertility rate was an average of 3 children per woman. Today it is a little over 2, and by 2030 it is expected to fall to 1.95— a rate that is below replacement. Taken together, these trends threaten the financial solvency and sustainability of this important program. (See fig. 1.) The combination of these trends means that labor force growth will begin to slow after 2010 and become negligible by 2050. (See fig. 2.) Relatively fewer workers will be available to produce the goods and services that all will consume. Without a major increase in productivity, low labor force growth will lead to slower growth in the economy and to slower growth of federal revenues. This in turn will only accentuate the overall pressure on the federal budget. This slowing labor force growth is not always considered as part of the Social Security debate. Social Security’s retirement eligibility dates are often the subject of discussion and debate and can have a direct effect on both labor force growth and the condition of the Social Security retirement program. However, it is also appropriate to consider whether and how changes in pension and/or other government policies could encourage longer workforce participation. To the extent that people choose to work longer as they live longer, the increase in the share of life spent in retirement would be slowed. This could improve the finances of Social Security and mitigate the expected slowdown in labor force growth. In addition to encouraging people to work longer, a second approach to addressing labor force growth would be to bring more people into the labor force. In domestic social policy, we have seen an increasing focus on encouraging those previously outside the labor force (i.e., welfare recipients, the disabled) into the workforce. Concern about the slowdown in the growth of the labor force may also lead to discussions about immigration and its role. Increased immigration, however, poses complex issues and is unlikely to be the sole solution. For example, according to a recent United Nations study, it would take more than a sustained tenfold increase in projected immigration to maintain the ratio of workers to retirees at recent levels. These are issues that the Congress may wish to explore further in the next few years. Because of the demographic trends discussed above, current estimates show that within 15 years benefit payments will begin to exceed program revenue, which is composed largely of payroll taxes on current workers. (See fig. 3.) Within the federal budget, Social Security—more properly, the Old-Age and Survivors Insurance and Disability Insurance programs (OASDI)—has two trust funds that authorize Treasury to pay benefits as long as the applicable trust fund has a positive balance. Currently, annual tax revenues to Social Security exceed annual benefit payments. The Trust Funds, by law, invest the resulting cash surplus in U.S. government obligations or securities that are backed by the full faith and credit of the U.S. government. At present, the Trust Funds’ assets are in the form of special, nonmarketable Treasury securities that are backed by the full faith and credit of the U.S. government and so carry no risk of default. Although the Trust Funds cannot sell their holdings in the open market, the Trust Funds face no liquidity risk since they can redeem their special Treasury securities before maturity without penalty. These securities earn interest credits at a statutory rate linked to market yields, and this interest from the Treasury is credited to the Trust Funds in the form of additional Treasury securities. I think it is useful to pause for a moment here and reflect on what the term “trust fund” means in the federal budget. Trust funds in the federal budget are not like private trust funds. An individual can create a private trust fund using his or her own assets to benefit a stated individual(s). The creator, or settlor, of the trust names a trustee who has a fiduciary responsibility to manage the designated assets in accordance with the stipulations of the trust. In contrast, federal trust funds are budget accounts used to record receipts and expenditures earmarked for specific purposes. The Congress creates a federal trust fund in law and designates a funding source to benefit stated groups or individuals. Unlike most private trustees, the federal government can raise or lower future trust fund collections and payments or change the purposes for which the collections are used by changing existing laws. Moreover, the federal government has custody and control of the funds. Under current law, when the Social Security Trust Funds’ tax receipts exceed costs—that is, when the Trust Funds have an annual cash surplus— this surplus is invested in Treasury securities and can be used to meet current cash needs of the government or to reduce debt held by the public. In either case, the solvency of the Trust Funds is unchanged. However, while the Treasury securities are an asset to the Trust Funds, they are a liability to the Treasury. Any increase in assets to the Trust Funds creates an increase of equal size in future claims on the Treasury. One government fund is lending to another. As a result, these transactions net out on the government’s consolidated books. The accumulated balances in a trust fund do not in and of themselves increase the government’s ability to meet the related program commitments. That is, simply increasing trust fund balances does not improve program sustainability. Increases in trust fund balances can strengthen the ability to pay future benefits if a trust fund’s cash surpluses are used to improve the government’s overall fiscal position. For example, when a trust fund’s cash surpluses are used to reduce debt held by the public, this increases national saving, contributes to higher economic growth over the long term, and enhances the government’s ability to raise cash in the future to pay benefits. It also reduces federal interest costs below what they otherwise would have been, thereby promoting greater fiscal flexibility in the future. According to the Trustees’ intermediate estimates, the combined Social Security Trust Funds will be solvent until 2041. However, our long-term model shows that well before that time program spending will constitute a rapidly growing share of the budget and the economy. Ultimately, the critical question is not how much a trust fund has in assets, but whether the government as a whole can afford the promised benefits in the future and at what cost to other claims on scarce resources. As I have said before, the future sustainability of programs is the key issue policymakers should address—i.e., the capacity of the economy and budget to afford the commitment. Fund solvency can help, but only if promoting solvency improves the future sustainability of the program. Today, the Social Security Trust Funds take in more in taxes than they spend. Largely because of the known demographic trends I have described, this situation will change. Under the Trustees’ intermediate assumptions, annual cash surpluses begin to shrink in 2006, and combined program outlays begin to exceed dedicated tax receipts in 2017, a year after Medicare’s Hospital Insurance Trust Fund (HI) outlays are first expected to exceed program tax revenues. At that time, both programs will become net claimants on the rest of the federal budget. (See fig. 4.) As I noted above, the special Treasury securities represent assets for the Trust Funds but are future claims against the Treasury. Beginning in 2017, the Trust Funds will begin drawing on the Treasury to cover the cash shortfall, first relying on interest income and eventually drawing down accumulated trust fund assets. Regardless of whether the Trust Funds are drawing on interest income or principal to make benefit payments, the Treasury will need to obtain cash for those redeemed securities either through increased taxes, spending cuts, increased borrowing from the public, or correspondingly less debt reduction than would have been the case had Social Security’s cash flow remained positive. Neither the decline in the cash surpluses nor the cash deficit will affect the payment of benefits. However, the shift affects the rest of the budget. The negative cash flow will place increased pressure on the federal budget to raise the resources necessary to meet the program’s ongoing costs. From the perspective of the federal budget and the economy, the challenge posed by the growth in Social Security spending becomes even more significant in combination with the more rapid expected growth in Medicare and Medicaid spending. This growth in spending on federal entitlements for retirees will become increasingly unsustainable over the longer term, compounding an ongoing decline in budgetary flexibility. Over the past few decades, spending on mandatory programs has consumed an ever-increasing share of the federal budget. Prior to the creation of the Medicare and Medicaid programs, in 1962 mandatory spending plus net interest accounted for about 32 percent of total federal spending. By 2002, this share had almost doubled to approximately 63 percent of the budget. (See fig. 5.) *Office of Management and Budget current services estimate. In much of the last decade, reductions in defense spending helped accommodate the growth in these entitlement programs. This, however, is no longer a viable option. Even before September 11, reductions in defense spending were no longer available to help fund other claims on the budget. Indeed, spending on defense and homeland security will grow as we seek to combat new threats to our nation’s security. Our long-term budget simulations continue to show that to move into the future with no changes in federal retirement and health programs is to envision a very different role for the federal government. Assuming, for example, that the tax reductions enacted last year do not sunset and discretionary spending keeps pace with the economy, by midcentury federal revenues may only be adequate to pay Social Security and interest on the federal debt. Spending for the current Medicare program—without the addition of a drug benefit—is projected to account for more than one- quarter of all federal revenues. To obtain balance, massive spending cuts, tax increases, or some combination of the two would be necessary. (See fig. 6.) Neither slowing the growth of discretionary spending nor allowing the tax reductions to sunset eliminates the imbalance. It is important as well to look beyond the federal budget to the economy as a whole. Figure 7 shows the total future draw on the economy represented by Social Security, Medicare, and Medicaid. Under the 2002 Trustees’ intermediate estimates and the Congressional Budget Office’s (CBO) most recent long-term Medicaid estimates, spending for these entitlement programs combined will grow to 14.1 percent of GDP in 2030 from today’s 8.3 percent. Taken together, Social Security, Medicare, and Medicaid represent an unsustainable burden on future generations. This testimony is not about the complexities of Medicare, but it is important to note that Medicare presents a much greater, more complex, and more urgent fiscal challenge than does Social Security. Unlike Social Security, Medicare growth rates reflect not only a burgeoning beneficiary population, but also the escalation of health care costs at rates well exceeding general rates of inflation. Increases in the number and quality of health care services have been fueled by the explosive growth of medical technology. Moreover, the actual costs of health care consumption are not transparent. Third-party payers generally insulate consumers from the cost of health care decisions. These factors and others contribute to making Medicare a much greater and more complex fiscal challenge than even Social Security. When Social Security redeems assets to pay benefits, the program will constitute a claim on real resources in the future. As a result, taking action now to increase the future pool of resources is important. To echo Federal Reserve Chairman Greenspan, the crucial issue of saving in our economy relates to our ability to build an adequate capital stock to produce enough goods and services in the future to accommodate both retirees and workers in the future. The most direct way the federal government can raise national saving is by increasing government saving. Ultimately, as this Committee recommended last fall, we should attempt to return to a position of surplus as the economy returns to a higher growth path. This would allow the federal government to reduce the debt overhang from past deficit spending, provide a strong foundation for future economic growth, and enhance future budgetary flexibility. Similarly, taking action now on Social Security would not only promote increased budgetary flexibility in the future and stronger economic growth but would also make less dramatic action necessary than if we wait. Perhaps the best way to illustrate this is to compare what it would take to achieve actuarial balance at different points in time by either raising payroll taxes or reducing benefits. Figure 8 shows this. If we did nothing until 2041—the year the Trust Funds are estimated to be exhausted—achieving actuarial balance would require changes in benefits of 31 percent or changes in taxes of 45 percent. As figure 8 shows, earlier action shrinks the size of the necessary adjustment. Thus both sustainability concerns and solvency considerations drive us to act sooner rather than later. Trust Fund exhaustion may be nearly 40 years away, but the squeeze on the federal budget will begin as the baby boom generation starts to retire. Actions taken today can ease both these pressures and the pain of future actions. Acting sooner rather than later also provides a more reasonable planning horizon for future retirees. As important as financial stability may be for Social Security, it cannot be the only consideration. As a former public trustee of Social Security and Medicare, I am well aware of the central role these programs play in the lives of millions of Americans. Social Security remains the foundation of the nation’s retirement system. It is also much more than just a retirement program; it also pays benefits to disabled workers and their dependents, spouses and children of retired workers, and survivors of deceased workers. Last year, Social Security paid almost $408 billion in benefits to more than 45 million people. Since its inception, the program has successfully reduced poverty among the elderly. In 1959, 35 percent of the elderly were poor. In 2000, about 8 percent of beneficiaries aged 65 or older were poor, and 48 percent would have been poor without Social Security. It is precisely because the program is so deeply woven into the fabric of our nation that any proposed reform must consider the program in its entirety, rather than one aspect alone. Thus, GAO has developed a broad framework for evaluating reform proposals that considers not only solvency but other aspects of the program as well. The analytic framework GAO has developed to assess proposals comprises three basic criteria: the extent to which a proposal achieves sustainable solvency and how it would affect the economy and the federal budget; the relative balance struck between the goals of individual equity and how readily a proposal could be implemented, administered, and explained to the public. The weight that different policymakers may place on different criteria will vary, depending on how they value different attributes. For example, if offering individual choice and control is less important than maintaining replacement rates for low-income workers, then a reform proposal emphasizing adequacy considerations might be preferred. As they fashion a comprehensive proposal, however, policymakers will ultimately have to balance the relative importance they place on each of these criteria. Historically, Social Security’s solvency has generally been measured over a 75-year projection period. If projected revenues equal projected outlays over this time horizon, then the system is declared in actuarial balance. Unfortunately, this measure is itself unstable. Each year, the 75-year actuarial period changes, and a year with a surplus is replaced by a new 75th year that has a significant deficit. This means that, changes that restore solvency only for the 75-year period will not hold. For example, if we were to raise payroll taxes immediately by 1.87 percentage points of taxable payroll today—which, according to the 2002 Trustees Report, is the amount necessary to achieve 75-year balance—the system would be out of balance next year. This is the case because actions taken to close the 75-year imbalance would not fully address the projected deficit in year 76 of 6.49 percent of taxable payroll. Reforms that lead to sustainable solvency are those that avoid the automatic need to periodically revisit this issue. As I have already discussed, reducing the relative future burdens of Social Security and health programs is essential to a sustainable budget policy for the longer term. It is also critical if we are to avoid putting unsupportable financial pressures on future workers. Reforming Social Security and federal health programs is essential to reclaiming our future fiscal flexibility to address other national priorities. The current Social Security system’s benefit structure strikes a balance between the goals of retirement income adequacy and individual equity. From the beginning, benefits were set in a way that focused especially on replacing some portion of workers’ pre-retirement earnings. Over time other changes were made that were intended to enhance the program’s role in helping ensure adequate incomes. Retirement income adequacy, therefore, is addressed in part through the program’s progressive benefit structure, providing proportionately larger benefits to lower earners and certain household types, such as those with dependents. Individual equity refers to the relationship between contributions made and benefits received. This can be thought of as the rate of return on individual contributions. Balancing these seemingly conflicting objectives through the political process has resulted in the design of the current Social Security program and should still be taken into account in any proposed reforms. Policymakers could assess income adequacy, for example, by considering the extent to which proposals ensure benefit levels that are adequate to protect beneficiaries from poverty and ensure higher replacement rates for low-income workers. In addition, policymakers could consider the impact of proposed changes on various subpopulations, such as low-income workers, women, minorities, and people with disabilities. Policymakers could assess equity by considering the extent to which there are reasonable returns on contributions at a reasonable level of risk to the individual, improved intergenerational equity, and increased individual choice and control. Differences in how various proposals balance each of these goals will help determine which proposals will be acceptable to policymakers and the public. Program complexity makes implementation and administration both more difficult and harder to explain to the public. Some degree of implementation and administrative complexity arises in virtually all proposed changes to Social Security, even those that make incremental changes in the already existing structure. However, the greatest potential implementation and administrative challenges are associated with proposals that would create individual accounts. These include, for example, issues concerning the management of the information and money flow needed to maintain such a system, the degree of choice and flexibility individuals would have over investment options and access to their accounts, investment education and transitional efforts, and the mechanisms that would be used to pay out benefits upon retirement. Harmonizing a system that includes individual accounts with the regulatory framework that governs our nation's private pension system would also be a complicated endeavor. However, the complexity of meshing these systems should be weighed against the potential benefits of extending participation in individual accounts to millions of workers who currently lack private pension coverage. Continued public acceptance and confidence in the Social Security program require that any reforms and their implications for benefits be well understood. This means that the American people must understand why change is necessary, what the reforms are, why they are needed, how they are to be implemented and administered, and how they will affect their own retirement income. All reform proposals will require some additional outreach to the public so that future beneficiaries can adjust their retirement planning accordingly. Yet the more transparent the implementation and administration of reform, and the more carefully such reform is phased in, the more likely it will be understood and accepted by the American people. With regard to proposals that involve individual accounts, an essential challenge would be to help the American people understand the relationship between their individual accounts and traditional Social Security benefits, thereby ensuring that any gaps in expectations about current or future benefits are avoided. In addition, increasing the public’s level of sophistication and understanding of how to invest in the market, the relationship between risk and return, and the potential benefits of diversification presents an education challenge that must be surmounted so that the American people have the necessary tools to secure their future. The Enron collapse helps to illustrate the importance of this, as well as the need to provide clear and understandable information so that the public can make informed retirement decisions. Early action to address the financing problems of Social Security yields the highest fiscal dividends for the federal budget and provides a longer period for future beneficiaries to make adjustments in their own planning. The events of September 11 and the challenges of combating terrorism do not change this. In fact, the additional spending that will be required to fight the war on terrorism and protect our homeland will serve to increase our long-range fiscal challenges. It remains true that the longer we wait to take action on the programs driving long-term deficits, the more painful and difficult the choices will become. Although the program does not face an immediate solvency crisis as it did in 1983, the fundamental nature of the program’s long-term financing challenge means that timely action is needed. The demographic trends recognized in 1983 are now almost upon us. It is these demographic trends—and their implications for both Social Security and Medicare—that lead to the conclusion that the program faces both a solvency and a sustainability problem. For the American people to understand why change is necessary, a public education campaign will be needed that focuses not just on Social Security but also on our long-range fiscal challenges. We will face many difficult choices in making Social Security sustainable. Focusing on comprehensive packages of reforms that protect the benefits of current retirees while achieving the right balance of equity and adequacy for future beneficiaries will help to foster credibility and acceptance. This will help us avoid getting mired in the details and losing sight of important interactive effects. It will help build the bridges necessary to achieve consensus. Today I have described the three basic criteria against which GAO thinks Social Security reform proposals may be measured. These may not be the same criteria every analyst would suggest, and certainly how policymakers weight the various elements may vary. However, if comprehensive proposals are evaluated as to (1) their financing and economic effects, (2) their effects on individuals, and (3) their feasibility, we will have a good foundation for devising agreeable solutions, perhaps not in every detail, but as an overall reform package that will meet the most important of our objectives. | Social Security not only represents the foundation of our retirement income system; it also provides millions of Americans with disability insurance and survivor's benefits. Although the Social Security Trustees now project that under the intermediate or "best estimate" assumptions the combined Social Security Trust Funds will be exhausted 3 years later than in last year's estimates, the magnitude of the long-term funding shortfall is virtually unchanged. Without reform, Social Security, Medicare, and Medicaid are unsustainable, and the long-term impact of these entitlement programs on the federal budget and the economy will be dramatic. Social Security reform is part of a larger and significant fiscal and economic challenge. Absent reform, the nation will ultimately have to choose between persistent, escalating federal deficits, significant tax increases, or dramatic budget cuts. Focusing on trust fund solvency alone is not sufficient. Aiming for sustainable solvency would increase the chance that future policymakers would not have to face, on a recurring basis, the difficult questions of whether the government will have the capacity to pay future claims or what else will have to be squeezed to pay those claims. Comparing the beneficiary impact of reform proposals solely to current Social Security promised benefits is inappropriate since all current promised benefits are not funded over the longer term. Reform proposals should be evaluated as packages. If the focus is on the pros and cons of each element of reform, it may prove impossible to build the bridges necessary to achieve consensus. Acting sooner rather than later helps to ease the difficulty of change. Waiting until Social Security faces an immediate solvency crisis will limit the scope of feasible solutions and could reduce the options field to only those choices that are the most difficult and could also delay the really tough decisions on Medicare and Medicaid. |
DOD may be unable to prevent an attack using chemical or biological weapons. Therefore, DOD has determined that servicemembers must be protected to survive and conduct effective military operations. Consequently, DOD supplies servicemembers with a protective ensemble consisting of a suit, mask with breathing filter, rubber boots, butyl gloves, and hoods as required. Figure 1 displays the components that comprise a protective ensemble. During Operation Desert Shield/Desert Storm, DOD noted that most of this equipment (1) could cause unacceptable heat stress to the wearer, (2) could limit freedom of movement and impair job performance, (3) is bulky, and (4) is not fully interoperable across the services. Furthermore, most of the existing suits (1) are no longer manufactured, (2) can be used for up to 14 years from the date of manufacture, and (3) will expire by 2007. To address these issues, DOD developed new, lightweight individual protective equipment such as the Joint Service Lightweight Integrated Suit Technology (JSLIST) trousers and coat to replace the current protective suits. DOD began procuring the JSLIST suits in 1997. An improved multipurpose overboot is in procurement and new protective gloves are under development to improve manual dexterity and/or reduce heat stress on the wearer. Similarly, since the existing masks may cause some breathing difficulty, DOD is developing a new mask but does not expect to begin procurement until fiscal year 2006. During fiscal years 2002 through 2007, DOD plans to spend about $5.7 billion on planning for chemical and biological defense, acquisition of defense equipment, facilities construction, and research and development. In 1999, we recommended that DOD develop a performance plan guided by outcome-oriented management principles embodied in the Government Performance and Results Act of 1993 (Pub. L. 103-62). DOD created a plan; however, performance goals and measures were being developed at the time of our review. DOD’s assessment process for determining the risk to military operations is unreliable, and, as a result, the Department’s current determination that the risk is generally low is inaccurate. Although the Department uses the Chemical and Biological Defense Program Annual Report to Congress to indicate its readiness for operations in a chemically and biologically contaminated environment, the 2000 report contains erroneous inventory data and understates equipment requirements. More important, the methodology for assessing the risk is flawed because it is not based on the number of complete ensembles needed and it obscures military service readiness by combining service data and reporting the results jointly. DOD’s criteria for assessing the risk of wartime shortages is to determine the numbers of protective suits, masks, breathing filters, gloves, boots, and hoods it has on hand, compare them against the requirements for those individual items, and then assign risk. (See table 1) In the draft fiscal year 2001 annual report to Congress provided to us in June 2001, DOD reported that it was generally at low risk for suits.However, the risk assessment process was flawed in part because DOD used erroneous data on protective suits. For example DOD made computational errors in comparing the older suits and JSLIST suits against a combined total suit requirement, the Air Force overreported its suit requirement by 801,167 suits, DOD reported it had 1,229,935 JSLIST suits on hand as of September 30, 2000, but overcounted its inventory by 782,232 JSLIST suits, the Navy included about 117,000 suits that had passed their expiration dates and were therefore unusable, and the Army underreported its suit stocks by an estimated 231,050 suits. These errors occurred in large part as the result of problems in DOD’s systems for managing protective suit inventories. We believe that the services collectively had no more than 4,348,999 suits of all types on hand as of September 30, 2000. When we included all the suits for wartime use and adjusted the numbers to account for the errors and miscounts, the risk category changed to high for suits, as shown in table 2. In February 2001, the military services informed the Joint Nuclear, Biological, and Chemical Defense Board that equipment requirements were actually much higher than those reported to Congress and included in the fiscal year 2001 Logistics Support Plan. The board subsequently accepted the new requirements. Based on these new requirements, the risk remained high for suits; changed to high for filters, boots, and hoods; and remained low for gloves and masks, as shown by comparing the risk level columns in tables 2 and 3. DOD has inaccurately assessed the risk to military operations by determining the number of individual items of equipment it has on hand and by combining the services’ inventories of individual items. Service guidance specifies that a total of 1,573,866 active and reserve servicemembers need protection to meet current operations plan requirements. DOD provides each deploying servicemember with up to four ensembles either at deployment or held in war reserve and distributed to theater operating forces when needed. The ensembles consist of five components: (1) four protective suits, (2) between four and eight pairs of gloves and boots, (3) between four and eight hoods, (4) up to four breathing filters, and (5) one mask. Because DOD does not report each service’s readiness based on the equipment it has on hand, but rather provides a joint assessment, critical service shortages or opportunities for cross-service assistance tend to be obscured. In fact, each service reported shortages of one component of the ensemble. Specifically, the Army reported critical shortages of hoods; the Air Force reported shortages of gloves; the Navy, shortages of suits; the Marine Corps, shortages of boots. When we compared the number of ensembles required by each service’s guidance and applied the DOD risk criteria, the risk was high for all four services. As a result, DOD cannot provide all the required ensembles for 682,331 servicemembers scheduled for wartime deployment, as shown in table 4. The risk posed by suit shortages is likely to worsen through 2007 due to increasing rates of older suits’ expiration and DOD’s plan not to replace all of them. As of October 1, 2000, DOD reported a shortage of about 1.7 million protective suits; it believes about 3.3 million, or 75 percent, of the current suit inventory will expire by 2006. JSLIST suits cost about $203 each compared to about $80 each for most of the existing suits, and DOD plans to buy only about 2.8 million JSLIST suits as replacements. Therefore, the shortage will increase to about 2.2 million suits by 2006. DOD’s plan to buy fewer new suits is also influenced by expiration of the suits and budgetary considerations. By replacing suits at a rate slower than the expiration rate, DOD plans to spread future suit purchases over more years to avoid a disproportionately large amount of suits expiring in any one year. This tactic allows greater dispersion of future suit expirations and replacement costs but is likely to also increase the short-term risk of wartime shortages. DOD is attempting to mitigate some of the shortages. For example, the Army plans to procure more than 500,000 hoods through fiscal year 2002, and the Defense Logistics Agency was procuring more of the existing generation of boots at the time of our report. Some opportunities also exist for one service to assist another. For example, the Army and Marine Corps reported significantly more gloves on hand than required and could transfer some to the Air Force to offset Air Force shortfalls since all the services use the same gloves. However, other available equipment is not interoperable and cannot be easily shared. For example, the Navy and Marine Corps suits are hooded, so they do not have separate hoods and therefore cannot help alleviate the Army’s shortage. If all goes according to plan, such interoperability problems should ease after fiscal year 2006, as all four services begin using the JSLIST suit and new joint masks, gloves, and boots. Shortcomings in DOD’s inventory management of chemical and biological protective equipment adversely affect the Department’s ability to accurately assess the readiness of the services to meet requirements for the equipment and mitigate the risk of shortages. DOD’s current inventory information on chemical and biological equipment is unreliable for making an accurate risk assessment because DOD and the services cannot easily link inventory records; lack data on suit expiration dates; cannot easily identify, track, and locate defective suits; and have miscalculated the requirements and the number of suits available. These shortcomings are consistent with long-term problems in DOD’s inventory management that we have consistently identified since 1990 as a high-risk area due to a variety of problems, including ineffective and wasteful management systems and procedures. The Defense Logistics Agency and the military services store war reserve inventories of chemical and biological protective suits and other equipment at a variety of depots, warehouses, and storage facilities and as noted earlier, use at least nine different inventory systems to manage the inventories. However, because these systems are not linked, DOD-wide oversight of the inventories is restricted, and the systems are not used to directly support the inventory data in the annual report to Congress and the Logistics Support Plan. Instead, DOD makes an additional effort to collect data theoretically already in the systems. The data collection requires units and depots that store the chemical protective equipment to provide separate data on the equipment annually and relies heavily on government and contractor personnel to manually compile the data. Although DOD has at least nine major inventory management systems, it cannot accurately determine the expiration rate of most of the older suits used by the Air Force and Army. These account for about 3.3 million of the current suit inventory and can be used for up to 14 years from the date of manufacture after which testing has determined that the suits cannot be used in a contaminated environment. Therefore, knowing the date the suits were manufactured is critical to estimating the suits’ expiration rate and the rate at which the suits must be replaced with JSLIST suits. However, neither DOD nor we can accurately determine the expiration rate of the old suits because the Defense Logistics Agency, the buyer of the suits, was unable to locate most of the relevant procurement records. Moreover, many of the inventory systems cannot be used to locate the actual expired suits in specific depots because the systems do not record equipment expiration dates or the manufacturers’ contract or lot numbers. Two examples or illustrations follow: The Army does not record suit expiration information in its primary inventory management system. To compensate, the Army has assumed an annual 20-percent expiration rate of its inventory through fiscal year 2005 and expects that all suits will expire by 2005. However, the Army’s assumption may be inaccurate. Records from a depot in Kentucky indicate that almost 80,000 suits would be serviceable after 2005 and some as late as 2008. The Navy does not know when its suits will expire because, according to the Naval Sea Systems Command, the Navy does not require inventory managers to include the expiration date in inventory records. Nonetheless, in June 2001, the Navy estimated that of 178,000 suits that it had on hand, only about 61,000 were actually serviceable because the rest had passed their expiration date. Our review of 19 Military Sealift Command ships, which help to sustain deployed U.S. forces, showed that most had severe suit shortages, due mostly to expirations. We found additional problems in 48 ships in the Atlantic and Pacific fleets. These ships currently report that they are missing one or more components of their ensembles and consequently cannot provide a complete ensemble for a single crewmember. The Air Force and Marine Corps use different inventory management systems that include contract, lot, and expiration information. Consequently, these two services can estimate suit expiration rates to manage their inventories effectively. Nonetheless, neither system is compatible with the other DOD systems. The majority of DOD’s and the services’ inventory systems cannot be used to identify, track, and locate defective suits that may be in current inventories because contract and lot numbers needed for the purpose are not always included in the inventory records. In September 1999, officials from one manufacturer pleaded guilty to selling 778,924 defective suits to the government. Since these defective suits were distributed to DOD war reserve and various other inventories, it was imperative that the suits be found. In May 2000, DOD directed units and depots to locate the defective suits and issue them for training use only. At the conclusion of our review, DOD had not found about 250,000 of these suits and did not know whether they had been used, were still in supply, or were sent for disposal. Finding the suits was difficult even when the storage depot was known. For example, the Defense Logistics Agency inventory system does not link the contract and lot number with the box or pallet number to allow ease in locating specific items. Consequently, during our review, the Agency resorted to using 19 reservists for up to 34 days to physically inspect all pallets and boxes containing about 1.3 million protective suits at its depot in Albany, GA. The reservists found about 347,000 defective suits. Figure 2 displays some of the boxes of these suits. Despite the problem in finding defective suits, the Defense Logistics Agency’s supply system remained unchanged at the time of our review. Agency officials acknowledged that they would have to physically reinspect depot stocks if specific lots of other suits need to be removed from the inventory before the end of their normal 14-year shelf life. Several questionable inventory management practices and related actions have further contributed to the generation of the inaccurate inventory data, which in turn affects the accuracy of DOD’s risk assessment process. These include miscalculating suit requirements, failing to count parts of the suit inventory, and counting suits as part of the inventory long before they are actually delivered from manufacturers. Some specifics regarding these counts are as follows: The Air Force double-counted a portion of its suit requirement by reporting a requirement for both 801,167 of the older suits and the same number of replacement JSLIST suits. The Army asked units that store suits to report the numbers being stored, but it did not tell them to include desert pattern suits, which are generally reserved for use in desert climates. As a result, units did not always include desert pattern suits in their reported inventories, and the Army believes it consequently underreported its desert pattern suit inventory by 10 percent of the total, or 231,050 suits. In the fiscal year 2001 Logistics Support Plan and draft Annual Report to Congress, the services reported they had 1,229,935 JSLIST suits on hand on September 30, 2000, but that included 782,232 suits not yet delivered. DOD procedures for compiling inventory data for these reports allow reporting suits expected to be delivered during the year as on hand. In March 2001, the Marine Corps Systems Command, which manages JSLIST suit distribution, acknowledged that DOD did not have 1,229,935 JSLIST suits on hand on September 30, 2000 but might reach that quantity a year later on September 30, 2001. Moreover, in the same two reports, DOD projected that it would reach 1.5 million suits by September 30, 2001, again overestimating JSLIST production. DOD’s inventory management practices tend to affect the suit inventory count. This count in turn can significantly affect the results of the risk assessment process, which is a comparison of requirements against the inventory on hand. Because the Department of Defense’s risk assessment process is flawed and unreliable, DOD inaccurately assessed the risk to servicemembers’ lives and military operations from potential wartime shortages of protective equipment as low. The Department underestimated the risk by analyzing requirements based on individual equipment items and not ensembles. Furthermore, DOD combined this service data into a consolidated DOD inventory position, which obscured service-specific shortages. As we discovered, the risk is currently higher than reported by DOD. Inadequate inventory management has contributed to increased risk. Because the Department has no integrated inventory system for managing protective equipment, it has no effective way to (1) gather the data needed for the annual report to Congress and Logistics Support plan, (2) determine the expiration dates of protective equipment, and (3) ensure that its data is correct. To further compound the problem, the services have counted equipment as on hand before it has been delivered, adding to the overcounting of equipment that they had in the inventory. Inaccurate risk assessment and inadequate inventory management could adversely affect readiness and prevent informed acquisition decisions that could undermine risk mitigation. To improve the Department of Defense’s ability to accurately assess the level of risk and readiness for operations in a contaminated environment, we recommend that the Secretary of Defense direct the Under Secretary of Defense for Acquisition, Technology, and Logistics to issue and implement guidance requiring each service to evaluate its risk on the basis of current inventory numbers of complete ensembles against wartime requirements; implement a fully integrated inventory management system to manage chemical and biological defense equipment and use it to prepare (1) the required annual report to Congress and (2) the annual Logistics Support Plan on chemical and biological defense; establish data fields in the inventory management system to show the contract, lot number, and expiration date of shelf life items; and cease counting equipment as on hand before delivery from the contractor. DOD provided written comments on a draft of our report and generally concurred with our recommendations. DOD partially concurred with our recommendation to conduct risk assessments on the basis of ensembles required in wartime and not just components of the ensemble and stated that the department will issue implementing guidance. DOD concurred with comment with our recommendations to (1) establish an integrated inventory management system; (2) include item contract, lot number, and expiration date information in the new inventory system; and (3) cease counting equipment as on hand before it is delivered and explained its plan to implement the recommendations. In addition, DOD provided technical comments, which we incorporated into our report as appropriate. DOD’s comments are printed in their entirety in appendix II along with our evaluation of their comments. We discuss our scope and methodology in detail in appendix I. We conducted our review from August 2000 to April 2001 in accordance with generally accepted government auditing standards. We will send copies of this report to interested congressional committees; the Secretaries of Defense, the Army, the Navy, and the Air Force; the Commandant of the Marine Corps; and the Director of the Office of Management and Budget. If you or your staff have any questions about this report, please contact me at (202) 512-6020. Additional contact and staff acknowledgments are listed in appendix III. We determined (1) whether DOD’s process for assessing the risk to military operations on the basis of wartime equipment requirements is reliable and (2) how DOD’s inventory management of chemical and biological protective gear has affected the risk level. We included in our scope chemical and biological protective suits, masks and breathing filters, gloves, boots, and hoods. To understand the process DOD uses to assess the risk, we determined how DOD performs risk assessments. We examined DOD’s fiscal years 1999, 2000, and 2001 Chemical and Biological Defense: Annual Report to Congress and Joint Service Nuclear, Biological, and Chemical Defense Logistics Support Plan: Readiness and Sustainment Status and service input to these reports. To understand equipment requirements, we interviewed an official from the Office of the Deputy Assistant to the Secretary of Defense, Chemical and Biological Defense; the Joint Nuclear, Biological, and Chemical Defense Board; the Joint Staff; and other organizations and obtained documents showing how many suits, masks, breathing filters, gloves, boots, and hoods are needed to support operations. We also obtained the Center for Army Analysis’ Joint Service Chemical Defense Equipment Consumption Rates IV, Volume II; briefing slides; guidance; directives; memorandums; cables; and other documents that specify requirements. We also used service guidance to determine the number of servicemembers scheduled for deployment who need protection. We did not evaluate the validity of the requirements. To calculate on-hand stocks, we obtained inventory records from war reserve or other depots in the United States, Japan, the Republic of Korea, the Netherlands, elsewhere in Europe, and aboard prepositioned ships in Guam to determine the size of the stockpile. As a result of the national security reviews under way at the time of our review, requirements for chemical defense equipment could change. If so, current risk assessments would need revision. To determine how DOD’s inventory management practices affected risk, we tried to verify the accuracy of inventory data reported by the services. We did this by (1) interviewing officials and obtaining documents showing how the inventory data were collected and verified, (2) obtaining Navy documents showing the number of suits still in the inventory that had not expired and comparing that number to the reported inventory, and (3) obtaining JSLIST suit production data. We also tried to determine how many of the older chemical protective suits DOD had bought and when, but the Defense Logistics Agency could not find most of its records documenting suit procurement. To determine the compatibility of the nine major supply systems, we interviewed the responsible DOD officials, compared system inventory procedures, checked records against physical inventories, and obtained relevant documents. To determine how long shelf life items can be used and to estimate equipment expiration rates, we interviewed officials from the Army’s Soldier Biological and Chemical Command in Maryland; Natick Soldier Center in Massachusetts; and Rock Island Arsenal in Illinois; the Naval Sea Systems Command in Virginia; the Air Force Headquarters Directorate of Supply in Washington, D.C.; and the Marine Corps’ Combat Development Command in Virginia and Materiel Command in Georgia. We also interviewed officials and obtained documents from the Defense Logistics Agency offices in Pennsylvania showing planned or actual procurement of JSLIST suits and other equipment. To determine how the services and depots identify which items will expire and need replacement, we inspected or inventoried chemical protective suits stored at the Bluegrass Army Depot in Richmond, KY; the Defense Logistics Agency’s war reserve depot in Albany, GA; the Air Force’s Mobility Bag Center in Avon Park and MacDill Air Force Base, FL; and aboard ships at the Norfolk Navy Base, Norfolk, VA. At these locations, we met with officials and obtained supply records and suit and other equipment expiration data. The following is our response to the Department of Defense letter dated September 18, 2001. 1. While DOD presents the data in the cited Annual Report to Congress and Logistics Support Plan annexes, the data is presented on an item-by- item basis and not an ensemble basis. Consequently, the information as presented does not give a fully reliable risk assessment. DOD acknowledges that it has scarce resources and must manage risk within those resource constraints. Consequently, DOD also indicated in its comments that it will rely on industrial surge capacity to make up any shortfall in required ensemble components. Nonetheless, the Department’s risk assessment is based on having 120 days of supply at the units or in war reserve. If the Department now plans to stock fewer than 120 days of supply and rely on industrial surge to make up the difference in a crisis, the risk level would be higher because the continuing shortages would be greater. 2. The Air Force has developed, and the Marine Corps is developing inventory systems, both of which include contract, lot number, and expiration date of equipment on hand. Adopting one of these systems DOD-wide could reduce or eliminate development costs associated with the Business System Modernization program, assure interoperability across the services, and meet the intent of our third recommendation. In addition to the contact named above, Brian J. Lepore, Raymond G. Bickert, Tracy M. Brown, and Sally L. Newman made key contributions to this report. Major Management Challenges and Program Risks: Department of Defense (GAO-01-244, Jan. 2001). Chemical and Biological Defense: Units Better Equipped but Training and Readiness Reporting Problems Remain (GAO-01-27, Nov. 2000). Chemical and Biological Defense: Program Planning and Evaluation Should Follow Results Act Framework (GAO/T-NSIAD-00-180, May 2000). Chemical and Biological Defense: Observations on Nonmedical Chemical and Biological R&D Programs (GAO/T-NSIAD-00-130, Mar. 2000). Chemical and Biological Defense: Chemical Stockpile Emergency Preparedness Program for Oregon and Washington (GAO/NSIAD-00-13, Oct. 1999). Chemical and Biological Defense: Observations on Actions Taken to Protect Military Forces (GAO/T-NSIAD-00-49, Oct. 1999). Chemical and Biological Defense: Program Planning and Evaluation Should Follow Results Act Framework (GAO/NSIAD-99-159, Aug. 1999). Chemical and Biological Defense: Coordination of Nonmedical Chemical and Biological R&D Programs (GAO/NSIAD-99-160, Aug. 1999). Chemical and Biological Defense: DOD’s Evaluation of Improved Garment Materials (GAO/NSIAD-98-214, Aug. 1998). Chemical and Biological Defense: Observations on DOD’s Plans to Protect U.S. Forces (GAO/T-NSIAD-98-83, Mar. 1998). Assuring Condition and Inventory Accountability of Chemical Protective Suits (D-2000-086, Feb. 25, 2000). M41 Protection Assessment Test System Capabilities (99-061, Dec. 24, 1998). Unit Chemical and Biological Defense Readiness Training (98-174, July 17, 1998). Inventory Accuracy at the Defense Depot, Columbus, Ohio (97-102, Feb. 27, 1997). Army Protective Mask Requirements (95-224, June 8, 1995). | The Department of Defense (DOD) believes it is increasingly likely that an adversary will use chemical or biological weapons against U.S. forces to degrade superior U.S. conventional warfare capabilities, placing servicemembers' lives and effective military operations at risk. To reduce the effects of such an attack on military personnel, DOD has determined the quantity of chemical and biological protective suits, masks, breathing filters, gloves, boots, and hoods that are needed based on projected wartime requirements. DOD's assessment process is unreliable for determining the risk to military operations. DOD's 2000 report is inaccurate because it includes erroneous inventory data and wartime requirements. Inadequate inventory management is an additional risk factor because readiness can be compromised by DOD's inventory management practices, which prevent an accurate accounting of the availability or adequacy of its protective equipment. |
STB is an independent agency administratively housed within the Department of Transportation. The successor to ICC, STB is responsible for the economic regulation of railroads and certain pipelines, as well as some motor carrier and water carrier activities. STB has fewer responsibilities and functions than ICC. STB’s fiscal year 1998 budget is nearly $15.9 million, and it employs about 135 full-time equivalent staff, compared with ICC’s fiscal year 1994 budget of $52.2 million and employment of 615 full-time equivalent staff. STB’s current authorization ends on September 30, 1998. Most of STB’s regulatory oversight centers on the rail industry. This oversight encompasses enforcement of the “common carrier” obligation (that the rates, services, and practices of carriers be reasonable), mergers and acquisitions, and the construction and abandonment of railroad lines. The ICC Termination Act eliminated, among other things, the regulatory requirement to file tariffs listing rates charged for transporting goods and requirements pertaining to contracts for the shipment of nonagricultural commodities. Rail issues constitute the majority of STB’s workload. In fiscal year 1997, STB dedicated 116 of its 131 full-time equivalent staff (89 percent) to rail issues. Of the 1,429 decisions STB issued in fiscal year 1997, 988 (about 70 percent) concerned rail issues. STB has jurisdiction over pipelines that provide interstate transportation of commodities other than oil, gas, or water. We identified 21 pipelines carrying five commodities—anhydrous ammonia, carbon dioxide, coal slurry, phosphate slurry, and hydrogen—that are subject to STB’s regulation. STB’s regulation of these pipelines includes ensuring that pipelines fulfill their common carrier obligations, including determining if the rates charged for these services are reasonable and nondiscriminatory. The ICC Termination Act limited STB’s role in regulating pipeline rates by specifying that STB can begin a pipeline rate investigation only in response to a complaint by a shipper or other interested party. The act also eliminated the sole reporting requirement for pipeline carriers—tariff filing. According to STB, over the past 10 years only five cases involving pipelines have come before STB or ICC; one is ongoing. Because of the limited caseload, STB issued only six decisions on pipeline cases in fiscal year 1997 and devoted the equivalent of about one full-time staff member to pipeline issues. STB also has regulatory authority over some motor carrier functions. This oversight includes regulating the rates of household goods carriers and disposition of motor carrier undercharge cases. The ICC Termination Act eliminated some requirements for motor carriers, including tariff filing for most carriers, and transferred responsibility for others, such as registration and insurance, to the Federal Highway Administration. Finally, STB has jurisdiction over domestic water carrier transportation to or from Alaska, Hawaii, or territories and possessions of the United States. This regulation is limited to tariff filing and rate regulation. In fiscal year 1997, STB dedicated about 12 full-time equivalent staff to motor carrier and water carrier issues, primarily motor carrier issues. STB issued 420 decisions (about 29 percent of its workload) on these issues, most related to motor carrier issues. Historically, the federal government has regulated industries engaged in interstate commerce—including common carrier pipelines—with inherent cost advantages that may limit competition from other pipelines as well as other modes of transportation. Specifically, because pipelines are expensive to build—but relatively inexpensive to operate—it is more efficient to build one large pipeline to transport a given amount of a commodity rather than two or more smaller pipelines. In addition, low operating costs may enable a pipeline to reduce its rates temporarily if faced with competition from other modes of transportation. The regulation of pipelines has been imposed to enforce the common carrier obligation, including ensuring that, in the absence of competition, pipeline carriers do not charge unreasonably high rates relative to their costs The federal economic regulation of interstate pipelines is provided by two agencies: the Federal Energy Regulatory Commission, which regulates oil and gas pipelines, and STB, which regulates the remaining pipelines. Regulation by the former covers more pipeline miles and involves more reporting requirements than the latter. For example, about 400,000 miles of oil and gas pipelines are under the jurisdiction of the Federal Energy Regulatory Commission, while fewer than 6,000 miles of pipelines are subject to STB’s jurisdiction. In addition, oil and gas pipeline carriers are generally required to file tariffs and annual reports, while pipeline carriers under STB are not. The ability of alternatives to pipelines—local production within the Midwest, as well as barge and rail transport from other areas of the United States—to compete with pipelines within local market areas in the Midwest depends on two factors. First, because storage terminals are key to the distribution of anhydrous ammonia in local midwestern market areas, alternatives must have access to storage terminals within market areas that are also served by pipelines. We found that, while some local market areas currently served by pipelines also have access to alternatives, other market areas may not. Second, alternatives to pipelines must have the ability to increase their supply of anhydrous ammonia to serve these markets. We found that alternatives may not offer effective competition to pipelines because they have limited ability to increase their supply of anhydrous ammonia without additional investments in capital. Because of the large number of local markets that exist along the two midwestern anhydrous ammonia pipelines, we were not able to definitively determine the number of markets that do or do not have competitive alternatives to pipelines. Two pipelines, one owned by Koch Pipeline Company, L.P., and one owned by MAPCO Ammonia Pipeline, Inc., carry anhydrous ammonia from Louisiana, Oklahoma, and Texas to the midwestern states. (See fig. 1.) These pipelines supplied 2.1 million tons (33 percent) of the estimated 6.4 million tons of anhydrous ammonia used in the Midwest in 1996. Three alternatives to pipelines—local production within the Midwest; barge shipments from Louisiana up the Mississippi, Illinois, and Ohio Rivers; and rail shipments primarily from other areas—also provide anhydrous ammonia to the Midwest. Local production accounted for about 3 million tons, or about 47 percent, of the total midwestern demand. Barge shipments accounted for 0.9 million tons (14 percent) and rail shipments accounted for 0.4 million tons (6 percent). The highly seasonal demand—lasting as little as 10 days each in spring and fall—for anhydrous ammonia applied directly to fields as a fertilizer makes it important to have large amounts of this product stored close to farms if farmers’ needs are to be met. Regardless of the means of transport, the most efficient way to meet this demand is to have large storage tanks (generally from 20,000 to 40,000 tons of anhydrous ammonia per tank) in terminals located close to fertilizer dealers and farmers throughout the Midwest. As a result, anhydrous ammonia markets in the Midwest appear to be fairly localized. Currently, 60 terminals throughout the Midwest—28 of which are located on pipelines—store anhydrous ammonia for peak-season use. Currently, more than half of the 28 terminals located on pipelines have no alternatives to the pipelines. (See tbl. 1.) The remaining terminals have access to alternatives that may limit the pipelines’ ability to charge high rates to deliver the product to that terminal. Some of the 32 terminals not on the pipelines may also be able to supply anhydrous ammonia to fertilizer dealers in a pipeline terminal’s market area and effectively limit pipelines’ ability to charge high rates. For example, if the price of anhydrous ammonia were to increase at a pipeline terminal in response to higher shipping rates on the pipeline, fertilizer dealers in the area could turn to cheaper sources of anhydrous ammonia—such as terminals served by barge, rail, or local production—if available. If these other sources could increase their supply without incurring significant increases in their transport and storage costs—thus enabling them to keep their prices steady—the pipeline might be forced to keep its rates reasonable in order to retain customers. However, the ability of these alternative sources to expand their supply of anhydrous ammonia beyond current levels without additional investment may be limited. It is unlikely that plants devoted to producing anhydrous ammonia as a first step in manufacturing other forms of fertilizer will switch to producing anhydrous ammonia for direct application. According to representatives from barge companies and barge terminals, the current fleet of barges is operating at or near capacity and terminals also have limited excess capacity. Fertilizer dealers and shippers were also skeptical about the ability of rail to expand capacity to compete with the volume of product currently provided by the pipelines. Expanding capacity in any of these modes could be expensive. For example, new barges are estimated to cost between $4 million and $5 million each, while new barge terminals cost approximately $15 million. As an alternative to the direct application of anhydrous ammonia, farmers could substitute other forms of nitrogen fertilizer. This action would lessen the need to have large amounts of anhydrous ammonia shipped to the Midwest. However, it is not clear that other nitrogen fertilizers are substitutable for anhydrous ammonia. For example, of the nitrogen fertilizers, anhydrous ammonia is best suited for fall application because it loses little of the nutrient during the winter compared with other forms of nitrogen fertilizers. In addition, anhydrous ammonia is relatively low cost and is the most concentrated form of nitrogen available. For example, in April 1997, the cost to farmers of nitrogen in anhydrous ammonia form—82-percent nitrogen content—was $369 per ton, while the cost of nitrogen in a liquid upgrade form—28- to 32-percent nitrogen content—was $533 per ton. Because an increase in pipeline transportation rates would represent only a small portion of the cost of anhydrous ammonia to farmers, a substantial increase in pipeline rates would be required before farmers would be likely to switch. No clear conclusions can be reached on whether the continued economic regulation of pipelines under STB’s jurisdiction is needed because such a determination requires the examination of competition in numerous local markets along 21 pipelines. Such an examination was not feasible for our study, nor was it feasible to address whether anhydrous ammonia pipelines are representative of all pipelines under STB jurisdiction. There will be several issues before the Congress as it decides whether to extend, modify, or rescind STB’s authority to regulate pipelines. These issues deal with whether to substantively change or leave in place how STB regulates pipelines. We have not addressed whether the current approach to the economic regulation of pipelines might remain substantially unchanged but carried out by another agency. First, do pipelines under STB’s jurisdiction lack effective competition in a significant number of market areas and have the potential to charge unreasonably high rates? Whether pipelines under STB’s jurisdiction have such power is uncertain. As discussed above, limited competition may exist in a number of anhydrous ammonia markets on two pipelines in the Midwest while other markets may have sufficient alternatives to constrain pipeline rates. However, according to a 1986 report from the Department of Justice, all markets along a pipeline do not necessarily have to be competitive in order to justify deregulation of the pipeline. Instead, Justice concluded that the number of markets along a pipeline that do not have competitive alternatives—and therefore require regulation—should be balanced against the costs of regulating that pipeline. Second, are the costs of regulation burdensome to pipeline carriers? The regulatory requirements imposed on pipeline carriers do not appear to be high. STB does not have the authority to initiate rate cases. In addition, STB does not require that pipelines file rate schedules, nor does it impose reporting requirements on pipelines wanting to start up or go out of business. In fiscal year 1997, STB devoted the equivalent of about one full-time staff member to pipeline issues. If a rate case is brought before STB, the cost to the pipeline carrier of defending the case could be substantial. The limited number of pipeline rate cases in STB’s and ICC’s history provides little basis for estimating the cost of these cases. However, STB officials told us that the cost of rail rate cases ranges from less than $50,000 up to about $1 million. Third, does the limited number of pipeline rate cases indicate there is no need for continued regulation? It is possible that the very limited number of rate cases brought before STB and its predecessor in the last 10 years is evidence of effective competition and therefore there is no need to continue pipeline regulation. However, some shippers we talked to contend that the mere existence of a federal regulatory agency with the authority to roll back rate increases acts as a deterrent to unfair rate increases. Finally, would shippers have recourse if STB’s economic regulation of pipelines was eliminated? Absent STB or any other regulatory body, shippers that believe they are being charged unfair rates would presumably complain to the Department of Justice or the Federal Trade Commission. However, neither of these agencies currently has the statutory authority to investigate complaints from shippers that believe they are being charged rates that are unreasonable or discriminatory, unless the complaint alleges a violation of antitrust laws. This concludes our statement. The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | GAO discussed the regulatory role of the Surface Transportation Board (STB), focusing on: (1) STB's responsibilities in regulating surface transportation; (2) the historical reasons for the economic regulation of pipelines; (3) the ability of alternatives to anhydrous ammonia pipelines to compete in the Midwest; and (4) issues before Congress as it decides to extend, modify or rescind STB's authority to regulate pipelines. GAO noted that: (1) STB is responsible for the economic regulation of railroads and certain pipelines, as well as some aspects of motor carrier and water carrier transportation; (2) the majority of STB's resources and workload are devoted to examining rail issues; (3) in fiscal year 1997, STB dedicated 89 percent of its staff years to rail issues and less than 1 percent to pipeline issues; (4) historically, the federal government has regulated the rates charged by interstate pipelines because pipelines have inherent cost advantages that may limit competition from other pipelines as well as from other modes of transportation; (5) two federal agencies--STB and the Federal Energy Regulatory Commission--regulate pipelines; (6) this regulation includes ensuring that all shippers have access to pipeline transportation services and that the rates charged by pipeline carriers for these services are reasonable and nondiscriminatory; (7) the ability of alternatives to anhydrous ammonia pipelines to compete with pipelines in the Midwest varies, depending on these alternatives' access to the market areas served by pipelines and their ability to increase their supply of anhydrous ammonia to compete within those market areas; (8) GAO's work showed that some market areas currently served by pipelines also have access to alternatives, while other market areas may not; (9) however, even where alternatives to pipelines are available, they may not offer effective competition because they have limited ability to increase their supply of anhydrous ammonia without additional investments in capital; (10) because of the large number of local markets that exist along the two midwestern anhydrous ammonia pipelines, GAO was not able to definitively determine the number of markets that do or do not have competitive alternatives to pipelines; (11) no clear conclusions can be reached on whether continued economic regulation of pipelines under STB's jurisdiction is needed because such a determination requires the examination of competition in numerous local markets along 21 pipelines; and (12) however, as Congress considers reauthorizing STB, pipeline regulation issues to consider include: (a) whether pipelines do not face effective competition in a significant number of market areas and subsequently have the potential to charge unreasonably high rates; (b) what the costs of regulating pipelines are; (c) whether the limited number of pipeline cases before STB and its predecessor indicates there is no need for regulation; and (d) whether shippers would have any recourse if STB's economic regulation of pipelines was eliminated. |
OJJDP, within the U.S. Department of Justice, is the primary federal agency responsible for the prevention and control of juvenile delinquency in the United States. This includes preventing, treating and controlling youth gang activity and violence through sponsored research, evaluation, and demonstrating new approaches for communities across the country. The initiatives are carried out by seven components within OJJDP: Research and Program Development Division, Training and Technical Assistance Division, Special Emphasis Division, State Relations and Assistance Division, Information Dissemination Unit, Concentration of Federal Efforts Program, and Child Protection Division. OJJDP annual appropriations have more than tripled from $162 million in fiscal year 1996, with 71 authorized staff positions, to $568 million in fiscal year 2000, with 87 authorized staff positions. In 1996, we reviewed the operations of OJJDP. We found that official grant files for discretionary grants generally contained monitoring plans, but little evidence that monitoring occurred. We reported that none of the grant files had documentation of telephone contacts, site visits, or product reviews. Also, no quarterly program and financial reports were included in 11 of the 78 files for which projects had been ongoing for at least 2 quarters at the time of our review, and one or more reports were missing from another 61 files. In addition, only 6 of the 78 files had program and financial reports for all quarters of work completed. In commenting on our testimony, the OJJDP Deputy Administrator said that the heavy workload of OJJDP staff may have resulted in a lack of monitoring records and that, as a result of the our 1996 review, OJJDP would take the steps necessary to improve records. To address the first objective, to report on OJJDP’s grant monitoring process, we reviewed applicable federal laws and regulations as well as OJP and OJJDP policies and procedures. We also interviewed staff in OJJDP and OJP’s Office of the Comptroller. To accomplish the second objective, to report on OJJDP’s documentation of its discretionary grant monitoring activities, we reviewed probability samples of official grant files and grant managers’ files using a data collection instrument to record whether required monitoring activity occurred—that is, that the progress reports, financial reports, and other required documents were included in the files. For each grant, we also reviewed the most recent award covering 12 months to examine specific monitoring requirements and activities, such as telephone calls and site visits. Specifically, we reviewed a random sample of 89 of 545 OJJDP demonstration discretionary grants, and 45 of 83 training and technical assistance discretionary grants that were active in all of fiscal years 1999 and/or 2000. The results of these samples are representative of the populations from which they were drawn. We express our confidence in the precision of our sample results as a 95-percent confidence interval. Unless otherwise noted, all percentage estimates have confidence intervals plus or minus 10 percentage points or less. To determine the timeliness of grantee progress reports and financial status reports, we reviewed all such reports for OJJDP demonstration and training and technical assistance discretionary grants from the initiation of the grant through the date of our review. For our review, we provided a list of the selected grants to OJJDP and OJP’s Comptroller. Because we did not always receive the grant files in the order requested, we reviewed them out of order. When files were missing or lost, we substituted replacement files. After our review was complete, some of the missing files were located. We did not review the missing files but used the substitute files that were a part of our random sample. Four grant files were not found in time to be included in our review. In addition, when only one of the two files (the official grant file or the grant managers’ file) was available, we used the available file to complete the data collection instrument. We have no reason to believe that the substitution of files affected the validity of our sample because they were a part of our random sample. To determine whether OJJDP grant closeout procedures were implemented in accordance with OJP policy, we reviewed grant files in our sample whose project periods expired before August 31, 2000. For our review, we focused on required closeout documentation, such as final progress or performance reports, final financial status reports, grant checklist, and closeout notification and grant adjustment notice. Our results should not be projected to the universe of cases in our sample. To address the third audit objective, to report on the processes OJJDP uses to oversee monitoring activities, we examined OJJDP’s monitoring and oversight of its grant managers’ compliance with applicable laws, regulations, and procedures for monitoring. We obtained information from OJJDP officials at headquarters and obtained data related to these activities, such as performance work plans. To determine the requirements in the performance work plans, OJJDP officials gave us a plan that they considered to be representative. Our work was done between March 2001 and September 2001 in accordance with generally accepted government auditing standards. After OJJDP awards discretionary grants, OJP policies require OJJDP to monitor the grants and related activities and document the monitoring results in the grant managers’ program files and OJP’s Office of the Comptroller official grant files. The monitoring is to ensure compliance with relevant statutes, regulations, policies, and guidelines; responsible oversight of awarded funds; implementation of required programs, goals, objectives, tasks, products, time lines, and schedules; identification of issues and problems that may impede grant implementation; and adjustments required by the grantee as approved by OJJDP. One of the OJJDP grant managers’ main responsibilities is project monitoring. Each grant manager prepares a monitoring plan as part of a grant manager’s memorandum recommending initial or continuation funding. The level of monitoring required is based upon the stated monitoring plan in the grant manager’s memorandum. The memorandum includes an overview of the project; a detailed description of what type of activities the grantee plans on implementing; a discussion of past monitoring activities and assessments, if the grant is a multiyear grant that is awarded yearly; an identification of monitoring activities to be performed for the current a discussion of the financial justification for the grant funds and of the cost-effectiveness evaluation of the application. Another part of the monitoring plan may list specific activities or milestones to be monitored by the grant manager. The 1991 OJJDP Monitoring Desk Book contains specific monitoring procedures that grant managers used for monitoring OJJDP’s discretionary grants. Monitoring activities documented included site visit reports and quarterly monitoring telephone calls as well as other significant telephone calls. Also, according to the OJP 1992 Handbook, all discretionary grantees are required to submit categorical assistance progress reports that summarize project activities and quarterly financial status reports for review by grant managers. The OJP Handbook also stated that when all of the grant funds are spent, all grant activities are completed, and all grantee and grant managers’ administrative requirements, including all paperwork and review of grant files, are completed, the grant managers have to close out the grant. (See app. I for a detailed discussion of the grant award processes.) The OJP Comptroller’s Control Desk is to maintain the official grant files and is responsible for tracking the receipt of all grant documents. The Control Desk receives the progress reports, files the original in the official grant file, and forwards a copy to the cognizant program office. The grant manager is responsible for the timely acceptance, review, and analysis of progress reports. The Comptroller’s Monitoring Division performs an in-house review of various program and financial documents contained in the official grant files. The Office of the Comptroller also performs risk-based on-site financial reviews of grantee organizations to monitor administrative and financial capability. The Comptroller applies risk-based criteria to a universe of grants to develop a sample for each fiscal year monitoring plan. The risk-based criteria include the dollar amount of the grant, new grantees, new grant programs, and programs with known problems. The Comptroller excludes from its sample any grantee organizations that were subject to its other on-site financial monitoring or to audit by the Office of the Inspector General during the last 2 years. According to OJP officials, the purpose of the reviews is to provide as much of a comprehensive financial review as possible without going on site. Our review of the results of the Comptroller’s in-house financial reviews of the grants in our sample disclosed that they identified a number of problems similar to our own observations. These included missing program and financial documents and program and financial monitoring not always being documented in a timely manner. In commenting on a draft of this report, the Acting Assistant Attorney General stated that there was no mention of delinquent documentation of financial monitoring during our review, nor was there any evidence cited in our draft report to support the contention that financial monitoring was not documented in a timely manner. However, contrary to the Acting Assistant Attorney General comment, our review found that there was no documentation that financial reports were generally reviewed and that many of the reports were not submitted in a timely manner. Specifically, we found one case where no progress or financial reports had been submitted during the life of a grant, and it was not discovered until almost a full year after the award date (see p. 15). In 1991, OJJDP issued a Monitoring Desk Book, which incorporated material from OJP’s 1987 Handbook, including a formula for determining the monitoring level requirements for the grant. According to OJJDP officials, this formula is applied to determine how often a grantee will be visited and officially contacted for monitoring purposes. This formula applies a weight to issues such as program priority (i.e., how important is the grant within the context of the OJJDP’s goals and objectives), size of grant, grantee’s need for assistance, and anticipated difficulties in implementing the approach. The grant manager assigns a weight or “score” to each of these criteria, which are added together for a total score. This score assigns the grant to a category that defines the specific number of on-site and off-site visits and the frequency of monitoring calls. In February 1992, OJP revised its 1987 Handbook and eliminated the score requirement to determine the monitoring level of a project. However, OJJDP grant managers still use the instructions to determine the monitoring level of a project. We found in our file reviews that 72 of the 89 demonstration grant files and 14 of the 45 training and technical assistance grant files had a score or level indicating monitoring requirements. Using the obsolete monitoring requirement to establish grant manager’s grant monitoring responsibilities may not result in the most efficient use of their time or in the needed monitoring to ensure grantee compliance, as pointed out by the Deputy Administrator in May 1996. He said that site visits did not occur because they are expensive and time-consuming. He added that OJJDP might need to revise some procedures, noting that a site visit once every 2 years and some interim telephone monitoring may be more appropriate than annual visits. Current OJP policy provides flexibility to determine monitoring needs (e.g., site visits as appropriate) consistent with resource availability as compared with a monitoring level that is associated with a specific monitoring requirement that may not be achievable with existing resources. Since April 2001, OJJDP has been engaged in a review of its monitoring practices. According to OJJDP officials, this includes a thorough review of its monitoring practices, including improvement of its grant records. OJJDP indicated that a Working Group on Monitoring was created and charged with reviewing its revised OJP Handbook: Policies and Procedures for the Administration of OJP Grants to identify new and modified policies and assess the need for training; to review OJJDP’s own policies and procedures to ensure compliance with those in the revised OJP Handbook; and to determine whether there is a continued need for the OJJDP Monitoring Desk Book (1991), and if so, how it should be updated. In addition, OJJDP recently embarked on an overhaul of its grant closeout practices to improve its monitoring records. In spring 2000, it launched a grant closeout effort jointly with the Office of the Comptroller that is intended to significantly reduce the backlog of grants requiring closeout. As part of this activity, grant files were reviewed, and relevant documentation was collected. OJJDP requires its grant managers to document their monitoring of discretionary grants. This documentation is to include the development and retention of a grant manager’s memorandum for each grant containing a plan to follow for monitoring the project. In addition, progress reports documenting progress achieved in relation to project milestones are required from all grantees. Required documentation also is to include quarterly financial reports from grantees. In addition, OJJDP requires certain documentation at the time of grant closeouts. While the grant files generally contained monitoring plans, OJJDP grant managers were not consistently documenting their monitoring efforts according to the monitoring plans. The grant files generally did not contain documentation showing required telephone contacts and site visits.Further, progress and financial reports did not cover the entire grant period over half of the time. For those grants in our sample that were closed, most did not contain documentation showing compliance with closeout procedures. We identified some of these same documentation- related problems in May 1996. We compared the proposed monitoring activities in the grant manager’s monitoring plan for 89 of 545 OJJDP demonstration discretionary grant files and 45 of 83 OJJDP training and technical assistance discretionary grant files to actual monitoring documentation. Our comparisons of the proposed to actual documented monitoring revealed that OJJDP grant managers were not consistently following the monitoring plan. Telephone Contacts. The grant managers’ monitoring plans for the most recent awards for 79 demonstration grants and 34 training and technical assistance grants had specific requirements for telephone contacts (e.g., monthly or quarterly). There was no documentation showing that the requirement for telephone contacts was met in 96 percent of the demonstration grants and in any of the training and technical assistance grants. Documentation was present that showed that some, but not all, of the required telephone contacts had been made for 49 percent of the demonstration grants and 41 percent of the training and technical assistance grants. Site Visits. The grant managers’ monitoring plans for the most recent awards for 66 demonstration grants and 20 training and technical assistance grants had a specified number of on-site visits to be made. There was no documentation showing that the requirement for site visits was met in 88 percent of the demonstration grants and 90 percent of the training and technical assistance grants. Documentation was present that showed that some, but not all, of the required on-site visits had been made for 6 percent of the demonstration grants and 10 percent of the training and technical assistance grants. When site visits are made to grantees, a written site visit report is to be prepared. Of the awards that had any evidence of site visits, such as memos or e-mails referring to the visit, 7 out of the 12 demonstration grants and 2 out of the 4 training and technical assistance grants had any of the OJP-required, written site visit reports. In commenting on a draft of this report, the Acting Assistant Attorney General informed us that most grant managers prepare site visit reports after an on-site review, as their travel vouchers are not forwarded for payment unless the reports are attached. She noted that the grant managers do not routinely forward copies of the reports to the official grant and program files. She said that a copy of the reports is filed in the grant manager’s travel folders that we did not review. However, according to the OJP Handbook, the original site visit report is to be placed in the official grant file. While the grant managers’ travel folders may contain site visit reports, the official grant files kept by the OJP Comptroller’s Control Desk are the official record of the history and activity of the grant. As such, we reviewed the official grant files as well as the grant managers’ files for required grant documents and not the travel voucher folders. In another comment to a draft of this report, the Acting Assistant Attorney General told us that monitoring plans that call for on-site visits by grant managers are contingent upon staffing capacity and administrative resources and not requirements. Our review indicated that, contrary to the Acting Assistant Attorney General’s comment, certain specific program monitoring requirements are included in the OJP Handbook, such as creating monitoring plans and writing site visit reports within 10 days of a site visit. To be useful, monitoring plans should be feasible. If they are not feasible, then they will not be helpful guidance for monitoring. While it is entirely reasonable to expect that some planned site visits will not take place (for a variety of reasons), the proportion of site visits specified to occur that were not documented seems to indicate that either they occurred but were not documented or monitoring plans for site visits were infeasible. Progress and Financial Reports. According to the OJP Handbook,recipients of discretionary grants are required to submit initial and then semiannual categorical assistance progress reports documenting progress achieved in relation to project milestones. Before December 31, 1995, progress reports were due every quarter. Since March 1, 1996, progress reports have been due every 6 months. This report must be submitted within 30 days after the end of the reporting period for the life of the award. A final report, which provides a summary of progress toward achieving the goals and objectives of the award, significant results, and any products developed under the award, is due 90 days after the end date of the award. We found that in 56 percent of the 89 demonstration grant files and 80 percent of the 45 training and technical assistance grant files reviewed, the progress reports did not cover the entire grant period. In addition, 7 percent of the demonstration grants files and 4 percent of the training and technical assistance grants had no progress reports. We also found that 56 percent of the progress reports submitted for demonstration grants were filed on time; for the other 44 percent, the number of days late ranged from 1 to 545, with 62 percent of the reports for demonstration grants being late up to 30 days and 24 percent between 31 and 90 days. We found that 42 percent of the training and technical assistance grant reports were filed on time; for the other 58 percent, the number of days late ranged from 1 to 297, with 76 percent being late up to 30 days and 15 percent between 31 and 90 days. According to the OJP Handbook, all discretionary grantees are required to submit quarterly financial status reports to OJP’s Office of the Comptroller Control Desk, which maintains the official grant files. Copies are disseminated to the appropriate financial analysts and cognizant grant managers. These quarterly reports are required for all active grants, even if there has been no financial activity during the reporting period. According to OJP officials, reports are considered delinquent, for payment purposes, if not received 45 days after the end of each calendar quarter. The final financial status report is due 90 days after the end of the grant period or extension. Future awards and fund drawdowns may be withheld if the progress and financial status reports are delinquent. We found that in 65 percent of the 89 demonstration grant files and 60 percent of the 45 training and technical assistance grant files reviewed, the financial status reports did not cover the entire grant period. We also found that 67 percent of the financial status reports submitted for demonstration grants were filed on time; for the other 33 percent, the number of days late ranged from 1 to 367, with 54 percent of the reports for demonstration grants being late up to 30 days and 34 percent between 31 and 90 days. We found that 70 percent of the training and technical assistance grant reports were filed on time; for the other 30 percent, the number of days late ranged from 1 to 353, with 72 percent being late up to 30 days, and 19 percent between 31 and 90 days. Further, our review found no documentation in the files that OJJDP grant managers regularly reviewed these reports, as required by OJP. Without such reviews, OJJDP does not have assurances that the grants are being properly administered. For example, one grant we reviewed had no documentation of required semiannual categorical assistance progress reports or quarterly financial reports in the file for the entire life of the award. The project and budget periods ran from October 1, 1999, to September 30, 2000. A $300,000 grant was awarded on March 7, 2000 (6 months after the start of the budget period). The grantee was required to submit an initial and then semiannual categorical assistance progress reports documenting progress achieved in relation to project milestones, starting June 30, 2000, and every 6 months thereafter. The grantee was also required to submit a quarterly financial status report within 45 days after the quarter ended, starting May 15, 2000, and every quarter thereafter. Although no progress or financial reports had been submitted during the life of the grant period, OJJDP did not discover this until almost a full year from the award date. Nevertheless, OJJDP extended the grant period for an additional year to allow the grantee to spend the funds for the project, which it had not drawn down during the grant period. Missing progress reports and quarterly financial reports increase the risk that OJJDP might not be able to identify and rectify grantee’s program and financial problems in a timely manner. In commenting on a draft of this report, the Acting Assistant Attorney General acknowledged this lack of documentation, but did not agree that it meant that OJJDP had not been monitoring grants to ensure they were being properly implemented. From our perspective, this lack of documentation suggests to us that OJJDP does not know the extent to which grant managers were carrying out their monitoring responsibilities to ensure that grants were being properly implemented. Despite the lack of evidence that grant managers were adhering to the established monitoring plans, we did find that some grant monitoring was occurring. For example, the files contained evidence that communication occurred between grantees and the grant managers in the form of reports, faxes, letters, and oral and e-mail communication on specific issues, problems, or requests for information. However, much of the communication between grantees and grant managers was not documented according to OJJDP requirements. Closeout Procedures. The OJP grant closeout process calls for an accounting of programmatic accomplishments in relation to planned activities; an inventory of all required financial, programmatic, and audit reports; and an accounting of all federal funds. According to the OJP Handbook, the grant manager is to complete an assessment report and other closure requirements within 180 days after the end date of the award or any approved extension. These requirements include a closeout checklist, overall assessment of grant performance, final progress reports, final financial status reports, and closeout notification and grant adjustment notice. Our assessment of the closeout process did not include enough cases to ensure that it is representative of all grant closeouts. However, our limited review showed that some grant files did not contain required closeout materials. At the time of our review, the following documents were missing from the files: The required case checklists were not prepared for two of the eight demonstration grants that expired and required closeout. Similarly, neither of the two training and technical assistance grants had the required case checklists. An overall assessment of grant performance was not prepared for two of the eight demonstration grants. Neither of the two training and technical assistance grants had this assessment. Final progress reports were missing from three of the eight demonstration grant files. Neither of the two training and technical assistance grants had final progress reports in the files. The final financial status report was missing from one of the eight demonstration grant files. Neither of the two training and technical assistance grants had a final financial status report in the files. And there was no documentation that grant managers had reviewed any of the reports or other official documents that were filed. A closeout notification and grant adjustment notice to deobligate funds or advise the grantee to retain records for 3 years was not prepared for three of the eight expired demonstration grants. Similarly, neither of the two training and technical assistance grants had a closeout notification and grant adjustment notice. In commenting on a draft of this report, the Acting Assistant Attorney General noted the programmatic and fiscal closure process might not have been completed for those grants until after our review. In our review of closeout procedures, we waited at least the required 180 days before reviewing the grant files to allow sufficient time for OJJDP to complete the grant closeout process. However, the files we reviewed did not contain the required closeout documents. The Comptroller General’s internal control standards require that all transactions and other significant events be clearly documented and that the documentation be readily available for examination. Appropriate documentation is an internal control activity to help ensure that management’s directives are carried out. Without such documentation, OJJDP has no assurance that grants are meeting their goals and funds are being used properly. In commenting on a draft of this report, the Acting Assistant Attorney General also said that almost 50 percent of the 89 Special Emphasis Division grants that we reviewed were awarded in fiscal years 1998 and 1999 under the Drug-Free Communities Support Program. Because of the time needed to recruit, hire, and train new program staff, OJJDP handled 217 Drug-Free Communities Support Program grants in those 2 years with less than adequate (i.e., two) staff. According to the Acting Assistant Attorney General, caseloads of this size make proper documentation a difficult goal to achieve. OJJDP does not systematically monitor grant managers’ compliance with its monitoring requirements or guidance nor does it assess the effectiveness of OJJDP grant monitoring practices. When we asked OJJDP officials about any oversight of OJJDP grant managers to determine the adequacy of their monitoring of OJJDP discretionary grants, they told us oversight takes several forms. They pointed out that the OJP Handbook requires staff performance to be carefully scrutinized against the performance work plans (i.e., the critical performance elements and standards pertaining to monitoring responsibilities). They said that OJJDP supervisory managers are to address the following questions to determine how well each grant manager with monitoring responsibilities is performing: (1) how informed is the grant manager about his or her project; (2) what specific oversight and review activities are being conducted; and (3) how are grantee performance issues, if any, being resolved? Our review of the performance work plan OJJDP considered to be representative found that it tended to contain information that made it difficult to precisely understand the nature of the monitoring to be conducted. The performance work plan, for example, referred to monitoring grants to ensure that projects meet their goals and milestones, with no clear expectations regarding monitoring requirements, such as site visits or telephone monitoring, and documentation of these contacts and reviews of grantees’ progress and financial reports. In addition, according to OJJDP officials, they do not have an officewide management information system that tracks monitoring activities specific to each grant and links these to grant monitoring staff. However, OJJDP officials said they take several actions to ensure that staff are adequately performing their monitoring responsibilities. They said the grant manager’s memorandum is carefully scrutinized to determine that the proper information regarding grant goals, objectives, tasks, program activities, and products are included. Division directors are to review requests for site visits on a quarterly or semiannual basis as part of their development of travel budgets. Site visit reports are reviewed at the time or within a reasonable period of submission of travel reimbursement requests. Finally, division directors are routinely consulted regarding corrective action on grants. In our May 1996 work, we reported that OJJDP’s files for discretionary grants generally contained monitoring plans but little documentation that monitoring occurred. OJJDP said that a heavy workload might have resulted in a lack of monitoring records, but that it would take the steps necessary to improve records. OJJDP has made limited progress in addressing the monitoring problems we identified in 1996. Our review showed that monitoring activities continue to be insufficiently documented, and because of this, neither OJJDP nor we can determine the level of monitoring being performed by grant managers as required by OJP, OJJDP, and the Comptroller General’s internal control standards. OJJDP also cannot rely on supervisory oversight of its grant managers to ensure that monitoring is being performed as required. The use of staff performance work plans seems problematic as a means of reviewing monitoring activities because the requirements are written in such general terms. Grant managers’ performance work plans do not include clear expectations and accountability for monitoring called for in grant managers’ monitoring plans as well as follow-up requirements to notify and assist grantees who consistently submit late or no progress or financial reports. Further, the approval of grant managers’ memorandums or preapproval of site visits does not ensure that required monitoring is taking place, and none of these oversight activities is designed to ensure that proper documentation of monitoring activity is taking place. Current activities to review monitoring policies and practices and to complete documentation for closed grants could result in positive changes. However, these activities are too new to be evaluated in relation to their potential for improving grant monitoring or its documentation. The current review of monitoring policies and practices affords an opportunity for OJJDP to assess monitoring in light of our recent findings. To facilitate and improve the management of program grant monitoring, we recommend that the Attorney General direct OJJDP, as part of its ongoing review of monitoring policies and practices, to determine whether the monitoring documentation problems that we identified were an indication of grant monitoring requirements not being met or of a failure to document activities in the official grant files that did, in fact, take place. If monitoring requirements are not being met, we recommend that the Attorney General direct OJJDP to determine why this is so and to consider those reasons as it develops solutions for improving compliance with the requirements. If it is determined that required monitoring is taking place but is not being documented, we recommend that the Attorney General direct OJJDP to develop and enforce clear expectations regarding monitoring requirements. This policy should, among other things, require supervisory review to ensure that monitoring activities are being carried out and documented as prescribed. Further, grant managers’ performance work plans should include clear expectations and accountability for monitoring called for in grant managers’ monitoring plans as well as follow-up requirements to notify and assist grantees who consistently submit late or no progress or financial reports. We provided a copy of this report to the Attorney General for review and comment. In a September 19, 2001, letter, the Acting Assistant Attorney General commented on a draft of this report. Her comments are summarized below and are presented in their entirety in appendix II. The Acting Assistant Attorney General said that overall, the report provides useful information in highlighting management and monitoring activities in need of improvement. Senior staff within OJJDP have been directed to address each of the issues identified in the draft report to focus on adjusting the nature and types of monitoring activities pursuant to these types of grants, thus ensuring that every active grant receives a programmatic assessment. OJJDP will also be considering ways to improve and streamline grant management monitoring activities to ensure that program monitoring efforts are documented and useful, including the development of a management information system and the use of private contractors to assist in some of these review functions. The Acting Assistant Attorney General also said that the draft report could have gone further in acknowledging the impact of OJJDP’s enormous growth of the past 5 years on its ability to monitor recipients. We recognized that the number of active OJJDP discretionary grants has more than tripled over the past 5 years. However, it is OJJDP’s responsibility to initiate action, such as to request additional funding to meet its monitoring responsibilities or revise its monitoring efforts to comport with its available resources. In commenting on our May 1996 testimony, the OJJDP Deputy Administrator recognized that OJJDP’s heavy workload might have resulted in a lack of monitoring records, but that he would take steps necessary to improve records. However, our current review disclosed that similar monitoring problems persist. Further, the Acting Assistant Attorney General said that we did not acknowledge OJJDP’s efforts in a major OJP-wide effort to improve grant monitoring practices and establish uniform policies. We recognized some of the efforts to review its monitoring practices, such as revising the OJP Handbook: Policies and Procedures for the Administration of OJP Grants and its recent embarking on an overhaul of its grant closeout practices to improve its monitoring records. Such efforts, we believe, should help address the monitoring problems we identified. The Acting Assistant Attorney General concurred with our recommendation that grant managers’ performance work plans include clear expectations and accountability for monitoring. She said performance work plans put into effect after February 2001 for grant managers in OJJDP’s Special Emphasis Division more clearly reflect monitoring responsibilities and provide for better supervisory controls. Further, she noted that OJJDP’s Special Emphasis Division has put into place a new process designating one staff member with responsibility for file “audits” to help ensure that files are in order and that documentation of monitoring is in place. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution of it until 30 days from the date of this letter. At that time, we will send copies to the Chairmen and Ranking Members of the Senate Judiciary Committee; Senate Subcommittee on Youth Violence; House Committee on Education and the Workforce; House Subcommittee on Early Childhood, Youth and Families; Attorney General; and Director, Office of Management and Budget. We will also make copies available to others upon request. If you or your staff have any questions about this report, please contact me or James M. Blume at (202) 512-8777. Key contributors to this report are acknowledged in appendix III. Each year, OJJDP publishes a proposed program plan seeking public comment about proposed discretionary funding opportunities for activities covered under parts C and D (all national programs and gang-related activities) of the Juvenile Justice and Delinquency Prevention Act. The proposed plan is published in the Federal Register. Program announcements and solicitations for discretionary funding opportunities are published as they become available. The announcements and solicitations provide details about specific funding opportunities, including eligibility requirements and deadlines. These include funding opportunities for tribal youth and drug-free communities as well as funding opportunities associated with parts C and D. Once discretionary grant applications are submitted, OJJDP said that its contractor conducts an initial review to eliminate any applications that do not comply with the specified grant instructions. After ineligible grant applications are eliminated, OJJDP has a peer review panel, which consists of three or more outside experts who evaluate and rank OJJDP’s grant applications. After the peer review panel has ranked the applications, OJJDP’s grant managers do an in-depth review of applications and make formal recommendations regarding the funding of individual grants to the OJJDP Administrator. As shown in figure 1, the key parts of OJJDP’s process for planning and announcing discretionary grants are issuance of a program plan, issuance of discretionary program announcements, and distribution of application kits. As shown in figure 2, according to OJJDP officials, the key parts of the grant application and review process include initial screening of an application by the OJJDP contractor, peer review of applications, and the final selection of applications for funding by the OJJDP Administrator or Assistant Attorney General. OJJDP’s proposed program plan describes OJJDP’s goals and priorities and the particular program activities it plans to carry out during the fiscal year under parts C and D of Title II of the OJJDP Act. OJJDP publicly announces its proposed program plan in the Federal Register. After a 45- day comment period, OJJDP is to publish a final plan that is to take into consideration comments received during the comment period. According to an official, the OJJDP program plan is generally drafted by the Administrator and senior managers and based on (1) congressional and departmental priorities, (2) their knowledge and expertise, and (3) input during the course of the year from a variety of sources. These sources include program reviews by OJJDP staff and comments from practitioners in the field, officials from Justice and OJP bureaus and offices, and other federal agencies. Following publication of the final program plan, OJJDP is to publish in the Federal Register a notice of discretionary grant programs and to announce the availability of an application kit for grants. The notice is to include the availability of funds, criteria for selection of grant applicants, procedures applicable to the submission and review of application for submission and review of applications for assistance, and information on how to obtain an application kit. Upon receipt of discretionary grant applications, the OJJDP contractor does an initial review of each application, using an application review checklist. The objective of this initial review is to determine whether applications are complete and eligible for federal funding. OJJDP contracts out the administrative arrangements (e.g., arranging panels and notifying reviewers) for the peer review of its discretionary grant applications. The contractor maintains a list of qualified consultants from which peer reviewers were selected by OJJDP. An OJJDP official estimated that there were 73 peer review panels in fiscal year 2000 for its 22 grant programs. The grant managers identify the particular expertise they want the peer reviewer to have. The contractor generates a list of potential reviewers with their resumes, background information, and other helpful information. The grant managers recommend peer reviewers on the basis of the contractor’s list and give the list to the Deputy Administrator for approval. After the Deputy Administrator approves the list, it is sent to the Administrator. Once the Administrator selects the peer reviewers, the contractor makes arrangements with the individuals selected and manages the peer review process. According to an OJJDP official, consultants performing the peer reviews are reimbursed at a flat rate of $150 a day. OJJDP guidance requires that the grant managers use the following criteria to help achieve balance on the peer review panel. Each reviewer should have expertise in or complementary to the subject area under review. Where possible, the peer review panel should be composed of a mix of researchers, practitioners, and academicians. Special attention should be paid to obtaining qualified representation from women and minority groups. OJJDP’s peer review process is advisory and is to supplement and assist OJJDP’s consideration of applications. However, an OJJDP official noted that, in practice, results of the peer review panels’ consideration of application were almost always accepted. After completion of the peer review panels, OJJDP’s grant managers do an in-depth review of the peer review panels’ ranked list. The OJJDP grant managers make recommendations concerning applications to receive funding in a memorandum to the Administrator, and the Administrator is to make tentative grant selections. The OJJDP Administrator makes final awards for research and evaluation grants, while final awards for all training and technical assistance and demonstration grants are made by the Assistant Attorney General. After the applications are approved, OJP’s Office of the Comptroller does a financial review of these applications to determine whether the applicant has the necessary resources and integrity to account for and administer federal funds properly and whether budget and cost data in the application were allowable, effective, and reasonable. The following are GAO’s comments on the Department of Justice’s September 19, 2001, letter. 1. According to the OJP Handbook: Policies and Procedures for the Administration of OJP Grants (Feb. 1992), official grant files kept by the Office of the Comptroller Control Desk are to contain documents relating to each grantee, including progress and financial reports and site visit reports. In addition, for documentation to be readily available for examination, as required by the Comptroller General’s internal control standards, keeping them in the official grant file seems appropriate. In addition to those named above, R. Rochelle Burns, Jill Roth, Michele Fejfar, Sidney H. Schwartz, Michele J. Tong, Jerome T. Sandau, and Ann H. Finley made key contributions to this report. | The Office of Juvenile Justice and Delinquency Prevention (OJJDP) provides block grants and discretionary funding to help states and communities prevent juvenile delinquency and improve their juvenile justice systems. OJJDP has specific program monitoring and documentation requirements for its discretionary grants. These monitoring requirements include having the grant manager make quarterly telephone calls, undertake on- and off-site grant monitoring visits, and review interim and final products. In a review of OJJDP's most recent award of grants active in fiscal years 1999 and 2000, GAO found that OJJDP's grant monitoring activities were not consistently documented. These findings are similar to those GAO reported in May 1996. |
Money laundering is the process used to transform monetary proceeds derived from criminal activities into funds and assets that appear to have come from legitimate sources. Although the magnitude of global money laundering is unknown, many estimates suggest annual ranges in the hundreds of billions of dollars. The process of money laundering generally takes place in three stages: placement, layering, and integration. In the placement stage, cash is converted into monetary instruments, such as money orders or traveler’s checks, or deposited into financial institution accounts. In the layering stage, these funds are transferred or moved into other accounts or other financial institutions to further obscure their illicit origin. In the integration stage, the funds are used to purchase assets in the legitimate economy or to fund further activities. All financial sectors and certain commercial businesses can be targeted during one or more of these stages. Many of these entities are required to report transactions with certain characteristics to law enforcement if they appear to be potentially suspicious. The transactions would generally fall within either the placement or layering stage if they proved to be involved in money laundering. Transaction reporting requirements are discussed further later in this report. Figure 1 shows the three stages of money laundering. Terrorist financing is generally characterized by different motives than money laundering and the funds involved often originate from legitimate sources. However, the techniques for hiding the movement of funds intended to be used to finance terrorist activity—techniques to obscure the origin of funds and the ultimate destination—are often similar to those used to launder money. Therefore, Treasury, law enforcement agencies, and the federal financial regulators often employ similar approaches and techniques in trying to detect and prevent both money laundering and terrorist financing. Agencies under the Departments of the Treasury, Justice, and Homeland Security are to coordinate with each other and with financial regulators in combating money laundering. Within Treasury, the Financial Crimes Enforcement Network (FinCEN) was established in 1990 to support law enforcement agencies by collecting, analyzing, and coordinating financial intelligence information to combat money laundering. In addition to FinCEN, Treasury components actively involved in anti-money laundering and antiterrorist financing efforts include the Executive Office for Terrorist Financing and Financial Crimes, the Office of International Affairs, and the Internal Revenue Service and its Criminal Investigation unit (IRS-CI). Department of Justice components involved in efforts to combat money laundering and terrorist financing include the Criminal Division’s Asset Forfeiture and Money Laundering Section (AFMLS) and Counterterrorism Section, the Federal Bureau of Investigation (FBI), the Drug Enforcement Administration (DEA), and the Executive Office for U.S. Attorneys (EOUSA) and U.S. Attorneys Offices. With the creation of DHS in March 2003, anti-money laundering activities of the Customs Service were transferred from Treasury to DHS’s Bureau of Immigration and Customs Enforcement (ICE). The financial regulators who oversee financial institutions’ anti-money laundering efforts include the depository institution financial regulators— the Federal Reserve Board (FRB), FDIC, Office of the Comptroller of the Currency (OCC), Office of Thrift Supervision (OTS), and the National Credit Union Administration (NCUA)—and also the Securities and Exchange Commission (SEC), which regulates the securities markets, and the Commodity Futures Trading Commission (CFTC), which regulates commodity futures and options markets. While OCC and OTS are bureaus within Treasury, the FRB, FDIC, NCUA, SEC, and CFTC are independent agencies that are not part of the executive branch. Figure 2 shows the primary agencies responsible for combating money laundering and terrorist financing before the creation of DHS. Figure 3 shows the primary agencies responsible for combating money laundering and terrorist financing after the creation of DHS. Given law enforcement’s mixed history of both productive partnerships and turf-protection battles, proponents of the Strategy Act envisioned that the implementation of an annual NMLS would inaugurate a new level of coordination and cooperation between law enforcement agencies. The NMLS also sought to coordinate the efforts of law enforcement agencies and financial regulators to ensure that financial institutions were sufficiently vigilant to detect possible money laundering and that they reported any suspicious activity to law enforcement agencies to assist in their efforts to investigate money laundering and, more recently, terrorist financing. The process for developing the NMLS has varied slightly from year to year, but it has generally involved Treasury working with other agencies to develop a draft. Treasury officials said that they have sometimes asked officials from other agencies to take the lead in drafting certain sections that pertain particularly to their efforts. In other instances, Treasury has drafted the sections and asked for participating agencies’ review and comments on the sections relevant to them. Upon completion of the draft NMLS, the relevant agencies are given the opportunity to clear the final document through the Office of Management and Budget’s clearance process, which requires that the agencies approve the document, that is, signify their agreement with its contents. Treasury officials told us that by approving the NMLS through this process, the agencies have agreed to it and should be held accountable to Congress and the public to complete their assigned responsibilities. The drafting process has generally resulted in a document that lists four to six broad goals, each containing objectives, which in turn contain a list of priorities. Over time, the goals have changed, sometimes in their wording or order, and other times to cover new threats. For example, in the wake of September 11, the 2002 NMLS added the goal, “Focus Law Enforcement and Regulatory Resources on Identifying, Disrupting, and Dismantling Terrorist Financing Networks.” As of September 24, the 2003 NMLS had not yet been issued. Table 1 shows the NMLS goals from 1999 through 2002 and the number of objectives and priorities. The Strategy Act also created an operating mechanism within which to enhance interagency coordination of law enforcement investigations— HIFCAs. In accordance with the Strategy Act and the 1999 NMLS: HIFCA designations would allow law enforcement to concentrate its resources in areas where money laundering or related financial crimes appeared to be occurring at a higher rate than average. An interagency HIFCA Designation Working Group would review requests for such designations and provide advice for selections to be made by the Secretary of the Treasury, in consultation with the Attorney General. A money laundering action team, where appropriate, would be created when a HIFCA was designated to spearhead a coordinated federal, state, and local anti-money laundering effort in the area, or an existing task force already on the ground would be mobilized. The 2001 NMLS specified that HIFCAs were to be operational and conduct investigations designed to result in indictments, convictions, and seizures, rather than focusing principally on intelligence gathering. Also, the 2001 NMLS reinforced the expectations that HIFCA task forces “will be composed of, and draw upon, all relevant federal, state, and local agencies, and will serve as the model of our anti-money laundering efforts” and that the Departments of the Treasury and Justice were to jointly oversee the HIFCA task forces. The Strategy Act mandated that the NMLS be submitted to Congress by February 1 of each year, 1999 to 2003. The Strategy Act also required that—at the time each NMLS was transmitted to the Congress (other than the first transmission of any such strategy)—the Secretary of the Treasury must submit a report containing an evaluation of the effectiveness of policies to combat money laundering and related financial crimes. The Treasury, Justice, and regulatory agency officials we interviewed said that the NMLS was initially beneficial but, over time, certain factors and events affected its development and implementation. Officials from each of the agencies endorsed the concept of having a national strategy for combating money laundering and related financial crimes. Generally, the officials said that the annual NMLS probably was more beneficial in the first 2 years (1999 and 2000) than in the subsequent years (2001 and 2002). As an initial benefit, for example, Treasury officials said that the NMLS was instrumental in focusing on the need to combat money laundering systemically and not solely on a case-by-case basis, encouraging multiple law enforcement agencies to work together, and raising general awareness of the importance of combating financial crimes. The NMLS enhanced their planning and communication when it was new because it served to formalize interagency communication in a way that had not existed before. Similarly, the officials noted that the early strategies were instrumental in expanding the perspectives of the regulatory agencies by refocusing them on the purposes underlying their BSA responsibilities. The early strategies renewed attention on the fight against money laundering that supports particular reporting or record keeping obligations. That is, due partly to the strategies, the financial regulators became more focused regarding ways in which criminals could be using financial institutions for money laundering activities. However, after the first couple of years, the benefit of the annual NMLS was affected by a number of factors and events, according to the Treasury, Justice, and regulatory agency officials we interviewed. One factor cited was that the principal agencies had significantly differing views about the appropriate purpose and structure of the strategy. For instance, Treasury preferred a document that covered the full breadth and scope of the federal government’s planned anti-money laundering efforts, while Justice preferred a more concise document that included only those priorities that realistically could be addressed during the year. Likewise, the regulatory agencies generally favored a more concise document. Several officials said that this fundamental difference in views resulted in less-than-full commitment or buy-in from some agencies, which lessened the overall benefit of the recent strategies. An event that affected the 2001 NMLS was the change in presidential administrations prior to the strategy’s issuance. Treasury and Justice officials explained that with the arrival of a new administration, it was necessary to revise a nearly complete NMLS to match the new administration’s vision for combating money laundering. This redrafting process caused the NMLS to be issued very late, leaving little time to implement any goals or objectives before drafting the 2002 NMLS. The officials said that the implementation of the recent strategies has been affected most significantly by external events—particularly September 11, 2001, and its aftermath, which included passage of the USA PATRIOT Act and the creation of DHS. Treasury and Justice officials said that the 2001 NMLS, which was issued on September 12, 2001, was virtually obsolete at issuance because the nature of the issues they faced had just changed dramatically. After September 11, combating terrorist financing became a major element of the federal government’s anti-money laundering efforts, but it was not part of the 2001 NMLS. The passage of the USA PATRIOT Act increased the requirements on many financial institutions for conducting due diligence, record keeping, reporting, and sharing information. Because implementing the USA PATRIOT Act became the main focus for the financial regulators in the 2002 NMLS, financial regulators attributed their efforts to the USA PATRIOT Act rather than the NMLS. The creation of DHS required the transfer of most of the law enforcement functions and staff from agencies formerly under Treasury to the new agency. Justice anti-money laundering components remained in Justice. Treasury and Justice officials said that the implementation of some 2002 NMLS priorities was delayed pending formation of the new department. They also said that issuance of the 2003 NMLS has been delayed by the same disruptions. As a mechanism for guiding the coordination of federal law enforcement agencies’ efforts to combat money laundering and related financial crimes, the NMLS has had mixed results and—according to the evidence we reviewed and the officials we contacted—generally has not been as useful as envisioned by the Strategy Act. For example, although expected to have a key role in the federal government’s efforts to disrupt and dismantle large-scale money laundering organizations, HIFCA task forces generally were not yet structured and operating as intended and had not reached their expectations for leveraging investigative resources or creating investigative synergies. Further, while Treasury and Justice made progress on some NMLS initiatives designed to enhance interagency coordination of money laundering investigations, most had not achieved the expectations called for in the annual strategies. Moreover, the 2002 NMLS did not address agency roles and interagency coordination procedures for conducting terrorist financing investigations. As envisioned by the Strategy Act, HIFCAs represent a major NMLS initiative and were expected to have a flagship role in the U.S. government’s efforts to disrupt and dismantle large-scale money laundering operations. They were intended to improve the coordination and quality of federal money laundering investigations by concentrating the investigative expertise of federal, state, and local agencies in unified task forces, thereby leveraging resources and creating investigative synergies. While neither the Strategy Act nor the annual NMLS specified a time frame for when designated HIFCAs were to become fully operational, we found that the task forces had made some progress but generally had not yet been structured and operating as intended. As of July 2003, Treasury and Justice were in the process of reviewing the HIFCA task forces to remove obstacles to their effective operations. The results of this review could provide useful input for an evaluation report on the HIFCA program, which the Strategy Act requires Treasury to submit to the Congress in 2004. In May 2003, we contacted each of the seven designated HIFCAs to obtain information on the status of their task forces (see table 2). At that time, two of the seven HIFCAs (the Southwest Border and Miami) had not started operations. Law enforcement officials in the Southwest Border area cited several reasons for the HIFCA’s nonoperational status, including (1) difficulty in coordinating activities in such a large area and (2) lack of funds to persuade state and local officials to participate. In Miami, federal law enforcement officials had met but had not reached agreement on how the HIFCA should be structured or how it should operate. For example, the officials had not agreed on whether the Miami HIFCA should conduct investigations or focus principally on intelligence gathering. In September 2003, in commenting on a draft of this report, Justice said that while the Southwest Border HIFCA has not worked out as intended, the participants in Texas and Arizona met on numerous occasions over the past 4 years in an attempt to find an organizational structure that could meet the needs of all of the participants. Justice also said that headquarters officials and participants in the Southwest Border area recently decided that the dual-state HIFCA was too ambitious and that the HIFCA should be limited to Texas and relocated to augment an existing task force. Although the 2001 NMLS specified that HIFCAs were to conduct investigations rather than principally gather intelligence, we found that two of the five operating task forces (Los Angeles and San Francisco) were primarily focusing on intelligence gathering activities—such as reviews of Suspicious Activity Reports (SAR) and other information required by the BSA—and had not established multiagency investigative units to act on the intelligence. HIFCA officials in Los Angeles told us they planned to locate task force participants together in the same area in mid- or late-2003, at which time multiagency investigative units would be established. In San Francisco, a HIFCA official told us their proposal to become a HIFCA specified that the task force would focus on intelligence and that there were no plans to establish multiagency investigative units within the HIFCA. Treasury and Justice officials responsible for overseeing the HIFCAs told us that headquarters was reluctant to require the task forces to establish multiagency investigative units, primarily because the Strategy Act did not provide additional funds or personnel to establish such units. The officials noted that even though the 2001 NMLS specified that HIFCAs were to conduct investigations, task forces that focus on intelligence gathering activities but do not conduct investigations do enhance interagency efforts to combat money laundering. Also, because the investigative activities of the three HIFCAs that had multiagency investigative units (Chicago, New York/New Jersey, and Puerto Rico) were based on task force structures already in place before the HIFCA designation, the overall effect of the NMLS on these task forces is unclear. For example, the New York/New Jersey HIFCA essentially represented a renaming of the well-established El Dorado Money Laundering Task Force, which had existed since 1992. As mentioned previously, a HIFCA task force could be (1) created when a HIFCA was designated or (2) based on an existing task force. Further, in some cases, federal law enforcement agencies had not provided the levels of commitment and staffing to the HIFCA task forces called for by the strategy. As shown in table 2, ICE and/or IRS-CI were designated the lead agency in each of the five operational task forces. We found that most of the HIFCAs did not have DEA or FBI agents assigned full-time to the task forces. For example, regarding the three HIFCAs with multiagency investigative units, DEA and the FBI were not members of the Chicago HIFCA, DEA was not a member of the New York/New Jersey HIFCA, and both DEA and the FBI had only part-time representation on the Puerto Rico HIFCA. As also shown in table 2, four of the five operating HIFCAs had little or no participation from state and local law enforcement agencies, with the notable exception being the New York/New Jersey HIFCA. The NMLS called for each HIFCA to include participation from all relevant federal, state, and local agencies. Justice headquarters officials said the main problem with supporting the HIFCA task forces was the absence of additional funds or personnel to offer law enforcement agencies in return for their participation. A DEA official told us that, because of differences in agencies’ guidelines for conducting undercover money laundering investigations, DEA will not dedicate staff to HIFCA task force investigative units but will support intelligence-related activities. FBI officials cited resource constraints as the primary reason why the bureau does not fully participate. Various task force officials mentioned lack of funding to compensate or reimburse participating state and local law enforcement agencies as a barrier to their participation in HIFCA operations. Further, Treasury and Justice officials noted that a key to the success of the HIFCA program is the ability to promote interagency cooperation by locating task force participants together in the same office space. Accordingly, the 2002 NMLS called for headquarters to examine how to fund the colocation of HIFCA task force participants absent funds appropriated specifically for that purpose. While we recognize that federal law enforcement agencies have resource constraints and competing priorities, we note that HIFCA task forces were expected to make more effective use of existing resources or of such additional resources as may be available. Without commitment and staffing from relevant federal, state, and local agencies, the task forces cannot fully leverage resources and create investigative synergies, as envisioned by the Strategy Act. Treasury and Justice have not provided the level of oversight of the HIFCA task forces called for by the NMLS. For example, in response to our initial inquiries and formal requests for information, Treasury and Justice officials responsible for overseeing the HIFCA task forces could not readily provide basic information, such as names of participating agencies and contact persons or the results of task force operations. Also, as shown in table 3, Treasury and Justice had not addressed various NMLS initiatives designed to oversee HIFCA operations, and many of the initiatives were still ongoing well past expected completion dates. Fully addressing these initiatives could help ensure accountability within the HIFCAs, as well as refine the operational mission, structure, and composition of the task forces. Treasury and Justice officials told us the primary reasons for not addressing or not yet completing the HIFCA initiatives were that headquarters (1) was reluctant to impose a structure or reporting requirement on the field without offering any new resources and (2) did not believe that a single structure could fit every HIFCA. The officials also said that the individual HIFCAs were in the best position to address their specific needs and problems. Further, the officials told us that, while most of the HIFCA initiatives had not been addressed or were not yet completed, the HIFCA structure at headquarters has provided a framework for regular interagency meetings to discuss money laundering trends and ways to improve interagency cooperation. As shown in table 3, although only one of the HIFCA initiatives was completed by the specified milestone or goal date, many of the initiatives were still ongoing. For example, the 2002 NMLS called for a review of HIFCA task forces to remove obstacles to their effective operation. Specifically, the initiative called for an interagency HIFCA team to (1) review the accomplishments of the HIFCA task forces to date; (2) examine structural and operational issues, including how to fund the colocation of participants in HIFCA task forces absent funds appropriated for that purpose; and (3) examine existing operations and make recommendations to ensure that each HIFCA is composed of all relevant federal, state, and local enforcement authorities, prosecutors, and financial supervisory agencies as needed. As of July 2003, the HIFCA review team was still in the process of assessing the HIFCAs. When completed, the team’s review could provide useful input for an evaluation report on the effectiveness of and the continued need for HIFCA designations, which is required by the Strategy Act to be submitted to the Congress in 2004. According to the 2002 NMLS, Treasury and Justice have conducted a substantial amount of fundamental, advanced, and specialized money laundering training to task forces, agencies, investigators, and prosecutors. For example, as included in the 2002 NMLS (see table 3), the Federal Law Enforcement Training Center, in cooperation with Treasury and Justice, have provided an advanced money laundering training course in six HIFCA locations. According to a Federal Law Enforcement Training Center official, approximately 900 to 1,000 agency representatives have participated in the 3-day training seminar. The official said that the training focused on numerous issues, including money laundering statutes, the impact of the USA PATRIOT Act, basic and international banking, asset forfeiture issues, and specific money laundering schemes and organizations. While Treasury and Justice made progress on some NMLS initiatives that were specifically designed to enhance coordination of federal law enforcement agencies’ money laundering investigations, most of the initiatives were not addressed or were still ongoing. In general, the failure to address or complete the initiatives indicates that interagency coordination was falling short of expectations. Treasury and Justice made progress in implementing some of the NMLS law enforcement coordination initiatives. For example, as called for in the 1999 and 2000 strategies, the departments took steps to (1) enhance the money laundering focus of interagency counter-drug task forces and (2) collect and analyze information on the money laundering aspects of the task forces’ investigations. More recently, the 2002 NMLS called for an interagency team to identify money laundering-related targets for coordinated enforcement action. The strategy noted that targets could be particular organizations or systems used or exploited by money launderers, such as the smuggling of bulk cash and unlicensed money transmitters. In August 2002, Treasury and Justice created an interagency team and identified a money laundering-related target and four cities in which to conduct investigations. In July 2003, Justice officials told us that U.S. Attorneys Office officials had agreed to participate in the targeting initiative and that the initiative was ongoing. Most of the annual strategy initiatives designed to enhance interagency coordination of law enforcement investigations were not addressed or were still ongoing. Three examples are as follows. First, the Customs Service created a Money Laundering Coordination Center in 1997 to (1) serve as the repository for all intelligence information gathered through undercover money laundering investigations and (2) function as the coordination and “deconfliction” center for both domestic and international undercover money laundering investigations. Both the 1999 and the 2000 NMLS contained an initiative to encourage all applicable federal law enforcement agencies to participate in the Money Laundering Coordination Center. During our review, Customs Service officials (before the agency was transferred to DHS) told us that, although Justice agencies (including DEA and FBI) were invited to use the center, these agencies were only occasional users and did not contribute information to the center. DEA and FBI officials told us that their agencies did not use the Money Laundering Coordination Center because they could not reach a satisfactory memorandum of understanding regarding participation, including controls over the dissemination of information. DEA officials added that the center does not meet DEA’s needs because it is used for deconfliction only. In August 2003, the DEA officials said that DEA had created and was testing a new database that is designed to be a single source for information on money laundering investigations related to drug money. The officials added that DEA has briefed Treasury and DHS about the new database, but as of August 2003, no other agencies were participating. Second, federal law enforcement agencies do not have uniform guidelines applicable to undercover money laundering investigations. According to the 2002 NMLS and our discussions with law enforcement officials, the lack of uniform guidelines inhibits some agencies from participating in investigations that have an international component. For example, a DEA official told us that DEA guidelines generally are more restrictive than guidelines used by Customs (as part of ICE) in regard to (1) obtaining approval to initiate and continue undercover investigations and (2) coordinating activities with foreign counterparts. Therefore, the officials noted that DEA generally could not participate in international undercover money laundering investigations led by Customs. The 2002 NMLS called for Treasury and Justice to develop uniform undercover guidelines by September 2002 to ensure the full participation of all applicable federal law enforcement agencies in undercover money laundering investigations. Treasury officials told us the initiative is still ongoing but has been put on hold, pending reorganizations associated with the creation of DHS. Third, Treasury and Justice have not yet fully implemented NMLS initiatives designed to establish SAR review teams in federal judicial districts. The 2001 NMLS contained an initiative that called for the creation of a SAR review team in each federal judicial district. Generally, each team—to be comprised of an Assistant U.S. Attorney and representatives from federal, state, and local law enforcement agencies—was expected to evaluate all SARs filed in their respective federal judicial district. The 2001 SAR initiative has been partially implemented. Treasury officials noted that Justice has primary responsibility for implementation because Justice provides guidance and direction to EOUSA and the U.S. Attorneys Offices. According to EOUSA officials, Justice, EOUSA, and the U.S. Attorneys Offices have actively encouraged the creation of SAR review teams and these efforts remain ongoing. At our request, in July 2003, EOUSA conducted an informal survey of U.S. Attorneys Offices and reported that at least 33 of the 94 federal judicial districts were actively using interagency SAR review teams. The 2002 NMLS had a more conservative SAR-related initiative, calling for the establishment of five additional review teams. Specifically, the 2002 NMLS initiative called for Treasury and Justice—by August 2002—to (1) identify a priority list of five federal judicial districts that do not have a SAR review team but could benefit from one and (2) work with EOUSA and the respective U.S. Attorneys Offices to encourage the creation of interagency review teams. As of July 2003, this initiative had not yet been completed, but efforts were still ongoing. Further, although not called for by the NMLS, the IRS has had a related initiative to create interagency SAR review teams. Specifically, IRS-CI data show that IRS has established 41 SAR review teams nationwide—with all 35 IRS field offices having at least one functioning team—and that most of the review teams had participation from other agencies. According to IRS CI officials, collectively, the 41 teams are to review every SAR filed in the 94 federal judicial districts. The officials said that at least 4 of the districts in which a HIFCA task force is located were using an interagency SAR review team. The officials noted that IRS review teams are not to duplicate SAR reviews already performed by existing task forces in federal judicial districts. Treasury officials told us that resource constraints and competing priorities were the primary reasons why many of the law enforcement coordination initiatives were not yet fully implemented. Also, the officials said that, over the past few years, Treasury has given higher priority to other parts of the annual strategy—such as international, regulatory, and terrorism-related initiatives—than to domestic law enforcement initiatives. Further, the officials said that Treasury generally took the lead in implementing the annual strategy but could not require other agencies to focus on specific initiatives or activities. In this regard, the officials said that other agencies frequently had their own priorities. Justice officials also said that the annual strategies have contained more initiatives than realistically could be pursued. The officials added that to the extent NMLS initiatives were not completed or target dates were missed, it was because of competing priorities or the lack of resources available for proper implementation of the strategy. The officials noted that there are complex issues involved in attempting to coordinate the resources, practices, and priorities of two (and sometimes more) departments and several law enforcement agencies, as well as U.S. Attorneys Offices throughout the country. Further, Justice officials told us that while NMLS initiatives to institutionalize coordination may not have been fully implemented, the efforts to do so and regular meetings have been continuing. In developing the 2002 NMLS, Treasury and Justice officials met to discuss the roles of the various investigative agencies involved in combating terrorist financing. However, the two departments could not reach agreement, and the 2002 strategy was published without addressing the agencies’ roles. In general, Justice’s position was that it had exclusive statutory authority to lead all terrorist financing investigations, while Treasury maintained that it also had the authority and the needed expertise to lead such investigations. In commenting on a draft of the 2002 strategy, the FBI noted the following: The strategy does not address the various agencies’ duplication of efforts to combat terrorist financing. By not specifically addressing and delineating the roles of the respective agencies, the strategy creates more confusion than it resolves and wastes limited resources. Moreover, the strategy section on U.S. government efforts to identify, disrupt, and dismantle terrorist financing networks did not mention or clarify roles of the three primary law enforcement task forces involved in investigating terrorist financing—Customs’ Operation Green Quest (OGQ) and the FBI’s Terrorist Financing Operations Section (TFOS) and Joint Terrorism Task Forces (JTTF). According to Treasury officials, the NMLS drafting process realistically could not have been expected to resolve the long-standing, highly challenging issues associated with the interagency jurisdictional dispute. While we agree that it may have been unrealistic to expect the drafting process to resolve the long-standing issues, we note that a primary role of the NMLS is to enhance interagency coordination and help resolve turf- protection battles. Because the issue was not addressed in the 2002 NMLS, the problem remained, thus leaving unresolved possible duplication of efforts and disagreements over which agency should lead investigations. In our view, any way the NMLS could have advanced resolution of the matter would have been beneficial. To help avoid overlapping investigations and duplication of efforts, it is important that agencies investigating terrorist financing have coordination mechanisms. At the policy level, a National Security Council policy coordination committee on terrorist financing is responsible for coordinating antiterrorist financing activities. This committee is to consider evidence of terrorist financing networks and coordinate strategies for targeting terrorists, their financiers, and supporters. At the operational level, we found that some interagency coordination of terrorist financing investigations existed between agency headquarters’ components. For example, OGQ and TFOS had assigned one agent to each other’s headquarters in Washington, D.C. The FBI also was to provide information on its activities to OGQ through daily downloads from the FBI’s terrorist financial database. Further, OGQ and FBI officials told us that local mechanisms existed around the country to deconflict investigations. While OGQ and the FBI task forces took steps to inform each other about the targets of their investigations, we found that the task forces did not fully coordinate their activities. For example, at the three locations we visited (Los Angeles, Miami, and New York City), OGQ and JTTF officials told us they generally were not aware of each other’s financial investigations and that the task forces generally did not share investigative information. Several officials indicated that there were problems with conflicting or competing investigations, including disagreements over which task force should lead investigations. Officials at all three locations noted that the government’s antiterrorist financing efforts could be improved if the task forces worked more closely with each other or were combined. Further, at the three locations we visited, IRS-CI officials who had agents assigned to the local OGQ and JTTF also indicated that the task forces were not fully operating in a coordinated and integrated manner. Specifically, in Miami and New York City, IRS-CI officials told us that having both OGQ and the JTTF doing the same type of antiterrorist financing work was a duplication of effort. IRS-CI officials in Los Angeles noted that communication between the two task forces could be better. Also, in response to our inquiry about interagency coordination, U.S. “With respect to the FBI’s Joint Terrorism Task Force (FBI-JTTF) and Customs’ Operation Green Quest, we would like to see increased cooperation and coordination between the agencies. Too often agents of the FBI and Customs are investigating terrorist financing independent of each other or overlapping in the targets of their investigations. Some of the barriers to greater interagency participation may be conflicting priorities of each of the agencies. Ongoing battles as to which agency is the ‘lead’ agency continues to be a problem…” In commenting on a draft of this report, Treasury said that it continues to believe that the dispute over who took the lead in investigating the financing of terrorism did not necessarily result in duplication of efforts. Treasury said that the issue was largely definitional, with the FBI leading terrorist investigations with an ancillary financial component versus Customs financial investigations that might have a terrorist-related connection. On May 13, 2003, the Attorney General and the Secretary of Homeland Security signed a memorandum of agreement regarding the antiterrorist financing roles of the respective departments and component agencies. In general, the agreement gives the FBI the lead role in investigating terrorist financing and specifies that DHS is to pursue terrorist financing investigations solely through its participation in FBI-led task forces, except as expressly approved by the FBI. Some excerpts from the May 2003 agreement are paraphrased substantially as follows: The FBI is to lead terrorist financing investigations and operations, utilizing the intergovernmental and intra-agency National JTTF at FBI headquarters and the JTTFs in the field. Through TFOS, the FBI is to provide overall operational command to the national JTTF and the field JTTFs. After June 30, 2003, DHS is to pursue terrorist financing investigations and operations solely through its participation in the National JTTF, the field JTTFs, and TFOS, except as expressly approved by TFOS. The Secretary of Homeland Security agreed that, no later than June 30, 2003, OGQ was to no longer exist as a program name. The Secretary agreed to ensure that any future DHS initiative or program to investigate crimes affecting the integrity and lawful operation of U.S. financial infrastructures would be performed through the financial crimes division at ICE. The May 2003 agreement also contained several provisions designed to enhance the coordination and integration of FBI and ICE financial investigations. For example, the agreement calls for the FBI and ICE to (1) detail appropriate personnel to each other’s task forces, (2) take steps to ensure that the detailees have full and timely access to data and other information, and (3) develop procedures to ensure effective operational coordination of FBI and ICE investigations. Further, the FBI Director and the Assistant Secretary for ICE were to provide a joint written report on the implementation status of the agreement 4 months after its effective date to the Attorney General, the Secretary of Homeland Security, and the Assistant to the President for Homeland Security. However, as of September 24, 2003, the report had not yet been issued. If successful, the May 2003 agreement could prove to be a significant step toward establishing a coordinated interagency framework for conducting terrorist financing investigations. At the time of our review, it was too early to assess the implementation of the agreement. Most financial regulators we interviewed said that the NMLS had some influence on their anti-money laundering and antiterrorist financing efforts but that it has had less influence than other factors. Officials said that, since September 11, a change in government perspective and additional requirements placed on financial institutions by the USA PATRIOT Act and its implementing regulations have been the primary influences on their efforts. Although the financial regulators said that the NMLS had minimal influence on establishing priorities for their anti-money laundering and antiterrorist financing activities, they have completed the tasks for which they were designated as lead agencies over the years, and most of those for which they were to provide support to Treasury. The 2002 NMLS noted that the financial regulators were responsible for implementing the parts of the USA PATRIOT Act that applied to the entities they regulate. Appendix III describes the anti-money laundering requirements set forth in the USA PATRIOT Act and the rules that have been implemented thereunder. Most financial regulators we interviewed said that the NMLS had some influence on their anti-money laundering efforts because it has provided a forum for enhanced coordination, particularly with law enforcement agencies, but that it has had less influence than other factors. Similarly, law enforcement agency officials told us that the level of coordination between the financial regulators and their agencies was good and that they received the assistance and information they needed from the regulators. They did not, however, attribute this to the strategy but, rather, to legal requirements. Financial regulators said that several other factors influenced their anti- money laundering efforts to a greater extent than the NMLS. These factors include working groups that had already developed as a result of BSA implementation, the impact of September 11 on raising awareness of the importance of fighting money laundering and terrorist financing, and the passage of the USA PATRIOT Act. The financial regulators said that they have been working on anti-money laundering issues for many years and generally initiate their own anti-money laundering activities. Bank regulators and SEC pointed out that the BSA was passed in 1970 and that they have been concerned with ensuring banks’ and broker-dealers’ compliance with its requirements ever since. The USA PATRIOT Act extended responsibility for implementing the BSA to additional financial regulators as well as increased anti-money laundering requirements for certain financial institutions. Additionally, most financial regulators participate in the BSA Advisory Group, in which the financial regulators coordinate and communicate among themselves and with financial institutions on enforcing BSA requirements. Other coordinating forums include the Federal Financial Institutions Examination Council, Financial Action Task Force, and USA PATRIOT Act working groups established to develop and implement regulations resulting from the passage of the USA PATRIOT Act. Although the NMLS provided a forum in which the financial regulators could better coordinate with law enforcement agencies, other avenues for cooperation are prescribed by law, and some existed before passage of the Strategy Act. For example, depository institutions have been required to file SARs since 1996. Since December 2002, securities brokers and dealers have been required to file SARs with FinCEN as a result of the USA PATRIOT Act and its implementing regulations. (See app. III.) Certain financial institutions are also required to file Currency Transaction Reports with FinCEN for transactions that involve $10,000 or more in currency. Like SARs, these reports are supposed to be analyzed to look for suspicious activity. Financial regulators said they oversee financial institutions’ programs for complying with these legal requirements because it is their statutory responsibility, not because it is included in the NMLS. They said they would do so with or without the strategy. Most officials said that September 11 greatly affected how the administration and Congress thought about money laundering because some of the techniques used to launder money, illicitly moving funds to avoid detection, are similar to those used to finance terrorist activity. Some officials said the new administration was more concerned with the burden anti-money laundering compliance placed on financial institutions prior to September 11, but that the events of September 11 changed this, resulting in more attention being paid to the importance of anti-money laundering compliance. Congress passed the USA PATRIOT Act, which, for example, increased the due diligence, reporting, and record keeping requirements for some financial institutions to guard against their being used by their customers to launder money or finance terrorist activity. Some officials noted that USA PATRIOT Act requirements reflected topics being discussed in the NMLS and other working group meetings that might still have been in the discussion phase had not September 11 motivated their inclusion in the USA PATRIOT Act, thus requiring Treasury and other agencies to issue regulations. Reflecting this change of emphasis, the 2002 NMLS discussed the need to adapt traditional methods of combating money laundering to unconventional tools used by terrorist organizations to finance their operations. According to the 2002 NMLS, the primary responsibility of the financial regulators was to participate in the drafting and issuance of USA PATRIOT Act regulations and to provide technical expertise on the operations of depository institutions and other financial institutions to Treasury. The regulators also worked to educate financial institutions and their own staff on the new requirements. The federal financial regulators have participated in the implementation of the NMLS from 1999 to 2002 in a variety of ways, including participation in working groups established by the NMLS and, in 2002, worked with Treasury to implement provisions of the USA PATRIOT Act. The federal financial regulators were expected to participate in NMLS initiatives, but Treasury, rather than the financial regulators, was usually designated as the lead agency responsible for implementation. Most federal financial regulators are independent federal agencies. Therefore, while the financial regulators have committed to work with Treasury and Justice on NMLS initiatives, they are not required to do so because, with the exception of OCC and OTS, they are not part of the executive branch. Previous strategies have called for the financial regulators to work with Treasury and Justice on several efforts, such as (1) coordinating on establishing policies for enhanced information sharing between law enforcement agencies and the regulatory agencies, (2) working with the financial services industry to develop guidance for financial institutions to enhance scrutiny of high-risk money laundering transactions and customers, and (3) developing a SAR requirement for broker-dealers. However, policies for enhanced information sharing were not finalized until the USA PATRIOT Act required that they be developed. For example, section 314 of the USA PATRIOT Act was designed to enhance cooperation among certain entities involved in the detection of money laundering. Section 314(a) encourages regulatory authorities and law enforcement authorities to share with financial institutions information regarding individuals, entities, and organizations engaged in or reasonably suspected based on reliable evidence of engaging in terrorist acts or money laundering activities. Section 314(b) encourages information sharing among financial institutions themselves. In addition, rules promulgated by FinCEN under section 314 allow law enforcement authorities to make requests to financial institutions through FinCEN of certain account information for individuals, entities, and organizations that may be engaged in terrorist acts or money laundering activities. Information is provided to FinCEN, who gives the law enforcement entities a comprehensive product. SEC worked with FinCEN on a proposed broker-dealer SAR requirement from 1999 to 2001. However, a final rule was not issued until 2002, when it was required under the USA PATRIOT Act. Each NMLS has called for the federal bank regulators as a group or OCC individually to lead a review of their bank examination procedures regarding anti-money laundering efforts and to implement the results of these reviews. While the financial regulators have been involved in a variety of different tasks and working groups in the NMLS, they served as leads only in these reviews. Table 4 lists annual NMLS initiatives to review bank examination procedures, the lead agency or agencies, and the status of the initiatives. The financial regulators have also worked with Treasury as the lead agency for the U.S. government’s international anti-money laundering efforts. Over time, the NMLS has called for the United States to strengthen international cooperation and collaboration and to work to strengthen the anti-money laundering efforts of other countries. Much of Treasury’s effort in this area has been done as part of multinational bodies, such as the Financial Action Task Force, and international financial institutions, such as the World Bank and the International Monetary Fund. Treasury’s efforts, working with these bodies, have focused on making anti-money laundering assessments a permanent part of the International Monetary Fund and World Bank surveillance and oversight of financial sectors and providing technical assistance and training to jurisdictions willing to make the necessary changes to their anti-money laundering regimes. Treasury officials involved in international anti-money laundering efforts said that the NMLS has served as a useful tool to plan and coordinate their international efforts that include the financial regulators, which provide technical assistance and participate in international meetings of these bodies. Officials from the FRB, OCC, FDIC, OTS, SEC, and CFTC all said that they had worked with Treasury on international anti-money laundering efforts, including the preparation for or participation in meetings of the Financial Action Task Force and of international financial institutions. In recent years, our work in reviewing national strategies for various crosscutting issues has identified several critical components needed for their development and implementation; however, key components have not been well reflected in the NMLS. These components include clearly defined leadership, with the ability to marshal necessary resources; setting clear priorities and focusing resources on the greatest areas of need, as identified by threat and risk assessments; and established accountability mechanisms to provide a basis for monitoring and assessing program performance. We identified a number of ways in which these critical components could be better reflected in the development and implementation of the annual NMLS, should it be reauthorized. Our past work in reviewing various national strategies has consistently concluded that having clearly defined leadership, with the ability to marshal necessary resources, is a critical component of any national strategy. For instance, our work has noted the importance of establishing a focal point or executive-level structure to provide overall leadership that would rise above the interests of any one department or agency. Regarding the annual NMLS, we found that the joint Treasury-Justice leadership structure generally has not been able to reach consensus in developing and implementing the strategies—particularly in recent years when the structure did not include representatives from the two departments’ top leadership. This has resulted in an inability to reach agreement on the appropriate scope of the strategy and ensure that target dates for completing strategy initiatives were met. The Strategy Act required the President, acting through the Secretary of the Treasury and in consultation with the Attorney General, to produce an annual NMLS. However, Treasury and Justice officials told us that the Strategy Act did not provide additional funding or otherwise enhance either department’s ability to develop and implement the annual strategies. Rather, development and implementation of the annual NMLS has been dependent largely on consensus-building efforts between Treasury and Justice—with Treasury having de facto lead responsibility. In this regard, Treasury officials told us that, while the department could request participation from other agencies, it had no incentives it could use to marshal necessary resources or compel participation in implementing initiatives or action items. The Treasury officials noted, for example, that the department’s inability to compel action by other agencies was a contributing factor to delays in producing each annual NMLS. As shown in table 5, none of the four annual strategies issued to date was submitted to the Congress by February 1 of each year, as required by the Strategy Act. As of September 24, the 2003 strategy had yet to be submitted. The initial NMLS (1999) established a joint leadership structure for implementing the strategy. Specifically, the strategy noted that overall implementation of the strategy would be guided by an interagency Steering Committee chaired by the Deputy Secretary of the Treasury and the Deputy Attorney General, with participation of relevant departments and agencies. The Steering Committee was to be responsible for overseeing action items and timelines and, as appropriate, making specific assignments. Also, with respect to action items that involved international aspects of anti-money laundering efforts, the National Security Council was to have a central role and was to advise the Steering Committee, as necessary. The 2000 NMLS also called for the Steering Committee to oversee implementation of initiatives, although the strategy did not mention a specific role for the National Security Council. According to Treasury officials, the Steering Committee was not reconvened to oversee the development and implementation of the 2001 NMLS, in part because of the change in administrations and the timing in making political appointments. Instead, overall responsibility for developing and implementing the 2001 NMLS was assumed by two lower- level officials—a Treasury Deputy Assistant Secretary (Money Laundering and Financial Crimes) and a Justice Criminal Division Section Chief (Asset Forfeiture and Money Laundering). The 2002 NMLS called for Treasury and Justice to reconvene the Steering Committee to provide coordination and cooperation among all the participating departments and agencies. However, according to Treasury and Justice officials, the Steering Committee was not reestablished. Treasury and Justice officials with responsibility for developing the strategy and overseeing its implementation at those departments said the benefits of the Steering Committee were that it brought together the officials who were needed to make decisions when those below them could not agree and that it could hold those responsible for implementing certain priorities accountable for getting things done. Moreover, the role of the National Security Council in overseeing implementation of the annual NMLS remains somewhat unclear. On the one hand, the National Security Council does have a designated policy coordination committee responsible for overseeing antiterrorist financing activities, including those related to implementation of the 2002 NMLS. On the other hand, Treasury and Justice officials told us that this policy coordination committee has no responsibility for addressing other aspects of the strategy. The officials said that they were unaware of any National Security Council component responsible for overseeing all aspects of NMLS implementation. Our past work in reviewing various national strategies has recognized the importance of identifying and prioritizing issues that require the most immediate attention. While each NMLS (1999 through 2002) identified some “top” priorities, each strategy contained more priorities—of seemingly equal importance—than could be realistically achieved. Our prior strategy work also has shown that threat and risk assessments can be useful in establishing priorities; however, none of the money laundering strategies issued to date was preceded or guided by such an assessment. The Strategy Act called for the NMLS to include comprehensive, research- based goals, objectives, and priorities for reducing money laundering and related financial crimes in the United States. The 1999 NMLS included a total of 66 priorities, which laid out actions to be taken by Treasury, Justice, and the financial regulators; the number decreased to 50 in the 2002 NMLS (see table 1). According to Treasury officials, Treasury’s vision for the annual strategies was to provide Congress and the public with a comprehensive document identifying current and planned anti-money laundering (and in 2002, antiterrorist financing) initiatives. The officials also said that the strategies did identify some top priorities for each respective year and that the most important priorities generally were discussed in the each strategy’s executive summary. Nonetheless, the officials acknowledged that, in retrospect, each strategy probably contained more priorities than realistically could have been completed during the annual strategy year. Similarly, Justice and regulatory officials told us that the annual strategies generally have been too long and included too many initiatives and priorities to deal with in a given year. The officials noted that the strategies looked good on paper and contained important issues and concepts but served more as reference documents than strategies. The officials said that the annual strategies generally did not affect how their agencies set policy direction or aligned resources. Also, Justice officials told us that the strategies generally did not affect field offices or how field agents conducted their work. Justice and regulatory officials told us they would prefer a broader, more conceptual and focused strategy with fewer priorities and more realistic goals that could be achieved during the year. Justice officials noted that target dates for completing strategy priorities generally were not met, because there were too many priorities and there was no funding or new resources provided to implement the plan. Justice officials said that by focusing on too many priorities, the strategy can divert resources from investigations and other law enforcement activities. “… a critical piece of the equation in decisions about establishing and expanding programs to combat terrorism is an analytically sound threat and risk assessment using valid inputs from the intelligence community and other agencies. Threat and risk assessments could help the government make decisions about how to target investments in combating terrorism and set priorities on the basis of risk; identify program duplication, overlap, and gaps; and correctly size individual agencies’ levels of efforts.” However, regarding the annual NMLS, none of the four strategies (1999 through 2002) issued to date was preceded or guided by such an assessment. Further, in response to our inquiries, Treasury and Justice officials indicated that the 2003 NMLS would not be based on a formal assessment of threats and risks. Law enforcement officials generally had favorable views on the need for the NMLS to be driven by some consideration of a threat and risk assessment. Justice officials noted that money laundering investigations take a lot of expertise, money, and time, and that, in their view, a formal assessment of threats and risks would help to set NMLS priorities and assist law enforcement in focusing its limited resources. Justice officials told us that they drafted a money laundering threat assessment in late 2002 and circulated it to other law enforcement agencies. The officials planned to use the assessment as a basis for setting 2003 NMLS priorities. Treasury officials generally agreed with the concept of a money laundering threat assessment to drive priorities, but told us that the assessment prepared by Justice was not useful. The officials added that, in their view, Justice’s threat assessment mostly contained information that was already widely known and, thus, probably was at least implicitly considered in setting priorities while drafting the 2003 strategy. Our past work in reviewing various national strategies has recognized the importance of establishing accountability mechanisms to assess resource utilization and program performance. The 2001 and 2002 strategies presented various initiatives designed to establish performance measures related to federal anti-money laundering efforts. As of July 2003, efforts were ongoing on many of them, while others had not been addressed. Another potential accountability mechanism required in the Strategy Act was annual reports to Congress on the effectiveness of anti-money laundering policies; however, Treasury has not provided such reports. Establishing and implementing performance measures for the NMLS would assist in monitoring and evaluating law enforcement and financial regulatory agencies’ anti-money laundering and antiterrorist financing efforts. The 2001 strategy was the first annual strategy to call for the creation of performance measures and indicators to evaluate results against stated goals. The 2002 NMLS continued on the work started under the 2001 strategy. Both strategies designated components of Treasury and Justice to co-lead the initiatives. As shown in table 6, the 2002 NMLS contained five initiatives to measure the effectiveness and results of federal anti-money laundering activities. As of July 2003, Treasury and Justice had not yet completed any of these initiatives, although efforts were still ongoing to complete some of them. Generally, the purpose of the 2002 NMLS measurement initiatives was to provide Congress and other policymakers a basis for (1) evaluating federal agencies’ anti-money laundering efforts and results and (2) deciding how to deploy limited public resources most effectively. For example, the traffic-light scorecard was intended to provide information on the overall performance of the federal government’s efforts to combat money laundering and assess how well the government was executing each of the six goals described in the 2002 strategy (and future strategies). Also, the 2002 NMLS notes that the initiative to review law enforcement and financial regulatory costs and resources devoted to anti-money laundering activities was designed to permit Congress and other policymakers to draw informed conclusions about the effectiveness of those activities. The 2002 NMLS noted that, while deceptively easy to articulate in the abstract, the task of developing meaningful performance measures for federal agencies engaged in combating money laundering has proven to be quite difficult. Treasury officials also told us that (1) the 2002 strategy was not published until July 2002, which did not leave much time for either implementation or evaluation and (2) several measurement initiatives were put on hold pending the reorganization associated with DHS. Further, the officials noted that Treasury generally had no plans to report on performance progress (results and accomplishments) made under the 2002 strategy. The 2002 strategy did provide, for the first time in an NMLS, some baseline facts and figures designed to help determine how well the federal government was succeeding in its efforts to detect, prevent, and deter money laundering. For example, the strategy published U.S. Sentencing Commission data for fiscal year 2000 regarding defendants sentenced in federal court for the principal offense of money laundering. The 2002 strategy noted that the Sentencing Commission data could be tracked over a period of years and, thereby, serve as one measure for evaluating progress in combating money laundering. The Strategy Act required that—at the time each NMLS was transmitted to the Congress (other than the first transmission of any such strategy)—the Secretary of the Treasury submit a report containing an evaluation of the effectiveness of policies to combat money laundering and related financial crimes. As of July 2003, Treasury had not submitted any effectiveness reports. Treasury officials said they did not see this as a requirement to submit a separate report and, in their view, the strategy itself has been used to report on the effectiveness of the government’s anti-money laundering efforts. The officials explained that the “accomplishment” sections that were added to the 2002 strategy were intended to meet the Strategy Act’s reporting requirement. We believe that this information does not fully meet the Strategy Act’s requirement, because the accomplishment sections generally provided descriptive information about initiatives rather than evaluations of the effectiveness of policies to combat money laundering and related financial crimes. For example, an accomplishment section in the 2002 strategy noted that HIFCA task forces initiated over 100 investigations in 2001, but the section did not address the effectiveness of the HIFCA concept or the task forces. We identified a number of ways in which the critical components for national strategies could be incorporated into the NMLS, should Congress decide to continue the requirement. To incorporate a more clearly defined leadership structure that has the ability to marshal resources for a coordinated effort against money laundering and terrorist financing, a high-level leadership mechanism could be reestablished or a single official could be designated to carry out this responsibility. The role of the leadership structure would be to marshal resources to ensure that the vision laid out in the strategy is achieved, resolve disputes between agencies, and ensure accountability for strategy implementation. This leadership mechanism would also be in a good position to evaluate annual progress and report such progress to Congress, as is currently required of Treasury. This is especially critical now that there are three principal departments with anti-money laundering and antiterrorist financing responsibilities, in addition to the federal financial regulators. One way to help set clear priorities and focus resources on the areas of greatest need would be to require that the strategy be linked to a periodic threat assessment. Such an assessment would outline what the lead agencies see as the most significant threats. This would provide a better basis to draft a strategy to address these threats. Performance could be measured by the level of progress made in combating these threats. One way to improve accountability for the agencies and regulators following the strategy would be for the strategy to set broad policy objectives that leave it to the principal agencies to develop outcome- oriented performance measures that are linked to the NMLS’s goals and objectives. These performance measures would be reflected in the agencies’ annual performance plans. However, our work showed that, throughout its history, the NMLS has tried to specify detailed priorities for each objective, many of which were not accomplished or, in the case of the financial regulators, would have been accomplished for statutory reasons even without a strategy. The annual NMLS has had mixed results in guiding the efforts of law enforcement and financial regulators in the fight against money laundering and, more recently, terrorist financing. Through our work in reviewing other national strategies, we have identified critical components needed for successful development and implementation; but, to date, these components have not been well reflected in the annual NMLS. We believe that incorporating these critical components into the NMLS would improve its development and implementation. For example, the current NMLS leadership structure has not reached consensus on the approach the strategy should take or ensured that goals and objectives are met, and has failed to issue any of the annual strategies on time. A clearly defined high-level leadership structure could better ensure that resources are appropriately marshaled for achieving the strategy’s vision and goals. Also, without an assessment of threats and risks, it is difficult to determine what the highest-priority activities should be. Linking the strategy’s development to a periodic assessment of threats and risks could help set priorities and ensure that resources are focused on the areas of greatest need. Moreover, such assessments could be helpful in tracking progress made in combating money laundering and terrorist financing. Furthermore, the establishment of accountability mechanisms could help to provide a basis for monitoring and assessing NMLS implementation. One possible mechanism would be linking the relevant agencies’ performance plans more closely to NMLS goals and objectives. Another mechanism would be to ensure that periodic progress reports are submitted to Congress, as currently required by the Strategy Act. In sum, if Congress decides to reauthorize the requirement for an annual NMLS, adoption of these critical components in the agencies’ future efforts could help to resolve or mitigate the deficiencies we identified. If Congress reauthorizes the requirement for an annual NMLS, we recommend that the Secretary of the Treasury, working with the Attorney General and the Secretary of Homeland Security, take appropriate steps to strengthen the leadership structure responsible for strategy development and implementation by establishing a mechanism that would have the ability to marshal resources to ensure that the strategy’s vision is achieved, resolve disputes between agencies, and ensure accountability for strategy implementation; link the strategy to periodic assessments of threats and risks, which would provide a basis for ensuring that clear priorities are established and focused on the areas of greatest need; and establish accountability mechanisms, such as (1) requiring the principal agencies to develop outcome-oriented performance measures that must be linked to the NMLS’s goals and objectives and that also must be reflected in the agencies’ annual performance plans and (2) providing Congress with periodic reports on the strategy’s results. We provided a draft of this report for review and comment to the Departments of the Treasury, Justice, and Homeland Security; seven federal financial regulatory agencies (FRB, FDIC, OCC, OTS, NCUA, SEC, and CFTC); and the National Security Council. In written comments, Treasury said that our recommendations for improving the process for creating the NMLS and enhancing accountability of all agencies with responsibility for combating financial crimes and the financing of terrorism are important, should Congress reauthorize the legislation requiring future strategies. Justice did not specifically address our recommendations but said that our observations and conclusions will be helpful in assessing the role that the strategy process has played in the federal government’s efforts to combat money laundering. For example, Justice concurred with our conclusion that linking the strategy’s development to a threat assessment could help set priorities and ensure that limited resources are focused on the areas of greatest need. DHS said that it would work with the Secretary of the Treasury as recommended and would do its part to implement necessary actions to address concerns raised in the report. Treasury, Justice, and DHS said that the lack of funds to finance NMLS development and implementation was an impediment and that the success of the HIFCA program in particular would be enhanced by an independent funding source. While we did not assess the participating agencies’ funding decisions regarding the NMLS or the HIFCA program, our report acknowledges that federal law enforcement agencies have resource constraints and competing priorities. We also note, however, that a primary purpose of the NMLS was to improve the coordination and quality of federal anti-money laundering investigations by concentrating and leveraging existing resources, including funding. Further, the report notes that HIFCA task force officials said that the lack of funding to compensate or reimburse participating state and local law enforcement agencies was a barrier to their participation. The 2002 NMLS called for an interagency team to examine how to fund the colocation of participants in HIFCA task forces absent funds appropriated for that purpose. At the time of our review, this initiative had not yet been completed. Treasury also said that it has satisfied the Strategy Act requirement that it submit a report to Congress—at the time the NMLS is submitted—on the effectiveness of policies to combat financial crimes. Treasury said that (1) evaluations of effectiveness have been contained in the NMLS itself and (2) any evaluation of effectiveness logically forms a part of the NMLS. While the annual strategies have contained some useful information to help Congress better understand programs to combat money laundering and terrorist financing, the strategies generally have provided descriptive information about NMLS initiatives rather than evaluations of the effectiveness of policies. As noted in our report, Treasury and Justice have efforts under way to measure performance that, when completed, could provide useful input into an overall evaluation of the effectiveness of policies to combat financial crimes. DHS highlighted the value of its Money Laundering Coordination Center, stating that the center has provided information to DEA, FBI, and other outside agencies on at least 46 occasions and that DEA was the most active outside agency user of the center, with at least 21 requests for assistance. While the sharing of relevant information is commendable, as mentioned in our report, DEA officials told us that the center does not meet DEA’s needs and that DEA has created a new database for information on money laundering investigations related to drugs. DHS also provided additional information on (1) methods used by ICE to coordinate terrorist financing investigations with other agencies and (2) steps taken by ICE and the FBI to implement the May 2003 memorandum of agreement between Justice and DHS regarding roles and responsibilities in investigating terrorist financing. The full text of Treasury’s, Justice’s, and DHS’s written comments are reprinted in appendix IV, V, and VI, respectively. The three departments also provided technical comments and clarifications, which have been incorporated in this report where appropriate. Of the seven federal financial regulatory agencies, four (FRB, FDIC, NCUA, and SEC) provided technical comments and clarifications, which have been incorporated in this report where appropriate. The other three agencies (OCC, OTS, and CFTC) had no comments. FDIC also said that, should a national money laundering strategy continue, annual goals should be achievable and roles and responsibilities clearly defined. The National Security Council did not respond to our request for comments. As arranged with your offices, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days after its issue date. At that time, we will send copies of this report to interested congressional committees and subcommittees. We will also make copies available to others on request. In addition, the report will be available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staffs have any questions about this report or wish to discuss the matter further, please contact Richard M. Stana at (202) 512- 8777 or by e-mail at stanar@gao.gov or Davi M. D’Agostino at (202) 512- 8678 or by e-mail at dagostinod@gao.gov. GAO contacts and key contributors to this report are listed in appendix VII. To determine agency perspectives on the benefit of the annual National Money Laundering Strategy (NMLS), we interviewed responsible officials at and reviewed relevant documentation obtained from the principal law enforcement components with anti-money laundering responsibilities at the Departments of the Treasury, Justice, and Homeland Security and the federal financial regulatory agencies. To determine whether the NMLS has served as a useful mechanism for guiding law enforcement agencies’ efforts, we (1) compared the structure and operation of High Intensity Money Laundering and Related Financial Crime Area (HIFCA) task forces to guidance provided in the strategies, (2) assessed whether the implementation of NMLS initiatives to enhance interagency coordination has met strategic goals, and (3) assessed the extent to which the 2002 NMLS addressed agency roles in combating terrorist financing. To do this, we interviewed responsible officials and reviewed documentation from the four primary agencies responsible for investigating money laundering and related financial crimes—Treasury’s Internal Revenue Service- Criminal Investigation (IRS-CI), Justice’s Federal Bureau of Investigation (FBI) and Drug Enforcement Administration (DEA), and Homeland Security’s Bureau of Immigration Control and Enforcement (ICE). For investigations of terrorist financing, we reviewed the roles and activities of and interviewed officials from ICE’s Operation Green Quest (OGQ) and two FBI components—Terrorist Financing Operations Section (TFOS) and Joint Terrorism Task Forces (JTTF). Our work with law enforcement agencies was conducted at the principal federal agencies’ headquarters in Washington, D.C., and at field office locations in three major U.S. financial centers (Los Angeles, Miami, and New York City). To determine the role of the NMLS in influencing the anti-money laundering activities of federal financial regulators, we reviewed their efforts to carry out the NMLS 2002 goal, “Prevent Money Laundering Through Cooperative Public-Private Efforts and Necessary Regulatory Measures,” and its earlier iterations. We gathered information on these agencies’ anti-money laundering and antiterrorist financing efforts— including efforts to implement provisions of the Uniting and Strengthening America by Providing Appropriate Tools Required to Intercept and Obstruct Terrorism Act of 2001 (USA PATRIOT Act)—and determined the influence of the NMLS on those efforts. We also examined the role the financial regulators played in supporting Treasury’s efforts under the NMLS goal to strengthen international cooperation to fight money laundering. To do this work, we interviewed responsible headquarters officials and reviewed documentation from the Commodity Futures Trading Commission (CFTC), Federal Deposit Insurance Corporation (FDIC), Federal Reserve Board (FRB), National Credit Union Administration (NCUA), Office of the Comptroller of the Currency (OCC), Office of Thrift Supervision (OTS), Treasury, and Securities and Exchange Commission (SEC). We also interviewed officials from the law enforcement agencies listed above to assess coordination between law enforcement and the financial regulators. To compare NMLS efforts to the components we have found are necessary for a successful strategy, we reviewed drafts of the strategies from 1999 to 2002, interviewed officials who had been involved in the development and implementation of the strategies, and compared the results from this work with findings from our past work reviewing national strategies and their implementation. We conducted our work from June 2002 to August 2003 in accordance with generally accepted government auditing standards. The U.S. government’s framework for preventing, detecting, and prosecuting money laundering has been expanded through additional pieces of legislation since its inception in 1970 with the Bank Secrecy Act (BSA). The BSA required, for the first time, that financial institutions maintain records and reports that financial regulators and law enforcement agencies have determined have a high degree of usefulness in criminal, tax, and regulatory matters. The BSA authorizes the Secretary of the Treasury to issue regulations on the reporting of certain currency transactions. The BSA had three main objectives: create an investigative audit trail through regulatory reporting standards; impose civil and criminal penalties for noncompliance; and improve detection of criminal, tax, and regulatory violations. The reporting system implemented under the BSA was by itself an insufficient response to money laundering because, under the BSA, anybody who satisfied the reporting requirements would not be subject to money laundering violations. Thus, Congress enacted the Money Laundering Control Act of 1986 (MLCA), which made money laundering a criminal offense separate from any BSA reporting violations. It created criminal liability for individuals or entities that conduct monetary transactions knowing that the proceeds involved were obtained from unlawful activity and made it a criminal offense to knowingly structure transactions to avoid BSA reporting. Penalties under the MLCA include imprisonment, fines, and forfeiture. Congress enacted the Money Laundering Prosecution Improvements Act of 1988 to enhance the provisions of the BSA and the MLCA and amended provisions in both statutes. The Improvements Act aimed to increase the cooperation that the government receives from financial institutions by imposing liability and fines on facilitators, such as negligent bankers. It also expanded the definition of a financial institution under the BSA and permitted government agencies to undertake sting operations. The Annunzio-Wylie Anti-Money Laundering Act of 1992 amended the BSA in a number of ways. It authorized Treasury to require financial institutions to report any suspicious transaction relevant to a possible violation of a law. It also authorized Treasury to require financial institutions to carry out anti-money laundering programs and create record-keeping rules relating to fund transfer transactions. Annunzio- Wylie also made the operation of an illegal money transmitting business a crime. As authorized by Annunzio-Wylie, in 1996, Treasury issued a rule requiring that banks and other depository institutions use a Suspicious Activity Report (SAR) form to report activities involving possible money laundering. During the same year, bank regulators issued regulations requiring all depository institutions to report suspected money laundering as well as other suspicious activities using this form. The bank regulators also placed SAR requirements on the subsidiaries, including broker-dealer firms, of the depository institutions and their holding companies under their jurisdiction. The Money Laundering and Financial Crimes Strategy Act of 1998 (Strategy Act) amended the BSA to require the President, acting through the Secretary of the Treasury, in consultation with the Attorney General and other relevant agencies, including state and local agencies, to coordinate and implement a national strategy, produced annually for 5 years beginning in 1999, to address money laundering. In addition, it requires the Secretary of the Treasury to designate certain areas as high- risk areas for money laundering and related financial crimes and to establish a Financial Crime-Free Communities Support Program. The purpose of demarcating areas as high risk is to designate the communities that experience severe problems with money laundering that need more help. The Strategy Act also authorizes federal funding of efforts by state and local law enforcement agencies to investigate money laundering activities. In 1999, Treasury consulted with 18 federal agencies, bureaus, and offices in developing the NMLS. By 2002, that number had increased to over 25. The Strategy Act provides that the NMLS should include: 1. Goals for reducing money laundering and related financial crimes in the United States. 2. Goals for coordinating regulatory efforts to prevent the exploitation of financial systems in the United States through money laundering. 3. A description of operational initiatives to improve the detection and prosecution of money laundering and related financial crimes and the seizure and forfeiture of the proceeds derived from those crimes. 4. The enhancement of partnerships between the private financial sector and law enforcement agencies with regard to the prevention and detection of money laundering and related financial crimes. 5. The enhancement of cooperative efforts between the federal government and state and local officials, including state and local prosecutors and other law enforcement officials; and cooperative efforts among the several states and between state and local officials, including state and local prosecutors and other law enforcement officials. In the wake of the September 11 terrorist attacks, Congress enacted the Uniting and Strengthening America by Providing Appropriate Tools Required to Intercept and Obstruct Terrorism Act (USA PATRIOT Act) on October 25, 2001. The passage of the USA PATRIOT Act was prompted, in part, by the enhanced awareness of the importance of combating terrorist financing as part of the U.S. government’s overall anti-money laundering efforts, because terrorist financing and money laundering both involve similar techniques. Title III of the USA PATRIOT Act, among other things, expands Treasury’s authority to regulate the activities of U.S. financial institutions; requires the promulgation of regulations; imposes additional due diligence requirements; establishes new customer identification requirements; and requires financial institutions to maintain anti-money laundering programs. In addition, title III adds activities that can be prosecuted as money laundering crimes and increases penalties for activities that were money laundering crimes prior to enactment of the USA PATRIOT Act. Further, title III amends various sections of the BSA, the MLCA, and other statutes. Appendix III contains a detailed summary of key provisions in title III of the USA PATRIOT Act. Due diligence (Section 312) Amends 31 U.S.C. § 5318 by requiring U.S. financial institutions to exercise due diligence and, in some cases, enhanced due diligence, when opening or operating correspondent accounts for foreign financial institutions or private banking accounts for wealthy foreign individuals. Also requires U.S. financial institutions to establish due diligence policies, procedures, and controls reasonably designed to detect and report money laundering through such correspondent and private banking accounts. Sets forth certain minimum due diligence and enhanced due diligence requirements and otherwise adopts a risk-based approach, permitting financial institutions to tailor their programs to their own lines of business, financial products and services offered, size, customer base, and location. Defers the application of Section 312 requirements and provides interim compliance guidance pending issuance of a final rule. Shell bank ban and record keeping (Sections 313 and 319(b)) Amends 31 U.S.C. § 5318 by prohibiting U.S. banks and securities firms from opening or maintaining accounts for foreign shell banks, meaning banks that have no physical presence anywhere and no affiliation with another, non-shell bank. Requires closure of any existing correspondent accounts for foreign shell banks by December 2001. Also requires U.S. firms to take reasonable steps to ensure no foreign bank client is allowing a foreign shell bank to utilize the foreign bank’s U.S. correspondent account. Also requires foreign banks with U.S. correspondent accounts to identify their owners and designate a U.S. resident to accept legal service of a government subpoena or summons. Allows U.S. to subpoena documents related to the foreign bank's U.S. account whether the documents are inside or outside the U.S. Allows the Attorney General or Treasury to require closure of a foreign bank's U.S. account if the foreign bank ignores a government subpoena or summons. Defines the scope of the application of the shell bank prohibition and record keeping requirement and provides certification form to aid covered financial institutions in complying with the rule. Public-private cooperation (Section 314) Enables and encourages two forms of information sharing to deter terrorism and money laundering: (1) among law enforcement, the regulators, and financial institutions (314(a)); and (2) among financial institutions themselves (314(b)). Creates a communication network to link federal law enforcement agencies with financial institutions so that information relating to suspected terrorists and money launderers can be exchanged quickly and without compromising pending investigations. Federal law enforcement agencies can provide names of suspected terrorists and money launderers to financial institutions through Treasury’s Financial Crimes Enforcement Network (FinCEN). After receiving the information, financial institutions are required to check accounts and transactions involving suspects and report matches. Law enforcement agencies can then follow up with the financial institution directly. Permits the sharing of information relating to individuals and entities suspected of money laundering or terrorism among financial institutions so long as financial institutions provide a yearly notice to FinCEN of their intent to share information under this provision. Clarifies that commodity firms and credit unions are "financial institutions" subject to Title 31's anti-money laundering provisions. Concentration accounts (Section 325) Amends 31 U.S.C. § 5318(h) by authorizing Treasury regulations to ensure that client funds moving through a financial institution's administrative accounts do not move anonymously, but are marked with the client’s name. Customer verification (Section 326) Amends 31 U.S.C. § 5318 by requiring all U.S. financial institutions to implement procedures to verify the identity of any person seeking to open an account and requires all clients to comply with such procedures. Requires Treasury jointly with financial regulators to implement rules requiring financial institutions to comply. Requires banks, savings associations, credit unions, private banks, and trust companies to implement procedures to verify the identity of any person seeking to open an account. Requires mutual funds to implement procedures to verify the identity of any person seeking to open an account. Requires futures commission merchants and introducing brokers to implement procedures to verify the identity of any person seeking to open an account. Requires broker-dealers to implement procedures to verify the identity of any person seeking to open an account. Requires certain banks lacking a federal functional regulator to implement procedures to verify the identity of any person seeking to open an account. Report to Congress on ways to improve the ability of financial institutions to identify foreign nationals. International cooperation (Section 330) Directs the United States to negotiate with other countries to increase anti-money laundering and terrorist financing cooperation, ensure adequate record keeping of transactions and accounts related to money laundering or terrorism. Anti-money laundering programs (Section 352) Amends 31 U.S.C. § 5318(h) to require all U.S. financial institutions to establish anti-money-laundering programs. Authorizes Treasury, after consulting with the appropriate federal regulators to prescribe minimum standards for these programs. Banks, savings associations, credit unions, registered brokers and dealers in securities, futures commission merchants, and casinos deemed in compliance with Section 352 if they establish and maintain anti-money laundering programs pursuant to existing BSA rules, or rules adopted by their federal functional regulator or self-regulatory organization. Required operators of a credit card system to establish programs reasonably designed to detect and prevent money laundering and the financing of terrorism. Required money services businesses to establish programs reasonably designed to prevent money laundering and the financing of terrorism. Required mutual funds to establish programs reasonably designed to detect and prevent money laundering and the financing of terrorism. Would require certain insurance companies (those offering life or annuity products) to establish programs reasonably designed to detect and prevent money laundering and the financing of terrorism. Prescribes minimum anti-money laundering standards applicable to insurance companies. Would require unregistered investment companies, such as hedge funds, commodity pools, and similar investment vehicles, to establish programs reasonably designed to detect and prevent money laundering and the financing of terrorism. Extends the provision that temporarily defers, for dealers in precious metals, pawnbrokers, loan or finance companies, private bankers, insurance companies, travel agencies, telegraph companies, sellers of vehicles, persons engaged in real estate closings, certain investment companies, commodity pool operators and commodity trading advisers, the requirement to adopt anti-money laundering programs. Would require dealers in precious metals, stones, or jewels to establish programs reasonably designed to detect and prevent money laundering and the financing of terrorism. Solicits comments on whether businesses engaged in vehicle sales should be required to adopt anti- money laundering programs under Section 352. Solicits comments on how to define travel agency and whether such persons should be required to adopt anti-money laundering programs under Section 352. Solicits comments on how to define persons involved in real estate closings and settlements and whether certain of these persons should be exempt from having anti-money laundering programs under section 352. Would require certain investment advisers that manage client assets to establish programs reasonably designed to detect and prevent money laundering and the financing of terrorism. Would delegate authority to examine certain investment advisers for compliance with such programs to SEC. Would require certain commodity trading advisors to establish programs reasonably designed to detect and prevent money laundering and the financing of terrorism. Would delegate the authority to examine such commodity trading advisers to the CFTC. Requires all U.S. securities firms to report suspicious financial activity to U.S. law enforcement under regulations to be published by July 1, 2002. Authorizes Treasury to issue regulations requiring suspicious activity reporting by commodity firms; and requires a report and recommendations by October 2002 on effective regulations applying anti-money laundering reporting and other requirements to investment companies. Requires brokers or dealers in securities to report suspicious transactions. Would require mutual funds to file suspicious activity reports. Would add futures commission merchants and introducing brokers in commodities to the regulatory definition of "financial institution'' and would require them to report suspicious transactions. Would require insurance companies to file suspicious activity reports. Report to Congress recommending additional regulations applicable to investment companies. Reporting of suspicious activities by underground banking systems (Section 359) Amends the definition of "financial institution" in 31 U.S.C.§ 5312(a)(2)(r) to include a licensed sender of money or any other person who engages as a business in the transmission of funds. Also directs the Secretary of the Treasury to issue a report to Congress on or before October 26, 2002, detailing the need for any additional legislation regarding the regulation of informal banking networks. Report to Congress discussing informal value transfer systems and suggesting methods for minimizing potential abuse of such systems. Adds new section 5331 to title 31 of the U.S. Code and requires any person who is engaged in a trade or business and who, in the course of such trade or business, receives more than $10,000 in coins or currency in one transaction (or two or more related transactions) to file a report with FinCEN. Deems the filing of form 8300 with the Internal Revenue Service to satisfy the requirement to file a currency report with FinCEN. Money laundering designations (Section 311) Amends 31 U.S.C. § 5318 by authorizing Treasury to designate specific foreign financial institutions, jurisdictions, transactions or accounts to be of "primary money laundering concern." Mandates special measures to restrict or prohibit access to the U.S. market. Would impose special measures against Nauru, requiring U.S. financial institutions to terminate all correspondent accounts involving Nauru. Amends 18 U.S.C. § 1956(c)(7) by expanding the list of crimes that can trigger U.S. money laundering prosecutions to include foreign corruption crimes such as bribery and misappropriation of funds. Also, expands the list of crimes that can trigger U.S. money laundering prosecutions to include weapons smuggling, export control violations, certain computer crimes, bribery, and other extraditable offenses. Antiterrorist forfeiture protection (Section 316) Authorizes any person whose property is confiscated as terrorist assets to contest the confiscation through civil proceedings in the United States. Long arm jurisdiction over foreign money launderers antiterrorist forfeiture protection (Section 317) Amends 18 U.S.C. § 1956(b) by giving U.S. courts jurisdiction over persons who commit a money laundering offense through financial transactions that take place in whole or in part in the United States, over foreign banks with U.S. accounts, and over foreign persons who convert to their personal use property that is the subject of a forfeiture order. Allows U.S. prosecutors and federal and state regulators to use court-appointed receivers in criminal and civil money laundering proceedings to locate and take custody of a defendant’s assets wherever located. Requires U.S. banks to respond within 120 hours to a request by a federal banking agency for money laundering information. Laundering money through a foreign bank (Section 318) Forfeiture of funds in United States interbank accounts (Section 319(a)) Amends 18 U.S.C. § 981 by closing a forfeiture loophole so that depositors’ funds in a foreign bank housing a U.S. bank account are subject to the same forfeiture rules as depositors’ funds in other U.S. bank accounts. Proceeds of foreign crimes (Section 320) Amends 18 U.S.C. § 981(a)(1)(B) to authorize the forfeiture of both the proceeds of, and any property used to facilitate, offenses listed in section 1956(c)(7)(B), if the offense would be a felony if committed within the jurisdiction of the United States. Corporation represented by a fugitive (Section 322) Amends 18 U.S.C. § 2466 by applying the fugitive disentitlement doctrine to claims filed by corporations if any majority shareholder, or individual filing the claim on behalf of the corporation is disqualified from contesting the forfeiture. It clarifies that a natural person who is a fugitive may not file, or have another person file, a claim on behalf of a corporation that the fugitive controls. Enforcement of foreign judgments (Section 323) Amends 28 U.S.C. § 2467 by allowing the government to apply for and the court to issue a restraining order to preserve the availability of property subject to a foreign forfeiture or confiscation judgment. Consideration of anti-money laundering record (Section 327) Amends 12 U.S.C. § 1842(e) by requiring U.S. bank regulators to consider when approving a bank merger or acquisition the anti-money laundering records of the banks involved. International cooperation on identification of originators of wire transfers (Section 328) Requires Treasury to consult with the U.S. Attorney General and the Secretary of State to take all reasonable steps to encourage foreign governments to require the inclusion of the name of the originator in wire transfer instructions sent to the United States. Criminal penalties (Section 329) Any person who is an official or employee of any federal agency who, in connection with administration of the anti-money laundering provisions in the Patriot Act, corruptly receives anything of value in return for being influenced in the performance of any official act will be fined, or imprisoned for 15 years, or both. Amendments relating to reporting of suspicious activities (Section 351) Amends 31 U.S.C. § 5318(g)(3) so that any financial institution that makes a voluntary disclosure of any possible violation of a law or regulation relating to money laundering is not liable to any other person for such disclosure. Penalties for violations of geographic targeting orders (Section 353) Amends 31 U.S.C. § 5321(a)(1), 5322, 5324(a), 5326(d). Under prior law Treasury has had the authority to issue orders requiring any domestic financial institution in a geographic area to perform additional record keeping and reporting requirements if reasonable grounds exist for concluding that additional requirements are necessary to carry out anti-money laundering requirements. These amendments expand civil and criminal penalties to include violations of geographic targeting orders issued and violations of regulations. Authorization to include suspicions of illegal activity in written employment references (Section 355) Amends 12 U.S.C. § 1828 to authorize certain depository institutions to disclose in a written employment reference information concerning a possible involvement in potentially unlawful activity. Banks secrecy provisions and activities of United States intelligence agencies to fight international terrorism (Section 358) Amends the Right to Financial Privacy Act of 1978, 12 U.S.C. § 3412(a), to allow law enforcement authorities to obtain financial data related to intelligence or counterintelligence activities, investigations, or analysis in an effort to protect against international terrorism. Increase in civil and criminal penalties for money laundering (Section 363) Amends 31 U.S.C. § 5321(a) & 5322 by increasing from $100,000 to $1,000,000 the maximum civil and criminal penalties for a violation of provisions added to the Bank Secrecy Act by sections 311 and 312 of the Patriot Act. Bulk cash smuggling (Section 371) Adds section 5332 to Title 31 of the U.S. Code, which makes smuggling large amounts of cash across U.S., borders a crime. Forfeiture in currency reporting cases (Section 372) Amends 31 U.S.C. § 5317(c), and allows forfeiture of undeclared cash whose source and intended use cannot be established. Illegal money transmitting businesses (Section 373) Amends 18 U.S.C. § 1960 -- which prohibits operation of an unlicensed money transmission business -- to abolish any requirement that the defendant be aware of the laws requiring money transmitting licenses. Counterfeiting domestic currency and 0bligations (Section 374) Amends 18 U.S.C. § 470, which prohibits the use of electronic images in counterfeiting. Counterfeiting foreign currency and obligations (Section 375) Amends 18 U.S.C. § 478 by providing that the penalties for counterfeiting are increased (generally to allow a maximum term of imprisonment of 20 years). Laundering the proceeds of terrorism (Section 376) Amends 18 U.S.C. § 1956(c)(7)(D) which provides material support to designated foreign terrorist organizations, as a predicate offense for a money laundering prosecution. Extraterritorial jurisdiction (Section 377) Amends 18 U.S.C. § 1029. Enhances the applicability of computer fraud by covering offenses committed outside the United States that involves an access device issued by a U.S. entity. Terrorism (Sections 801 to 817) Modernizes anti-terrorism criminal statutes by, among other provisions, making it clear the crime includes bioterrorism, mass transit terrorist acts, cyberterrorism, harboring of terrorists and support for terrorists; that all terrorist crimes serve as predicate offenses for money laundering prosecutions; and that anti-money laundering provisions apply to all terrorists assets, including legally obtained funds, if intended for use in planning, committing or concealing a terrorist act. Exclusion of aliens (Section 1006) Permits the United States to exclude any alien engaged in money laundering from the United States and requires establishment of a money laundering watch list for officials admitting aliens into the United States. (3) Required Reports by Treasury Report and recommendation (Section 324) Requires Treasury within 30 months of enactment of the Patriot Act to make a report on operations respecting the provisions relating to international counter-money laundering measures and any recommendations to Congress as to advisable legislative action. Anti-money laundering strategy (Section 354) Amends 31 U.S.C. § 5341(b) by directing the Secretary of the Treasury to consider data regarding the funding of terrorism and efforts directed at the prevention, detection and prosecution of such funding as topics for the Anti-Money Laundering Strategy. Special report on administration of bank secrecy provisions (Section 357) Treasury must submit a report to Congress relating to the role of the IRS in the administration of the records and reports on monetary instrument transactions within 6 months of enactment of the Patriot Act. Report submitted on April 26, 2002. Efficient use of currency transaction report system (Section 366) Directs Treasury to review the cash transaction reporting system to make it more efficient, possibly by expanding the use of exemptions to reduce the volume of reports, and to submit a report by October 25, 2002. Report submitted on October 25, 2002. (4) Miscellaneous Provisions Use of authority of United States executive directors (Section 360) Allows Treasury to instruct the United States Executive Director of each international financial institution to use the voice and vote of the Executive Director to support loans and use of funds of respective institutions or public and private entities within the country if the President determines that a foreign country has taken actions supporting the United State's effort to combat terrorism. Financial crimes enforcement network (Section 361) Amends 31 U.S.C. § 310 by specifying the responsibilities of FinCEN's director, expanding the duties of FinCEN, and, giving it statutory authority to perform its functions. Establishment of highly secure network (Section 362) Directs Treasury to establish within FinCEN a highly secure electronic network through which reports (including SARs) may be filed and information regarding suspicious activities warranting immediate and enhanced scrutiny may be provided to financial institutions. Uniform protection authority for Federal Reserve facilities (Section 364) Amends 12 U.S.C. § 248 by allowing law enforcement officers to protect and safeguard Federal Reserve facilities. In addition to those named above, Allison Abrams, Thomas Conahan, Eric Erdman, Barbara Keller, Marc Molino, Jan Montgomery, Robert Rivas, Barbara Roesmann, and Sindy Udell made key contributions to this report. Internet Gambling: An Overview of the Issues. GAO-03-89. Washington, D.C.: December 2, 2002. Interim Report on Internet Gambling. GAO-02-1101R. Washington, D.C.: September 23, 2002. Money Laundering: Extent of Money Laundering through Credit Cards is Unknown. GAO-02-670. Washington, D.C.: July 22, 2002. Money Laundering: Oversight of Suspicious Activity Reporting at Bank- Affiliated Broker-Dealers Ceased. GAO-01-474. Washington, D.C.: March 22, 2001. Suspicious Banking Activities: Possible Money Laundering by U.S. Corporations Formed for Russian Entities. GAO-01-120. Washington, D.C.: October 31, 2000. Money Laundering: Observations on Private Banking and Related Oversight of Selected Offshore Jurisdictions. GAO/T-GGD-00-32. Washington, D.C.: November 9, 1999. Private Banking: Raul Salinas, Citibank, and Alleged Money Laundering. GAO/T-OSI-00-3. Washington, D.C.: November 9, 1999. Private Banking: Raul Salinas, Citibank, and Alleged Money Laundering. GAO/OSI-99-1. Washington, D.C.: October 30, 1998. Money Laundering: Regulatory Oversight of Offshore Private Banking Activities. GAO/GGD-98-154. Washington, D.C.: June 29, 1998. Money Laundering: FinCEN’s Law Enforcement Support Role Is Evolving. GAO/GGD-98-117. Washington, D.C.: June 19, 1998. Money Laundering: FinCEN Needs to Better Manage Bank Secrecy Act Civil Penalties. GAO/GGD-98-108. Washington, D.C.: June 15, 1998. Money Laundering: FinCEN’s Law Enforcement Support, Regulatory, and International Roles. GAO/T-GGD-98-83. Washington, D.C.: April 1, 1998. Money Laundering: FinCEN Needs to Better Communicate Regulatory Priorities and Timelines. GAO/GGD-98-18. Washington, D.C.: February 6, 1998. Private Banking: Information on Private Banking and Its Vulnerability to Money Laundering. GAO/GGD-98-19R. Washington, D.C.: October 30, 1997. Money Laundering: A Framework for Understanding U.S. Efforts Overseas. GAO/GGD-96-105. Washington, D.C.: May 24, 1996. Money Laundering: U.S. Efforts to Combat Money Laundering Overseas. GAO/T-GGD-96-84. Washington, D.C.: February 28, 1996. Money Laundering: Stakeholders View Recordkeeping Requirements for Cashier’s Checks As Sufficient. GAO/GGD-95-189. Washington, D.C.: July 25, 1995. Money Laundering: U.S. Efforts to Fight It Are Threatened by Currency Smuggling. GAO/GGD-94-73, Washington, D.C.: March 9, 1994. Money Laundering: Characteristics of Currency Transaction Reports Filed in Calendar Year 1992. GAO/GGD-94-45FS. Washington, D.C.: November 10, 1993. Money Laundering: Progress Report on Treasury’s Financial Crimes Enforcement Network. GAO/GGD-94-30. Washington, D.C.: November 8, 1993. Money Laundering: The Use of Bank Secrecy Act Reports by Law Enforcement Could Be Increased. GAO/T-GGD-93-31. Washington, D.C.: May 26, 1993. Money Laundering: State Efforts to Fight It Are Increasing but More Federal Help Is Needed. GAO/GGD-93-1. Washington, D.C.: October 15, 1992. Money Laundering: Civil Penalty Referrals for Violations of the Bank Secrecy Act Have Declined. GAO/T-GGD-92-57. Washington, D.C.: June 30, 1992. Tax Administration: Money Laundering Forms Could Be Used to Detect Nonfilers. GAO/T-GGD-92-56. Washington, D.C.: June 23, 1992. Money Laundering: Treasury Civil Case Processing of Bank Secrecy Act Violations. GAO/GGD-92-46. Washington, D.C.: February 6, 1992. Money Laundering: The Use of Cash Transaction Reports by Federal Law Enforcement Agencies. GAO/GGD-91-125. Washington, D.C.: September 25, 1991. Money Laundering: The U.S. Government Is Responding to the Problem. GAO/NSIAD-91-130. Washington, D.C.: May 16, 1991. Money Laundering: Treasury’s Financial Crimes Enforcement Network. GAO/GGD-91-53. Washington D.C.: March 18, 1991. | Money laundering is a serious crime, with hundreds of billions of dollars laundered annually. Congress passed the Money Laundering and Financial Crimes Strategy Act of 1998 to better coordinate the efforts of law enforcement agencies and financial regulators in combating money laundering. This act required the issuance of an annual National Money Laundering Strategy for 5 years, ending with the issuance of the 2003 strategy. To help with deliberations on reauthorization, GAO determined (1) agency perspectives on the benefit of the strategy and factors that affected its development and implementation, (2) whether the strategy has served as a useful mechanism for guiding the coordination of federal law enforcement agencies' efforts, (3) the role of the strategy in influencing the activities of federal financial regulators, and (4) whether the strategy has reflected key critical components. Treasury, Justice, and financial regulatory officials with whom GAO spoke said that the National Money Laundering Strategy was initially beneficial but that, over time, certain factors and events affected its development and implementation. They endorsed the concept of a strategy to coordinate the federal government's efforts to combat money laundering and related financial crimes. They also said that the strategy initially had a positive effect on promoting interagency planning and communication, but different agency views emerged over the scope and commitment required, and other events affected the strategy, such as the September 11 terrorist attacks and the creation of the Department of Homeland Security. The strategy generally has not served as a useful mechanism for guiding the coordination of federal law enforcement agencies' efforts to combat money laundering and terrorist financing. While Treasury and Justice made progress on some strategy initiatives designed to enhance interagency coordination of money laundering investigations, most have not achieved the expectations called for in the annual strategies. Also, the 2002 strategy did not address agency roles in investigating terrorist financing, thus, it did not help resolve potential duplication of efforts and disagreements over which agency should lead investigations. In May 2003, Justice and Homeland Security reached an agreement aimed at resolving these problems. Most financial regulators GAO interviewed said that the strategy had some influence on their anti-money laundering efforts because it provided a forum for enhanced coordination, particularly with law enforcement agencies. However, they said that it has had less influence than other factors. They described several other influences on their efforts, particularly their ongoing oversight responsibilities in ensuring compliance with the Bank Secrecy Act and, more recently, the USA PATRIOT Act, which was passed in October 2001 to fight terrorist financing and increase anti-money laundering efforts. GAO's work reviewing national strategies has identified several critical components needed for development and implementation; however, key components have not been well reflected in the strategy. The first is clearly defined leadership, with the ability to marshal necessary resources. However, the leadership for the strategy has not agreed on the strategy's scope or ensured that target dates for completing initiatives were met. The second is clear priorities, as identified by threat and risk assessments, to help focus resources on the greatest needs. Each strategy contained more priorities than could be realistically achieved and none of the strategies was linked to a threat and risk assessment. The third is that established accountability mechanisms provide a basis for monitoring and assessing program performance. While later strategies contained several initiatives designed to establish performance measures, as of July 2003, none had yet been completed. Officials attributed this to the difficulty in establishing such measures for combating money laundering. |
Education administers three programs—the TEACH Grant, Stafford Teacher Loan Forgiveness, and Perkins Loan Teacher Cancellation—that may help attract and retain high-quality teachers in high-need schools and subjects by helping teachers pay for school. TEACH Grant. Individual colleges and universities elect whether to participate in the program and choose the grade levels and education programs in which students may receive the grant. Students participating in the TEACH Grant program may generally receive up to $4,000 a year while in school and must fulfill teaching service requirements after graduation or the grant converts to an unsubsidized Direct Loan. Stafford Teacher Loan Forgiveness. Teachers who teach for 5 consecutive years in a low-income school can receive up to $5,000 in loan forgiveness or up to $17,500 if also teaching in certain subjects. Perkins Loan Teacher Cancellation. Eligible teachers can have up to 100 percent of their Perkins loans cancelled over the course of 5 years of qualifying service. The maximum annual amount a student can borrow under the Perkins loan program is generally $5,500 for undergraduates and $8,000 for graduate and professional students. Table 1 describes program eligibility, service requirements, benefit amounts, and participating schools for these three programs, which are the subject of our review. 20 U.S.C. §§ 1078-10, 1087j. program requirements, may have implications for teacher eligibility for and benefits under the Stafford Loan Forgiveness program. PSLF was enacted in 2007 and is meant to encourage individuals to enter and continue to work full-time in public service jobs, such as teaching, by offering loan forgiveness to qualified public service employees. To benefit from these three programs, grant recipients and borrowers must periodically submit paperwork that verifies eligible employment (see fig. 1). Until their service requirement is fulfilled, TEACH grant recipients are required—per an Agreement to Serve that they sign when in school—to annually certify that they intend to teach or that they are currently teaching in a qualified position and school, otherwise their grant will convert to an unsubsidized Direct Loan. If that happens, the grant recipient must pay back the full amount of the grant award, plus interest, which is calculated from the date the TEACH grant was disbursed (see table 2).requirements. The TEACH Grant and the loan forgiveness programs are administered by FSA. For all three programs, FSA develops policies and regulations, provides oversight, and monitors compliance by participants, including schools, financial entities, loan servicers, and students. FSA is also responsible for outreach on how to apply for and receive aid. Additionally, FSA is to develop borrower guidance and maintain StudentAid.gov—a centralized source for information about federal student aid programs—to help borrowers and grant recipients understand the repayment process and the terms and conditions of the programs. In 1998, the FSA Ombudsman was established to resolve complaints from student loan borrowers and to make recommendations for improving service within FSA. According to FSA’s 2014 annual report, the Ombudsman has received more than 300,000 customer contacts since it began operations in 1999. FSA contracts with loan servicers to assist in administering the TEACH Grant and Stafford loan programs. These loan servicers are independent entities that provide billing and repayment services, inform borrowers about their repayment options, and respond to customer service inquiries. Much of the TEACH Grant program is administered by a single servicer who manages all active grants and most grants that have been converted to loans.when their employment certification paperwork is due and send quarterly interest notices informing grant recipients of the amount they would owe, including interest, if their grant were to convert to a loan. A borrower’s Stafford loans are serviced by 1 of 11 FSA-contracted companies. Among other duties, Stafford servicers process applications for forbearance and loan forgiveness once a borrower has completed their service obligation. Among other duties, the servicer is to remind grant recipients Servicers review applications for eligibility and forward recommendations for approval to FSA for final review. In contrast, while FSA is responsible for overall management of the Perkins loan program, it is campus-based and administered directly by participating schools. Perkins loans are made directly by schools using a combination of federal and institutional funds, and borrowers repay these loans to their school.universities are generally responsible for determining eligibility, processing applications for benefits and loan deferment, disbursing benefits and collecting repayments, and advising and providing technical assistance to borrowers, among other responsibilities. About one-fourth of the more than 410,000 teacher aid recipients benefited from TEACH grants. Between school year 2008-2009, when TEACH grants were first awarded, and October 2014, more than 112,000 students were awarded grants from participating colleges and universities, according to National Student Loan Data System (NSLDS) figures. While TEACH Grant participation is decreasing for both undergraduates and graduates, participation rates among graduate students are declining at a slightly higher rate, falling 17 percent since Education conducts analysis for internal award year 2011 (see fig. 2).purposes, but it has not assessed program participation, including why participation is declining, as part of any evaluation or study as discussed later in the report. About three-fourths of the more than 410,000 teacher aid recipients have received loan forgiveness or cancellation. Over the past decade, more than 298,000 teachers have participated in the Stafford Teacher Loan Forgiveness and the Perkins Loan Teacher Cancellation programs. Participation in the Stafford Teacher Loan Forgiveness program has increased each year, benefiting 53 times more participants in fiscal year 2014 than in fiscal year 2004. In contrast, participation in the Perkins Loan Teacher Cancellation program has decreased significantly—69 percent since fiscal year 2004. The steady decrease from fiscal years 2011 through 2014 may be due at least in part to a lapse in funding.example, according to Education budget documents, Congress last appropriated federal reimbursements to participating schools for Perkins loan cancellations in fiscal year 2009. As no estimates exist of the potentially eligible populations for the TEACH Grant and Stafford Teacher Loan Forgiveness programs, we developed estimates by reviewing program requirements and information on students and teachers from the Integrated Postsecondary Education Data We System (IPEDS) and the School and Staffing Survey (SASS). estimated that 19 percent and 0.8 percent of the potentially eligible population for the TEACH Grant and Stafford Teacher Loan Forgiveness programs, respectively, are benefiting from these programs. For the TEACH Grant program, we estimated that in school year 2012-2013 approximately 188,000 students graduated or completed a potentially eligible program from participating schools and 36,326 students (19 percent) received grants.school year 2011-2012 approximately 3.4 million teachers taught in low- income schools, however some of those teachers may not have eligible For the Stafford program, we estimated that in loans or may not yet have taught the required 5 years to receive Stafford loan forgiveness. According to Education’s data, 27,023 teachers benefited from the program in fiscal year 2012, representing 0.8 percent of the estimated 3.4 million teachers in low-income schools that year. For more information on our methodology, see Appendix I. Participation in the TEACH Grant and the loan forgiveness programs might be higher if more teachers and students were aware of them, according to college administrators in our focus groups, students, and other stakeholders we interviewed. For example, officials from one higher education association said that awareness of these programs is low and that Education does not promote them to potentially eligible students and teachers. While college administrators in all four of our focus groups with officials from schools that offer the grants said that colleges inform students about the TEACH Grant, typically in advising or recruiting sessions or by speaking directly with qualifying students, more than half of the 23 students and recent graduates we interviewed said they received little to no information about the grant, including 9 who attended schools that offered the program. Awareness of the loan forgiveness programs appears to be similarly limited. Officials from both teacher unions, and 6 of 23 students and recent graduates we interviewed, cited a lack of awareness about teacher loan forgiveness programs. For example, one recent graduate said she did not know that her Perkins loan could have been cancelled until she had already paid it off. Participants in two of the four focus groups that contained financial aid officers also said that students may not be aware that a Perkins loan can be cancelled. Education officials stated that they provide information on the TEACH Grant and the loan forgiveness programs online and through loan servicers who are in contact with participants during repayment. Specifically, Education officials said that students are made aware of these programs via resources such as: its website (StudentAid.gov); the master promissory note all students sign when applying for a loan; and exit counseling, which is conducted by schools when students graduate. Education requires institutions and lenders to provide certain information at various intervals throughout the loan repayment process. Further, while the Stafford loan servicers we interviewed said they are not required to advertise loan forgiveness programs, they provide program information to borrowers on their websites. While Education makes information on these programs available, awareness among potential participants continues to be a challenge, according to college administrators in our focus groups, students, and other stakeholders we interviewed. Education’s FSA has a stated goal to actively inform all eligible individuals of their funding options. However, a senior Education official contended that conducting outreach to promote these programs to students and teachers may be difficult because the agency does not capture information on who is teaching and therefore is unaware of teachers who may be eligible. Another senior Education official noted that while Education is responsible for awarding student loans, colleges and teachers’ workplaces also have a role in informing borrowers about these programs. Nevertheless, senior Education officials stated that they will continue to consider ways to increase awareness of teacher loan forgiveness benefits. In the absence of additional outreach about all three teacher aid programs, however, students and teachers may be unaware of opportunities to receive financial assistance, and Education may be missing opportunities to expand the pool of highly qualified teachers, particularly for low-income schools where retention has been a challenge. While most TEACH grant recipients are still enrolled in an eligible teacher training program or otherwise are fulfilling requirements, about one-third of recipients have had their grants converted to unsubsidized Direct Loans because they did not satisfy requirements or requested their grant be converted, according to data from the TEACH grant servicer (see fig. 3). According to servicer data from August 1, 2013—when the servicer began tracking this information—through September 30, 2014, approximately 86 percent of the 12,648 recipients who had their grants converted to loans during this time frame were involuntary conversions— that is, initiated by the TEACH Grant servicer. Involuntary conversions occur because recipients do not meet grant requirements, such as not teaching in a qualified school or subject or not submitting required documentation verifying eligible employment. The remaining approximately 14 percent of these recipients voluntarily requested that their grants be converted to loans. According to servicing officials, recipients voluntarily requested conversions primarily because they no longer intended to teach or were not teaching in an eligible school or subject area. Despite the significant number of recipients who have not met TEACH Grant requirements, Education has not taken steps, such as conducting studies or evaluations, to determine why recipients are not successfully completing the program. Officials said they conduct analysis for internal purposes, but said the TEACH Grant program is too new and data is insufficient for an accurate and beneficial formal evaluation or study. However, Education could leverage existing data or collect new information to illuminate the challenges recipients face in meeting requirements. For example, the TEACH Grant servicer asks all recipients who voluntarily convert online to provide information on why they chose to do so (see fig. 4). Analyzing this type of information could better position Education to make informed decisions about how it might wish to address these challenges. The servicer also tracks some information on conversions that it initiates, such as when a recipient does not respond to certification requests or submits an incomplete certification form. However, the servicer does not collect information as to the underlying reasons why a recipient does not certify. Without collecting this type of information, the servicer and Education are limited in understanding challenges recipients face meeting grant requirements and taking steps, to the extent possible, to improve their experience. Further, Education has a stated goal to take a data- driven approach to better understand its customers and has made teacher recruitment and retention a priority. However, absent data to better understand these issues, it will be difficult for Education to improve student and teacher outcomes and it is likely missing opportunities to broaden the pool of highly qualified teachers available to teach in high- need subjects and high-need schools. The TEACH Grant and the loan forgiveness programs provide benefits to students, teachers, and colleges, according to college administrators in our focus groups, as well as students and other stakeholders we interviewed. These benefits include: Helping to recruit teachers, particularly to low-income schools. College administrators in half of our focus groups with colleges and universities that offer the TEACH Grant (two of four) said the program helps them recruit students. One focus group participant noted that the grant program may be particularly beneficial for education programs struggling to attract students. Additionally, college administrators in three of these four focus groups said the program helps recruit graduates to teach at low-income schools. Helping students pay for school. Officials from more than half of the higher education associations we interviewed (four of seven), and officials from both teacher unions said that the TEACH Grant program benefits students by helping them fund their education. The loan forgiveness programs can also reduce student loan debt because a portion of that debt is forgiven after a participant teaches for the specified number of years in an eligible school or subject area. This particularly resonated with financial aid officers in two of the four focus groups that included financial aid administrators. They said the Perkins cancellation program in particular is well designed because a portion of a borrower’s Perkins loan can be cancelled after only 1 year of service. While TEACH Grants help students pay for school, recipients may face challenges satisfying and understanding grant requirements after they graduate according to college administrators in our focus groups, as well as students and other stakeholders we interviewed, and our review of FSA Ombudsman complaint data. Finding and keeping an eligible teaching position. College administrators in three-fourths of our focus groups (six of eight) said school districts in their states had limited hiring, which could limit a graduate’s ability to find a qualifying teaching position. Officials from five of seven higher education associations, representatives from both teacher unions, and more than half of the students and recent graduates we interviewed (12 of 23) expressed related concerns about finding and keeping an eligible teaching position. For example, one student said that at times it can take a year or two to find a permanent position, and that the position may not be at a school that meets TEACH Grant eligibility requirements. Paperwork requirements. TEACH grant recipients may not understand annual employment certification requirements, according to college administrators in our focus groups and our review of FSA Ombudsman complaint data, putting them at risk of not meeting grant requirements. College administrators in half of our focus groups with schools that offer the TEACH Grant program (two of four) said that students may not fully comprehend the paperwork requirements associated with the grants after they graduate. Our review of customer complaint data from the FSA Ombudsman from October 2011 through March 2014 corroborated these concerns. Specifically, of the 212 requests for assistance the Ombudsman received on the TEACH Grant program, the majority of these—64 percent—cited problems submitting annual certification paperwork. The more common challenges cited included recipients not receiving reminders that their paperwork was due—as the servicer is required to send—or asserting that their grant was converted through no fault of their own. For instance, one participant told the Ombudsman that the servicer did not process the paperwork she submitted and her grant was converted to a loan. She said she was told she could file an appeal but that she should not bother because most appeals are rejected. Another participant told the Ombudsman she submitted the paperwork after her first year of teaching but forgot after her second year. As a result, despite working as a special education teacher in an eligible school, her grant was converted to a loan. As a result of challenges that TEACH grant recipients and schools can face in either meeting grant requirements or administering the program, some colleges and universities restrict participation or do not offer the program at all, according to college administrators in our focus groups, and other stakeholders we interviewed. As freshman and sophomores tend to change majors more frequently, college administrators in more than half of the focus groups that offer the TEACH Grant program (three of four) said that some colleges and universities restrict TEACH Grant participation to juniors, seniors, or graduate students. Other colleges and universities have opted not to offer the program at all due to concerns with their students’ ability to satisfy grant requirements. College administrators in all four of our focus groups with schools that do not offer the program said they did not do so because they feared it would put students at risk of accumulating additional debt if they were unable to satisfy grant requirements. For instance, one college administrator said that Education’s early estimates that 75 to 80 percent of TEACH grant recipients would not meet grant requirements influenced their decision not to offer the program.some schools considered this rate, among other things, when deciding not to participate. College administrators also noted concerns about the grant converting to a loan if a recipient changed to a non-eligible major or could not find an eligible job after graduation. Officials from a higher education association representing college financial aid offices said some schools chose not to offer the program because it was too burdensome for their schools to administer and expressed concern about the grants converting to loans if recipients do not fulfill requirements. College A senior Education official also confirmed that administrators from two of the four focus groups who offer TEACH grants also said the program can be burdensome to administer. For instance, one focus group participant said additional staff time is needed to provide TEACH Grant specific counseling to students. Another participant expressed concern about having to provide counseling to students each year they received a TEACH grant as opposed to other student aid programs that do not require students to receive counseling with each disbursement. Loan forgiveness program requirements can be confusing, making it difficult for teachers to participate in or maximize the benefits from the Perkins cancellation and Stafford forgiveness programs, according to stakeholders we interviewed. For example, officials from three institutions and two Perkins loan servicers said that graduates may not understand they have to submit an application annually to have 100 percent of their Perkins loans cancelled. Graduates may also not be aware that they will lose Perkins cancellation benefits if they choose to consolidate their Perkins loans with other federal student loans, according to participants in half of our focus groups with financial aid administrators (two of four). A Stafford loan servicer said that teachers may also be unaware that Stafford loans disbursed before 1998 are ineligible for forgiveness. This concern was corroborated by our review of FSA complaint data: it showed an estimated 21 percent of the 825 customer contacts received by the Ombudsman on teacher loan forgiveness programs from October 2011 through March 2014 involved a teacher’s not understanding that an outstanding balance on a Stafford loan or the age of a loan could affect his or her eligibility for the Stafford forgiveness program. Teachers may also not know how requirements for the Stafford loan forgiveness program affect other programs that may be available to them, such as Public Service Loan Forgiveness (PSLF), which may make it difficult to decide which is most beneficial to participate in. For example, due to federal law governing Stafford loan forgiveness, teachers cannot count the same teaching service towards both Stafford loan forgiveness and PSLF. Therefore, qualifying teachers must choose between a set amount of forgiveness through the Stafford program after 5 years, or wait and receive a potentially greater amount through PSLF after 10 years. Loan repayment options can also affect the total amount of loan forgiveness a teacher could receive across the programs. Stafford loan borrowers are automatically placed in a standard 10-year repayment plan, the same length of time needed to qualify for PSLF. Consequently, a teacher who pays off the loan through the Standard plan would have repaid the loan in full at the time he is eligible to receive PSLF. If, however, borrowers choose a repayment plan that spreads the payments over a longer period of time, they are likely to have a balance to be forgiven through PSLF at 10 years, provided the borrower qualifies for lower payments because of income and family size. Further, decisions about repayment plans and whether to participate in Stafford loan forgiveness or PSLF, which have to be made soon after graduation and throughout the life of the loan, can affect a teacher’s total payment toward the loan, according on our analysis of Education data.For example, a typical teacher at a low-income school with a master’s degree incurs about $41,500 in federal student loan debt, and would pay about $57,300 in total loan payments under the default standard 10-year repayment plan. However, the “typical teacher” in our example could pay more or less than $57,300 over the life of the loan depending on the repayment plan and loan forgiveness option he chooses. For example: A teacher who stays in the standard repayment plan and participates in Stafford teacher loan forgiveness, would pay about $50,500 but would have repaid the loan in full at the time he was eligible to receive PSLF. A teacher who chose the Income-Based Repayment (IBR) plan, which has a repayment period up to 25 years, and participates in the Stafford loan forgiveness program would pay about $56,400. A teacher who chose the IBR plan and forgoes Stafford teacher loan forgiveness after 5 years to instead take PSLF after 10 years of teaching, would pay about $48,500. As mentioned previously in this report, although information on the loan forgiveness programs is available on Education’s website and in resources provided to borrowers while in school, borrowers continue to struggle to understand how these programs work. Students and teachers, therefore, may not fully understand their options or the tradeoffs of participating in one program over another. The Consumer Financial Protection Bureau (CFPB) has also asserted that the complexity of financial aid program requirements may act as a deterrent for many To help teachers and borrowers weighing a career in public service.others navigate the various loan forgiveness options, CFPB created a toolkit to help public sector employers educate their staff about loan forgiveness programs including how the programs interact. Education monitors the servicer contracted to manage the TEACH Grant program including ensuring the servicer regularly communicates with recipients, tracking recipients’ progress towards satisfying their grant requirements, and ensuring the servicer correctly converts TEACH grants into loans if recipients do not satisfy them, according to Education officials.reports generated by the servicer and addresses recipient complaints including disputes regarding grant-to-loan conversion. Education monitors program participation through monthly From August 2013 through September 2014, Education and its contracted TEACH Grant servicer discovered 2,252 recipients had their grants erroneously converted to loans by the current and previous servicer due to servicer error. The erroneous conversions were identified when recipients contacted the current servicer or Education to dispute their conversion, or when, after investigating specific errors related to particular loans—some of which were discovered when recipients disputed their loan conversions—the loan servicer performed system queries to more systemically investigate these particular errors. The overall scope of the problem is unknown. Our analysis of servicer data shows that the majority of the erroneous conversions—56 percent— occurred because the servicer did not give the recipient 30 days from the final notification to certify that the recipient was teaching or intended to teach (see fig. 5). Rather, the servicer converted the grant to a loan within the 30-day window when a recipient is permitted to respond. According to the current servicer, recipients did not get the full 30 days to submit the paperwork because of time required to process and mail the letter. The servicer classified another 19 percent as erroneously converted because, according to the servicer, a recipient did not understand the terms of the grant and certification requirements—including paperwork needed to document teaching service—or the servicer provided inaccurate, unclear, confusing, or misleading explanations of program and certification requirements to the recipient. As a result of the errors discovered to date, Education officials said they plan to review accounts for all of the approximately 36,000 TEACH grant recipients who had grants converted to loans by the current and previous servicer since the program’s inception. Education officials said the review will include transferring the approximately 2,600 TEACH loans held by other servicers to the current servicer, but according to officials, discussions are ongoing and the agency has not yet established a time frame for when the transfer or overall review will be completed. Federal internal control standards call for establishing time frames as part of ongoing monitoring and such time frames would allow Education and others to assess Education’s progress towards reviewing all recipient accounts including those currently held by other servicers. The current servicer managing the program has also taken steps to address servicer error moving forward. Specifically, the servicer reported that it now conducts additional system checks and manually reviews all accounts flagged for conversion to determine whether the required certification forms were submitted by the recipient according to regulation. Over the course of 1 month (May 2014), for example, this manual review prevented erroneous conversions for 108 recipients. Education also expanded the servicer’s authority, so that it can reconvert loans to grants in some circumstances without elevating a dispute to Education. According to the servicer officials, this allowed the servicer to evaluate the cases more efficiently and contributed to the number of loans reviewed and reconverted to grants over the past year. According to Education officials, the high number of recent reconversions from loans back to grants is the result of addressing errors that occurred in the past under the previous servicer, and officials do not expect this volume to continue. In October 2012—about 4 years after the first TEACH grants were awarded and before the new servicer began managing the program— Education initiated quarterly monitoring of a sample of grants to ensure they had been appropriately converted to loans. Three of these four quarterly monitoring reports in fiscal year 2013 did not include observations of or issues with servicing error, although 541 of the 2,252 (24 percent) erroneous conversions discovered as of September 30, 2014 occurred with the previous servicer. Further, Education provided us with incomplete documentation regarding how it provided oversight prior to October 2012. For example, officials said they received quarterly briefings from the previous servicer on the status of the TEACH Grant portfolio, including information on grant and loan volume, but they could not locate documentation for fiscal years 2011 and 2012. While agencies have flexibility in how they oversee contractor performance, Office of Management and Budget best practices for contract administration cite the importance of documenting monitoring activities. Education officials said that in July 2013, they established a more formal and comprehensive monitoring process when the new servicer began managing the program, which includes monthly reports and weekly meetings. According to officials, Education became aware of the errors through this increased monitoring, including more detailed system checks independently initiated by the current servicer. Further, while the current servicer initiated the majority of the erroneous conversions, some of these errors may have occurred as a result of inaccurate data provided by the previous servicer. Officials with the current servicer stated that data received from the previous servicer contained a large number of anomalies that resulted in confusion and servicing issues when they began servicing the contract. As a result, the current servicer said it proposed a 60-day hold on grant-to-loan conversions (July 23, 2013, through October 10, 2013) to allow the servicer to work with the data and give recipients more time to submit required paperwork. To date, Education and the current servicer have identified, for each corrected erroneous conversion, whether the error was caused by a data issue or a communication issue with recipients. However, Education has not assessed why the data or communication issues occurred or how the errors went undetected by Education, nor has it outlined plans to do so. Therefore, we are unable to assess whether the changes Education made to its monitoring process are sufficient to address any systemic problems. Further, federal internal control standards highlight the need for ongoing monitoring and our body of work on performance measures and evaluations has shown that successful organizations conduct periodic or ad hoc program evaluations to examine how well a program is working. These types of evaluations allow agencies to more closely examine aspects of program operations, factors in the program environment that may impede or contribute to its success, and the extent to which the program is operating as intended. Absent a review examining the underlying cause of the erroneous conversions or plans to conduct such a review, Education cannot provide reasonable assurance it has taken steps to minimize the risk of erroneous conversions from occurring in the future. Education has increased its communication with TEACH grant recipients since the program transitioned to the new servicer in July 2013, including the servicer sending additional notifications to recipients and providing more assistance to schools, according to Education officials. Officials said that while they did not identify specific communication issues with the previous servicer, they used the transition to a new servicer as an opportunity to provide recipients with information on the program that is easier to understand. The current servicer also standardized the certification form and established a web portal where students can certify their intent to teach. Education has also established a process whereby TEACH grant recipients may appeal to the servicer or Education to dispute a conversion to a loan, and Education will reverse it if servicer error is found. However, we found that Education and the servicer provide incomplete and inconsistent information to recipients about the availability of and criteria for disputing a grant-to-loan conversion. The servicer’s website, for example, states that recipients can contact the servicer if they believe a grant was converted to a loan in error, but does not include the reasons such a conversion would be deemed erroneous, how the problem would be rectified, or the criteria considered in the adjudicating process. Further, all correspondence and policy documents provided to TEACH grant recipients—including the Agreement to Serve, exit counseling at graduation, and notification from the servicer that the grant has converted to a loan—state that once a TEACH grant is converted to a loan it cannot be reconverted to a grant, which is inconsistent with Education’s grant-to-loan conversion dispute process. This incomplete and unclear communication with TEACH grant recipients about the dispute process is inconsistent with federal internal control standards. These standards include establishing open and effective external communication. A senior Education official acknowledged information about the dispute process and what constitutes servicer error could be valuable to share with grant recipients to ensure they understand how the program works. Absent clear, consistent, and complete information about whether loan conversions can be disputed and what constitutes servicer error, recipients are unlikely to understand how to navigate the dispute process, criteria on which the dispute decisions are made, or whether a dispute process even exists. Education collects participation data on the loan forgiveness programs for teachers and the TEACH Grant program and periodically reviews students’ persistence in pursuing an education degree and teaching position, but the agency has not established performance measures for the TEACH Grant or the loan forgiveness programs nor used available data to systematically evaluate them. Education officials said the TEACH Grant program is too new to provide meaningful outcome data and specific metrics for the program are still being determined. Officials added that data on the loan forgiveness programs is too small for in-depth analysis. While no cohort of TEACH grant recipients has completed the 8- year service period, available data indicate that one-third of participants have not met grant requirements or requested their grants be converted. These programs support Education’s strategic goals of access to and persistence in higher education programs, and its goal to recruit and retain effective teachers. Without interim measures, such as evaluating participation against established benchmarks, it will be difficult for Education to assess whether the programs are meeting these goals or gauge performance on an ongoing basis. GAO-13-518. effective use of high-quality and timely data, including evaluations and performance measures, throughout the life cycle of policies and programs. Absent performance measures and efforts to assess progress towards them, it will be difficult for Education to gauge the programs’ success or use data to improve program administration and participant outcomes. The U.S. Department of Education plays a critical role in ensuring that children have access to high-quality teachers—particularly in low-income schools where teacher retention can be a challenge—by managing financial aid programs that help teachers pay for school. However, while the TEACH Grant and the loan forgiveness programs could be key drivers in Education’s efforts to increase teacher recruitment, a current lack of program awareness among teachers and students may undermine the agency’s teacher recruitment goals. Without exploring and implementing ways to increase awareness of these programs, Education is missing an opportunity to widen the pool of prospective educators and reduce levels of student debt that may drive teachers away from the profession. By accepting a TEACH grant award, students make a significant service commitment to teach in schools and subjects that have the greatest need for qualified and dedicated teachers. In recognition of this commitment, Education could take additional steps to ensure that the program is well- managed and give new teachers every opportunity to succeed. While over 112,000 students have committed to teaching in America’s neediest schools by accepting a TEACH grant award, a significant portion are not meeting the terms of the grant. Without a full understanding of why teachers are not able to meet TEACH Grant service requirements, Education is hindered from taking mitigating actions to reduce the number of grant-to-loan conversions and bolster program completion. The number of erroneously converted grants also signals a clear need for improved program management, particularly with regard to the grant-to- loan conversion dispute process. Examining the underlying cause of the erroneous conversions is an important step that will provide Education with additional assurance the actions it has taken to improve program management are sufficient. In addition, establishing time frames for transferring the approximately 2,600 TEACH loans held by other servicers to the current servicer, and reviewing the accounts for the approximately 36,000 recipients who had their grants converted to loans would allow Education and others to track the agency’s progress toward more effectively managing the program. Providing clear, consistent information about the option to dispute a grant-to-loan conversion, and the criteria for which a dispute would be considered, would also help teachers who may have felt deprived of due process better understand the process. A better understanding of these issues may also encourage colleges and universities to consider offering TEACH grants to their students. Finally, without performance measures it will be difficult for Education to assess whether the TEACH Grant and the loan forgiveness programs are helping to recruit and retain teachers to high-need schools and subject areas. To enhance participation in and strengthen management of federal student aid programs for teachers, we recommend the Secretary of Education direct Federal Student Aid’s Chief Operating Officer to take the following five actions: Explore and implement ways to raise awareness about the TEACH Grant and the loan forgiveness programs. Take steps to determine why participants are not able to meet TEACH Grant service requirements and examine ways to address those challenges. Review the underlying cause of the known erroneous conversions to ensure steps Education has taken are sufficient to address the problem, and establish time frames for transferring the approximately 2,600 loan conversions currently with other loan servicers. Review the TEACH grant-to-loan conversion dispute process and disseminate to appropriate audiences clear, consistent information on it, including that recipients have an option to dispute, how to initiate a dispute, and the specific criteria considered in the adjudicating process. Establish program performance measures for the TEACH Grant and the loan forgiveness programs to assess against established goals and to inform program administration. We provided a draft of the report to Education for review and comment. Education provided us with their written comments, which are reproduced in Appendix II. Education concurred with our five recommendations and noted steps it is taking or plans to take to implement them. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies to the Secretary of Education, relevant congressional committees, and other interested parties. In addition, the report is available at no charge on the GAO website at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (617) 788-0580 or nowickij@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in Appendix III. The objectives of our report were to examine: (1) how many students and teachers are potentially eligible for and participate in the three teacher aid programs, and for the Teacher Education Assistance for College and Higher Education (TEACH) Grant program, the extent to which recipients are satisfying grant requirements; (2) what selected schools, teachers, and students have identified as the benefits of and challenges with program participation; and (3) to what extent the U.S. Department of Education (Education) has taken steps to effectively manage and evaluate these programs. We conducted this performance audit from November 2013 to February 2015 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Federal Agencies, Laws, and Regulations To inform all of our objectives, we reviewed relevant federal laws, regulations and documents, and interviewed officials with Education’s Office of Postsecondary Education; Office of Planning, Evaluation, and Policy Development; Budget Service Office; and Federal Student Aid (FSA). In addition to interviews with agency officials, to assess the extent to which Education has taken steps to effectively manage and evaluate these programs, we interviewed officials from the loan servicing company contracted by Education to administer the TEACH Grant program, the two largest Federal Stafford loan servicers in terms of the number of borrowers, and two of the larger Federal Perkins loan servicers in terms of the number of colleges and universities they serve, according to a higher education association representing schools that offer Perkins loans. We also reviewed agency policy and planning documents, guidance provided to various offices within the agency and to the loan servicers, documentation of servicer monitoring activities, and information provided to prospective and current participants; we reviewed documentation from the servicers on program administration and management and correspondence and other documentation provided to program participants; and we interviewed officials from the Consumer Financial Protection Bureau to discuss their work on student debt and loan forgiveness programs available to teachers and other public sector employees. We compared this information to federal internal control standards, GAO’s prior work on performance measurement and evaluation leading practices, and Education’s departmental and FSA- specific strategic planning documents. To assess participation in the TEACH Grant, Stafford Teacher Loan Forgiveness, and Perkins Loan Teacher Cancellation programs, we reviewed data from Education’s National Student Loan Data System (NSLDS) from the last decade for the Stafford Forgiveness and Perkins Cancellation programs, and from school year 2008-2009 for the TEACH Grant—the first year Education awarded the grants. For the TEACH Grant program specifically, we determined the extent to which grant recipients were satisfying requirements by analyzing participation data from the contracted servicer who administers the program. These data included information on the status of TEACH grant recipients such as the number of recipients still in school and those who started the service period. For those in the service period, we reviewed the number of recipients who have satisfied the grant requirements, those who had their grant converted to a loan, and those who were teaching or had certified intent to teach. We compared this and other information about the TEACH Grant program to Education’s departmental and FSA-specific strategic planning documents, which outline goals and objectives both for federal financial aid programs and teacher recruitment and retention. To develop estimates for the population that is potentially eligible for these programs, we analyzed data from Education’s Integrated Postsecondary Education Data System (IPEDS) and Schools and Staffing Survey (SASS). For the TEACH Grant program, we analyzed IPEDS academic program completion data to obtain the number of students who graduated or completed a potentially eligible program in participating schools. Specifically, we matched the Office of Postsecondary Education Identification number of schools that offered TEACH Grants during the 2013 though 2014 academic year to IPEDS completion data for the prior year because this is the most recent IPEDS data available. We reviewed the Classification of Instructional Programs codes—which identifies instructional program specialties within educational institutions—in the IPEDS Completion Data file for academic year 2012-2013. We identified 96 potentially eligible program types. Schools have discretion to choose which programs are eligible to participate and the federal government, states, and local educational agencies have discretion to designate high- needs subjects; therefore we identified programs that could be included in teacher preparation programs such as general education and secondary education and teaching. We also included those with a concentration in math, science, special education, foreign language, bilingual education, and reading specialists because those programs are specifically listed in TEACH Grant statute as qualifying subjects. In developing our estimate, we included those who graduated with qualifying bachelor’s or master’s degrees and those who completed post-baccalaureate certificate programs since students graduating with these degree types are also eligible to receive TEACH Grants. Our analysis of the TEACH Grant eligible population is limited in a number of ways: (1) schools have discretion on which students and grade levels are eligible to participate in the program, therefore not all who graduated from eligible programs could necessarily apply; (2) students who graduated may also have not reached or maintained the 3.25 GPA often needed to participate. Further, we only counted those who graduated from eligible programs, not those who were enrolled in any given year. We focused on graduating seniors and graduate students because, according to our focus groups and interviews, some schools limit eligibility to juniors, seniors, and graduate students. To estimate the population of teachers potentially eligible for the Stafford Teacher Loan Forgiveness program, we reviewed program eligibility requirements and analyzed SASS data to identify the number of full-time teachers in public and private low-income schools during the 2011 through 2012 academic year; the most recent year of SASS data available. Because this is a probability sample, estimates based on this survey are subject to sampling error. Unless otherwise noted, percentage estimates based on this survey have 95 percent confidence intervals within +/- 5 percentage points of the estimate itself. We determined that NSLDS, SASS, IPEDS, and TEACH Grant servicer data are sufficiently reliable for the purposes of this report by testing it for accuracy and completeness, reviewing documentation about systems used to produce the data, and interviewing agency officials. To illustrate a typical teacher’s loan repayment and forgiveness options and the choices teachers must make to take advantage of the forgiveness programs, we used data from Education’s Baccalaureate and Beyond Longitudinal Study (B&B) from the 2008 cohort to create a typical teacher in terms of debt and salary with a master’s degree teaching at a low- income school. The study included approximately 19,000 sample members interviewed in 2009 and 2012. For the purposes of our analysis, we focused on responses from the 2012 follow-up for those teachers teaching at Title I schools who earned their master’s degree because this survey year includes the most recent debt and salary information available. Because this is a probability sample, estimates based on this survey are subject to sampling error. Unless otherwise noted, percentage estimates based on this survey have 95 percent confidence intervals within +/- 5 percentage points of the estimate itself. To determine the debt amount for this typical teacher, we analyzed the cumulative total amount borrowed in federal loans through 2012 assuming the teacher would consolidate the federal loans they incurred. We assumed half of their debt was incurred as an undergraduate and half pursing their master’s degree. Therefore, when we applied the general Stafford subsidized loan limits to the overall debt amount ($23,000 for undergraduate and $65,500 for graduate) the entire debt amount for the typical teacher in our scenario was subsidized. Because the typical teacher in our example works at a low-income school, we assumed the teacher would receive $5,000 forgiveness under the Stafford Loan Forgiveness Program. We determined the salary amount for those teachers by examining the average initial salary in 2012 and assumed a 5 percent salary growth rate and a 3.3 percent growth of the poverty threshold to coincide with assumptions made by Education as part of its repayment calculator. We also used the Income-Based Repayment (IBR) plan thresholds for loans made before July 1, 2014; for new loans made on or after July 1, 2014 new IBR thresholds apply. Under these new IBR thresholds, a teacher with similar income and debt amounts could see larger benefits. We also assumed a 6.8 percent interest rate because teachers responding to the B&B survey would have undergraduate and graduate Direct Subsidized Loans disbursed between July 1, 2006, and June 30, 2012, when that interest rate applied. Teachers could have had additional loans disbursed before July 1, 2006, but for ease of reporting, we assumed a consistent interest rate. We also assumed the teacher would not receive loan forbearance. We assumed that the teacher was single and without children (which affects the poverty threshold and the income based amount due). After discussion with loan servicers, we assumed that even after the teacher loan forgiveness was applied, the monthly amount due did not change. This had the effect of shortening the term of the loan. Finally, we applied the observed loan level and income data to the various loan forgiveness programs to simulate what the total payments would be under the different programs. To the extent possible, we validated our results against the online repayment estimator provided on Education’s website. These simulations are intended for illustrative purposes only, as they do not incorporate the experiences of individual borrowers. For example, individual borrowers may experience periods of unemployment (reducing income temporarily), or marry and form families (changing the appropriate poverty threshold). We determined that B&B data are sufficiently reliable for the purposes of this report by testing it for accuracy and completeness and reviewing documentation about systems used to produce the data. To learn more about the challenges participants face, and the steps Education has taken to resolve them, we analyzed customer complaint data from Education’s FSA Ombudsman. We reviewed all 212 customer requests the Ombudsman received for assistance related to the TEACH Grant program from October 2011 through March 2014. For the loan forgiveness programs, we reviewed a simple random sample of 100 customer requests out of a population of 825 such requests the Ombudsman received from October 2011 through March 2014. The loan forgiveness sample is generalizable to the population of requests. We requested data covering this time period because it spanned both Because we servicers that managed the TEACH Grant program.followed a probability procedure based on random selections, our sample is only one of a large number of samples that we might have drawn. Since each sample could have provided different estimates, we express our confidence in the precision of our particular sample’s results as a 95 percent confidence interval (e.g., plus or minus 10 percentage points). This is the interval that would contain the actual population value for 95 percent of the samples we could have drawn. In this report we present confidence intervals for estimates based on this sample along with the estimate itself. Using content analysis, we developed categories to characterize the types of assistance TEACH grant recipients and borrowers requested or complaints they expressed about the programs. An analyst then reviewed the requests and recorded each request by category. These categorizations were verified by a second analyst and any discrepancies were resolved by both analysts. To describe program benefits and challenges as identified by stakeholders, we conducted eight focus groups with administrators from schools of education and financial aid offices from 58 colleges across the public, private, and for-profit sectors. The Stafford Teacher Loan Forgiveness program is administered by servicers contracted by Education; therefore we did not ask school officials about challenges administering the program. Four focus groups were with college administrators whose institutions offered the TEACH Grant program to their students while the remaining four were with college administrators whose institutions did not offer the program. These sessions involved structured small-group discussions designed to gain more in-depth information about (1) the benefits and challenges associated with the TEACH Grant and Perkins Loan Teacher Cancellation programs, (2) outreach and guidance the institutions provided to students or others about these programs, and (3) any guidance Education provided to these administrators about the programs. We also asked about program improvements and to what extent these programs succeed in recruiting and retaining teachers in high-need subjects and low-income schools. Consistent with typical focus group methodologies, our design included multiple groups with varying characteristics but shared one or two homogenous characteristics. As the primary homogenous characteristic, each focus group contained either representatives of colleges of education or financial aid administrators. All but one of the focus groups involved 6 to 10 participants while the remaining group had 5 participants. By including college of education representatives and financial aid administrators from institutions that do and do not offer TEACH grants, we intended to gather a range of perspectives about how well these programs are working or why institutions chose not to participate. To select focus group participants, we obtained membership lists from professional associations of colleges of education and financial aid administrators and invited representatives from 235 institutions of higher education to participate. The primary selection criteria included whether the institution offered TEACH grants, and, for those that did, the number of students that received them. To obtain input from a wide variety of institutions, we also considered the size of the school in terms of total enrollment and whether the institution was public, private, or for-profit. As all focus groups were conducted via teleconference, we were able to include participants from across the United States. Focus group discussions were structured and guided by a moderator who used a script with a standardized list of questions to encourage participants to share their thoughts and experiences. During the sessions, we informed participants that their names would not be used in the published report. Each of the eight focus groups was recorded and transcriptions were created, which served as the record for each group. Using content analysis, we then evaluated these transcripts to develop our findings. Our analysis was conducted in three steps. In the first step, two analysts developed a set of themes and categories to track the incidence of various responses during focus group sessions. In the second step, each transcript was tagged by an analyst and recorded by theme and category. In the third step, these tags were verified by a second analyst and any discrepancies were resolved by both analysts agreeing to the tagged theme and category. Methodologically, focus groups are intended to generate in-depth information about the reasons for the focus group participants’ attitudes on specific topics and to offer insights into their concerns about and support for an issue. Focus groups are not designed to (1) demonstrate the extent of a problem or to generalize results to a larger population, (2) develop a consensus to arrive at an agreed-upon plan or make decisions about what actions to take, or (3) provide statistically representative samples or reliable quantitative estimates. The projectability of the information produced by our focus groups is limited for several reasons. First, the information includes only the responses from college of education and financial aid administrators from among the eight selected groups. Second, while the composition of the groups was designed to ensure a range of stakeholders from institutions of varying size and geographic diversity, the groups were not randomly sampled. Third, participants were asked questions about their experiences with the programs, and other college of education representatives and financial aid administrators not in the focus groups may have had other experiences. Because of these limitations, we did not rely entirely on focus groups, but rather used several different methods to corroborate and support our findings and conclusions. In-depth interviews with individual schools. We supplemented information obtained through focus groups with non-generalizable in- depth interviews with selected representatives of four institutions of higher education that offered the TEACH Grant program. Specifically, we spoke with representatives from each of the institutions’ (1) college of education, (2) financial aid office, and (3) bursar’s office to obtain their perspectives, particularly regarding the benefits and challenges of administering the TEACH Grant program. Institutions were primarily chosen based on whether they offered TEACH grants and, if so, the number of students who received the grant and whether the school participated in the Perkins loan program. Secondary selection criteria included whether the institution was public, private, or for-profit. In advance of each interview, we reviewed materials each selected institution made publicly available via their websites including information on the TEACH Grant program. During interviews, we discussed how these institutions administer and conduct outreach on the TEACH Grant and Perkins cancellation programs and the benefits and challenges associated with the programs. Students and recent graduates. To obtain the students’ perspective particularly around awareness of the TEACH Grant and the loan forgiveness programs and their benefits and challenges participating in them, we spoke with 23 students enrolled in or recently graduated from teacher training programs at 21 institutions, 14 of whom had attended or had graduated from institutions that offer TEACH grants. To select participants, the National Education Association contacted students or recent graduates on a list of state and national student leaders they shared with us and requested that each participate in one of two conference calls. Evidence gained through these group interviews is not generalizable to all students. Higher education associations and teacher unions. To gain further insights on teacher financial aid programs, we interviewed officials from seven higher education associations and organizations, and the two largest teacher unions on the benefits and challenges of these programs. We selected higher education associations and organizations representing a range of public and private degree granting institutions and organizations that represent colleges of education and financial aid administrators including: the American Association of Colleges for Teacher Education; National Association of Student Financial Aid Administrators; National Direct Student Loan Coalition; Student Aid Alliance; American Association of State Colleges and Universities; Coalition of State University Aid Administrators; and Coalition of Higher Education Assistance Organizations. We also interviewed representatives from the two largest national teacher unions: the National Education Association and the American Federation of Teachers. While information obtained from these interviews are not generalizable, they provide valuable insights from college administrators, teachers, and other stakeholders about these programs. In addition to the contact named above, Janet Mascia, Assistant Director; Rachel Beers, Analyst-in-Charge; Justin Dunleavy, and Robin Marion made significant contributions to this report. Also contributing to this report were Susan Aschoff, Deborah K. Bland, Ben Bolitzer, Alex Galuten, Stuart Kaufman, John Mingus, Mimi Nguyen, Mark F. Ramage, Kathleen van Gelder, Walter Vance, and Amber Yancey-Carroll. | Education estimates 430,000 new teachers will be needed by 2020. It administers three programs that may help attract and retain qualified teachers by helping them finance their education. However, little is known about the efficacy of these programs. GAO was asked to examine the TEACH Grant and two loan forgiveness programs. This report examines (1) the number of current and potential participants in the three teacher aid programs and the extent to which TEACH Grant recipients satisfy grant requirements; (2) what selected schools, teachers, and students identified as benefits and challenges of program participation; and (3) the extent to which Education has taken steps to effectively manage and evaluate these programs. GAO reviewed applicable federal laws, regulations, and documents; analyzed participation data for the past decade; and interviewed stakeholders including agency officials, loan servicers, and students. GAO also held eight non-generalizable focus groups with officials from 58 colleges representing a range of sizes. GAO also reviewed Ombudsman data covering the former and current TEACH Grant servicers from October 2011 to March 2014. More than 410,000 students and teachers have participated in financial aid programs for teachers over the past decade, though GAO estimates 0.8 and 19 percent of the potentially eligible population participates in the Stafford Teacher Loan Forgiveness and Teacher Education Assistance for College and Higher Education (TEACH) Grant programs, respectively. GAO did not develop an estimate for Perkins Loan Teacher Cancellation because U.S. Department of Education (Education) budget documents indicate that federal funds for cancellations were last appropriated in fiscal year 2009. About 36,000 of the TEACH Grant's more than 112,000 recipients have not fulfilled grant requirements, according to GAO's analysis of servicer data, and have had their grants converted to loans, known as grant-to-loan conversions, as required by regulation. Education has a stated goal to take a data-driven approach to better understand its customers, but does not collect information on why recipients do not meet requirements. Absent this data, Education is hindered in taking steps to reduce grant-to-loan conversions and improve participant outcomes. Key benefits of the TEACH Grant and the two loan forgiveness programs are helping to recruit needed teachers and helping teachers pay for their education, while key challenges include participants' lack of knowledge about the programs' requirements, according to GAO's focus groups with college officials and interviews with other stakeholders. Regarding challenges, college officials said TEACH recipients may have difficulty finding and keeping an eligible teaching position and that annual certification requirements are confusing. GAO's review of data from Education's Federal Student Aid Ombudsman corroborates these challenges: 64 percent of the 212 requests for TEACH assistance from October 2011 through March 2014 cited problems submitting certification paperwork. Further, some college administrators said a key reason their schools do not participate in the program is the grant-to-loan conversion issue. Education tracks participation in all three programs, but lacks clear, consistent guidance to help recipients understand the TEACH grant-to-loan conversion dispute process. As of September 2014, GAO's analysis of TEACH servicer data shows that 2,252 grants were erroneously converted to loans. Education officials said they now monitor the servicer more closely and plan to review all of the nearly 36,000 of the program's grant-to-loan conversions, but the agency has not systemically reviewed the cause of the errors. Federal internal control standards emphasize ongoing monitoring and absent a review, Education lacks reasonable assurance that it has taken steps to minimize future erroneous conversions. Education established a dispute process to address concerns about TEACH grants converted to loans in error; however, GAO found that Education and the servicer provide incomplete and inconsistent information to recipients about the availability of and criteria for disputing conversions. This is inconsistent with federal internal control standards that highlight effective external communication. Absent clear and complete information, recipients are unlikely to understand the dispute process. Education also has not established performance measures for the three programs nor used available data to systematically evaluate them. Managing for results includes setting meaningful performance goals and measuring progress toward them. Absent those, Education is unlikely to be able to use data to improve program administration and participant outcomes. GAO recommends, among other things, that Education assess TEACH Grant participants' failure to meet grant requirements, examine why erroneous TEACH grant-to-loan conversions occurred, disseminate information on the TEACH grant-to-loan dispute process, and establish program performance measures. Education agreed with GAO's recommendations. |
Total Medicare spending for physician services depends on actual payment rates, the volume of services provided, and the mix of those services. Medicare spending goes up when the price paid to physicians for each service increases, when the number of services provided rises, or when more intensive, and therefore more expensive, services replace less intensive ones. Since 1992, Medicare has paid for physician services using a fee schedule. The fee for each service is a dollar conversion factor, adjusted to reflect the resources required for that service relative to the resources required to provide all other physician services, and the differences in the costs of providing services across geographic areas. Along with the fee schedule, the Congress enacted a system of spending targets designed to control growth in total spending for physicians’ services. The Sustainable Growth Rate (SGR) system was created in the Balanced Budget Act of 1997 and revised in the Medicare, Medicaid, and SCHIP Balanced Budget Refinement Act of 1999 (BBRA). It replaced the first system of spending targets, implemented in 1992, known as the Volume Performance Standard. The SGR system sets spending targets for physician services and adjusts payment rates to bring spending in line with those targets. The SGR target for total spending is based on spending in an initial, or base, year and the estimated growth in real per capita GDP each year and three other factors that affect overall spending on physician services—the changes in the cost of inputs used to produce physicians’ services (as measured by the Medicare Economic Index (MEI)), the number of Medicare beneficiaries in the traditional fee-for-service program, and expenditures that result from changes in laws or regulations. The spending target for physician payments is applied by incorporating it into the adjustment to the conversion factor that determines the payment amount per service. The conversion factor is determined annually by adjusting the previous year’s conversion factor by the change in the MEI, to account for the cost of inputs for physician services, and adjusting this product on the basis of the relationship between the cumulative SGR target and Medicare physician spending. The conversion factor update is greater than the MEI when physician spending has been below the targets and is less than the MEI when physician spending has been higher. In response to escalating Medicare expenditures, the Congress made major changes in Medicare payment policies, beginning first by enacting the hospital inpatient prospective payment system, which was implemented in 1983, and then the Medicare physician fee schedule, implemented in 1992. When enacting the fee schedule, the Congress recognized that setting fees alone would not sufficiently restrain physician spending growth. Despite some constraints on physician fees since the 1970s, spending on physician services had grown dramatically in the 1980s as a result of increases in the volume and intensity of services provided. The Congress, therefore, provided that annual physician fee increases would depend upon whether total Medicare physician spending exceeded or fell short of cumulative spending targets. Since the implementation of the fee schedule and spending targets, the rise in Medicare spending for physician services has slowed significantly, reflecting lower growth in the volume and intensity of these services. Before the physician fee schedule was implemented, Medicare payments for physicians’ services were largely based on historical charges. Although during the 1970s the Congress introduced some controls on annual payment rate increases, Medicare spending for physician services continued to rise. This was also true in the 1980s—between 1980 and 1990, for example, Medicare spending per beneficiary for physician services grew at an average annual rate of more than 12 percent, tripling from $278 to $890 (see fig. 1). Much of the spending growth resulted from increases in the volume of services provided to each beneficiary and the substitution of more intensive and expensive services for less intensive and expensive ones. The Physician Payment Review Commission, which was charged with advising the Congress on Medicare physician payment issues, observed, “y the late 1980s. . . volume and intensity growth had become the primary cause of higher program spending. In fact, from 1986 until 1992, while physician payment rates grew by less than 2 percent annually, the volume and intensity of services rose by almost 8 percent per year.” The Congressional Budget Office in 1986 stated that “oth the price and the volume of services must be controlled to constrain costs . . ..”Spending targets were needed to limit growth in volume and intensity of physician services. In 1989 testimony, Health and Human Services Secretary Louis W. Sullivan said “Medicare physician spending has increased at compound annual rates of 16 percent over the past 10 years. And in spite of our best efforts to control volume and reign in expenditures, Medicare physician spending is currently out of control. . . An expenditure target. . . sets an acceptable level of growth in the volume and intensity of physician services.” The Congress introduced spending targets for physician services in conjunction with the physician fee schedule in 1992 to help constrain the rise in Medicare spending for physician services. The targets incorporated limited growth in the volume and intensity of services and were revised each year based on estimates of changes in the number of Medicare beneficiaries and physician input prices. If actual spending exceeded the targeted amounts, future payment rates would be reduced, relative to what they would have been if actual spending had equaled the targets, to offset the excess spending. If actual spending fell short of the targets, future payment rates would be increased. Since 1992, the growth in the volume and intensity of physicians’ services per Medicare beneficiary has moderated (see fig. 2). Between 1992 and 2000, the average annual increase in Medicare spending due to changes in volume and intensity of services per beneficiary was about 2 percent. In contrast, between 1985 and 1991, immediately before the introduction of spending targets, volume and intensity of services per beneficiary increased at an average annual rate of about 8 percent. The application of the SGR system in 2002 resulted in a 5.4 percent reduction in physician payment rates, despite an estimated 2.6 percent increase in the costs of inputs used to provide physician services. The reduction occurred because estimated cumulative physician services spending since 1996 exceeded the target for cumulative spending by approximately $8.9 billion, or 13 percent of projected 2002 spending. In part, the payment update reflects adjustments made to the spending targets for previous years for revisions in GDP estimates and for more accurate actual spending statistics. Correcting these errors in previous years’ targets and spending totals to reflect more recent data resulted in larger physician payment increases in those years than if accurate data had been used, and they contributed to the size of the reduction in payments in 2002. The SGR system sets spending targets for physician services and adjusts payment rates to bring spending in line with those targets. Conceptually, if spending equals the targeted amount, physician payment rates are updated to keep pace with the percentage change in input prices as measured by the MEI. If spending exceeds the target, the change in payment rates is smaller than the change in input prices. If spending falls short of the target, payment rates are allowed to grow faster than the rise in input prices. By adjusting payment rates when prior-year spending has been too high, the SGR system moderates the growth in Medicare outlays for physician services. The SGR adjustments to the input price update are determined by how much the cumulative physician spending since 1996 differs from the cumulative spending target since then. Spending and targets must both be estimated from information available each November when payment rates are set for the following year. Previously, those estimates were then used in subsequent years. Based on requirements in the BBRA, however, HCFA implemented a process for revising the most recent two years of spending and target estimates. Because the annual targets are determined by changes in four factors—the number of fee-for-service beneficiaries, real per capita GDP, input costs, and the effect of changes in laws or regulations—a revision to any of those factors, or to estimates of prior spending, can change the spending estimate. The SGR adjustments to the input price update can then take effect because of growth in the volume or intensity of services delivered, resulting in spending deviating from targets, or because of revised estimates for prior years’ targets and spending. In setting payment rates for 2002, CMS updated its estimates of past spending and recalculated past targets. The net effect of both revisions indicated that Medicare outlays were an estimated $8.9 billion too high. The SGR is designed to recover this excess amount by lowering payment rates in 2002 and future years. CMS’ original estimates of spending since 1998 were too low, in part because the agency had not included all appropriate claims in making these estimates. The original spending targets for 2000 and 2001 were too high, largely because the Bureau of Economic Analysis in the Department of Commerce revised its GDP and GDP growth estimates for those years. To some extent, the reduction in payment rates this year corrects for inaccuracies in previous estimates that produced physician fees that were too high in 2000 and 2001. In both years, payment rates increased by more than the change in input prices because the information available at the time those rates were established suggested that physician spending had been held below the targets. In 2000, payment rates increased 5.4 percent, while input costs increased 2.4 percent; in 2001, payment rates increased 4.5 percent, while input costs increased by 2.1 percent. The reason that 2002 payment rates fall 5.4 percent while input prices increase by 2.6 percent is that the revised estimates revealed that spending exceeded the targets in previous years. The reduction would have been almost 4 percentage points greater, but the SGR system limits how much fees can be adjusted for the differences between actual and target spending. Several measures could moderate the fluctuations in physician payment rates. Although these modifications to the SGR could mitigate the possible impact of rate instability or reductions in beneficiary access to needed services, doing so could also lessen the ability of spending targets to encourage fiscal discipline. Available data indicate that access is adequate, but more timely and detailed information is critical to promptly recognize potential deteriorations in access. The SGR system is designed to limit the fluctuations in payment rates, but its design could be modified to achieve greater rate stability. The BBRA specified that adjustment to realign spending with the targets cannot cause payments to fall by more than 7 percentage points below, or increase by more than 3 percentage points above, the percentage change in input prices. In addition, spending deviations from past targets are not corrected in a single year but are spread over several years. Greater rate stability could be achieved by narrowing the range over which rates could change from one year to the next. Similarly, the corrections for spending deviations could be spread over longer periods of time. Modifying how spending targets are set could also reduce year-to-year fluctuations in rates. Currently, the changes in GDP for a single year are used to establish the spending target. The difficulty in accurately estimating GDP has contributed to the problem of fluctuations in the target. In addition, linking annual changes in the targets to annual changes in GDP ties the target to the business cycle. GDP growth rates are higher during periods of prosperity and lower during downturns—a commonly used definition of a recession is a decline in real GDP for two successive quarters. But health care needs of Medicare beneficiaries are not cyclical. Neither significantly lowering spending targets during a downturn nor unduly increasing them in a period of prosperity is appropriate. Linking the determination of spending targets to average levels of GDP over several years would help eliminate much of the cyclical variation. Any changes to the SGR must balance the desire for greater rate stability with the need for fiscal discipline. Spending targets create a feedback mechanism between physicians’ behavior and payment rate increases. However, spending targets do not create direct incentives for any individual physician, since it is the collective behavior of all physicians that determines the payment adjustments that result from missing the spending targets. The primary value of spending targets in instilling fiscal discipline is the signal they send that affordability of the program is an important concern in establishing Medicare policies. Limiting the size of the payment adjustment for missed spending targets or to corrections in prior years’ targets, and lengthening the time over which those adjustments are incorporated, could partially mute the signal targets send and erode some of the fiscal discipline they encourage. On the other hand, excessive rate fluctuations can be difficult for providers and may ultimately affect beneficiaries’ access to physician services if rate fluctuations cause too many providers to decline to participate in the program. Ensuring that the use of spending targets does not compromise appropriate access to services is a key concern. Spending targets that are updated principally by the growth in GDP and other factors may not reflect fully changes in medical care and the markets for these services. It is therefore important to monitor service use to assess whether appropriate access for beneficiaries is secured, especially if fees are reduced. Such monitoring needs to involve recent experience so that if spending targets need adjustments, those adjustments are done promptly to ameliorate any problems. Information on physicians’ willingness to see Medicare patients is dated but overall does not indicate access problems. Data from the 1990s show that virtually all physicians treated Medicare beneficiaries or if accepting new patients, accepted those covered by Medicare. According to data from the American Medical Association (AMA), 96.2 percent of all nonfederal physicians (excluding residents and pediatricians who do not normally serve Medicare patients) treated Medicare beneficiaries in 1996, an increase from the 94.2 percent AMA reported in 1994. A 1999 survey sponsored by MedPAC found that 93 percent of physicians who were accepting any new patients were accepting new patients covered by Medicare. Payment rate decisions should not be made in a data vacuum. As health needs change, technology improves, or health care markets evolve, spending targets and resulting payment rates may need to be adjusted periodically, not by a formula designed for annual updates, but by specifying a new base year target calibrated to ensure appropriate access. Payment rates that are too low can impair beneficiary access to physician services, while payment rates that are too high add unnecessary financial burdens to Medicare. Informed decisions about appropriate payment rates and rate changes cannot be made unless policymakers have detailed and recent data on beneficiaries’ access to needed services. The SGR mechanism uses information about physician spending in relation to cost increases, changes in the number of beneficiaries, and growth in the overall economy to impose fiscal discipline on Medicare outlays for these services. This mechanism provides a signal when spending threatens to grow out of control and in that sense is analogous to what the Comptroller General has called for in testimony on several occasions with regard to the entire Medicare program. The demographic changes facing the nation require policymakers to look ahead and assess both current and future Medicare spending in relation to the entire federal budget and the economy—first to understand the urgent need for fiscal discipline, then to make choices to ensure the sustainability and affordability of the program. A mechanism like the SGR provides a benchmark for assessing the trend in physician spending and can prompt actions to bring that spending in line with overall program goals. In assessing the options for updating physician payments, the program’s prospects for long-term sustainability should be paramount. Meeting that challenge will involve difficult decisions that will likely affect beneficiaries, providers, and taxpayers. This concludes my prepared statement. I would be happy to answer any questions that you or Members of the Subcommittee may have. For more information regarding this testimony, please contact me at (202) 512-7114 or Laura A. Dummit at (202) 512-7119. James Cosgrove, Kathryn Linehan, Lynn Nonnemaker, and Hannah Fein also made key contributions to this statement. | Congress implemented a physician fee schedule and a fee update formula to moderate spending growth relative to specified Medicare spending targets. These spending targets increase annually to reflect higher costs for physician services, the growth in the overall economy, and changes in the number of Medicare beneficiaries. Physician fees are adjusted for changes in the costs of providing services and on actual cumulative spending compared to the cumulative targets. Physician fees are updated to reflect higher costs to provide services. These updates are adjusted up or down on actual spending either falling below or exceeding the targets. In November 2001, the Centers for Medicare and Medicaid announced that updating Medicare's fees will decline 5.4 percent from what was paid in 2001, despite an estimated 2.6 percent increase in the cost of physician inputs. This reduction accounts for historical cumulative spending that exceeded the target by $8.9 billion, or 13 percent of estimated 2002 spending. Several factors contributed to the disparity between actual and targeted spending, including the correction of substantial errors in past spending estimates and the revision of targets for prior years. The current update mechanism could be modified to moderate fluctuations in physician fees and to ensure adequate payments, while retaining the fiscal discipline created by having a spending target. Such modifications would need to balance concerns about preserving fiscal discipline on physician spending with the need to maintain adequate payment rates to ensure that beneficiaries have access to physician services. Because the paramount consideration in setting payment rates is ensuring appropriate beneficiary access to services, timely and detailed data on Medicare beneficiary service use are essential to achieving this balance. |
Labor required states to implement major provisions of WIA by July 1, 2000, although some states began implementing provisions of WIA as early as July 1999. WIA replaced the Job Training Partnership Act (JTPA) program and requires that many federal programs provide employment and training services through one-stop centers. Services funded under WIA represent a marked change from those provided under the previous program, allowing for a greater array of services to the general public. WIA is designed to provide for greater accountability than under previous law: it established new performance measures and a requirement to use Unemployment Insurance (UI) wage data to track and report on outcomes. Program services provided under WIA represent a marked change from those provided under JTPA. When WIA was enacted in 1998, it replaced the JTPA programs for economically disadvantaged adults and youth and for dislocated workers with three new programs—Adult, Dislocated Worker, and Youth—that provide a broad range of services to the general public, no longer using income to determine eligibility for all program services. The WIA adult and dislocated worker programs no longer focus exclusively on training, but provide for three tiers, or levels, of service: core, intensive, and training. Core services include basic services such as help with job searches and providing labor market information. These activities may either be self-service or require some staff assistance. Intensive services include such activities as comprehensive assessment of jobseekers’ skill levels and service needs and case management—activities that typically require greater staff involvement. Training services include such activities as occupational skills development or on-the-job training. Labor’s guidance specifies that monitoring and tracking for the adult and dislocated worker programs should begin when jobseekers receive core services that require significant staff assistance. Jobseekers who receive core services that are self-service or informational in nature are not counted in the performance measures. In addition to those services provided by the three WIA funded programs, WIA also requires that states and local areas use the one-stop center system to provide services for many other employment and training programs. Seventeen categories of programs funded through four federal agencies are now required to provide services through the one-stop center under WIA. Table 1 shows the programs that WIA requires to provide services through the one-stop centers (also known as mandatory programs) and the federal agencies that administer these programs. WIA is designed to provide for greater accountability than its predecessor program by establishing new performance measures, a new requirement to use UI wage data to track and report on outcomes, and a requirement for Labor to conduct at least one multi-site control group evaluation. According to Labor, performance data collected from the states in support of the measures are intended to be comparable across states in order to maintain objectivity in determining incentives and sanctions. The performance measures also provide information to support Labor’s performance goals under the Government Performance and Results Act (GPRA), the budget formulation process using the Office of Management and Budget’s (OMB) Program Assessment Rating Tool (PART), and for program evaluation required under WIA. In contrast to JTPA, under which data on outcomes were obtained through follow-ups with job seekers, WIA requires states to use UI wage records to track employment-related outcomes. Each state maintains UI wage records to support the process of providing unemployment compensation to unemployed workers. The records are compiled from data submitted to the state each quarter by employers and primarily include information on the total amount of income earned during that quarter by each of their employees. Although UI wage records contain basic wage information for about 94 percent of workers, certain employment categories are excluded, such as self-employed persons, independent contractors, federal employees, and military personnel. According to Labor’s guidance, if a program participant does not appear in the UI wage records, states may then use supplemental data sources, such as follow-up with participants and employers, or other administrative databases, such as U.S. Office of Personnel Management or U.S. Department of Defense records, to track most of the employment-related measures. However, only UI wage records may be used to calculate the earnings change and earnings replacement performance measures. (See table 2 for a complete list of WIA performance measures.) Unlike JTPA, which established expected performance goals using a computer model that took into account varying economic and demographic factors, WIA requires states to negotiate with Labor to establish expected performance levels for each measure. States, in turn, must negotiate performance levels with each local area. The law requires that these negotiations take into account differences in economic conditions, participant characteristics, and services provided. To derive equitable performance levels, Labor and the states use historical data to develop their estimates of expected performance levels. These estimates provide the basis for negotiations. WIA holds states accountable for achieving their performance levels by tying those levels to financial sanctions and incentive funding. States that meet their performance levels under WIA are eligible to receive incentive grants that generally range from $750,000 to $3 million. Nineteen states were eligible to apply for incentive grants in program year 2003. States that do not meet at least 80 percent of their WIA performance levels are subject to sanctions. If a state fails to meet its performance levels for 1 year, Labor provides technical assistance, if requested. If a state fails to meets its performance levels for 2 consecutive years, it may be subject to a 5 percent reduction in its annual WIA formula grant. No states received financial sanctions in program year 2003. Labor determines incentive grants or sanctions based on the performance data submitted by states each October in their annual reports. States also submit quarterly performance reports, which are due 45 days after the end of each quarter. In addition to the performance reports, states submit updates for their Workforce Investment Act Standardized Record Data (WIASRD) in mid-October. WIASRD is a national database of individual records containing characteristics, activities, and outcome information for all enrolled participants who receive services or benefits under WIA. All three submissions primarily represent participants who have exited the WIA programs within the previous program year. The process of collecting and reporting WIA data involves all three levels of government. Participant data are typically collected by frontline staff in local areas and entered into a state or local IT system. In some states, local area staff may enter data directly into a statewide IT system; in other states, local areas may use their own individualized IT system to enter data, from which staff can extract and compile the necessary information for state submission. After the state receives data from local areas, this information is compiled and formatted for various submissions to Labor, including the state’s WIASRD file, quarterly report, and annual report. During the data compilation process, state agencies administering WIA typically match participant records to their state’s UI wage record system to obtain wage records and employment status. In addition, states may use the Wage Record Interchange System (WRIS) to match participant records to other state’s UI wage records or use other databases such as that of the U.S. Office of Personnel Management to fill gaps in the UI wage records. States may also link participant records to partner programs’ IT systems to track activities across programs or to determine outcomes such as attaining high school diplomas, degrees, and certificates. For the quarterly and annual report, states use software to calculate their performance measures. States generate the required WIA performance reports and electronically submit them to Labor’s regional offices using the Enterprise Business Support System (see fig. 1). Internal controls comprise the plans, methods, and procedures an organization uses to meet its missions, goals, and objectives. Internal controls used by government agencies may include guidance that defines the specific data to be collected and any documentation needed to support the data and safeguards to ensure data are secure. Some key aspects of internal controls for collecting and reporting data include: Guidance: Guidance should clearly and consistently define all data elements required for reporting, and effectively communicate this information to states and local areas. If definitions are vague or inconsistent, then program staff may interpret them incorrectly, resulting in more errors to the data. Additionally, any guidance and documentation from the national office to states and local areas must be clear and free of any conflicting or contradictory instructions. If reporting instructions are misinterpreted by program staff, then the data may not be useful to assess program performance. Data entry procedures and edit check software: Data entry procedures and edit check software can help ensure data entering the designated reporting system are accurate and consistent. Written guides establishing who is responsible for each step in data creation and maintenance, and how data are transferred from initial to final formats can ensure data are consistently reported. Additionally, using electronic data management and processing software programs to conduct automated checks on data values and integrity can limit errors when data are reported at a later date. Monitoring: Monitoring can ensure reported data are accurate and complete. Common monitoring practices may include formal on-site reviews of individual case files and source documentation at both the state and local levels, and assessments of issued guidance to ensure that information collected nationwide is consistent with existing policies and in compliance with laws and regulations. Three key issues—flexibility in federal guidance, major changes to states’ information technology (IT) systems and limited monitoring efforts—have compromised states’ early efforts to collect and report WIA performance data. The guidance available to states at the time of implementation allowed flexibility in key definitions and contributed to inconsistency in the way the data are collected and reported. The transition from JTPA to WIA required states to make major changes to their IT systems and in some cases, the transition led to problems with the data. States used a variety of strategies to make the necessary system changes, some used the software they had used to report under JTPA, and others developed new software for WIA. More than three-fourths of the states told us that they had made major modifications to their WIA IT systems since implementation. One-third of these states reported that when these modifications were made, they experienced significant problems that affected the quality of the data. Lack of oversight at the local, state, and federal levels made it difficult to ensure that early WIA performance data were accurate. The guidance available to states at the time of implementation was open to interpretation in key terms and contributed to inconsistency in the way that data are collected and reported. Labor allowed states and local areas flexibility in determining when to register a jobseeker in WIA and when participants leave the program (see table 3). Registration. When and who is registered affects all WIA performance measures for adults and dislocated workers because performance data are only collected for those job seekers who are registered under WIA—a process that occurs when they begin receiving services that require significant staff assistance. Labor has provided detailed written guidance to states on who should be registered under WIA and when this registration should occur, but the guidance is open to interpretation in some areas. The guidance provides examples of when to register job seekers, but it sometimes requires staff to make subtle and subjective distinctions. For example, those who receive initial assessment of skill levels and the need for supportive services are not to be registered; those requiring comprehensive assessment or staff-assisted job search and placement assistance are to be registered. In an earlier report, we found that local areas differed on when they registered WIA jobseekers, raising questions about both the accuracy and comparability of states’ performance data, and we recommended that Labor provide clearer guidance. Inconsistencies in when states register participants could lead some states to register fewer participants than others do, which could affect the reported outcomes. Exit. Determining when a participant leaves the program—or exits— affects nearly all WIA performance measures because jobseekers must exit the program in order to be counted in the performance measures. While Labor’s guidance explains when an exit occurs, it also has allowed two different kinds of exits—the hard exit and the soft exit. A hard exit occurs when a participant has a specific date of case closure, program completion or known exit from WIA-funded or one-stop partner-funded services. A soft exit occurs when a participant does not receive any WIA-funded or partner-funded service for 90 days and is not scheduled for future services except follow-up. Furthermore, Labor’s guidance on WIA did not clearly specify which services are substantial enough to delay exiting a participant, and local areas define these services differently. In a recent review we found considerable variation in exit practices at the state and local levels. For example, one local area defined exit as occurring when participants are finished with their WIA services; another local area defined exit when participants have found a new job and the wages for their new job are considered acceptable (regardless of the number of days that have passed since their last service). In addition to allowing states the flexibility to define some performance elements, the initial guidance failed to specify other key elements necessary to ensure data quality. For example, the guidance did not specify which source documentation was to be collected and maintained to support entries into the IT system. In the absence of guidance, some states continued to collect source documentation similar to that collected under JTPA, other states moved to paperless systems and did not collect and retain any source documentation. Without consistent source documentation, there is no assurance that the data in the IT system are accurate. The transition from JTPA to WIA required states to make significant changes to their IT systems, and in some cases, problems during the transition led to data errors. For example, several data elements required in WIASRD—the file of individual exiters that states submit to Labor every year—were similar to those collected under JTPA, but the data definitions were slightly changed. This sometimes led to miscoded or missing data— especially for those participants who were carried over from JTPA into WIA. In addition, new data sources were used to measure outcomes, and the calculations for the measures were complex. Some states integrated their IT systems so that the system that is used for WIA data collection is used for tracking participation in other partner programs as well. These changes required major modifications to the IT systems. States used a variety of strategies to make the necessary system changes, often facing challenges in fully implementing WIA’s requirements. For example, 22 states reported that they used the same software they had used under JTPA to report on WIA performance, but 15 of these states later converted to different software for WIA. Twenty-six states used new software for WIA at implementation, but almost one-third of them replaced that system when it became clear that the new system was not sufficient to meet WIA reporting requirements. The time needed to make system changes varied across states. While nearly half of the states reported that they were able to implement their IT system changes in 1 year or less, the other half reported that it took more than one year, and as long as 3 years (see fig. 2). m. m. . . . . Thirty-nine states reported to us that they had made major modifications to their WIA IT systems since implementation, such as converting to Internet-based systems or adding new capabilities such as case management tracking. Thirteen of these states reported that when they made these modifications, they experienced significant problems that affected the quality of the data, including lost data and difficulties in combining or reconciling data from the multiple systems they had used. While 8 of the states reported that these issues have been resolved, 5 told us that they are still trying to resolve these data quality concerns. Some of the remaining 11 states that did not report making major changes to their IT systems since WIA implementation reported that they made minor changes, such as adding or deleting data elements and adding reporting capabilities. In addition to collecting and reporting the performance data, IT systems must also be able to calculate the performance measures. However, states are not all using the same methodology to calculate these measures. The calculations for the measures are complex and sometimes confusing. For example, in calculating some of the measures for the adult program, states must consider (1) whether the jobseeker is employed at registration, (2) whether he or she is employed at both the first and third quarters after exit, and (3) what data were used to confirm employment. This information results in 14 different ways that adult participants can be grouped together in order to calculate the measures. Labor does not mandate which software package states must use to calculate their performance measures, and at the 5 states we visited, each used a different approach—commercially available software, software developed by the state, or one of two different software packages developed under contract with Labor. These software packages can use slightly different formulas to calculate the measures and, as a result, produce differences in the outcomes reported. Lack of oversight at the local, state, and federal levels made it difficult to ensure that early WIA performance data are accurate or verifiable. During the first year of WIA implementation, Labor’s Inspector General (IG) found insufficient documentation of verification procedures at the state and local levels. The same report questioned the lack of formal federal monitoring to gauge the progress of state efforts to ensure the quality of the data. Furthermore, the report noted that Labor and states lacked adequate monitoring procedures and little was being done to monitor performance data at the case file level. In a previous study, we reported that Labor did not have a standard data monitoring guide in place, and regional officials—who have primary responsibility for monitoring—followed various oversight procedures. Table 4 summarizes WIA’s data quality issues. States have made efforts to address data quality concerns and improve the quality of WIA performance data. Most states have taken actions to clarify Labor’s guidance to help local areas determine who should be tracked in the performance measures. Almost all states reported on our survey that they have controls for IT systems, such as edit checks or reports to help screen for errors or missing data. In addition, most states reported to us that they monitor local areas to ensure data quality and consistency by assessing local procedures and policies. States have taken some steps to provide additional clarity to help local areas adhere to federal guidance. Over 40 states reported to us that they provide guidance to help local areas determine which jobseekers should be tracked—or registered—for WIA and when participants leave—or exited—services, and therefore get counted in the performance measures. For example, a West Virginia state official said the state developed a list of staff-assisted services that should trigger registration under WIA. Most states also provide technical assistance and training on registration and exit policies (see fig. 3). Some states take other steps to help local areas adhere to federal policies. For example, California state officials attempt to prevent local areas from keeping participants enrolled in the program once they have exited services by incorporating a capability in their IT system that will automatically exit a person who has not had any service for 150 days. States have made efforts to reduce the errors in their WIA performance data. Almost all states reported on our survey that they have controls for IT systems, such as edit checks or reports to help screen for errors or missing data. Forty-six states screen for missing values, and 44 states screen for errors such as data logic inconsistencies (see fig. 4). For example, if an individual is registered in the youth program, but the birth date indicates that the person is 40 years old, this case would be flagged in an error report checking for inconsistencies between these two data elements. Some of the states we visited told us they allow local areas flexibility in deciding who should enter data and how it gets done. In some locations, a case manager who works with the participant may enter data, and sometimes the case manager completes forms that are given to a data entry specialist. Despite these differences, most states have implemented edit checks and other controls in their IT systems to detect and control for errors. For example, state officials we met with in West Virginia said that the state created screen edits and drop-down menus to guide case managers as they enter data. If a case manager does not enter the necessary data, the system will not let the data entry process go forward until the data are entered. State officials acknowledged that people entering data can still make mistakes if they choose the wrong option on a drop down menu, but they told us they try to minimize these mistakes by conducting training sessions to acquaint staff with the right techniques. States also address data entry errors by running error reports. In New York, state officials told us that they produce error reports for each local area to show where data are missing, meeting with local officials to discuss these reports every 6 weeks. Labor officials in most of Labor’s six regions told us that states have made improvements to their IT systems since WIA was first implemented. For example, Labor officials in one region said that they identified data quality issues related to states’ IT systems in 10 of the 11 states in that region in program year 2000 and found similar issues in only 5 states in the region between program years 2002 and 2004. A Labor official in another region told us that initial data collection efforts were poor because states were largely focused on getting WIA up and running and had not developed adequate IT system instructions. Now, most states have developed IT system manuals with clear instructions. Some regional officials told us that they provided technical assistance and closely monitored states that had early problems with their IT systems. Most states told us they monitor local areas to ensure data quality and consistency by assessing local procedures and policies. Thirty-eight states reported to us that they monitor data collection at the local level. At least 33 states also reported to us that they conduct monitoring of local policies and procedures on registrations and exits and data entry (see fig. 5). State officials at the sites we visited generally said that they conduct annual monitoring visits to local areas or one-stop centers, and some conduct more frequent monitoring visits. Texas state officials we visited told us that the state monitors each local area once a year that includes reviewing participant files to assess eligibility decisions and ensure that outcomes are documented. In New York, state officials said that they have monitoring teams located in five regions across the state who visit the local areas within their regions about once a month. Initially, these visits focused on program compliance, but they have recently been expanded to include data quality. Labor recently began addressing data quality issues, however, some data quality issues remain. In 2004, Labor addressed some data quality concerns by implementing new data validation requirements that called for states to review samples of participant files and provided software to help states ensure that the performance measures are computed accurately. Most states reported on our survey that Labor’s new requirements are having positive effects on states’ and local areas’ attention to data quality. However, Labor does not currently have methods in place to review states’ data validation efforts and hold states accountable to the data validation requirements. Labor’s guidance requiring states to implement common performance measures on July 1, 2005, clarified some key data elements that had been problematic with regard to the WIA performance measures, but it does not address all the issues. Further, Labor has some federal monitoring processes in place but lacks a standard monitoring guide to address data quality. To address data quality concerns, Labor required states to implement new data validation procedures for WIA performance data in October 2004. This process requires states to conduct two types of validation: (1) data element validation—reviewing samples of WIA participant files, and (2) report validation— assessing whether states’ software accurately calculated performance outcomes. These requirements addressed a gap in earlier guidance by providing instructions for collecting and retaining source documentation to verify that the reported data are accurate. This includes specifying which documentation is acceptable and what should be maintained in participant files. For example, to document that a participant is placed in post program employment, states must show that the information was obtained from the UI wage records or Wage Record Interchange System or other sources such as a pay stub, a 1099 form, or telephone verification with employers. Labor’s data validation process requires states to monitor local areas to compare data elements that were reported to the state against source documentation to verify that the data are accurate. Labor selected data elements for validation based on factors such as feasibility and risk of error. For example, self-reported data elements, such as race and ethnicity, are not validated because it is not feasible to locate the participant to verify these items. Data elements needing independent documentation, such as the use of supplemental data sources to determine employment, are assumed to be at higher risk of error than from using the UI wage records. Labor provided software to help states select a sample of files to be validated that includes participants from each group reported on in the performance measures—adults, dislocated workers, older youth, and younger youth. States are required to conduct monitoring visits to the local areas selected for validation and compare data elements for each participant in the sample to source files to ensure accuracy, but Labor does not have a standard process to verify that states did this correctly. State monitors record whether each data element is supported by source documentation, and therefore passes, or whether the documentation shows the element was incorrect or was not supported with source documentation, and, therefore, fails the element. States use Labor’s software to total error rates for each population group and states submit these data to Labor. To address inconsistencies in calculating the performance measures, Labor’s report validation software verifies the accuracy of outcomes reported by states. States can use Labor’s software in two ways: they can use the software to compute the state’s performance measures or they can use the software to check the calculations computed by their state’s software to make sure that the measures were calculated accurately. According to Labor, about 20 states are currently using its software to compute their states performance measures. The remainder of states use their own or commercially available software to compute outcomes. These states must submit validation reports to Labor to show any differences between their calculations and the outcomes computed with Labor’s software. Since initiating data validation, Labor made a number of modifications to its software, and states reported on our survey that they experienced some challenges in using the software. Most states reported that they experienced only minor difficulties or had no problems in using Labor’s software for both data element validation and report validation (see fig. 6). However, some states did report major difficulties. For example, seven states reported that they initially had major difficulties with report validation, such as resolving discrepancies or errors in Labor’s software. States also reported concerns that they were not always informed when Labor made updates to the software and did not always receive adequate time to work with the software before the results were due to Labor. In addition, some states reported on our survey that conducting data element validation was time consuming. Half the states that were able to estimate the time it took to complete data element validation said it took 60 days and half said that it took more than 60 days. The majority of states told us that Labor’s guidance, training, and technical assistance on data validation were sufficient (see fig. 7). It is too soon to fully assess whether Labor’s efforts have improved data quality, however, at least 46 states reported on our survey that Labor’s new requirements have helped increase awareness of data accuracy and reliability at the state and local level (see fig. 8). A New York state official told us that the federal requirements helped local staff better understand the connection between the data that get entered and how these data affect performance levels. In addition, over 30 states said that the new requirements have helped them in their monitoring of outcomes and eligibility. Some states and local areas we visited reported finding errors in their data through the data validation process and have made modifications to state and local procedures to enhance data quality as a result. For example, a local area in California started doing monthly spot checks of files to identify and correct errors on an ongoing basis. In New York, a local area told us that it added a new staff person, developed new forms and procedures, and centralized data entry to have more control over data quality as a result of the federal data validation process. While either centralized or decentralized data entry may be effective, experts in WIA performance data told us that one of the most important factors to avoid human error is for program managers and staff who enter data to understand how the data are used. While Labor’s data validation requirements are having some positive effects on states and local areas, Labor currently has no mechanism to hold states accountable for complying with the data validation requirements. Labor has plans to develop accuracy standards for report validation and to hold states accountable to these standards in about 3 years. Initially, Labor planned to use program year 2003—July 1, 2003 until June 30, 2004—as a base year for developing accuracy standards on report validation. However, as a result of reporting changes for the common measures, Labor has postponed the development of these standards until program year 2006, beginning July 1, 2006. At this time, Labor does not have plans to develop accuracy standards for the data element validation portion of its requirements. In addition, Labor does not conduct its own review of a sample of WIA participant files verified by states as part of data validation to ensure that states did this process correctly. Table 5 provides a summary of data quality concerns and how Labor’s data validation efforts affect these concerns. In response to an OMB initiative, Labor recently began requiring states to implement common performance measures for WIA programs. OMB established a set of common measures to be applied to most federally funded job training programs that share similar goals. Labor further defined the common measures for all of its Employment and Training Administration programs and required states to implement these measures beginning July 1, 2005. In addition, Labor is replacing the definitions for the WIA measures that are similar to the common measures with the new definitions for common measures (see table 6). Moving to the common measures may increase the comparability of outcome information across programs and make it easier for states and local areas to collect and report performance information across the full range of programs that provide services in the one-stop system. Many federal job training programs had performance measures that track similar outcomes but have variations in the terms used and the way the measures are calculated. For example, WIA’s adult program uses a different time period to assess whether participants got a job than the Wagner-Peyser funded Employment Service does. WIA’s adult program looks at whether participants get a job by the end of the first quarter after exit, whereas the Employment Service looks at whether participants get a job in the first or second quarter after registration. Under common measures, both programs use the same time period for this measure. Labor’s new guidance for common measures requires states to collect a count of all WIA participants who use one-stop centers. This can help provide a more complete picture of the one-stop system, but it does not clarify when participants should be registered for WIA and tracked in the performance measures. Therefore, it raises questions about both the accuracy and comparability of WIA’s outcomes for adults and dislocated workers. Under common measures, states are being required to begin collecting and reporting a quarterly count of all jobseekers who receive services at one-stop centers. To track these jobseekers, Labor suggested that states collect a valid Social Security number, but allowed states to exclude individuals who do not wish to disclose their Social Security numbers. In addition, Labor is encouraging states to voluntarily report performance information on all jobseekers that are counted in one-stops. However, it is not clear how many states have the capability to track jobseekers who receive only self-service and informational activities. While 30 states reported on our survey that they have a state system to track all jobseekers, some officials we visited told us they do not require local areas to collect and report this information to the state. Given this, implementing the new requirement may take time and early data collection efforts may be incomplete. Labor’s guidance on common measures provides for a clearer understanding of when WIA participants should be exited from the program than did earlier WIA guidance. First, the guidance provides a more uniform definition of exit. In the past, local areas could use a hard exit—when a participant has a known date of completion or exit from services or a soft exit—when a participant has not received any services for 90 days. Under the new guidance, only soft exits will be allowed and states will no longer be able to report a hard exit. Second, Labor clarified that some services are not substantial enough to keep a participant from being exited from WIA. For example, if a case manager is only making phone calls to the participant to see if he or she has a job or needs additional services or income support payments, those phone calls are not considered a service (see table 7). This new clarification may help prevent local areas from keeping WIA participants enrolled long after they have completed their last valid service. In a previous study, however, we cautioned that rushed implementation of these reporting changes may not allow states and local areas enough time to fully meet the requirements and could negatively affect the data quality of the information reported. In addition to data validation, Labor has some limited federal monitoring processes in place to oversee state and local performance data. Labor’s regional offices—with primary responsibility for oversight—conduct a limited review of WIA report data to review quarterly and annual WIA performance reports. This generally involves identifying outliers or missing data and comparing the data with data in previous reports. If Labor regional officials identify basic problems with the data, they contact states to reconcile concerns. Labor’s headquarters implemented an electronic system to manage grant oversight and track activities throughout the program year—called Grants E-Management System (GEMS). This system provides automated tools for conducting grant monitoring activities, including performing risk assessments and generating reports. Labor developed the risk assessment to help determine the programs and grant projects most in need of monitoring. The risk assessment assigns a risk level to each state based on past performance and other criteria. For example, for the WIA program, a state may be considered at risk if it failed to meet its performance levels in the prior year. However, regional officials can override the risk assessment if they are aware of other information that may not be captured in GEMS. Labor also implemented a core monitoring guide in spring 2005 to ensure that certain basic parameters are being followed during monitoring visits across all regions, but this guide does not provide for a standard analysis of data quality issues. According to Labor officials, they are developing program supplements for this guide that will address other issues specific to various programs. One regional office developed an extensive monitoring guide to review state and local guidance, procedures used for data entry, IT systems, and other data quality factors. This guide has been used since 2003 to review the eight states in its region. In addition, Labor officials said that several regional offices are using this guide and they plan to develop a similar guide that will be used across all regions. WIA overhauled the way federally funded employment and training services are provided to jobseekers and employers, and introduced changes that significantly affected the way performance data are collected and reported for WIA. Making this shift has taken a long time and some trial and error on the part of Labor, states, and localities. The magnitude of changes required considerable retooling of states’ IT systems, which had a negative effect on the integrity of WIA performance data during the initial years of implementation. Since then, states have made progress in addressing challenges they faced in modifying or developing new IT systems and have invested considerable effort establishing controls for IT systems to minimize data errors. In addition, Labor’s recent efforts to implement common performance measures across many of the WIA partner programs and its revised WIA reporting requirements have helped to address the concerns about when participants complete services and should be tracked in the performance measures. The new requirement for states to capture limited data on all WIA participants is an important step to better determine the full reach of WIA. However, this change still does not address the long-standing challenge Labor has faced in clearly defining which participants should be counted in the performance measures. Without clear guidance, the WIA performance data will continue to be inconsistent, even if the other data quality safeguards in place at the federal, state, and local level improve the quality of each state’s and local area’ s data. Labor’s implementation of new data validation requirements is a major step toward addressing concerns about data quality resulting from the limited guidance and monitoring of WIA performance data in the past. By providing additional guidance and software to help states calculate the performance measures in a more uniform manner and requiring states to compare data reporting with participant case files, Labor has gone a long way toward helping ensure the consistency and comparability of the data. Most notably, these requirements have significantly raised awareness of data quality at the state and local levels, which is an essential part of ensuring data quality. However, more time is needed to fully assess the impact these new requirements are having on data quality. In addition, Labor does not currently review a sample of the participant files verified by states, nor does it have a mechanism to hold states accountable for meeting the data validation requirements. Further, Labor has not developed a standard monitoring guide to more uniformly assess state and local data collection and processing to ensure data quality. Without a standard monitoring guide and a means to hold states accountable to the data validation requirements, it will be difficult to assure decision makers that the data is of sufficient quality for applying incentives and sanctions, and making budget decisions. To address the inconsistencies in determining when participants should be registered and counted in the performance measures, we recommend that the Secretary of Labor determine a standard point of registration and monitor states to ensure that the policy is consistently applied. To enhance the data validation requirements, we recommend that the Secretary of Labor: conduct its own review of the WIA participant files validated by states to ensure that states did this correctly, and ensure that steps are taken to hold states accountable to both the report validation and data element validation requirements. To address variations in federal monitoring practices, we recommend that the Secretary of Labor develop a standard comprehensive monitoring tool for WIA performance data that is used across all regions, including monitoring the new guidelines for determining when participants end services. We provided a draft of this report to Labor for review and comment. Labor agreed with our findings and recommendations. Labor agreed that the lack of a standard point of registration and exit prevents comparisons across states and leads to performance outcome information that is arbitrary and inconsistent. Labor also agreed that steps are needed to increase the integrity of the data validation requirements and to improve the completeness and consistency of oversight. A copy of Labor’s response is in appendix II. In response to our recommendations, Labor noted that it plans to implement a policy prior to the start of program year 2006 to clarify the point of registration and exit. In addition, Labor plans to modify the current data validation procedures to begin reviewing a sample of states’ validated files and plans to hold states accountable for data validation results by program year 2006. Further, Labor told us that it is taking steps to develop a comprehensive monitoring guide for performance data and plans to provide training on this new guide to help improve the completeness and consistency of oversight. We are sending copies of this report to the Secretary of Labor, relevant congressional committees, and others who are interested. Copies will also be made available to others upon request. The report is also available on GAO’s home page at http://www.gao.gov. If you or members of your staff have any questions about this report, please contact me at (202) 512-7215. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributors to this report are listed in appendix III. We examined (1) the data quality issues that have affected states’ efforts to collect and report Workforce Investment Act (WIA) performance data; (2) states’ actions to address them; and (3) the actions the Department of Labor (Labor) is taking to address data quality issues, and the issues that remain. To learn more about states’ experiences implementing data collection and reporting system changes for WIA, their implementation of Labor’s data validation requirements for WIA, and state and local efforts to address the quality of WIA data, we conducted a web-based survey of state workforce officials and conducted site visits in five states, where we interviewed state officials and visited two local areas or one-stop centers in each state. We also collected information on the quality of WIA data through interviews with Department of Labor officials in headquarters and all six regional offices, nationally recognized experts, and reviewed relevant research literature. Our work was conducted between June 2004 and September 2005 in accordance with generally accepted government auditing standards. To determine the factors that affect the quality of WIA performance data, we conducted a Web-based survey of state workforce officials. These officials were identified using a GAO-maintained list of state WIA officials. We e-mailed the contacts, and they confirmed that they were the appropriate contact for our survey or identified and referred us to another person at the state level. Survey topics included (1) the changes made to data collection and reporting during the transition from the Job Training Partnership Act to WIA, (2) the current status of WIA data collection and reporting systems, (3) implementation of the U.S. Department of Labor’s data validation requirements, and (4) state and local efforts to ensure the accuracy and reliability of WIA data. The survey was conducted using a self-administered electronic questionnaire posted on the Web. We contacted respondents via e-mail announcing the survey, and sent follow- up e-mails to encourage responses. The survey data were collected between February and May 2005. We received completed surveys from all 50 states (a 100 percent response rate). We did not include Washington, D.C.,and U.S. territories in our survey. We worked to develop the questionnaire with social science survey specialists. Because these were not sample surveys, there are no sampling errors. However, the practical difficulties of conducting any survey may introduce errors, commonly referred to as nonsampling errors. For example, differences in how a particular question is interpreted, in the sources of information that are available to respondents, or how the data are entered into a database can introduce unwanted variability into the survey results. We took steps in the development of the questionnaires, the data collection, and data analysis to minimize these nonsampling errors. For example, prior to administering the survey, we pretested the content and format of the questionnaire with several states to determine whether (1) the survey questions were clear, (2) the terms used were precise, (3) respondents were able to provide the information we were seeking, and (4) the questions were unbiased. We made changes to the content and format of the final questionnaire based on pretest results. In that these were Web-based surveys whereby respondents entered their responses directly into our database, possibility of data entry errors was greatly reduced. We also performed computer analyses to identify inconsistencies in responses and other indications of error. In addition, a second independent analyst verified that the computer programs used to analyze the data were written correctly. We visited five states—California, New York, Texas, West Virginia, and Wyoming,—and traveled to two local areas or one-stop centers in each of these states. We selected these states because they represent a range of IT systems—statewide comprehensive systems versus local systems with a state reporting function, include single and multiple workforce areas, and are geographically diverse. From within each state, we judgmentally selected two local boards. In the case of our single workforce area state, we visited two one-stop centers (see table 8). In each state visited, we obtained general information about the state’s implementation of WIA, an overview of the state’s WIA administrative structure, the management information system and reporting processes in place to meet the federal requirements, data quality practices at the state and local levels, implementation of Labor’s data validation requirements. We interviewed state officials responsible for local areas’ WIA programs and analyzing and reporting on the state’s WIA performance data, as well as other state WIA and information technology (IT) officials and staff of the state’s Workforce Investment Board. At the local areas, we interviewed WIA officials and staff, including service providers, staff responsible for performance management issues, IT staff, case managers and other frontline staff, as well as staff of the local area Workforce Investment Board. The state and local interviews were administered using a semi- structured interview guide. Information that we gathered on our site visits represents only the conditions present in the states and local areas at the time of our site visits, from August 2004 through March 2005. We cannot comment on any changes that may have occurred after our fieldwork was completed. Furthermore, our fieldwork focused on in-depth analysis of only a few selected states and local areas or sites. On the basis of our site visit information, we cannot generalize our findings beyond the states and local areas or sites we visited. Dianne Blank, Assistant Director Laura Heald, Analyst-in-Charge In addition, the following staff made major contributions to this report: Melinda Cordero, Vidhya Ananthakrishnan, and Leslie Sarapu served as team members; Jennifer Miller assisted with early data collection. Carolyn Boyce advised on design and methodology issues; Susan Bernstein advised on report preparation; Jessica Botsford advised on legal issues; Avrum Ashery and Robert Alarapon provided graphic design assistance; and Bill Hutchinson and Daniele Schiffman verified our findings. Workforce Investment Act: Substantial Funds Are Used for Training, but Little Is Known Nationally about Training Outcomes. GAO-05-650. Washington, D.C.: June 29, 2005 Unemployment Insurance: Better Data Needed to Assess Reemployment Services to Claimants. GAO-05-413. Washington, D.C.: June 24, 2005 Workforce Investment Act: Labor Should Consider Alternative Approaches to Implement New Performance and Reporting Requirements. GAO-05-539. Washington, D.C.: May 27, 2005 Workforce Investment Act: Employers Are Aware of, Using, and Satisfied with One-Stop Services, but More Data Could Help Labor Better Address Employers’ Needs. GAO-05-259. Washington, D.C.: February 18, 2005 Workforce Investment Act: States and Local Areas Have Developed Strategies to Assess Performance, but Labor Could Do More to Help. GAO-04-657. Washington, D.C.: June 1, 2004. Workforce Investment Act: Labor Actions Can Help States Improve Quality of Performance Outcome Data and Delivery of Youth Services. GAO-04-308. Washington, D.C.: February 23, 2004. Workforce Investment Act: One-Stop Centers Implemented Strategies to Strengthen Services and Partnerships, but More Research and Information Sharing Is Needed. GAO-03-725. Washington, D.C.: June 18, 2003. Older Workers: Employment Assistance Focuses on Subsidized Jobs and Job Search, but Revised Performance Measures Could Improve Access to Other Services. GAO-03-350. Washington, D.C.: January 24, 2003 Workforce Investment Act: Youth Provisions Promote New Service Strategies, but Additional Guidance Would Enhance Program Development. GAO-02-413. Washington, D.C.: April 5, 2002. Workforce Investment Act: Better Guidance and Revised Funding Formula Would Enhance Dislocated Worker Program. GAO-02-274. Washington, D.C.: February 11, 2002. Workforce Investment Act: Improvements Needed in Performance Measures to Provide a More Accurate Picture of WIA’s Effectiveness. GAO-02-275. Washington, D.C.: February 1, 2002. | Federal programs carried out in partnership with states and localities continually balance the competing objectives of collecting uniform performance data with giving program implementers the flexibility they need. Our previous work identified limitations in the quality of performance data for the key employment and training program--the Workforce Investment Act (WIA). WIA relies on states and localities to work together to track and report on participant outcomes, and it changed the way outcomes are measured. Given the magnitude of changes and the impact such changes can have on data quality, we examined (1) the data quality issues that affected states' efforts to collect and report WIA performance data; (2) states' actions to address them; and (3) the actions the Department of Labor (Labor) is taking to address data quality issues, and the issues that remain. Three key issues--flexibility in federal guidance, major changes to states' information technology (IT) systems, and limited monitoring--compromised states' early efforts to collect and report WIA performance data. Labor's initial guidance allowed states and local areas flexibility in deciding which jobseekers to track and when jobseekers leave services and get counted in the measures. As a result, states and local areas have differed on whom they track and for how long. States took various approaches to implement IT systems for meeting WIA reporting requirements. Thirty-nine states reported to us that they made major modifications to their IT systems since WIA was first implemented in 2000. Thirteen of them said the changes resulted in problems affecting data quality, and 5 states are still trying to resolve these problems. In addition, oversight of WIA performance data was insufficient at all levels during early implementation. Almost all states have made efforts to improve the quality of WIA performance data--at least 40 states have controls in their IT systems that capture WIA performance data, such as edit checks or exception reports to help screen for errors or missing data. Forty-three states have taken actions to clarify Labor's guidance and help local areas determine who should be tracked in the performance measures. In addition, most states said they monitor local areas by assessing local procedures and policies. Labor recently began addressing data quality issues, however, some issues remain. In 2004, Labor addressed some data quality concerns by requiring states to validate their data and ensure the accuracy of their performance outcomes. Most states told us that Labor's requirements have increased awareness of data quality at the state and local level. However, Labor does not have methods in place to review states' validation efforts or hold states accountable for complying with its requirements. Labor issued guidance requiring states to implement common performance measures on July 1, 2005, which clarified some key data elements, but does not address all the issues. Labor has some federal monitoring processes in place but lacks a standard monitoring guide to address data quality. |
In 2006, more than 42,600 people were killed in motor vehicle crashes. Overall, this represents a 1.5 percent increase in the number of fatalities from 1997 to 2006, although the fatality rate—fatalities per 100 million vehicle miles traveled (VMT)—declined by approximately 14 percent, from 1.65 in 1997 to 1.41 in 2006 (see fig. 1). Through SAFETEA-LU, Congress authorized nearly $2.2 billion for 4 years, from fiscal years 2006 through 2009, for seven programs to provide safety grants to assist states in their efforts to reduce traffic fatalities (see table 1). This represents an increase of $172 million annually from the authorization levels under the Transportation Equity Act for the 21st Century (TEA-21). States’ highway safety offices use these funds to reimburse selected state and local organizations—nonprofit organizations, universities, hospitals, and law enforcement agencies—for conducting traffic safety improvement activities that have been approved by the state. Two of these grant programs—State and Community Highway Safety and Occupant Protection—were largely unchanged by SAFETEA-LU. The State and Community Highway Safety grant provides highway safety funds for states through a formula based on each state’s population and public road miles. All states are eligible to receive this grant after submitting a performance plan that establishes goals and performance measures to improve highway safety in the state, as well as a highway safety plan that describes activities to achieve those goals. The Occupant Protection grant provides incentive funds for states to adopt and implement programs to reduce deaths and injuries from riding “unrestrained” or “improperly restrained.” To be eligible for this grant, states must meet four out of six criteria, some of which are also criteria for other safety incentive grants. One of these criteria is to pass a safety belt law providing for primary enforcement, which allows law enforcement officers to stop a driver for not wearing a safety belt. Transportation safety experts generally consider primary safety belt laws to be the most effective countermeasure to prevent traffic fatalities and injuries to vehicle occupants, and states with primary enforcement laws generally have lower fatality rates than states that do not have such laws. States are required to provide matching funds for both grants; the state share required for the State and Community Highway Safety grant is at least 20 percent of the total program cost, while the state share for the Occupant Protection grant is at least 25 percent for the first and second years (beginning in 2003), 50 percent for the third and fourth years, and 75 percent for the fifth and sixth years. The remaining five grant programs—Safety Belt Use, Child Safety and Child Booster Seat Use, Alcohol Impaired Driving Countermeasures, Motorcyclist Safety, and State Traffic Safety Information Systems Improvement—were modified or added by SAFETEA-LU. States’ ability to qualify for grants is exclusively determined by whether they meet the statutory grant qualification criteria established by SAFETEA-LU. Safety Belt Use: This one-time grant encourages states to enact and directly enforce safety belt use laws. States can use grant funds for a range of highway safety activities, including public education programs or construction to improve a hazardous roadway. To be awarded a grant, a state can qualify in one of three ways: Enact a primary safety belt law after January 1, 2003, and certify that the law is enacted and will be enforced; these “new law states” receive priority in the award process and are awarded a one-time grant equal to 475 percent of the amount they were apportioned for their fiscal year 2003 State and Community Highway Safety grant. Beginning in fiscal year 2008, certify that the state has achieved at least an 85 percent safety belt use rate in the two preceding calendar years. If a state does not meet either of the first two criteria, and if funds remain after grants have been awarded to all states meeting those criteria, states that had a primary safety belt law in effect prior to 2003, and certify the law is enacted and being enforced are eligible to receive a one-time grant equal to 200 percent of the amount they were apportioned for their fiscal year 2003 State and Community Highway Safety grant. In addition, NHTSA will allocate any grant funds remaining available on July 1, 2009, among all states that have in effect and are enforcing a primary safety belt law for all passenger motor vehicles as of that date. These funds will be allocated among the states in accordance with the formula used to determine the amount of the State and Community Highway Safety grant. The Safety Belt Use grant program authorized in SAFETEA-LU includes more stringent criteria than the corresponding TEA-21 incentive grant, which awarded grants to states based solely on improvements in safety belt use rates. Congress authorized $500 million over 5 years for the safety belt incentive grant authorized in TEA-21 and $498 million over 4 years for the program authorized in SAFETEA-LU. There are no requirements for state matching funds. Child Safety and Child Booster Seat Use (Child Safety and Booster Seat): This grant is designed to encourage states to enact and enforce booster seat laws. These grant funds may be used only for child restraint programs, including programs to enforce laws or train child safety professionals and parents on the proper use of child safety and booster seats. Up to 50 percent of the funds that a state receives under this grant may be used to purchase and distribute child restraints—both child safety and booster seats—for low-income families. To qualify for this grant, states must enact and enforce a law requiring any child riding in a passenger motor vehicle who is under the age of 8 to be secured in an appropriate child restraint system, unless the child weighs more than 65 pounds or is 4 feet 9 inches or taller. Similar to the Safety Belt Use grant program, the criteria for the Child Safety and Booster Seat grant program adopted in the wake of SAFETEA-LU are more stringent than the corresponding pre-SAFETEA-LU grant program; states had to apply and then use the grant for child passenger protection education activities. Congress authorized $15 million for the child passenger protection education grant under TEA-21 and $25 million for the Child Safety and Booster Seat grant under SAFETEA-LU. States are also required to provide matching funds of at least 25 percent during the first 3 years and 50 percent during the fourth year. Alcohol Impaired Driving Countermeasures (Impaired Driving): This grant is designed to encourage states to implement enforcement, education, training, and other countermeasure activities to reduce alcohol- impaired driving. States can use the grant to implement these activities, such as training for law enforcement officers and advertising and educational campaigns that publicize sobriety checkpoints. States may also purchase equipment—such as blood alcohol content (BAC) testing devices—to assist officers in enforcement activities. States can qualify to receive this grant in three ways: (1) achieving an alcohol-related fatality rate of 0.5 or less per 100 million VMT, (2) being 1 of the 10 states with the highest alcohol-related fatality rate, or (3) meeting a minimum number of the eight programmatic criteria—three in fiscal year 2006, four in fiscal year 2007, and five in fiscal years 2008 and 2009. The programmatic criteria are as follows: Implement a program to conduct high-visibility enforcement campaigns using checkpoints or saturation patrols, along with paid and earned media; Implement a program to educate judges and prosecutors about prosecuting and adjudicating offenders; Implement a program to increase the rate of BAC testing of drivers involved in fatal crashes; Enact legislation imposing stronger sanctions or additional penalties for high-risk drivers whose BAC is 0.15 or more; Implement a program to rehabilitate repeat or high-risk offenders or refer them to a state-sanctioned DWI court; Develop a strategy to prevent underage drivers from obtaining alcoholic beverages and to prevent persons of any age from making alcoholic beverages available to persons under 21; Implement a program to suspend or revoke licenses for drivers who were apprehended while driving under the influence; or Implement a “self-sustaining impaired driving prevention program” in which a significant portion of DWI fines or surcharges collected are returned to communities for activities to reduce alcohol-impaired driving. While the Impaired Driving grants authorized by SAFETEA-LU are similar to the impaired driving grants authorized in TEA-21, the number of criteria a state must meet differs. Specifically, states had to meet more criteria for the TEA-21 grants, but the SAFETEA-LU grant criteria are more stringent. For example, the TEA-21 criteria required states to implement programs, whereas the SAFETEA-LU criteria may include legislative requirements. SAFETEA-LU more than doubled the funding—from $219.5 million to $515 million—authorized for the Impaired Driving grant program. States are also required to provide matching funds of at least 25 percent the first and second years and 50 percent the third and fourth years. Motorcyclist Safety: SAFETEA-LU established a new incentive grant program to encourage states to adopt and implement programs to reduce the number of crashes involving motorcyclists. To be eligible to receive this grant, a state must meet one of six criteria in the first fiscal year and two of the criteria in the second and subsequent years. The criteria are as follows: Implement a statewide training program for motorcycle riders; Implement a program to promote motorcyclist awareness; Achieve a reduction in fatalities and crashes involving motorcycles in Implement a statewide impaired-driving program, including measures to reduce impaired motorcycle operation; Achieve a reduction in fatalities and crashes involving impaired motorcyclists in the preceding year; or Use all fees collected from motorcyclists—such as motorcycle licensing and registration—for motorcycle programs. Funds under this grant program may be used for motorcyclist safety training and motorist awareness programs, including improvement of training curricula, delivery of training, recruitment or retention of motorcyclist safety instructors, and public awareness and outreach programs. States are not required to provide matching funds. State Traffic Safety Information Systems Improvement (Traffic Safety Information Systems): This grant program provides funding for states to adopt and implement programs to improve the timeliness, accuracy, completeness, uniformity, integration, and accessibility of state data needed to identify priorities for national, state, and local highway and traffic safety programs. The purpose of the grant program is to improve the compatibility and interoperability of states’ data systems with each other and with national data systems, and enhance NHTSA’s ability to analyze national trends in traffic safety. To qualify for a grant for the first year, a state must establish a multidisciplinary highway safety data and traffic records coordinating committee; develop a multiyear safety data and traffic records strategic plan, approved by the coordinating committee and containing performance- based measures; and certify that it has adopted and is using the model data elements included in the Model Minimum Uniform Crash Criteria and National Emergency Medical Service Information System determined by the Secretary to be useful, or certify that grant funds will be used toward adopting and using the most elements practicable. To qualify for a grant in subsequent years, a state must certify that an assessment or audit of the state traffic records system has been conducted or updated within the preceding 5 years, certify that the coordinating committee continues to operate and specify how the grant funds and any other state funds will support the demonstrate measurable progress toward achieving the goals and objectives identified in the plan, and submit a report showing measurable progress in implementing the plan. The eligibility requirements for this grant program are more stringent than under prior law. For example, states must demonstrate measurable progress to qualify for subsequent-year funding, whereas previously they only had to certify they had a traffic records committee and activities were funded with the grant regardless of the impact of the projects. Congress authorized $32 million in TEA-21 compared with $138 million in SAFETEA-LU, a significant increase in funding. States are also required to provide matching funds of at least 20 percent of the total federal and state program costs. In fiscal years 2006 and 2007, NHTSA awarded to eligible states about $576 million for the five safety grants—about $11 million less than the authorized amount of $587 million. According to NHTSA officials, the total amount of funds authorized was not awarded because fewer states were able to pass laws to become eligible for certain grants than anticipated. With the exception of the safety belt grant program, the number of states receiving the grants remained constant or increased from fiscal year 2006 to 2007, although the extent to which states qualified for the different grants varied, with fewer states receiving grants that require passing a law. NHTSA uses several oversight processes to determine the extent to which states are meeting safety-related performance goals that grant-funded activities are designed to address and monitor how the states spend grant funds. NHTSA’s oversight processes have evolved over time in response to state concerns about restrictive NHTSA oversight and our prior recommendations that NHTSA ensure more consistent use of management and oversight tools, provide guidance to its regions on use of those tools, and provide a more consistent means of measuring progress toward highway safety goals. The DOT Office of Inspector General and we currently have reviews under way to evaluate the effectiveness of NHTSA’s performance-based oversight approach. In fiscal years 2006 and 2007, NHTSA awarded about $576 million for the five safety grant programs—about $11 million less than the authorized amount of $587 million—to states meeting statutory grant criteria (see fig. 2). The unawarded funds were primarily for the Child Safety and Booster Seat grants in fiscal year 2006—$3.4 million, or over one-half of the authorized amount of $6 million—and for the Safety Belt Use grant in fiscal year 2007—$4.2 million. NHTSA officials determined that a number of states did not meet the statutory criteria contained in SAFETEA-LU for the Child Safety and Booster Seat grant. Consequently, the amount of awards for the Child Safety and Booster Seat grant in fiscal year 2006 was less than the amount authorized. The remainder of the $11 million was due to an across-the-board 1 percent rescission in fiscal year 2006. From fiscal year 2006 to 2007, the number of states that received the five grants generally remained constant or increased, although the number of states varied by grant. With the exception of the Safety Belt Use grant, states maintained or increased participation in the grant programs. Specifically, the number of states eligible for the Child Safety and Booster Seat grant more than doubled from 5 to 13 states, the number awarded the Impaired Driving grant held steady at 50 states, and states awarded the Motorcyclist Safety grant and the Traffic Safety Information Systems grant increased from 44 states to 47 and 49 states, respectively. Overall, 8 states have enacted and are enforcing a primary safety belt law to qualify for Safety Belt Use grants since December 31, 2002—the date specified in SAFETEA-LU. The 8 states represent a 50 percent increase in the number of states that previously had primary safety belt laws. The decline in the number of states receiving Safety Belt Use grants—from 22 states in fiscal year 2006 to 18 states in fiscal year 2007—reflects the fact that states passing a primary safety belt law for the first time received grants in a single year, while 16 states that had the law in place prior to 2003, received the grant over 2 years: In fiscal year 2006, 6 states qualified for grants by enacting laws to implement and enforce primary safety belt laws, in addition to the 16 states that had primary safety belt laws prior to 2003; these 6 states received the full amount of their grants in fiscal year 2006, while the 16 states received a first installment of their grants. In fiscal year 2007, 2 states enacted laws and received the full amount of their grants, and the 16 states received a second and final installment of their grants. The extent to which states received the different grants varied, with fewer states receiving grants that require passing a law. For example, as shown in appendixes II and III, fewer than half of the states qualified for grants that require legislation—the Safety Belt Use and the Child Safety and Booster Seat grants, while all states qualified for Impaired Driving grants and the majority qualified for the Motorcyclist Safety and Traffic Safety Information Systems grants (see apps. IV through VI). In contrast to the Safety Belt Use and Child Safety and Booster Seat grants, states could qualify for the Impaired Driving, Motorcyclist Safety, and Traffic Safety Information Systems grants by submitting plans for varied highway safety activities to be funded under those grants, such as motorcyclist training, alcohol enforcement activities, or new or enhanced information communication systems, without having to enact new legislation. NHTSA awards grants annually under the five safety incentive programs between June and September because the various grant programs require that states meet criteria by completing activities or enacting laws by a specified date in order to be eligible (see fig. 3). NHTSA assists states with applications—including reviewing applications—to meet application deadlines throughout the year. The application deadlines for all five grants are scheduled from June through August of the fiscal year to allow states sufficient time to meet eligibility requirements and receive awards in that fiscal year. For example, the Child Safety and Booster Seat Use grant program requires states to enact a law or revise an existing statute to require that children be restrained in a child safety seat meeting requirements established by the Secretary, which in turn requires that children under 8 years old be restrained properly. To qualify, states must have enacted a qualifying booster seat law by June 30. Consequently, NHTSA set a July 1 application deadline for these grants, at which time states applying for this grant had to certify that the law met NHTSA’s requirements and submit a copy of the law for NHTSA’s review. For the Motorcyclist Safety grant program, NHTSA established an August 1 application deadline and required states to submit an application in the first year and in subsequent years provide certifications that its program meets the grant criteria, such as having a statewide motorcyclist training course or a statewide motorcyclist awareness program. NHTSA officials indicated they needed to give states time to meet the additional criteria in subsequent years and noted that most of these activities would be completed by June 30 of a given year. ! ! ! ! ! ! ! ! The award letter notifies the state it has been awarded a grant. Grant funds are usually available to states shortly after they receive their award letter. If safety incentive grant funds are not used in the fiscal year in which they are awarded, they can be carried over and used in the following fiscal year. Within 2 to 3 months following the receipt of the applications, NHTSA reviewed grant applications and awarded the grants to the states. For 5 of the 10 application periods over both fiscal years, NHTSA reviewed grant applications and awarded the grants to the states within 2 months following the application deadline. For example, in fiscal years 2006 and 2007, NHTSA reviewed applications for the Impaired Driving and Motorcyclist Safety grants, issued letters notifying the states of the amounts awarded under each grant, and made funds available by September 22 of each year following the application deadline of August 1. However, NHTSA took 3 months to review applications and award grants for the Safety Belt Use grants in fiscal year 2007 and the Child Safety and Booster Seat grant in both years. According to NHTSA officials, reviewing the individual state laws took more time than expected because of the large number of applications and the need to determine whether states met the statutory criteria. In some instances, the same NHTSA staff were involved in preparing documentation and reviewing application approval packages for multiple grant programs. Additionally, while applications for the Traffic Safety Information Systems grant program were due June 15, NHTSA awarded these grants in September. According to NHTSA officials, the NHTSA team reviewing the applications for these grants had to request additional or clarifying information from states based on their applications in an effort to assist as many states as possible to qualify. As part of the award process, NHTSA also assists states in applying for grants and meeting application deadlines. NHTSA issued guidance for three of the grant programs within 6 months of the passage of SAFETEA- LU in August 2005, while it issued the implementing regulations for the Impaired Driving and Motorcyclist Safety grant programs within 8 to 11 months (see fig. 4). NHTSA officials explained that they chose the most expedient route for issuing guidance to the states in order to get the guidance out to the states as quickly as possible within relatively tight time frames. Because it takes longer to publish regulations, NHTSA issued implementing guidelines for three of the grant programs. State officials we spoke with generally found NHTSA’s guidance helpful. They told us that NHTSA headquarters and the regions provided information to help the states prepare applications. For example, NHTSA’s regional offices provided the states with information on national priorities and letters indicating traffic safety areas on which states could focus their efforts to support the national goals. Some NHTSA regions also shared grant application review checklists with the states to help them complete their applications as much as possible before sending them to the region for review. In addition, for the Traffic Safety Information Systems grant, NHTSA hired contractors to meet with state officials to help them understand what NHTSA expected in a grant application. However, for the Traffic Safety Information Systems grant program, the guidance from NHTSA headquarters and regions was inconsistent, which resulted in some states having to provide additional information or revise their applications. Some state officials explained they were frustrated by the application process because NHTSA headquarters frequently requested they provide additional or clarifying information to demonstrate measurable progress. However, according to officials from NHTSA headquarters and regions, the team in headquarters that reviewed the applications had different expectations from the NHTSA regions and states. When state officials worked closely with NHTSA regional officials to develop their applications, they believed that the applications were complete and that they qualified for the grants. After the review from NHTSA headquarters, they learned that additional work was needed on their application to qualify. According to NHTSA headquarters officials, in many instances, the states were asked to submit additional information and clarification because their initial applications lacked valid or accurately calculated performance measures or, in some cases, included no performance measures. NHTSA officials initially attributed these problems to the grant’s criteria and more stringent requirements than under prior law; this grant requires that states clearly incorporate how they will measure progress and demonstrate quantifiable results in order to qualify for the grant in subsequent years. However, similar problems also occurred in the application process for fiscal year 2007 when states again received questions from NHTSA and had to clarify information or provide additional information to the review team. To address this issue, NHTSA has begun providing training to NHTSA regional officials so that they provide consistent guidance. In addition, several regions have established regional grant review teams similar to NHTSA headquarters grant review teams to ensure grants receive more complete reviews at the regional level. NHTSA uses several processes to determine the extent to which states are meeting safety-related performance goals that grant-funded activities are designed to address and monitor how the states spend grant funds. Specifically, NHTSA uses a performance-based approach to assess state progress toward meeting safety goals and complements this assessment with oversight processes that monitor whether states are accomplishing the tasks that will allow the state to achieve its goals. NHTSA’s performance-based approach primarily involves assessing states’ progress in meeting safety-related performance goals by comparing state planning documents with annual reports on the state’s performance and conducting special reviews of states not making adequate progress toward their goals. These activities create the opportunity for NHTSA to be involved throughout the lifecycle of state grants, allowing the agency to provide input on the development of state safety goals and performance measures and to oversee the implementation of safety programs and use of federal funds. Establishing state safety goals and conducting activities to achieve the goals: Early in the calendar year, each state initiates a planning process to prioritize safety goals and develop a performance plan and highway safety plan for the upcoming fiscal year. These plans establish performance goals and objectives based on safety problems identified in each state and include activities the state funds with NHTSA’s traffic safety grants that will help the state reach its goals. For example, Nevada’s fiscal year 2008 plan identified impaired driving as the most common cause of fatal crashes and established a performance goal of reducing the number of fatalities to 5.75 per 100,000 people by 2008 (which would be down from 6.31 in 2005). To achieve this goal, Nevada plans to devote about 28 percent of its federal funding to impaired-driving activities such as conducting highly publicized driving under the influence (DUI) enforcement activities and training prosecutors on DUI cases. To help states identify safety priorities and develop performance goals for the year, NHTSA provides each state with an analysis of state-level traffic safety data, such as fatality rates, seat belt use, and alcohol-related fatalities. The agency also shares information on countermeasures that address safety problems that are specific to the state. For example, NHTSA regional offices may recommend that states develop a statewide media plan regarding seat belt use, with specific emphasis on heightening media exposure in high-risk counties. NHTSA also informs states of countermeasures that other states have used through activities like regional conferences. NHTSA regional staff meet regularly with state highway safety staff during the planning process, providing guidance on national priorities and other technical assistance to help the state develop a plan that addresses key safety issues. States must submit performance plans and highway safety plans to NHTSA by September 1, and NHTSA responds with a formal letter documenting the agency’s analysis of the plans and providing additional feedback. Review of state progress toward state safety goals: In December of each year, states must submit to NHTSA regional offices a report on the previous fiscal year’s program activities, including the programs funded by the safety grants. The regional offices assess these reports to determine state progress toward achieving the goals and performance measures identified in the state’s performance plan. These assessments allow NHTSA to track state performance in improving safety outcomes and to provide feedback to states on strengths and weaknesses in the programs. Assisting states not making progress toward safety goals: For states that are not making adequate progress in NHTSA’s priority areas of reducing alcohol-related deaths or improving safety belt use, NHTSA conducts special management reviews. To identify states for impaired driving reviews, NHTSA analyzes national fatality data and compares state performance with national levels; to identify states for occupant protection reviews, NHTSA analyzes state-reported seat belt use rates from observational surveys, and compares those rates to the VMT- weighted average seat belt use of all 50 states, the District of Columbia, and Puerto Rico. States with performance consistently below the national average are candidates for a special management review. These reviews, which are currently only conducted for impaired driving and occupant protection, focus on state management of a specific program area—such as those funded by the safety grants—to identify barriers to progress and make formal recommendations on strategies that could help the state improve safety outcomes. Special management reviews involve an on-site review lasting several days and include interviews with state staff, reviews of program files, a formalized report developed by NHTSA, and potentially creating a “performance enhancement plan,” which establishes target dates for the completion of recommendations made in the review. As of September 2007, NHTSA conducted 29 special management reviews for fiscal years 2005 through 2007. Oversight processes to track state use of funding: Throughout the year, NHTSA monitors state grant management activities, including whether states are expending funds in a timely fashion and directing funds appropriately to the programs identified in the state’s highway safety plan as having potential to address the state’s safety goals. The primary tool NHTSA uses to monitor state spending is its Grants Tracking System (GTS), a Web-based application that allows NHTSA to track, approve, and release to states the grant funds available for highway safety programs. At the beginning of each fiscal year, states enter accounting information into GTS reflecting the total amount of funding the state may obligate that year and how much the state indicated in its highway safety plan it would spend on different programmatic areas. Then during the year, states enter data into GTS to indicate how grant funding is being spent, and NHTSA regional offices use GTS to monitor state spending by periodically reviewing financial data to ensure states are making progress in expending funds, meeting financial requirements, and using funding for the programs indicated in their plans. NHTSA monitors whether states deviate from their planned spending—for example, by shifting funding from an impaired driving program to a program addressing a safety goal that state data suggest is a lower priority—and can request additional justification if states make changes. GTS contains the requirements of each grant—for instance, the percentage of funding that states must match—and allows states to submit vouchers for reimbursement. Because the data entered into GTS include the total spent on a given grant, NHTSA regional offices also conduct on-site reviews of vouchers to examine documentation supporting these expenditures and determine whether the expenses were allowable, reasonable, related to the project, and expended within the grant year. Assessing state grant management: In addition to monitoring throughout the year, NHTSA conducts more comprehensive on-site “management reviews” for each state once every 3 years to assess state operational practices to ensure efficient administration and effective planning, programming, implementation, and evaluation of the state’s highway safety programs. This oversight approach has evolved over time in response to agency processes to identify areas for improvement, congressional and state concerns, and our prior recommendations. Prior to fiscal year 1998, NHTSA regulations required each state to submit a highway safety plan to NHTSA for approval. This plan detailed—down to the project level—the activities the state proposed to implement with federal grant funding. However, in response to congressional and state concerns that NHTSA’s project-by-project approval process was too restrictive, NHTSA adopted a performance-based approach in 1998. Specifically, states were required to submit a performance plan that identified key highway safety problems in the state and established goals and performance measures to address these problems. States still submitted the highway safety plan, but NHTSA no longer approved or disapproved individual projects contained in the plan unless those projects were not allowed under statutory limitations imposed on the various grants; rather, the agency determined whether the state submitted the plan in compliance with regulations. In a 2003 report, we identified problems with NHTSA’s performance-based approach, noting inconsistencies among the NHTSA regional offices in the level of guidance to states on how to use safety grant funding. We recommended that NHTSA provide more specific guidance to regional offices on when to conduct reviews of state safety programs and how to measure progress toward meeting safety goals. In response to our recommendations and new direction from Congress, NHTSA clarified and revised its guidance to states on how best to craft highway safety plans, as well as the process for how the agency would conduct regular reviews of state use of grant funding. Both the Department of Transportation Inspector General (DOT IG) and GAO are required by law to evaluate NHTSA’s oversight process since the agency initiated changes in 2005. The DOT IG’s review will verify whether NHTSA developed and followed policies and procedures for conducting management reviews and special management reviews in accordance with our April 2003 recommendation, determine the extent to which NHTSA’s reviews addressed states’ safety performance measures, and identify best practices for improving NHTSA’s oversight activities. To accomplish their review, officials from the IG told us that they accompanied NHTSA regional office staff on several reviews to observe the process and analyze management review reports, as well as underlying workpapers that supported the review’s findings and conclusions. Our review will describe how NHTSA oversees state safety grants and the role of the State and Community Highway Safety Grant Program in state highway safety programs. Our report will be released in July 2008, and as of mid-February 2008, the DOT IG planned to release its report in the spring of 2008. Using grant funds, states are planning and implementing safety improvement activities to address the key traffic safety issues in their states, but the structure of the grant programs has created eligibility difficulties for states, as well as management difficulties for governors’ highway safety offices that administer the grants. State safety officials agreed that the grants are helping them address key safety issues but have found it challenging to meet eligibility requirements for some grant programs, particularly those requiring states to pass laws. In addition, state safety officials have faced management challenges involving multiple grant applications and deadlines. State safety officials also noted concerns about the timing of the grant awards and limitations on the amount of flexibility they have to use the funds. NHTSA officials agree with state officials’ concerns, but noted that they cannot take action to address them because these issues are required by law. Any changes to the grant structure would have to be undertaken by Congress. For each grant, state highway safety officials are planning and implementing activities to address key traffic safety problems in their states. These activities generally fall within five categories of activities— education and training, media and public information, enforcement, data and technology, and infrastructure improvements (see table 2). Safety officials we spoke with in seven selected states agree that the safety incentive grant programs are assisting states in implementing activities that address key safety issues. State and local chapters of associations such as Mothers Against Drunk Driving (MADD) and state sheriffs and chiefs of police also indicated that the issues the grants target are critical safety problems in their states. Furthermore, states are using their grants to address goals and performance measures established in state highway safety plans. For example, although New Jersey had a 90 percent safety belt usage rate in 2006, the state’s 2007 highway safety plan indicates the state plans to continue efforts to increase safety belt usage rates by 2 percentage points, from 90 percent in 2006 to 92 percent in 2007. The state budgeted $1.3 million for safety belt enforcement to help reach this goal. Similarly, although California has made significant progress in reducing the frequency of impaired driving and related injuries and fatalities, alcohol remains the number one primary collision factor in fatal crashes in California. Consequently, the state has allocated $3.6 million of the Impaired Driving grant for DUI enforcement, education, and public information activities to 22 law enforcement agencies to achieve its goal of decreasing the number of persons killed in alcohol-involved collisions by 2 percent by December 31, 2008. Also, given the rise in the rate of motorcycle fatalities, Illinois and several other states have set goals to reduce statewide motorcycle fatalities. Specifically, Illinois has allocated $104,000 toward developing a motorcycle awareness program to address its goal of reducing motorcycle fatalities from 9.8 percent of total fatalities in 2003 to 7 percent by January 2008. Additionally, as Massachusetts faces incomplete statewide data on injuries and fatalities, it has designated a portion of the Traffic Safety Information Systems grant for the development of an ambulance record information system in its 2007 highway safety plan. Because the structure of the Safety Belt Use grant program provides more flexibility in how the funds may be used compared with the other grants, activities that states are planning and conducting with the grant encompass all five key areas as well as other activities, such as conducting activities to reduce alcohol-impaired driving by teens or conducting statewide safety belt use surveys. While this is the only grant program that can be used to fund infrastructure improvements, states have allocated a large portion of these grants toward programs designed to influence safety. For example, Illinois allocated 75 percent ($22.3 million) of the $29.7 million grant for behavioral projects and 25 percent ($7.4 million) for infrastructure projects. California used all of the $19.4 million it received in fiscal year 2006 for behavioral activities. Education and training, media and public information, and enforcement activities funded by this grant program are generally linked to the national high-visibility “Click It or Ticket” campaign. Some states, such as South Carolina, have allocated most of the grant to data and technology activities to improve traffic safety information systems. These activities may include developing systems to enable electronic crash reporting and data transmission or identifying top locations for aggressive driving. Specifically, South Carolina received $10.6 million and is using the majority of this amount— about $8 million—for traffic records improvement. Finally, states have also used these grants for other activities, such as infrastructure safety measures, including upgrading and improving locations where pedestrian and motor vehicle collisions occur or videotaping and assessing county roadway systems. Because the goal of the Child Safety and Booster Seat grant program is to encourage greater use of child safety and booster seats, the activities that states are planning and conducting are limited to education and training, media and public information, and other activities such as car seat purchases. States are planning to use these grants to increase training for child safety seat technicians and instructors and to support additional safety checkpoints and clinics where parents learn how to properly install safety seats. Safety advocacy groups have indicated that this funding is important because the population of first-time parents is constantly changing. For example, New Jersey funded the Safe C.A.R.G.O Program at the Monmouth County Sheriff’s Office to help educate parents, child-care services, and other caregivers on how to properly install and use child safety seats. States are planning to fund media and public information to promote awareness of child passenger safety. Vermont, for instance, plans to educate parents through public awareness campaigns. The Monmouth County Sheriff’s Office also partnered with local social service agencies to determine which families are eligible for seats and provided vouchers to be redeemed at Safe C.A.R.G.O. inspection locations. States have used the Impaired Driving grant to focus their efforts on education, training, media, and public information. For example, Georgia and other states have used the funding to promote Students Against Destructive Decisions, which educates youth on alcohol and drug education and prevention. States are planning to fund training for local law enforcement officers and promote outreach programs to prosecutors and judges. Our analysis also indicated that states have used the grants to fund awareness programs on the impact of impaired driving and target messages to teen drivers, as well as to fund a variety of high-visibility enforcement activities, such as the national “Over the Limit, Under Arrest” campaign. For example, 15 California police departments participated or plan to participate in the traditional holiday enforcement campaigns where the departments set up DUI checkpoints and conduct billboard campaigns urging residents to drive sober during this period. The Illinois State Police uses grant funds to target establishments that sell alcoholic beverages to minors. States have also used the grant to purchase new equipment, such as breath alcohol testing vans, diesel-powered light towers, and variable message boards for law enforcement to use at checkpoints. In addition to these activities, states have used this grant for activities to enhance courts’ and prosecutors’ ability to prosecute impaired driving, as well as to encourage legislation imposing stronger sanctions and penalties for impaired driving. For example, Arkansas plans to use the funding to provide a traffic safety resource prosecutor who will serve as a resource to prosecutors in the state on impaired driving and other traffic cases. In addition, NHTSA officials indicated that states have used this grant to promote legislation to impose sanctions and penalties for impaired driving. The Motorcyclist Safety grant program’s focus on training for motorcyclists and increasing other motorists’ awareness of motorcyclists has provided states with a new source of funding for education and training, media and public information, and activities such as purchasing motorcycles for training courses. States are using the funds to train more rider education coaches and add more classes because capacity for these classes has been limited and classes have filled up quickly in many states. For example, in Alaska, the Juneau American Bikers Aimed Toward Education (ABATE) plans to expand the number of Motorcycle Safety Foundation-certified rider education coaches in southeast Alaska. Nevada plans to allocate $20,000 to provide training to maintain current instructor levels and add new instructors in fiscal year 2007. Additionally, states have funded or plan to fund campaigns to increase other motorists’ awareness of motorcyclists and promote motorcycle training courses. For example, Utah funded motorcycle safety and “Share the Road” public awareness campaigns, while Kentucky and several other states plan or have begun to publicize and promote training courses. In addition, states purchased or plan to purchase additional motorcycles for training courses. The Traffic Safety Information Systems grant program allows states to focus activities specifically on data and technology activities. This program assists states in identifying and addressing the quality of information concerning crashes, drivers’ licenses, injury surveillance, roadways, enforcement and adjudication, and vehicles. For example, Arkansas plans to use part of its funding to improve the timeliness and uniformity of its crash data, while Indiana plans to develop an electronic citation system to allow the electronic issuance, collection, and court processing of citation data. Similarly, Michigan plans to use the grant to create a statewide emergency medical system and trauma database, while New Mexico plans to use the funds to enhance its DWI records. In addition, NHTSA officials indicated that California plans to upgrade its crash system to include GPS coordinates. While state safety officials we contacted during site visits agreed that the grants are helping states address key safety issues, they also noted difficulties in passing laws to meet eligibility requirements for some grants, as well as managing grant applications, deadlines, and timing. While about half of the states have passed primary safety belt laws to qualify for the Safety Belt Use grant, other states have not enacted primary safety belt laws principally because of state legislatures’ or governors’ opposition to mandating safety belt use laws that could infringe on individuals’ personal freedom. For example, according to Montana traffic safety officials and others involved in traffic safety, the state Senate passed a primary safety belt bill that the Governor supported, but the bill failed in the state House by a narrow margin. In Vermont, a traffic safety official told us that the state House passed a primary safety belt bill, but the Governor did not support it, and it failed in the state Senate. An analysis of state legislative activity indicates states have faced challenges in enacting primary safety belt laws. Of the 29 states that introduced primary safety belt bills from 2003 through 2007, 8 passed the bills (see table 3); 16 states had primary safety belt laws in effect before 2003. Similarly, about one-fourth of the 50 states have passed laws to qualify for Child Safety and Booster Seat grants. As our analysis of state legislative activity indicates, from 2003, when states became aware of some of the provisions that would likely be included in the reauthorization legislation, through 2007, 24 states considered requiring children to use booster seats up to age 8. In total, 5 states passed new laws or modified existing laws so that they qualified for the grant in fiscal year 2006. An additional 8 states passed laws to qualify for the grant in fiscal year 2007 (see table 4). Although many states already had booster seat laws in effect, these laws vary in terms of the age, height, and weight requirements. For example, some states require children to use booster seats up to ages 5, 6, or 7 but not age 8. Other states use height or weight requirements. According to traffic safety officials and safety advocates, the variations occurred because, over time, NHTSA has changed the criteria concerning age, height, and weight for determining who should be in booster seats. According to NHTSA officials, these changes were based on evolving research and understanding on how to best protect children. However, once a state has a booster seat law in effect—even one not meeting the grant’s requirement—state safety officials, safety advocates, and others familiar with traffic safety legislation are often reluctant to attempt to upgrade it because of fears that the current law could be revisited and the safety provisions could be lost, according to an NCSL official and representatives of organizations involved in child passenger safety. Although all 50 states met the eligibility requirements for the first and second years of the Impaired Driving grant program, NHTSA regional and governors’ highway safety officials have expressed concerns about states’ ability to meet the eligibility criteria in the future. According to these officials, the criteria will be more difficult to meet because states may need to pass laws that impose stronger sanctions against those convicted of drunken driving, such as installing ignition interlock devices and suspending or revoking drivers’ licenses. As table 5 illustrates, states have not attempted to qualify for the Impaired Driving grant using three of the grant’s programmatic criteria, which may require legislation; the criteria are high-risk driver program, administrative license suspension or revocation system, and self-sustaining impaired-driving prevention programs. Federal and state officials have expressed concern that, as with efforts to pass laws related to the Safety Belt Use and Child Safety and Booster Seat grants, some states will find it more difficult than others to pass laws imposing stronger sanctions. In addition, officials are concerned that states that cannot meet the eligibility criteria would have difficulty continuing their efforts to reduce alcohol-impaired driving. The multiple grant applications and deadlines presented challenges for the seven states we visited. As illustrated in figure 5, the five applications are due within a period of 1-1/2 months between June 15 and August 1. According to state highway safety officials, each application requires extensive amounts of staff time and resources. The application process requires states to submit to NHTSA the application, a certification of compliance, and additional information, depending on the grant. Although the application process is similar for each grant, having to complete it several times within a few months presents administrative challenges for states. Several states expressed concerns about the demands the application process placed on their staff, even though states with larger safety programs have more staff and resources available to manage grant applications than states with smaller safety programs. For example, Illinois officials said they have had difficulty meeting NHTSA’s administrative requirements to apply for the grants because the state Division of Traffic Safety has had staff shortages resulting from cutbacks and are considering an electronic grants process to enable them to meet the requirements. Similarly, Montana officials indicated they have a small staff and are burdened with grant paperwork. According to NHTSA and state highway safety officials, smaller state highway safety offices—such as that in Vermont—struggled to manage their grants. According to NHTSA, the application requirements reflect SAFETEA-LU’s requirements to award the grants in the same year in which the state’s legislative status and fatality-rate performance are measured. State officials also expressed concern over the delay in receiving grant awards and the associated impact on states’ ability to expend the funds to address traffic safety concerns. States received most of the grants in September—at the end of the federal fiscal year—in both fiscal years 2006 and 2007, from 1 to 3 months after the states submitted the grant applications. According to data provided by NHTSA, states received $250 million late in fiscal year 2006, and expended 14.1 percent of the combined funds in fiscal year 2006. Most of the fiscal year 2006 funds carried forward to fiscal year 2007. In fiscal year 2007, states received $274 million; the states carried forward $261 million and expended $157 million, or 29.4 percent of the combined funds (see table 6). One NHTSA official attributed the low rate of expenditures for the fiscal year 2006 grants to the fact that SAFETEA-LU was not passed until August 2005, giving state officials little time to plan how to spend the funds they received beginning in fiscal year 2006. State officials indicated that, once they received the awards, they needed time to assess applications from subgrantees and award the grants. Additionally, the amount of funding a state receives in a given year for a grant depends in part on the number of other states that also receive the grant, which can make it difficult to predict the amount a state is going to receive or can result in a state receiving more than state officials had anticipated in a given year. For example, according to NHTSA officials, Michigan received more Safety Belt Use grant funding in fiscal year 2006 than anticipated. Similarly, the states that received the Child Safety and Booster Seat grant in fiscal year 2006 received much more than they expected because so few received the grant. As a result, they did not have plans for using the increased level of funding. NHTSA officials indicated they were aware that the states received the funds late in the fiscal year and that this affected states’ ability to use the funds, but noted that states will be able to carry the funds over to the next fiscal year. One NHTSA regional official indicated that, while state officials do not know the exact amount of the grants prior to receiving the money, after the first year of the grants, state officials should have a general idea of how much they will receive and therefore can begin planning how to use the funds. State officials also noted they would have preferred having more flexibility in how they use the grants. For example, the Motorcyclist Safety grant program allows the federal funding to be used only for training and to increase other motorists’ awareness of motorcyclists. Officials in Montana noted they would like to use the funds to expand training sites to provide more training opportunities. However, the grant does not allow the funds to be used for this purpose. Additionally, officials in New Jersey noted that the Child Safety and Booster Seat grant they received in fiscal year 2006 was much larger than they expected. They would have preferred to have used the additional funding for other areas, such as the state’s traffic safety information systems. NHTSA officials agree with state officials’ concerns but noted they cannot take action to address these issues because the challenges stem from the grant requirements built into the law itself. The officials indicated they are currently in the preliminary stages of developing a proposal to reauthorize funding for these programs but have not developed it sufficiently to provide specific information on what they are considering. NHTSA plans to develop more comprehensive performance measures to evaluate the results of these grant programs, but state performance is not tied to the receipt of the grants; the absence of such performance accountability mechanisms, as well as issues described above that states face in using grants, raise implications for reauthorization. The changes made by SAFETEA-LU to the safety incentive grant programs have not been in place long enough to assess their impact on fatalities, injuries, and crashes, although the grants have resulted in states enacting laws and implementing activities to meet grant criteria. NHTSA officials indicated they plan to rely on performance measures to determine the results of the grant programs. However, these performance measures are not comprehensive. Additionally, three of the five grants do not include performance accountability mechanisms that would tie states’ eligibility for the grants or the amount states receive to performance. These issues as well as those that states have faced in applying for and using the grants raise implications for Congress to consider in reauthorizing the surface transportation program. Although changes made by SAFETEA-LU to the safety incentive grant programs have not been in place long enough to allow for an evaluation of results, such as improvements in fatality rates, NHTSA officials indicated they plan to rely on performance measures to help determine the results of the programs and other safety initiatives (see figure 6). NHTSA has awarded grants for 2 fiscal years, and states are currently implementing activities using the grant funds. According to a NHTSA official, the grant awards made in late fiscal year 2006 could not have been expected to impact performance measures such as fatality rates until calendar year 2007 at the earliest. Moreover, because of the time required to start up projects that states fund with the grants, the impact of the activities might not be realized until 2008. Data on fatality rates for 2007 will be available in the fall of 2008. The measures currently used by NHTSA include DOT- wide measures that reflect the overall goal of reducing traffic fatalities, such as the passenger vehicle occupant fatality rate and motorcycle fatality rate. In addition to DOT’s performance measures, NHTSA also has intermediate outcome measures to track behaviors that influence traffic safety, such as safety belt use, improperly licensed motorcycle riders in fatal crashes, and safety restraint use for children under age 8. However, the performance measures are not comprehensive for the traffic safety areas covered by the grant programs. NHTSA’s intermediate outcome measures do not include measures to track behaviors that influence alcohol-related fatalities. Such measures could include the numbers of impaired-driving citations issued, arrests, and convictions. Currently, states vary in the extent to which the data to track such measures are collected. NHTSA recognizes the need to improve these measures and, in partnership with GHSA, has hired a contractor to develop a common set of performance measures that federal, state, and local governments could use. The objective is to establish intermediate outcome measures for a broad range of traffic safety areas, including safety belts and child passenger safety, impaired driving, and motorcycles, that can reliably track progress toward reducing the safety problems in each area. NHTSA plans to use these measures to track progress at the national level and encourage states to consider these measures in the highway safety planning process. The contractor’s analysis of performance measures is expected to be completed in August 2008. While more comprehensive performance measures could improve NHTSA’s ability to measure national and individual state progress toward traffic safety goals, the receipt of grant funds by states are generally not directly linked to the results—or performance—of states’ activities to achieve those goals. For example, one way for states to qualify for Safety Belt Use grants, and the only way for noncomplying states to qualify for the Child Safety and Booster Seat grants, has been to pass laws. The Motorcyclist Safety grants require states to conduct certain activities to initially receive the grants. A majority of the states also qualify for the Impaired Driving grant by conducting activities. To receive grants in the future, states must continue these activities and meet additional criteria, which could include conducting additional activities or, in the case of the Impaired Driving grant, pass additional legislation. States do not need to achieve a particular performance level—or reduction in fatality rate— in order to continue receiving the grants. In contrast, the fifth grant program—Traffic Safety Information Systems—includes a performance accountability mechanism requiring that states demonstrate measurable improvement in their systems, as well as continue to meet other eligibility criteria, in order to receive the grant in subsequent years. Additionally, the Impaired Driving grant includes a performance accountability mechanism because states that achieve an alcohol-related fatality rate of 0.5 or less per 100 million VMT receive a grant based on the amount of the State and Community Highway Safety grant. In fiscal years 2006 and 2007, 19 states qualified for grants in this manner. However, these states are not eligible to receive funds based on the programmatic criteria specified for this grant and receive these grants after the states with the highest alcohol-related fatality rates receive their Impaired Driving grants. We have previously reported that such performance accountability mechanisms could improve the design and implementation of federal grants. Regarding transportation-related grants, we have raised concerns about insufficient links between state performance and receipt of grants. In addition, transportation and other experts on a panel convened by GAO’s Comptroller General in May 2007 stated that the nation’s overall transportation goals—including safety goals—need to be linked to performance measures that measure what the respective programs and polices are designed to accomplish. Congress changed the traffic safety incentive grant programs in SAFETEA- LU to encourage states to undertake activities tied to safety areas that Congress had identified as being high priority. These programs set forth specific criteria for states to qualify for grants, as well as specific requirements for how states can use the funding provided. However, this structure does not always allow states flexibility to direct funding toward safety priorities as identified in highway safety plans. Furthermore, while the current grant structure ensures that states are directing their efforts toward congressionally established priorities, it does not necessarily ensure that the grants are in proportion to the extent of the traffic safety problems. As noted previously, the current structure also presents some eligibility and management issues for states and does not include performance accountability mechanisms for all grants. NHTSA has not developed its proposal for the next highway safety authorization bill when the current authorization, expires in 2009, but NHTSA officials noted that DOT’s 2003 reauthorization proposal included features that would address these issues. For example, NHTSA’s proposal included performance-based grants within the State and Community Highway Safety grant program that could address the issues that states have faced in applying for and using the current safety incentive grants. The basis for awarding the grants under this proposal would have been data driven; as states improved the performance of their highway safety programs by decreasing fatalities, their grant amounts would have increased. States would not have faced the issues concerning multiple applications because they would not have had to apply for the grants. Additionally, according to the proposal, NHTSA would have awarded the grants by December 31 of each year—much earlier than states currently receive them. Finally, states would have had flexibility in using the funds because the grants would have been part of the State and Community Highway Safety grant program; this program permits flexibility in how grant funds are used. In December 2007, the National Surface Transportation Policy and Revenue Study Commission proposed a National Safe Mobility Program with a performance-based structure similar to NHTSA’s proposal. In the proposed program, DOT would define safety performance metrics to be used by all federal, state, and local agencies, as well as work with states to define specific goals for individual states. States would then develop strategies for reaching these goals, including safety projects within the highway safety plan. The commission recommended that a national plan for safety be developed that leads to transportation investments undertaken purely for safety purposes and that the federal share of the funding for qualifying safety projects be 90 percent of the project cost. According to the report, qualifying safety projects could include projects designed to change safety behaviors, such as safety belt use and impaired driving, as well as projects to improve the safety of surface transportation infrastructure. DOT’s 2003 reauthorization proposal also included performance accountability mechanisms for traffic safety incentive grants that would link performance or progress toward achieving safety goals with grant awards. Specifically, NHTSA proposed that the amount of an individual state’s grant award depend on the state’s performance related to various crash fatality rates, safety belt use, and safety belt laws. Since NHTSA’s current oversight approach is based on state performance in achieving safety goals states establish in their highway safety plans, establishing such performance accountability mechanisms for traffic safety incentive grants would be more consistent with NHTSA’s oversight approach and is an option that could be considered when Congress reauthorizes the surface transportation program. (See app. VII for a more complete description of NHTSA’s 2003 proposal.) As Congress considers the reauthorization of funding for these programs, it will face the decision of retaining the current grant structure or moving toward a more performance-based, data-driven grant structure. While a performance-based, data-driven grant structure is more consistent with NHTSA’s current oversight approach, such a structure would depend on the quality of state data systems because states would need to be able to report on fatalities, crashes, and other traffic safety characteristics in a timely and accurate manner. In 2004, we reported that states vary considerably in the extent to which their traffic safety data systems meet recommended criteria used by NHTSA to assess the quality of crash information. We reviewed systems in nine states and found, for example, that some states entered crash information into their systems in a matter of weeks, while others took a year or more. While some systems were better than others, all had opportunities for improvement. Furthermore, a performance-based grant structure would also depend on a robust oversight approach for NHTSA to ensure that states are establishing appropriate traffic safety goals and making sufficient progress toward those goals. As previously noted, we are currently examining NHTSA’s oversight approach and expect to issue a report on it in July 2008. We provided a draft of this report to DOT for its review and comment. DOT officials, including the Deputy Administrator of NHTSA, generally agreed with the findings of the report and offered technical corrections that we incorporated, as appropriate. We are sending copies of this report to interested congressional committees and the Secretary of Transportation. We will also make copies available to others upon request. In addition, the report will be available at no cost on GAO’s Web site at www.gao.gov. If you or your staffs have any questions about this report, please contact me at (202) 512-6570 or siggerudk@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix VIII. We were asked to evaluate five safety incentive grant programs included in the Safe, Accountable, Flexible, Efficient Transportation Equity Act (SAFETEA-LU), specifically the Safety Belt Use, Child Safety and Child Booster Seat Use, Alcohol-Impaired Driving Countermeasures, Motorcyclist Safety, and State Traffic Safety Data Systems Improvement grant programs. This report addresses (1) the National Highway Traffic Safety Administration’s (NHTSA) status in awarding and overseeing states’ use of these grant programs, (2) activities states have conducted using the grants and issues they have faced in applying for and implementing them, and (3) how NHTSA plans to evaluate the results of the grant programs and implications for reauthorizing the programs. To determine the progress NHTSA has made in awarding and overseeing the use of the grants, we reviewed past GAO reports on NHTSA’s oversight and analyzed applicable laws, regulations, and other guidance for awarding and overseeing grants. We interviewed NHTSA officials to determine guidance it provided to states on grant applications, guidelines in overseeing the grants, and grant award data. We also interviewed Federal Highway Administration (FHWA) and Federal Motor Carrier Safety Administration officials within the U.S. Department of Transportation (DOT) and representatives from professional groups, including the Governors Highway Safety Association, National Conference of State Legislatures (NCSL), National Safety Council, American Association of State Highway Transportation Officials, and Advocates for Auto and Highway Safety on the grants process and measures NHTSA has taken to improve the process. In addition, we interviewed national safety and law enforcement associations, including Mothers Against Drunk Driving (MADD), the American Automobile Association (AAA), Safe Kids, International Association of Chiefs of Police, and National Sheriffs’ Association. We also requested and received data regarding grant awards from NHTSA. To obtain state information on NHTSA’s progress in awarding and overseeing the five grants, we conducted case studies in seven states—California, Illinois, Missouri, Montana, New Jersey, South Carolina, and Vermont—to discuss NHTSA’s progress in awarding and overseeing the grants with state officials. We selected the states based on a combination of characteristics, including fatality rates, funding, and geographic distribution. For fatality rates and funding, we selected states to include those with high- and low-fatality rates and high and low use of highway safety grants. We considered highway safety grants during our state selection process to mirror the multitude of safety factors that may be reflected in fatality rates. We also ensured our selection covered at least three states for each incentive grant to provide sufficient coverage of each grant. However, the results of our case study analyses are not generalizable to all 50 states because the states selected are not necessarily representative of the grants each state is eligible to receive and activities other states would implement. To identify activities states are funding with the grants and issues they have faced in applying for and implementing the grants, we interviewed officials and analyzed data for the seven states we visited. In each state we visited, we interviewed governors’ highway safety representatives or their designees and reviewed documentation describing those states’ traffic safety programs and the activities states have funded with the grants. In addition, we reviewed the 2007 state highway safety plans and 2006 annual reports for all 50 states to identify activities states are funding with the grants and categorized them into five key areas—education and training, media and public information, enforcement, data and technology, and infrastructure improvements. For each state we visited, we also interviewed the applicable NHTSA regional and FHWA division officials. In general, we also interviewed representatives of state and local chapters of national safety associations, such as MADD, AAA, and Safe Kids. Additionally, we interviewed representatives from law enforcement, such as state associations of chiefs of police and sheriffs’ associations about the activities states have funded and challenges they faced. We also examined NCSL’s legislative tracking database of primary safety belt and booster seat legislative activity from 2001 to 2007 to assess states’ progress in meeting legislative requirements for the Safety Belt Use and Child Safety and Booster Seat grant programs. In addition, we requested and received data regarding states’ expenditures of the grant funds. To determine how NHTSA plans to evaluate the results of the grant programs and the implications for reauthorizing them, we interviewed NHTSA officials and reviewed a variety of documents related to performance measures, and reauthorization. Specifically, we reviewed DOT’s and NHTSA’s performance measures, as well as GAO reports on performance measures and federal grant management. We also reviewed and considered the National Surface Transportation Policy and Revenue Study Commission report and NHTSA’s 2002 reauthorization proposal related to safety incentive grant programs. Total includes the District of Columbia, Puerto Rico and the territories of American Samoa, the Commonwealth of the Northern Marianas Islands, Guam, and the Virgin Islands. FY 2007 high- fatality grant A state can qualify in one of three ways for an Impaired Driving grant: as a “low-fatality” rate state, a programmatic state, or as a “high-fatality” rate state (being among the 10 states with the highest fatality rates). It is possible for a high-fatality state to be awarded both a programmatic grant and a high-fatality rate grant in the same year. FY 2007 total includes Puerto Rico, and FY 2006 total includes the District of Columbia and Puerto Rico. FY 2006 grants (44 states) FY 2006 grants (44 states) FY 2006 grants (44 states) FY 2006 grants (44 states) In 2003, DOT submitted to Congress its reauthorization proposal for the surface transportation program, which included NHTSA’s grant programs. The proposal, as it related to traffic safety, included grant programs that had both similarities to and differences from the structural changes that Congress included in SAFETEA-LU. Several of the programs included in that proposal are relevant to the five safety incentive grant programs included in our review. Specifically, the relevant programs included in the proposal are as follows: A program to award performance grants based on states’ performance in three categories: motor vehicle crash fatalities; alcohol-related crash fatalities; and motorcycle, bicycle, and pedestrian crash fatalities. The proposal would have required NHTSA to determine measures using fatality data and vehicle miles traveled (VMT) and establish goals based on those measures. States would have received awards by December 31 of each year in amounts based on the extent to which they achieved the goals or made progress toward them, creating a performance accountability mechanism that tied the grant amount to states’ performance. The proposal included these grants within the State and Community Highway Safety grant program. This program was not adopted. A program to award safety belt performance grants designed to encourage states to adopt and enforce primary safety belt laws and increase the rate of safety belt use. Each state that had already enacted and was enforcing a primary enforcement safety belt law would have received a one-time grant. Additionally, states passing primary enforcement safety belt laws after the reauthorization bill became law, or demonstrating a safety belt use rate of 90 percent or more, would have subsequently been eligible for a one-time grant. This proposal was similar to the safety belt use grant program requirement, as adopted by SAFETEA-LU, with the difference being that, as adopted, the law provides different grant amounts for states depending on when their primary safety belt laws became effective. In addition, under DOT’s proposal, each state would have received an annual award based on the state’s prior year’s safety belt use rate, regardless of whether the states passed a primary safety belt law. This feature would have created a performance accountability mechanism by tying a portion of the grant amount to states’ performance. States would have received these grants by December 31 of each year. These grant proposals were also within the State and Community Highway Safety grant program. The proposal for annual grants based on safety belt use rates was not adopted. An amended Traffic Safety Information Systems grant program that was similar to the Traffic Safety Information Systems grant program adopted by SAFETEA-LU. A discretionary impaired driving program designed to reduce impaired driving in states with a high number of alcohol-related fatalities and a high rate of alcohol-related fatalities relative to VMT and population. This program would have required NHTSA to develop, demonstrate, and evaluate comprehensive state programs to reduce impaired driving in states with high alcohol-related fatality rates. NHTSA would have chosen participating states based on an application specifically developed for this grant. This proposal was not adopted. In addition to the contact named above, other key contributors to this report were Sara Vermillion (Assistant Director), Elizabeth Curda, Colin Fallon, Lynn Filla-Clark, Chir Jen Huang, Bert Japikse, Tom James, Terry Richardson, Beverly Ross, and Don Watson. | In 2005, the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (SAFETEA-LU) included authorizations of nearly $2.2 billion for safety incentive grant programs to assist states in their efforts to reduce traffic fatalities. Administered by the Department of Transportation's (DOT) National Highway Traffic Safety Administration (NHTSA), five of these programs provide incentive grants to states to implement legislation governing the use of safety belts and child safety seats, and promote activities to reduce alcohol-impaired driving, improve motorcycle training and awareness, and improve traffic safety information systems. To help Congress prepare for the reauthorization of the surface transportation programs in 2009, this report provides information on (1) NHTSA's status in awarding and overseeing states' use of these five grants programs, (2) activities states have conducted using the grants and issues they have faced in applying for and implementing the grants, and (3) how NHTSA plans to evaluate the results of the grant programs and implications for reauthorizing the programs. To conduct this work, GAO interviewed DOT and state officials, analyzed safety reports from 50 states, and analyzed grant data from DOT and 7 selected states. DOT officials generally agreed with the findings of the report and offered technical corrections that were incorporated, as appropriate. In fiscal years 2006 and 2007, NHTSA awarded about $575 million to states for the five safety incentive grant programs; NHTSA uses several oversight processes to determine the extent to which states are meeting safety-related performance goals and to monitor how the states spend grant funds. The number of states receiving the grants generally remained constant or increased from fiscal year 2006 to 2007, although the extent to which states qualified for the different grant programs varied. For example, in 2006, 22 states received the Safety Belt Use grant and 5 states received the Child Safety and Child Booster Seat Use grant because not all states were able to pass the laws that the grant programs required, while the majority of states received the other grants. To oversee states' use of grants, NHTSA uses a performance-based approach to assess state progress toward meeting safety goals and complements this assessment with oversight processes that monitor whether states are accomplishing the tasks that will allow the state to achieve its goals. This approach allows NHTSA to be involved throughout the lifecycle of state grants. In response to a mandate to evaluate the effectiveness of NHTSA's oversight process, GAO plans to issue a report in July 2008. States are planning and implementing safety improvement activities using grant funds, but the structure of the grant programs has created eligibility and management difficulties for states. The activities generally fall within five categories--education and training, media and public information, enforcement, data and technology, and infrastructure improvements. Safety officials GAO spoke with in seven selected states agree that the safety incentive grant programs are assisting states in implementing activities that address key safety issues and meeting goals and performance measures established in state highway safety plans. However, state safety officials also noted difficulties in passing laws to meet eligibility requirements for some grant programs, as well as managing grant applications, deadlines, and timing. For example, not all states have passed a primary safety belt law, which allows law enforcement officers to stop a driver for not wearing a safety belt and is required to qualify for a Safety Belt Use grant. The selected states have also had difficulty managing the multiple grant applications, which are all due within a 3-month period. NHTSA officials acknowledge state officials' concerns but noted they cannot address the concerns because the difficulties stem from the grant requirements established in SAFETEA-LU. NHTSA plans to develop additional performance measures to evaluate the results of these grant programs, but state performance is generally not tied to receipt of the grants; the absence of such performance accountability mechanisms as well as issues described above that states face in using grants raise implications for reauthorization. Congress will be faced with deciding whether the grant programs could be designed differently to allow states more flexibility in using grant funds and to focus more specifically on performance accountability, as some have advocated. However, these changes would require improved safety data and a robust accountability system. |
DOJ prosecutors cited the Principles of Federal Prosecution of Business Organizations as a major factor in their decision on entering into a DPA or an NPA, and considered other factors, such as the Federal Sentencing Guidelines, in determining the terms of these agreements. Prosecutors also said that they generally negotiated these decisions with companies. However, in making these decisions, prosecutors differed in their willingness to use DPAs or NPAs. In addition, prosecutors’ different perspectives on the definitions of DPAs and NPAs led to inconsistencies in how they labeled the agreements. DOJ plans to determine the need to require consistency in the use of the labels DPA and NPA. Prosecutors in all 13 DOJ offices we included in our review consistently said that they based their decision on whether to enter into a DPA or NPA rather than prosecute the company or decline to do so on the Principles of Federal Prosecution of Business Organizations. First issued in 1999, these principles are DOJ’s guidance to federal prosecutors on investigating, charging, and negotiating a plea or other agreement with respect to corporate crimes. The principles instruct prosecutors to consider nine factors when determining how to treat a corporation suspected of criminal misconduct and provide a number of actions prosecutors may take, including declining to prosecute, entering into a DPA or NPA, or criminally prosecuting, the corporation. The principles also include guidance on when the nine factors most appropriately apply. The factors, and examples of the manner in which they influence prosecutors’ choice of action, are shown in figure 1 below. While the prosecutors with whom we spoke said that many of these factors may have influenced their decision on entering into a DPA or NPA in each case, they most frequently cited the company’s cooperation with the investigation, the collateral consequences of a criminal prosecution, and any remedial measures the company had taken or planned to take as most important in their decision on entering into a DPA or NPA. For instance, one prosecutor told us that the company’s cooperation is an important factor in cases involving violations of the Foreign Corrupt Practices Act because obtaining the evidence from foreign countries in these types of cases is a cumbersome and lengthy process that could take up to 10 years. However, with the company’s cooperation, which may entail assisting DOJ in tracing bribe payments through multiple overseas accounts, DOJ may be able to obtain the evidence it needs in a matter of weeks. With regard to collateral consequences, some DOJ prosecutors explained, for example, that the potential harm that prosecution and conviction of health care companies can have on innocent third parties may be a key factor in their decision on entering into a DPA or NPA with these kinds of companies. Federal law provides for health care companies convicted of certain crimes to be debarred from—or no longer eligible to participate in—federal health care programs. Prosecutors in one office said that they chose to enter into DPAs and an NPA simultaneously with five orthopedic device companies that provided kickbacks to physicians because, combined, these companies comprised the vast majority of the market for hip and knee replacements; therefore, conviction and debarment of these companies would have severely limited doctor and patient access to replacement hips and knees. In terms of remedial measures, prosecutors cited enhancements companies made to their compliance programs, the termination of employees responsible for the wrongdoing, and the company’s willingness to make payments to the victims of the crime as influential in their decision on entering into a DPA or NPA, rather than prosecute. Our preliminary analysis suggests that officials from many of the DOJ offices we met with have made efforts to be transparent about the basis for their decisions on entering into DPAs or NPAs. For example, 10 of the 13 DOJ offices issued press releases explaining how they applied the Principles of Federal Prosecution of Business Organizations when deciding whether to enter into these agreements. According to an official in the Criminal Division’s Fraud section, its policy is to issue press releases upon entering into DPAs and NPAs with companies related to the Foreign Corrupt Practices Act, which helps to increase transparency. As part of our ongoing review, we will determine the extent to which DOJ offices have additional policies—including supervisory review and documentation of the reasons for their decisions to enter into a DPA or NPA—that promote transparency and accountability regarding these agreements. DOJ’s reliance on the Principles of Federal Prosecution of Business Organizations was also apparent to many of the companies involved in the DPAs and NPAs. Ten of the 17 company officials with whom we spoke as of June 5, 2009, said that they were aware that DOJ based its decision on whether to enter into a DPA or NPA on the factors articulated in the Principles of Federal Prosecution of Business Organizations. Moreover, officials from 6 of these 10 companies reported making presentations to DOJ based on the nine factors in order to influence prosecutors’ decisions on using agreements in their cases, although companies generally reported that the prosecutors made the ultimate decision about whether to enter into a DPA or an NPA. DOJ prosecutors also made decisions about which of these agreements— DPA versus NPA—the office would enter into. A commonly accepted distinction between these two types of agreements is that a DPA involves the filing of a charging document with the court, while, for an NPA, charges are not filed with the court. Officials from 12 of the 17 companies with whom we spoke preferred an NPA, largely because they viewed NPAs as more advantageous from a public relations perspective for the company. Some of these officials explained that, because a charge is not filed in court in association with an NPA, companies are able to report that they were not charged or prosecuted in the case; a DPA, on the other hand, involves the filing of charges in court, which can result in greater negative publicity for the company. In choosing between a DPA and an NPA, prosecutors most frequently reported considering the same factors they did when deciding whether to enter into an agreement at all—namely, cooperation, collateral consequences, and the companies’ remedial actions. For example, prosecutors at 6 of the 13 DOJ offices said that they considered the company’s cooperation in their investigation when deciding between a DPA and an NPA. Prosecutors from one DOJ office said that once the company learned it was the target of the office’s investigation, its lawyers immediately called the office seeking to cooperate and continued to cooperate extensively throughout the office’s ensuing 3-year investigation, remaining in daily contact with the office and assisting in its investigation. As a result, the DOJ office chose to enter into an NPA rather than a DPA with the company. Not all of the 13 DOJ offices we included in our review reported entering into both types of agreements. For instance, 3 of the 13 DOJ offices we included in our study, including one section of the Criminal Division, exclusively entered into DPAs with companies. A prosecutor from one of these offices said that he did not consider entering into NPAs in any of its cases because he viewed NPAs as too lenient on the company. We will continue to assess this issue as part of our ongoing work. Officials from 11 of the 17 companies with whom we spoke said that the decision between a DPA and an NPA was exclusively made by DOJ, and officials from 4 of these companies reported that DOJ’s reasons for choosing between a DPA and an NPA were not made clear. On the other hand, officials from 4 other companies said that the decision was a result of negotiations between DOJ and the company. Companies’ opinions varied on whether guidelines for choosing between a DPA and an NPA would be beneficial. Officials from 5 of the 17 companies we interviewed said that such guidelines would assist the companies in negotiating between a DPA and an NPA with DOJ, whereas officials from three companies believed that guidelines would make DOJ’s decision between a DPA and an NPA more transparent to the company. Officials from 6 companies cited reasons why guidelines may not be useful, such as concerns that such guidelines may not address the unique circumstances of each case, would not be binding on DOJ prosecutors, and were not necessary because DOJ’s rationale for choosing a DPA versus an NPA was made clear to the company. Prosecutors at 4 of the 13 offices we spoke with stated that these guidelines would not be beneficial because they need the flexibility to choose between a DPA and an NPA based on the unique circumstances of each case. In addition, prosecutors differ in whether they called their agreements DPAs and NPAs. For example, prosecutors from 2 of the 13 offices with whom we spoke told us that they are reluctant to file agreements in court because of their understanding that some judges do not want the case to be open on their dockets for the length of the deferral period. While prosecutors from one of these offices called the agreements it did not file in court NPAs, the other office still labeled its agreements DPAs because it viewed DPAs as agreements in which the company admits guilt, regardless of whether charges are filed in court. Recognizing the inconsistent use of the labels DPA and NPA, in March 2008, then Acting Deputy Attorney General Craig Morford issued a memorandum—also known as the “Morford Memo”—which stated that a DPA is typically predicated on the filing of both a formal charging document and the agreement with the appropriate court, while an NPA is an agreement maintained by the parties, rather than being filed with the court. The Morford Memo also states that clear and consistent use of these terms will help DOJ more effectively identify and share best practices and track the use of DPAs and NPAs. However, based on our analysis of the agreements entered into after DOJ issued this guidance, not all the agreements were labeled in accordance with the definitions provided. Of the 27 agreements entered into after DOJ issued this guidance, 20 were labeled as DPAs or NPAs in the agreement or the press release announcing the agreement. Of these 20 agreements, 3 were not labeled in accordance with the definitions in the guidance. The remaining 7 agreements were labeled as agreement, case disposition agreement, or pretrial diversion agreement. One reason for the differences in the manner in which agreements are labeled is that not all prosecutors believe that the use of the definitions of DPAs and NPAs in the guidance is mandatory. For instance, a prosecutor at one office told us that the office believed that the definitions were provided only for the purposes of reading the Morford Memo and not as guidance for labeling DPAs and NPAs going forward, while a prosecutor at another office believed that the Morford Memo was intended as mandatory guidance on the use of the definitions of DPAs and NPAs in the future. According to the Office of the Deputy Attorney General, DOJ intends for the definitions in the Morford Memo to be mandatory and followed consistently by prosecutors for the purpose of internal reporting and tracking of these agreements. However, DOJ does not intend for the definitions to inhibit prosecutors’ ability to externally label these agreements in accordance with the unique circumstances of a particular case or the practices and preferences of a particular DOJ office, company, or judge. For instance, the company may prefer that an agreement be labeled as “agreement” rather than “deferred prosecution agreement” because companies believe this label is less severe. Thus, the prosecutor may negotiate with the company over the external label. Regardless of the external label on the agreement, DOJ intends for prosecutors to track the agreement either as a DPA or NPA in accordance with Morford Memo definitions. In addition, DOJ is aware that there may be agreements that share some of the elements of DPAs and NPAs but may not readily fit the Morford Memo definitions—for instance, the Office of the Deputy Attorney General explained that in one case the company had already been indicted on some of the criminal charges associated with the agreement prior to the agreement being reached, but had not been indicted on other charges associated with the agreement, and therefore it was not clear whether the agreement fit the definition of a DPA—in which charges are filed—or an NPA—in which charges are not filed. Taking into account external circumstances such as these, DOJ plans to determine whether there is a need to take additional steps to require the use of the definitions, to ensure consistency in the use of labels across offices. Prosecutors in 11 of the 13 offices and officials from 14 of the 17 companies with whom we spoke reported that they negotiated at least one of the terms in their DPAs and NPAs, including monetary payments to victims or the government, the duration of the agreement, or compliance program requirements, as well as additional terms, such as monetary donations to foundations or educational institutions. Furthermore, according to prosecutors in all 13 DOJ offices, they considered other factors, such as guidance provided in the Federal Sentencing Guidelines or the terms included in other DPAs or NPAs as examples, when determining the terms of their agreements. Monetary payments: Of the 57 DPAs and NPAs we reviewed, 45 required monetary payments—which may include restitution to victims of the crime, forfeiture of the proceeds of the crime, and monetary penalties imposed by DOJ—ranging from $30,000 to $615 million. While the remaining 12 agreements did not require such payments, in 3 agreements the companies were required to make payments to organizations or individuals that were not directly affected by the crime; for 7 agreements the company had already agreed to make payments as part of a separate agreement with another agency or DOJ division, such as the Securities and Exchange Commission or DOJ’s Civil Division; and for 1 agreement, two of the company’s subsidiaries had already agreed to make monetary payments as part of a plea agreement and a DPA. In the remaining agreement, the company was not required to make a payment and did not enter into a civil settlement in order to obtain release from its civil liability in the case. In setting the payment amounts in DPAs and NPAs, prosecutors reported that they considered the following: (1) the section of the Federal Sentencing Guidelines on determining fines for business organizations, which includes consideration of the seriousness of the offense, culpability of the organization, and the company’s cooperation, among other factors; (2) monetary gains to the company or losses to its victims as a result of its crime; and (3) the company’s ability to pay. Prosecutors in 6 of the 12 offices with whom we spoke whose DPAs and NPAs included monetary payments reported that they negotiated the monetary payments with the other party. While representatives of 7 of the 13 companies we interviewed that were required to make monetary payments told us that they were able to negotiate the monetary payment with DOJ, representatives of 4 companies told us that they were not able to negotiate the payment. Representatives from 2 of these companies did not express concern over the lack of negotiation—1 said that DOJ’s reasons for setting the payment were made clear to the company, while the other said that the company had no reason to question the payment figure DOJ set. One of these companies reported that DOJ did not provide its rationale for the monetary payment, and the remaining company did not provide opinions about the process by which the payment was set. Duration: The durations of DPAs and NPAs have ranged from 3 months to 5 years. Prosecutors at 9 of the 13 DOJ offices with whom we spoke based the duration of the agreement on the amount of time they believed was necessary for the company to correct the problems underlying the criminal conduct. For instance, one prosecutor said that the company was replacing its old computer billing system, which had overbilled a federal agency, resulting in the criminal conduct underlying the DPA. The prosecutor set the duration at 27 months in order to allow the company to install the new billing system and ensure it was functioning appropriately, and not continuing to overbill the agency. Prosecutors at 5 of the 13 offices we visited also reported that they negotiated with companies over the duration of the agreement. On the other hand, companies that had agreements with 5 other DOJ offices told us that they did not negotiate the duration, although none of these companies expressed concern over the duration of the agreement. For instance, an official from one of these companies said that the company would have preferred a shorter duration, but was satisfied with the duration DOJ set. Prosecutors in 3 DOJ offices also told us that they considered the duration of other DPAs or NPAs as examples when setting the duration of their agreements. Compliance program requirements: Forty-five of the 57 DPAs and NPAs we reviewed included requirements that the company improve or enhance its compliance program, while 12 did not include this type of requirement. According to prosecutors in 6 of the 13 DOJ offices we met with, they required companies to enhance or implement a compliance program in order to reform the company, prevent further misconduct, or help establish and publicize a compliance program standard for the industry. In deciding not to include compliance requirements, prosecutors reported that they considered whether the company that committed the wrongdoing could engage in such criminal conduct again. For instance, one prosecutor said that a compliance program was not required as part of an agreement because the company’s violations occurred during its participation in the United Nation’s Oil-for-Food Program, which was no longer in existence when the agreement was signed. In addition, prosecutors were aware that 2 of the companies involved in DPAs or NPAs that did not include compliance program requirements had entered into agreements with other regulatory agencies that did include such requirements. When developing compliance requirements in DPAs and NPAs, prosecutors most commonly (8 of 13 offices) worked with regulatory agencies with relevant jurisdiction over the companies—such as Immigration and Customs Enforcement for issues related to the hiring of illegal immigrants, the Environmental Protection Agency for environmental crimes, or the Securities and Exchange Commission for issues involving accounting and financial fraud—to develop the compliance requirements included in the agreement. Several prosecutors and company officials also reported that they negotiated over the compliance requirements in the DPA or NPA. For instance, one company official said that DOJ initially developed the compliance program requirements, but when the company raised concerns about the practicality and effectiveness of the requirements, DOJ worked with the company to revise them. In the end, the official felt that the company’s enhanced program was a best practice in the industry. Extraordinary restitution: DPAs and NPAs have also included additional terms, such as payments or services to organizations or individuals not directly affected by the crime; these payments are sometimes referred to as extraordinary restitution. Of the 57 DPAs and NPAs we reviewed, 4 included such terms. Prosecutors and companies with whom we spoke about these provisions generally reported that the provisions were determined through negotiations between the two parties. In addition, these prosecutors were supportive of including extraordinary restitution provisions in DPAs and NPAs because, for example, they believe such terms can help improve the availability of services in the community and prevent similar misconduct from occurring in the future, not just within the company, but in a larger context. For instance, 1 DPA required the organization to provide uncompensated medical care to the state’s residents, while an NPA required the company to provide funding for a not-for-profit organization to support projects designed to improve the quality and affordability of health care services in the state. Another DPA required a company that had not complied with water treatment regulations to provide an endowment of $1 million to the U.S. Coast Guard Academy for the purposes of enhancing the study of maritime environmental enforcement, with an emphasis on compliance, enforcement, and ethics issues. In May 2008, DOJ issued guidance prohibiting the use of terms requiring payments to charitable, educational, community, or other organizations or individuals that are not the victims of the criminal activity or are not providing services to redress the harm caused by the criminal conduct because the use of such terms could create actual or perceived conflicts of interest or other ethical issues. Based on our preliminary analysis, none of the 25 DPAs and NPAs that were entered into since this guidance was issued required companies to make payments or perform services for individuals or organizations that were not directly harmed by the crime. While most company officials stated that they had input into, or were able to negotiate over, whether to enter into a DPA or NPA and the terms of the agreements, officials from nine of these companies reported that DOJ had greater power in the negotiations than the company because, for instance, if the negotiations were not successful, DOJ could have proceeded with prosecution. However, prosecutors at 4 of the 13 offices with whom we spoke noted that if companies had concerns about the terms of their DPAs or NPAs, they could express them to their office, or appeal them to a higher level within DOJ. Representatives from six companies expressed reluctance to appeal any concerns they had with the terms of the agreement. Officials from two of these companies explained that appealing to a higher level in DOJ could negatively affect their interactions with the prosecutors involved in the case. On the other hand, officials from four companies told us that they would have been comfortable appealing the terms, if needed. As part of our ongoing review, we will continue to assess the extent of the companies’ role in setting the terms of the agreements and obtain DOJ’s perspective on this issue. In 26 of the 57 DPAs or NPAs we have reviewed to date, prosecutors required that the company hire, at its own expense, an independent monitor to assist the company in establishing a compliance program, review the effectiveness of a company’s internal control measures, and otherwise meet the terms of the agreements. In the remaining cases, DOJ coordinated with the relevant regulatory agency already monitoring or overseeing the company, or used other means, such as requiring companies to certify their compliance, to ensure the terms were met. When deciding whether a monitor was needed to help oversee the development or operations of a company’s compliance program, DOJ considered factors such as the availability of DOJ resources for this oversight, the level of expertise among DOJ prosecutors to monitor compliance in more technical or complex areas, and existing regulatory oversight. Of the 13 DOJ offices we met with, 10 utilized monitors. Prosecutors in four of these nine offices cited as a reason for requiring an independent monitor the limited time and resources their offices had to oversee a company’s compliance program, make appropriate recommendations, and reform the company’s compliance behavior, whereas monitors often have an entire staff available to them to perform these activities. Prosecutors in five of the nine DOJ offices we met with that had utilized monitors, cited as a reason for requiring an independent monitor the limited expertise the office had in overseeing company compliance in a particular area of misconduct. For example, prosecutors in one office stated that part of the company’s wrongdoing dealt with commodities trading, and while they did not have this background, the monitor selected by the office had commodities trading experts on his staff. Other prosecutors cited the need for technical expertise regarding misconduct in a particular geographic region to oversee company compliance effectively—resources and skills which DOJ prosecutors did not have—as the reason to require that a company hire a monitor. In 22 of the 26 agreements requiring an independent monitor, the monitor was required to file written reports with DOJ prosecutors. The frequency of reporting to DOJ prosecutors varied by agreement, with 13 monitors required to report every 3 or 4 months; 2 monitors required to file semiannual reports; 5 monitors required to file annual reports or an initial report with annual or semiannual follow-up reports; 1 monitor required to report within 120 days of entering into the agreement; and 1 monitor required to report no later than 45 days and 90 days after the commencement of the agreement, on or before 90 prior to termination of the agreement and at such other times as designated by DOJ. For two of the three agreements overseen by an independent monitor where the agreement did not specifically require written reports, the prosecutors we spoke with said that they typically met frequently with the monitor themselves to discuss the company’s progress towards fulfilling the agreements. We have not assessed whether the monitors’ reports were filed in a timely fashion or covered the elements required by the agreements, but plan to obtain information on monitor reporting as part of our ongoing review. In one instance, the district court judge also received the reports filed with federal prosecutors by the independent monitor because, in that district, the office typically involved the court in the selection of the independent monitor, and the judge had issued an order requiring quarterly reporting to the court. We are in the process of collecting information from federal judges who have been involved with DPAs to determine the extent to which judges received monitor reports, or assessments of these reports provided by DOJ, in their oversight of DPAs. In 18 of the 57 agreements we reviewed to date, there was a requirement for companies to make improvements to existing ethics and compliance programs or implement new programs, but there was no requirement for companies to hire an independent monitor to review the effectiveness of these programs or the companies’ compliance with the terms of the agreement. In 4 cases, the company had signed a civil or administrative agreement with a federal regulatory agency as part of a settlement related to the underlying criminal misconduct, which required the company to hire an independent consultant, review organization or compliance officer. In such cases, DOJ officials said that they depended on the reports of these regulatory monitors or the regulatory agency to assure themselves of companies’ compliance in part to avoid unnecessary duplication. In the other 14 cases, where the company had not signed a settlement agreement with a regulatory agency requiring an independent monitor, DOJ officials stated that they used other methods to determine companies’ compliance with the agreement. In 9 of the 14 cases, they stated that they depended on the regulatory agency to inform them if, in the course of its regulatory oversight, the agency discovered the company was violating any of the provisions of the agreement. For example, in 2 DPAs we reviewed where financial institutions failed to maintain effective anti-money laundering programs, DOJ prosecutors said that they communicated frequently with financial regulators, reviewed reports submitted to the regulators, and spoke to the regulators before the agreements were completed. In the remaining 5 cases, the prosecutors said they reviewed documents submitted by the company or depended on the companies to self-certify that they had complied with the provisions of the agreement. For the remaining 12 of the 57 agreements that did not require companies to improve or expand ethics and compliance programs, DOJ offices conducted oversight through various mechanisms, including: Assuring that monetary penalties or restitution payments were paid in full. For example, an accounting firm agreed to make restitution payments to a fund established to repay wronged investors, and to pay an administrator to administer the fund. The administrator provided reports to the office on the names of victims that received payments from the fund, and the amount received. Assuring that the company cooperated with DOJ in continuing investigations, including responding to information requests from federal prosecutors. For example, an energy trading company in a DPA with one office agreed to continue to cooperate with federal prosecutors by providing information relevant to ongoing investigations in the natural gas industry. Requiring the company to certify that it had followed certain requirements in the agreement. For example, one pharmaceutical company was required to certify that it had not filled prescriptions for off-label uses of one of its drugs. In that case, the prosecutors stated that it would be easy to examine the company’s prescription records at the end of the agreement to determine if the certification was accurate, and if not, the company would additionally be liable for falsely certifying compliance. We reviewed 26 agreements that required the company to hire a monitor. Although DOJ was not a party to the contracts between companies and monitors, DOJ generally took the lead in approving the monitors. Specifically, according to officials in the 10 DOJ offices we contacted that entered into DPAs and NPAs that required monitors, DOJ had the final say in selecting the monitor for all but one of these agreements. However, according to these officials, the monitors were not selected by any one individual; rather, the decision was made among several DOJ officials and, in most instances, companies were able to provide input to DOJ on who the monitor should be, although the extent of company involvement varied. For 12 of the agreements we reviewed, DOJ prosecutors said that the companies proposed a single monitor or a list of several monitors from which DOJ could choose. In all of these cases, DOJ officials said they were able to select an appropriate monitor for the DPA or NPA based on the company’s suggestions. For three of the agreements we reviewed, DOJ prosecutors said that they and the company developed separate lists of monitor candidates, shared their lists with one another, and worked together to choose the monitor. For seven of these agreements, DOJ prosecutors said that they chose the monitor. For five of the seven agreements, according to DOJ officials, the prosecutors selected the monitors and later provided the companies with the opportunity to meet with the selected individual. According to the prosecutors, they gave companies the option to object to DOJ’s monitor selection, but none of the companies did so. However, our preliminary work suggests that at least one company reported that they did not have this opportunity. For one of these agreements, DOJ officials said that they sought the companies’ input on monitor qualifications before making their selection. For another of these agreements, it was unclear whether the company had any discussion with DOJ regarding monitor qualifications before DOJ selected the monitor. For the agreements we reviewed where DOJ officials identified monitor candidates, the selection processes employed across these offices were similar. DOJ officials generally stated that in these instances, they identified monitor candidates based on their personal knowledge of individuals whose reputations suggest they would be effective monitors, or through recommendations from colleagues or professional associates who were familiar with requirements of a monitorship. After identifying several candidates, the prosecutors established a committee, which generally consisted of individuals such as the prosecutors involved in the case, the DOJ office section chief, and sometimes the Chief Assistant U.S. Attorney or a Deputy U.S. Attorney. The committees were responsible for evaluating the candidates and selecting a monitor. Prosecutors said they evaluated candidates based on whether they had any conflicts of interest with the company and their qualifications and expertise in a particular area. Officials from the five companies we interviewed who identified monitor candidates for DOJ approval used a similar process as DOJ. For example, officials from one company reached out to their associates who they believed could help them identify individuals who would be effective monitors. Company officials said that they were looking for a monitor with experience working with DOJ and knowledge of the specific area of law that the company violated. From these suggestions, the company developed a list of candidates to interview, and based on the results of the interviews, generated a shorter list of candidates from which DOJ would choose the monitor. In selecting the monitors, DOJ sometimes sought input from federal regulatory agencies. According to prosecutors in DOJ’s Criminal Division, it is not uncommon for the division to collaborate with agencies such as the Securities and Exchange Commission to select a monitor to serve under agreements both agencies have reached with a company, particularly if the agreements contain similar requirements for the company. The prosecutors said having two different monitors could be cost-prohibitive and result in duplication of effort. Courts were rarely involved in monitor selection. Of the 26 agreements we reviewed that had monitor requirements, 2 required court approval of the selected monitor. One of the 13 DOJ offices included in our review has a formal monitor selection policy. According to the prosecutors in this office, court involvement in monitor selection limits the possibility of favoritism in monitor selection by the office. The policy requires prosecutors to compile a list of potential monitor candidates and submit the list to the court, where a district judge would then appoint a monitor from this list. We plan to solicit input on court involvement from the judiciary as a part of our ongoing review. When we asked DOJ officials, company representatives, and monitors about other methods to prevent the appearance of favoritism in monitor selection, such as developing a national list of prescreened monitors from which DOJ would make its selection, they identified both advantages and disadvantages. Some of the advantages identified were (1) assurance that the monitors have been prescreened and are considered qualified by the government, (2) increased consistency in the monitor selection process, and (3) the ability to expedite the monitor selection process. The disadvantages they cited were (1) not all of the monitors on the list would have the specific expertise required for certain cases, such as commodities trading expertise; (2) based on their own experiences searching for monitors, it is likely that many of the monitors on a prescreened list will have conflicts of interest with the companies—such as the monitor having previously provided services for the company in an unrelated matter; (3) use of the list would limit company input in monitor selection; and (4) use of the list may actually increase the likelihood of favoritism because DOJ officials could populate the list with their associates, and could exclude other qualified monitor candidates. As a part of our ongoing work, we will continue to identify other models that aim to reduce favoritism in monitor selection. For example, one company official with whom we spoke cited the International Association of Independent Private Sector Inspectors General (IAIPSIG) as a possible model for developing a national pool of monitors. Members of this association are individuals or private sector firms with legal, auditing, investigative, and management skills who are available to be employed by an organization to ensure compliance with relevant laws and regulations. According to IAIPSIG, members—who may be retained by the government to prevent fraud in contracting and by private firms conducting internal investigations—must also adhere to the principles and standards in IAIPSIG’s code of ethics which require, among other things, that its members remain independent of both the monitored entity and the entity to which it is reporting, and refrain from accepting or performing work involving an actual or potential conflict of interest. In March 2008, the Acting Deputy Attorney General issued the Morford Memo to help ensure that the monitor selection process is collaborative, results in the selection of a highly qualified monitor suitable for the assignment, avoids potential and actual conflicts of interest, and is carried out in a manner that instills public confidence. The guidance requires U.S. Attorneys Offices and other DOJ litigation divisions to establish ad hoc or standing committees, consisting of the office’s ethics advisor, criminal or section chief, and at least one other experienced prosecutor to consider the candidates for each monitorship. DOJ components are also reminded to follow federal conflict of interest guidelines and to check monitor candidates for potential conflict of interest relationships with the company. In addition, the names of all selected monitors must be submitted to the Office of the Deputy Attorney General for final approval. According to the Senior Counsel to the Deputy Attorney General, this approval is required in order to ensure public integrity in the monitor selection process. While the Morford Memo established policies and guidance for the selection of independent monitors, including that the Office of the Deputy Attorney General approve the monitor selection, the memo does not require documentation of the process used and the reasons for selecting a specific monitor. Standards for internal control in the federal government state that all transactions and significant events, which could include the selection of monitors, should be clearly documented and that the documentation be readily available for examination. In addition, our prior work suggests that documenting the reasons for selecting a particular monitor helps avoid the appearance of favoritism and verify that selection processes and practices were followed. Since the release of the Morford Memo, we have identified two DPAs and NPAs that DOJ entered into for which monitors have been selected. According to the Office of the Deputy Attorney General, which is responsible for approving monitor selections, the United States Attorneys Offices involved in these two cases submitted e-mails to predecessors in the Office of the Deputy Attorney General regarding their proposed monitor selections. DOJ provided us with a summary of the correspondence from the prosecutors seeking Deputy Attorney General approval. While the correspondence in one case included information describing how prosecutors adhered to the processes required by DOJ guidance, the correspondence in the other case did not. For instance, the correspondence did not describe the membership of the committee that considered the monitor candidate. In addition, because the approval of one of the monitors was relayed via telephone and no documentation was readily available at the Office of the Deputy Attorney General, DOJ officials had to reach out to the individuals who were involved in the telephone call to obtain information regarding the monitor’s approval. As this example demonstrates, without requiring documentation of the process used and the reasons for selecting a particular monitor, it may be difficult for DOJ to validate whether its monitors have been selected and approved across DOJ offices in a manner that is consistent with the Morford Memo, which established monitor selection principles intended to instill public confidence. In commenting on a draft of this report in June 2009, the Office of the Deputy Attorney General agreed that documenting the process used and reasons for monitor selection would be beneficial. However, because the office has not had to approve any monitor selections since the presidential transition in January 2009, the office did not believe it was in a position to determine exactly what internal procedures should be adopted to document the monitor selection process until it had reviewed more selection proposals. From January 2009 through May 2009, DOJ had four ongoing agreements that required the appointment of a monitor where, to date, the monitors have not yet been selected. We expect that when the Office of the Deputy Attorney General reviews the monitor proposals for these agreements, once they are submitted, the office will be in a better position to establish procedures for documenting monitor selection decisions. Of the 12 companies we have met with so far for which DOJ required a monitor, 6 told us that they did not have any concerns about the rate charged by the monitor, 3 expressed concern that the monitor’s rate was high, and the remaining 3 did not comment on the monitor’s rate. Officials from 6 of the 12 companies perceived that the monitors were either charging their customary rates or, in two additional cases, lower rates because the companies could not afford the customary rates. While the companies we met with generally did not express concern about the monitors’ rates, they reported concerns with other aspects of the monitorship that affected the overall compensation to the monitor. Specifically, 6 of the 12 companies raised concerns about the scope of the monitor’s responsibilities or the amount of work completed by the monitor; and four of the six companies reported that they did not feel they could adequately address their concerns by discussing them with the monitors. For instance, 1 company said that the monitor had a large number of staff assisting him on the engagement, and he and his staff attended more meetings than the company felt was necessary, some of which were unrelated to the monitor responsibilities delineated in the agreement. As a result, the company believes that the overall cost of the monitorship was higher than it needed to be. While the company reportedly tried to negotiate with the monitor over the scope of work and number of staff involved, the company stated that the monitor was generally unwilling to make changes. The company did not feel that there was a mechanism at DOJ whereby it could raise concerns regarding monitor costs because the costs were not delineated in the agreement. Instead, the costs were identified in an agreement between the company and the monitor and, therefore, DOJ was not responsible for overseeing the costs of the monitorship. Another company reported that its monitor did not complete the work required in the agreement in the first phase of the monitorship, which necessitated the monitor completing more work than the company anticipated in the final phase of the monitorship. This led to unexpectedly high costs in the final phase. The company official believed it was DOJ’s responsibility, not the company’s, to address this issue because the monitor had failed to complete the requirements DOJ had delineated in the agreement. As part of our ongoing review, we plan to obtain the perspectives of DOJ officials and monitors, in addition to companies, regarding the amount and scope of the monitors’ work and the most appropriate mechanisms companies can use to address any concerns they may have related to this issue. Two company officials reported that they had little leverage to negotiate fees, monitoring costs, or the monitor’s roles and responsibilities with the monitor because the monitor had the ability to find that the company was not in compliance with the DPA or NPA. Officials from three companies suggested that DOJ should play a larger role in helping companies address concerns with their monitors. For example, one company official said that DOJ may need to develop a mechanism for companies to raise issues regarding their monitors without fear of retribution, while another company official suggested that DOJ meet routinely with the company to allow for a conversation between the company and DOJ about the monitoring relationship. Two companies felt that having a sense of the potential overall costs at the beginning of the monitorship, such as developing a work plan and estimated costs, would be beneficial for companies. For instance, one of these officials said that this would help establish clear expectations for the monitor and minimize unanticipated costs. DOJ has taken some actions which may address these concerns. In 2 of the 26 DPAs or NPAs we discussed with DOJ that had monitoring requirements, the monitor was required to submit a work plan prior to the monitor’s first review of the company. Additionally, an official in the Criminal Division Fraud Section said that it is the section’s general practice to meet with the monitor to discuss the monitor’s work plan. The Morford Memo also instructs DOJ prosecutors to tailor the scope of the monitor’s duties to address the misconduct in each specific case, which the memo indicates may align the expense of the monitorship with the failure that led to the company’s misconduct covered by the agreement. However, we have not yet been able to evaluate how these actions may address companies’ concerns. We will continue to obtain information on the ways in which company concerns regarding the monitors’ responsibilities and workload can be addressed. We are conducting a survey of companies to solicit more comprehensive information on monitors’ fees, total compensation and roles and responsibilities, as well as the companies’ perceptions of the monitor costs in relation to the work performed. We will integrate these survey results into our final report. In addition, we are continuing to assess the potential need for additional guidance or other improvements in the use of DPAs and NPAs in our ongoing work. One of DOJ’s chief missions is to ensure the integrity of the nation’s business organizations and protect the public from corporate corruption. DOJ has increasingly employed the tools of DPAs and NPAs in order to carry out this mission, and has recognized the potential long-term benefits to the company and the public of assigning an independent monitor to oversee implementation of a DPA or NPA. On the other hand, DOJ has also acknowledged concerns about the cost to the company of hiring a monitor and perceived favoritism in the selection of monitors, and thus the resultant need to instill public confidence in the monitor selection process. DOJ has made efforts to allay these concerns by issuing guidance requiring prosecutors to create committees to consider monitor candidates; evaluate potential conflicts of interest the monitor may have with the government and the company; and obtain approval of selected candidates from the Office of the Deputy Attorney General. Nevertheless, more could be done to avoid the appearance of favoritism. Requiring that the process and reasons for selecting a specific monitor be documented would assist DOJ in validating that monitors were chosen in accordance with DOJ’s guidance that is intended to help assure the public that monitors were chosen based on their merits and through a collaborative process. We are continuing to assess the potential need for additional guidance or other improvements in the use of DPAs and NPAs in our ongoing work. To enhance DOJ’s ability to ensure that monitors are selected according to DOJ’s guidelines, we recommend that the Deputy Attorney General adopt internal procedures to document both the process used and reasons for monitor selection decisions. We requested comments on a draft of this statement from DOJ. DOJ did not provide official written comments to include in the statement. However, in an email sent to us on June 18, 2009, DOJ stated that the department agreed with our recommendation. DOJ also provided technical comments, which we incorporated into the statement, as appropriate. For questions about this statement, please contact Eileen R. Larence at (202) 512-8777 or larencee@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Individuals making key contributions to this statement include Kristy N. Brown, Jill Evancho, Tom Jessor, Danielle Pakdaman, and Janet Temko as well as Katherine Davis, Sarah Kaczmarek, Amanda Miller, Janay Sam, and Mandana Yousefi. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Recent cases of corporate fraud and mismanagement heighten the Department of Justice's (DOJ) need to appropriately punish and deter corporate crime. Recently, DOJ has made more use of deferred prosecution and non-prosecution agreements (DPAs and NPAs), in which prosecutors may require company reform, among other things, in exchange for deferring prosecution, and may also require companies to hire an independent monitor to oversee compliance. This testimony provides preliminary observations on (1) factors DOJ considers when deciding whether to enter into a DPA or NPA and setting the terms of the agreements, (2) methods DOJ uses to oversee companies' compliance, (3) processes by which monitors are selected, and (4) companies' perspectives regarding the costs and role of the monitor. It also includes the results of GAO's recently completed work on DOJ's efforts to document the monitor selection process (discussed in objective 3). GAO reviewed DOJ guidance and 57 of the 140 agreements negotiated from 1993 (when the first 2 were signed) through May 2009; and interviewed DOJ officials, officials from 17 companies, and 6 monitors. While not generalizable, these results provide insight into decisions about DPAs and NPAs. Prosecutors in all 13 DOJ offices with whom GAO spoke said that they based their decision on whether to enter into a DPA or NPA on DOJ's principles for prosecuting business organizations, particularly those related to the company's willingness to cooperate, collateral consequences to innocent parties, and remedial measures taken by the company. However, prosecutors differed in their willingness to use DPAs or NPAs. In addition, prosecutors' varying perceptions of what constitutes a DPA or NPA has led to inconsistencies in how the agreements are labeled. In March 2008, DOJ issued guidance defining DPAs and NPAs, but this guidance is not consistently followed, in part because not all DOJ offices view it as mandatory. DOJ plans to determine the need to take additional steps to require consistency in the use of the labels DPA and NPA. While DOJ and companies generally negotiated the terms of DPAs and NPAs--such as monetary payments and compliance requirements--DOJ also considered other factors in its decisions, such as monetary gains to the company as a result of the criminal misconduct. To ensure that companies were complying with the terms of the DPAs and NPAs, DOJ employed several oversight mechanisms, including the use of independent monitors, coordination with regulatory agencies, and other means. Of the 57 agreements GAO reviewed, 26 required the company to hire, at its own expense, an independent monitor. In the remaining agreements, DOJ relied, among other things, on reports from regulatory agencies or from monitors hired by companies under separate agreements with these agencies, and company certifications of compliance. For the DPAs and NPAs GAO reviewed, even though DOJ was not a party to the contracts between companies and monitors, DOJ typically selected the monitor, and its decisions were generally made collaboratively among DOJ and company officials. Monitor candidates were typically identified through DOJ or company officials' personal knowledge or recommendations from colleagues and associates. In March 2008, DOJ issued guidance stating that for monitor selection to be collaborative and merit-based, committees should consider the candidates and the selection must be approved by the Deputy Attorney General. However, because DOJ does not require documentation of the process used or the reasons for particular monitor selection decisions, it will be difficult for DOJ to validate whether its monitor selection guidance-which, in part, is intended to instill public confidence-is adhered to. Some company officials GAO spoke with reported that they had little leverage to address concerns about the amount and scope of the monitors' work and, therefore, would like DOJ to assist them. GAO in its ongoing work will assess this and other issues about the use and oversight of DPAs and NPAs. |
FDA’s mission is to protect public health by ensuring the safety, efficacy, and security of human and veterinary drugs, biologic products, medical devices, our nation’s food supply, cosmetics, and products that emit radiation. The agency is also responsible for advancing public health by helping to speed innovations that make medicines and foods more effective, safer, and more affordable and by helping the public get the accurate, science-based information it needs to use medicines and foods to improve health. FDA carries out its regulatory mission primarily through five main centers and its Office of Regulatory Affairs: Center for Biologics Evaluation and Research. Regulates and evaluates the safety and effectiveness of biological products, such as blood and blood products, vaccines and allergenic products, and protein-based drugs. Center for Devices and Radiological Health. Ensures that new medical devices are safe and effective before they are marketed and that radiation- emitting products, such as microwave ovens, TV sets, cell phones, and laser products meet radiation safety standards. Center for Drug Evaluation and Research. Promotes and protects the health of Americans by ensuring that all prescription and over-the-counter drugs are safe and effective. Center for Food Safety and Applied Nutrition. Ensures the safety of 80 percent of food consumed in the United States (it is responsible for everything except meat, poultry, and some egg products, which are regulated by the U.S. Department of Agriculture). Center for Veterinary Medicine. Helps to ensure that animal food products are safe; also evaluates the safety and effectiveness of drugs used to treat more than 100 million companion animals. Office of Regulatory Affairs. Works to ensure that FDA’s health standards are properly implemented and adhered to through inspections, lab analysis, and public outreach. The agency relies extensively on IT to fulfill its mission and to support related administrative needs. FDA has systems dedicated to supporting the following major mission activities: Reviewing and evaluating new product applications, such as for prescription drugs, medical devices, and food additives. These systems are intended to help FDA determine whether a product is safe before it enters the market. For example, the Document Archiving Retrieving and Regulatory Tracking System is intended to manage the drug and therapeutics review process. Overseeing manufacturing sites and production supply chains to ensure that products comply with regulatory requirements. For example, the Field Accomplishments and Compliance Tracking System supports inspections, investigations, and compliance activities. Monitoring the safety of products on the market by collecting and assessing adverse reactions to FDA-regulated products, such as illnesses due to food or negative reactions to drugs. For example, the Vaccine Adverse Event Reporting System accepts reports of adverse events that may be associated with U.S.-licensed vaccines from health care providers, manufacturers, and the public. In addition, the agency has systems performing administrative processes, such as payroll administration and personnel systems. All these systems are supported by an IT infrastructure that includes network components, critical servers, and multiple data centers. Appendix III provides additional details on the agency’s mission-critical systems and infrastructure. The information that FDA receives is growing in volume and complexity. According to FDA, from 2001 to 2006, the number of import shipments that the agency inspected for admission into the United States increased from about 7 million imports reviewed annually to about 18 million. During this period, the number of adverse event reports and generic drug applications more than doubled. Advances in science and the increase in imports are also factors affecting the complexity of information that FDA receives. The ability of the agency’s IT systems and infrastructure to accommodate this growth will be crucial to FDA’s ability to accomplish its mission effectively. FDA’s IT has been the subject of numerous reports and studies, both by the agency itself and by others (see app. IV for a list of major reports and studies related to limitations of the agency’s IT). These reports have noted limitations in a number of key areas, including data availability and quality, IT infrastructure, ability to use technology to improve regulatory effectiveness, and IT management. Data availability and quality. Issues with the quality and availability of FDA’s data have been raised in several studies. In 2007, the FDA Science Board issued FDA Science and Mission at Risk, a broad assessment of challenges facing the agency. This study found that information was not easily and immediately accessible throughout the agency (including critical clinical trial data that were available only in paper form), hampering FDA’s ability to regulate products. Data and information exchange was impeded because information resided in different systems that were not integrated. The Science Board also reported that FDA lacked sufficient standards for data exchanges, both within the agency and between the agency and external parties, reducing its capability to manage the complex data and information challenges associated with rapid innovation, such as new data types, data models, and analytic methods. In 2007, FDA commissioned Deloitte Consulting, LLP, to examine ways the agency could better meet increased demand for information and make decisions more quickly and easily. Deloitte noted that FDA’s former decentralized approach to IT, in which the centers developed their own systems, led to duplicative work efforts, tools, and information. Noting that the agency had begun moving toward a more enterprisewide approach, Deloitte recommended further steps, including establishing enterprisewide information standards and incorporating data exchange standards into its day-to-day processes and applications in order to achieve interoperability with external partners. Our previous work also has identified issues related to the availability and quality of the agency’s data. For example, our 1998 study of FDA’s foreign drug inspection program cited evaluations that essential data for foreign inspections were not readily available, and that FDA did not have a comprehensive, agencywide, automated system for managing foreign inspection of manufacturers. Further, in a series of products (most recently in September 2008) on FDA inspections of foreign establishments, we reported that the agency’s databases on these establishments contained incorrect information and that different databases had differing information. IT infrastructure. Issues raised regarding FDA’s infrastructure include aging and redundancy. According to the FDA Science Board’s 2007 report, the agency’s IT infrastructure was outdated and unstable, and it lacked sufficient controls to ensure continuity of operations or to provide effective disaster recovery services. For example, as many as 80 percent of the network servers were more than 5 years old and had exceeded their recommended service life. In addition, the report stated that outages were occurring in other systems as well; for example, e-mail problems occurred during an E. coli food contamination investigation. Further, critical network components did not reside in data centers that provided the necessary security, redundancy, and continuity of operations assurances. In addition, after assessing the agency’s legacy applications, FDA’s contractor, High Performance Technologies, Inc., issued a report in 2008 that identified many systems that were redundant and could be combined with each other, as well as systems that could be retired. Ability to use technology to improve regulatory effectiveness. According to the FDA Science Board report, advances in science and technology have been outpacing the capabilities of FDA’s IT infrastructure and systems. For example, although genetics and genome-wide association analyses are an increasingly important technique in drug reviews, the agency had minimal IT infrastructure to support genomics-focused efforts, which generate large data sets. To implement the real-time acquisition and sharing of genomics data would require the development of appropriate data storage, mining, analysis, and risk evaluation tools for FDA scientists. IT management. Issues with FDA’s IT management have been found in several areas, including human capital, enterprise architecture, governance, and information security. In assessing IT human capital, the Science Board stated that the agency did not have sufficient IT staff with skills in such areas as capital planning/investment control and enterprise architecture, that processes for recruitment and retention of IT staff were inadequate, and that the agency did not invest sufficiently in professional development. Deloitte’s study also commented on IT management, stating that FDA needed to develop both a common enterprise information management architecture and an IT architecture to facilitate both short-term operational gains such as improved information access, as well as long- term gains in strategic flexibility. In another study, the Breckenridge Institute examined the process being used to develop requirements for the agency’s adverse event reporting system and found that FDA’s management of requirements development did not follow proper IT methodology, such as documenting the reasons for changes to system requirements. Finally, in October 2008, an HHS inspector general report concluded that FDA had made progress implementing an infrastructure to support the security management program. However, the Inspector General also noted that the agency had not fully implemented a security program infrastructure and was not performing all the activities required to integrate security into applications. Driven in part by the various studies that the agency has performed or sponsored (as discussed previously), as well as legislative requirements, FDA has been transitioning to an enterprisewide approach to IT management. For example, in February 2006 the agency created the Bioinformatics Board to replace center-specific investment review boards, in order to better coordinate its IT investment decisions from an agencywide perspective. According to the agency’s Chief Information Officer (CIO), this broader perspective led to an increased emphasis on the need for FDA to treat its information as a strategic corporate asset and manage it accordingly. Among the steps taken to help achieve this goal were centralizing the IT organization and consolidating IT infrastructure. In May 2008, the agency transferred responsibility for managing IT from individual components (centers and the Office of Regulatory Affairs) to a new centralized Office of Information Management (OIM), headed by the CIO. The CIO reports to the agency’s Chief Operating Officer. As head of OIM, the CIO is responsible for managing IT, creating a foundation to enhance the interoperability of its systems, and managing more than 400 staff assigned to this office. OIM has five divisions to carry out its responsibilities: Division of Business Partnership and Support. Acts as liaison and provides management and technical consultation resources regarding IT to FDA offices, centers, and other stakeholders, including parties outside the agency. Systems Division. Manages design, development, implementation, and maintenance of agency software applications and systems, as well as their integration with other entities. Infrastructure Division. Manages design, development, implementation, and maintenance of the agency’s IT infrastructure. Division of CIO Support. Oversees internal IT management controls, such as its enterprise architecture, investment management, and human capital management. Division of Technology. Reviews and evaluates the appropriateness of new and emerging information technologies for potential benefits. As part of its centralization efforts, FDA is transferring IT staff and assets from its components to the new centralized organization, and it is consolidating its IT infrastructure. Under one initiative, Information and Computer Technology for the 21st Century (ICT21), the agency is, for example, consolidating its data into two new data centers, one to host its production and preproduction systems and information, and the other to host system testing, development, and scientific computing needs. FDA’s fiscal year 2009 budget totals about $2.67 billion and is derived both from the agency’s annual appropriations and user fees. The appropriated budget authority is about $2.05 billion or 77 percent of funding, and user fees account for about $613 million or 23 percent of funding. FDA collects user fees primarily from companies that produce certain human drug and biologic products, as authorized by the Prescription Drug User Fee Act of 1992 (PDUFA). FDA’s fiscal year 2009 IT budget is approximately $364 million, which is about 14 percent of the agency’s total budget. The IT budget includes funds of $308.4 million for projects and systems and $55.2 million for federal employee salaries and expenses. The funding for projects and systems is derived from annual appropriations of $246.1 million and user fees of $62.3 million. The funding for federal employee salaries and expenses is derived from annual appropriations of $44.4 million and user fees of $10.8 million. According to data provided by FDA officials, the portion of FDA’s fiscal year 2009 IT budget that funds IT projects and systems has increased from previous years. As shown in table 1, from fiscal year 2005 to fiscal year 2009, funding for projects and systems increased from $202.3 million in annual funding to $308.4 million. According to the agency’s CIO, during fiscal years 2008 and 2009, IT expenditures have focused on addressing limitations, such as updating the infrastructure, and on problems that could be immediately addressed, such as eliminating duplicative databases related to adverse event reporting. He added that in the future, FDA plans to focus on more long- term modernization projects for supporting the agency’s regulatory responsibilities. Key to an agency’s success in modernizing its IT systems, as our research and experience at federal agencies has shown, is institutionalizing a set of interrelated IT management capabilities, among which are strategic planning to describe an organization’s goals, strategies it will use to achieve desired results, and performance measures; developing and using an agencywide enterprise architecture, or modernization blueprint, to guide and constrain IT investments; establishing and following a portfolio-based approach to investment implementing information security management that ensures the integrity and availability of information; and building and sustaining an IT workforce with the necessary knowledge, skills, and abilities to execute this range of management functions. Figure 1 shows these capabilities, which are critical to enable organizations to manage IT effectively. The Congress and OMB have recognized the importance of these and other IT management controls. The Clinger-Cohen Act, for example, provides a framework for effective IT management that includes systems integration planning, human capital management, and investment management. In addition, the Paperwork Reduction Act requires that agencies have strategic plans for their information resource management, and the E- Government Act of 2002 contains provisions for improving the skills of the federal workforce in using IT to deliver government information and services. Further, OMB has issued guidance on integrated IT modernization planning and effective IT human capital and investment management. Establishing IT management capabilities involves carrying out specific practices. For example, human capital management requires assessing present and future agency skills needs and making a plan to fill gaps. We have developed methods of evaluating agencies’ progress on these management capabilities, such as our IT Investment Management (ITIM) framework, Enterprise Architecture Management Maturity Framework, and framework for strategic human capital management. These frameworks list specific practices that an agency should use. We have observed that without these types of capabilities, organizations increase the risk that system modernization projects will (1) experience cost, schedule, and performance shortfalls and (2) lead to systems that are redundant and overlap. They also risk not achieving such aims as increased interoperability and effective information sharing. As a result, technology may not effectively and efficiently support agency mission performance and help realize strategic mission outcomes and goals. FDA is pursuing numerous initiatives to modernize its IT systems and infrastructure, including at least 16 enterprisewide initiatives. However, it does not yet have a comprehensive IT strategic plan, with well-defined goals, strategies, milestones, and measures, to guide these efforts. According to the Chief Operating Officer, the agency must resolve many near-term planning activities and strategic investment decisions before it can complete long-term plans. Without a strategic plan to sequence and synchronize these initiatives based on a comprehensive picture of its strategic IT goals, the agency increases the risk that its modernization efforts will not be effective. Of FDA’s numerous modernization initiatives, some began as a result of federal law and guidance (such as initiatives associated with PDUFA), and others in response to urgent mission requirements, including those pointed out in the various analyses of FDA’s IT systems and infrastructure previously described. Table 2 lists 16 major modernization projects with an enterprisewide focus that are under way or planned. As the table shows, many of these projects are still in the early stages of the life cycle (that is, planning and requirements development). In addition to these system and infrastructure development projects, FDA is taking actions to develop and enhance its IT management capabilities. That is, the agency is taking actions such as beginning to develop its enterprise architecture, gathering information on needed IT skills, and seeking contract support to improve application security and to analyze skills gaps. (FDA’s IT management capabilities are further discussed later in this report.) However, even as it undertakes these various initiatives and activities, FDA does not yet have the necessary planning in place to guide its efforts. Although agency officials identified two high-level planning documents that address different aspects of the agency’s IT environment, FDA lacks a comprehensive IT strategic plan, which is a foundation for effective modernization and is required by federal guidance. As we have previously reported, such a plan is to serve as the agency’s IT vision or roadmap and help align its information resources with its business strategies and investment decisions. The plan might include the mission o the agency, key business processes, IT challenges, and guiding principles. A strategic plan is important to enable an agency to consider the resources, including human, infrastructure, and funding, that are needed to manage, support, and pay for projects. For example, a strategic plan that identifies what an agency intends to accomplish during a given period helps ensure that the necessary infrastructure is put in place for new or improved capabilities. In addition, a strategic plan that identifies interdependencies within and across individual IT systems modernization projects helps ensure that the interdependencies are understood and managed, so that projects—and thus system solutions—are effectively integrated. In summary, an IT strategic plan would provide a comprehensive picture of what the organization seeks to accomplish, identify the strategi use to achieve desired results, provide results-oriented goals andperformance measures that permit it to determine whether it is succeeding, and describe interdependencies that these can be understood and managed. within and across projects so However, FDA has not yet developed such a plan, although it does have two high-level planning documents—the agency’s Strategic Action Plan and the PDUFA IV IT Plan (PDUFA plan). Even in combination, however, the two plans do not have the scope and depth of an IT strategic plan: the first does not treat IT initiatives in depth, and the second is not an agencywide plan. Although these two plans include some elements of an IT strategic plan, they do not include all. FDA’s Strategic Action Plan, approved in fall 2007, does not include all IT projects or their associated performance measures, milestones, and interdependencies, although it does include strategic goals and objectives. Specifically, the plan describes four major strategic goals for the agency along with subsidiary implementation objectives, some of which identify IT initiatives (table 3 shows these major goals, objectives, and initiatives). As an overall agency plan, the Strategic Action Plan includes initiatives related to the agency’s major strategic goals, but it does not include performance measures or milestones for those initiatives. In addition, it does not include certain IT initiatives; for example, the PREDICT initiative, described in table 2, is a major initiative not mentioned in the Strategic Action Plan. Further, it does not identify interdependencies within and across individual IT modernization projects to ensure that they are understood and managed appropriately. For example, FDA has several ongoing projects that are developing data standards, including Regulated Product Submission, Harmonized Inventory, and Automated Laboratory Management. A well-designed IT strategic plan would document any interdependencies in such related projects. The PDUFA plan, published in July 2008, does focus on IT, and it provides details on goals, initiatives, and milestones, as well as performance measures. The plan includes several sections addressing current FDA IT goals and strategies. For example, it discusses detailed measures to create data standards to be used throughout the agency for regulatory submissions, and it describes the responsibilities of a Data Standards Council, which coordinates standards with data provider organizations. However, this document is not a comprehensive plan for the agency’s IT because it addresses only those IT initiatives that are related to user fee programs (which cover drugs and biologics). Further, it does not include an assessment of interdependencies among projects. Thus, although the Strategic Action Plan and PDUFA plan contain elements that would be included in an IT strategic plan, neither provides the comprehensive coverage of FDA’s goals and activities that a well- crafted IT strategic plan would provide. FDA officials agreed that the current plans do not include all the elements required for an IT strategic plan. The CIO said that the agency is aware of the importance of having such a plan and intends to develop one. However, according to the Chief Operating Officer, the agency must resolve many near-term planning activities and strategic investment decisions before it can complete long-term systems development plans. He stated that FDA is still working on its vision for modernizing IT infrastructure and services and how to incorporate that vision into an IT strategic plan. Accordingly, FDA has not defined either milestones or a completion date for an IT strategic plan. As reflected by its projects and high-level plans, FDA intends to address most of the limitations in its IT systems and infrastructure that had been previously identified by the agency’s Science Board, its contractors, and us. Table 4 provides an overview of the limitations along with related projects and activities that the agency is planning or currently undertaking. The table also shows which identified limitations are discussed in the two high-level planning documents mentioned earlier (the agency’s Strategic Action Plan and the PDUFA plan). Addressing these limitations in plans and projects does not guarantee that the limitations will be successfully overcome, but it does indicate that they are receiving management attention. As the table shows, FDA intends to address most of the previously identified limitations in its IT systems, infrastructure, and management. That is, of the 17 limitations in the table, 14 are associated with projects, activities, or plans. For example, to address IT infrastructure limitations, the ICT21 project is, among other things, replacing outdated data centers. To address limitations in the agency’s ability to handle data and make the data available, the Common Electronic Document Room project is to digitize data formerly available only in paper form, as well as establish a single repository for all regulatory documents (replacing separate document repositories at FDA’s centers). Further, to increase the agency’s ability to use technology to improve regulatory effectiveness, the PREDICT project is to provide the capability for predictive, risk-based surveillance of imported food. That is, it is to assist FDA inspectors in deciding which shipments of imported food to inspect by using a rule- based expert system to assess information from multiple sources and determine which shipments carry the highest risk. However, FDA is not addressing 3 of 17 limitations. For example, the agency does not have projects, activities, or plans to address its inability to perform inspections, remote monitoring, or sensing for contaminants in regulated products at manufacturing sites or in transportation vehicles. According to FDA officials, an initial investigation of the possible use of RFID (radio frequency identification) tags to allow remote monitoring to prevent drug counterfeiting was not successful. Agency officials indicated that remote sensing was currently not a high priority. In addition, the agency does not plan to address two previously identified limitations in IT management (this topic is discussed in the next section). Further, although these projects, activities, and high-level plans are intended to address most of the limitations, successfully overcoming the limitations depends in part on the agency’s developing and implementing appropriately detailed plans. FDA is taking steps to respond to the need to modernize its IT systems and infrastructure, but the number and range of its activities are further evidence of the importance of a comprehensive IT strategic plan to guide and coordinate them. Such a plan would allow FDA to integrate the planning for all of its modernization projects, including setting priorities, allocating resources, and accounting for dependencies. At the same time, it would provide a roadmap for improving FDA’s IT management capabilities, which would decrease the risk that the agency’s modernization initiatives will not achieve their goals or deliver planned capabilities on time and within budget. An agency’s chance of success in modernizing its IT systems is improved if it institutes critical IT management capabilities, including strategic planning (discussed in the previous section), investment management, information security, enterprise architecture, and human capital. Although FDA is making progress in these areas, it has considerable work to do. It is building necessary capabilities in investment management and information security, but it continues to have information security deficiencies, and important elements of its enterprise architecture are not in place. Finally, it is not effectively managing its IT human capital. Without these management capabilities in place, FDA increases the risk that its modernization efforts will not deliver required system capabilities and expected mission value on time and within budget. IT investment management links investment decisions to an organization’s strategic objectives and business plans. The Clinger-Cohen Act requires an agency to, among other things, select and control IT projects as investments in a manner that minimizes risks while maximizing the return. Projects are seen as investments and are selected and managed on the basis of cost, benefit, risk, and organizational priorities by an investment board made up of senior agency managers. To select an investment, the organization (1) identifies and analyzes each project’s risks and returns before committing significant funds to any project and (2) selects those IT projects that will best support its mission needs. The selection process should take account of the specific business needs addressed by each project and should use the agency’s enterprise architecture. Once a project is under way, the organization manages project schedules, costs, benefits, and risks to ensure that the project meets mission needs within cost and schedule expectations. Our ITIM framework for assessing investment management maturity includes foundational processes for selecting projects and for managing them at the project level, such as establishing an investment review board, developing an investment selection process, and overseeing the progress of individual projects. FDA has made progress in implementing selected foundational processes, as described below. Selecting IT investments. FDA has put in place several important practices cited in our ITIM framework, including establishing an investment review board and developing an investment selection process: In February 2006, the agency created an IT investment review board—the Bioinformatics Board. The board has broad responsibilities, including approving all IT budget execution decisions; overseeing business decisions on priority, planning, and execution of agency cross-cutting automation projects; directing the related business process analyses; and overseeing planning activities to ensure coordination. Members of the board are senior officials: It is co-chaired by two Deputy Commissioners—the Chief Operating Officer and the Chief Medical Officer. FDA has established Business Review Boards, representing core agencywide business areas, as standing subcommittees of the Bioinformatics Board. The Business Review Boards, among other things, act as the agencywide “business sponsor” of new systems development, provide oversight and direction of the work being performed on IT systems and projects within their defined areas, and prepare and present proposals to the Bioinformatics Board for review and approval. FDA has documented criteria for evaluating prospective projects, such as public health impact, cost savings, and whether the project is agencywide. Bioinformatics Board members told us that the Business Review Boards use these criteria and others specified by the Bioinformatics Board, such as budget considerations. Oversight and project management. As part of an effective IT investment process, an agency must be able to control its investments—manage its projects—so that they finish predictably within established schedule and budget. To accomplish this, agencies should have policies and procedures for oversight and should provide adequate resources, such as managers and staff responsible for monitoring projects. In the absence of predictable, repeatable, and reliable investment control processes, investments will be subject to a higher risk of failure. FDA’s Business Review Boards and Bioinformatics Board are responsible for overseeing projects. The Business Review Boards are responsible for day-to-day oversight of projects, for providing status reports, and for elevating problems to the Bioinformatics Board as needed. In the oversight area, the Bioinformatics Board reviews status reports and makes decisions on problems elevated by the Business Review Boards. FDA also has put in place a policy framework to manage its projects effectively. For example: FDA has created a project management office to assess and improve project management, standardize project management practices, improve communication so that senior executives and stakeholders know program and project status, and centralize and coordinate the management of IT programs and projects. The agency also has a staff of trained project managers and has assigned project managers to most of its modernization projects. FDA has a documented project monitoring and control process intended to track progress so that appropriate corrective actions can be taken when the project’s performance deviates significantly from the baseline project management plan. It defines tasks to be performed by the project manager—such as tracking progress and managing risk—and identifies supporting tools. This process, if appropriately implemented, provides FDA with a foundation for an effective project management capability. Information security is critically important for federal agencies, where the public’s trust is essential, and poor information security can have devastating consequences. Since 1997, we have identified information security as a governmentwide high-risk issue in each of our biennial reports to the Congress. Concerned by reports of significant weaknesses in federal computer systems, the Congress passed the Federal Information Security Management Act of 2002 (FISMA), which requires agencies to develop and implement an information security program, evaluation processes, and annual reporting. FDA’s most recent FISMA results indicate that the agency has made progress on information security but that problems remain. The 2008 FISMA audit by the HHS Inspector General found that FDA continued to make progress in implementing an infrastructure to support security management. However, the report cited 78 deficiencies in seven categories, including infrastructure, integrating security into applications, network management, and personnel security. In response to the Inspector General’s report, FDA’s CIO reported that the agency has conducted a comprehensive security review and made major changes to its information security program. According to the CIO, it has a new IT security program that is consolidated at the agency level and will provide consistent, centralized support across the agency. In addition, the agency has awarded a new contract for security services, and it is taking steps to address the Inspector General’s specific concerns. However, FDA is not addressing all of the Inspector General’s findings, because it believes it already meets the requirements for several of the controls found to be deficient. Security issues could be a challenge for FDA’s modernization plans; the Common Electronic Document Room, for example, will need to securely keep confidential records, trade secrets, and classified materials. Effective information security is essential to prevent data tampering, disruptions in critical operations, fraud, and unauthorized access or disclosure of sensitive information. An agency’s enterprise architecture describes both its business operations and the technology it uses to carry out those operations. It is a blueprint for organizational change defined in models that describe (in both business and technology terms) how an entity operates today and how it intends to operate in the future; it also includes a plan for transitioning to this future state. For example, it discusses interrelated business processes and business rules, information needs and flows, and work locations and users. Technical topics include hardware, software, data, communications, security attributes, and performance standards. It provides these perspectives both for the enterprise’s current or “as is” environment and for its target or “to be” environment, as well as a transition plan for moving from the “as is” to the “to be” environment. We have developed our Enterprise Architecture Management Maturity Framework to provide federal agencies with a common benchmarking tool for planning and measuring their efforts to improve enterprise architecture management. Like the ITIM, it provides a five-stage hierarchy of core management elements that agencies should perform to manage enterprise architecture development, maintenance, and implementation. The initial core elements for building the enterprise architecture foundation focus on building a management foundation; for example, one of these core elements is the organization’s recognizing that an enterprise architecture is a corporate asset by vesting accountability for it in an executive body that represents the entire enterprise. At this stage, an organization also assigns management roles and responsibilities and establishes plans for developing enterprise architecture products and for measuring program progress and product quality; it also commits the resources necessary for developing an architecture—people, processes, and tools. In addition, the organization develops a documented enterprise architecture program management plan, describing in detail the steps to be taken and tasks to be performed in managing the program, including a detailed work breakdown and estimates for funding and staffing. According to FDA, it has taken several initial steps toward building an enterprise architecture management foundation, such as establishing a committee or group representing the enterprise that is responsible for enterprise architecture, establishing a program office responsible for enterprise architecture, and designating a Chief Architect. However, according to the chief architect, FDA has not developed the program management plan that our framework characterizes as essential to ensuring that the enterprise architecture is effectively and efficiently developed. Beyond establishing an enterprise architecture management foundation, FDA has not yet developed architecture artifacts at the depth and breadth associated with a well-defined enterprise architecture. According to FDA’s Chief Architect and other officials, they are currently modeling the agency’s existing business processes and the data exchanges among existing processes as part of an HHS-wide modeling effort. Further, the agency has a listing of its current systems and the business processes that they support. However, no other “as is” artifacts were available. For the “to be,” the Chief Architect stated that they have developed an initial version of the “to be” architecture and have completed a transition plan for moving from the “as is” to the “to be.” However, they could not provide either the “to be” architecture artifacts that we requested or the enterprise transition plan. According to relevant guidance and best practices, the transition plan should provide a road map for moving from the “as is” to the “to be” environment. To facilitate its enterprise architecture efforts, FDA is using an approach called segment architecture. A segment architecture allows for the details needed to implement an enterprise architecture to be built in pie by piece. First a corporate layer of architecture is built that sufficient ly reflects, among other things, those policies, rules, and standards that apply across the whole enterprise; then the more specific content needed to implement the enterprise architecture on a segment-by-segment basis is added. The segment architecture extends the enterprisewide layer, providing additional detail and depth needed to implement project and IT solutions. Accordingly, segment architectures do not stand alone. FDA has begun building segments before it has a well-defined enterprise architecture and before it has prioritized its segments. According to the Federal Enterprise Architecture Practice Guide, prioritizing segments should precede building them. Once prioritization is completed, the agency should define (1) the scope and strategic intent of each segment, (2) business and information requirements, and (3) the conceptual solution architecture. FDA has identified 26 segments in all (for example, product safety, risk analysis, scientific analysis, and external partnerships), but it has not yet prioritized them. According to FDA, its enterprise architecture staff are currently working to define a standard set of criteria that the Bioinformatics Board is to use to set priorities for the remaining segments. Although FDA has not prioritized its segments, it has, according to officials, completed the architecture for one segment—product safety— including an “as is,” “to be,” and transition plan. According to the Chief Architect, the completed product safety segment architecture describes the scope and strategic intent of the segment, defines business and information requirements, and includes a description of the solutions architecture. According to FDA officials, this architecture has been sent to HHS for approval. However, they could not provide documentation of the completed segment. Attempting to define and build major IT systems without first completing either an enterprisewide architecture and, where appropriate, the relevant segment architecture is risky. According to the Federal Enterprise Architecture Practice Guide, prioritizing segments should precede building them, and developing the segment architecture should take place before an agency executes projects. FDA has identified three modernization projects as being within the product safety segment: MedWatch Plus, FAERS, and Harmonized Inventory. Thus, the other 13 major modernization projects are proceeding without the guidance and constraint of an enterprise or segment architecture. For example, some projects outside the product safety segment—such as the Common Electronic Document Room and PREDICT—that will need to use data from multiple sources may not be able to exchange data seamlessly with future systems. Similarly, a recent FDA study to identify existing applications with potential for agencywide use said it could not make definitive recommendations without a “to be” architecture. Also, going forward, further development of a “to be” enterprise architecture could be hindered by the lack of an IT strategic plan, since an enterprise architecture must align with an organization’s strategic planning. As long as the architectural context for its enterprise architecture and segment architectures lags behind its modernization projects, FDA increases the risk that its modernization solutions will not be defined, developed, and deployed in a way that promotes interoperability, maximizes shared reuse, and minimizes overlap and duplication. The success or failure of federal programs, like those of other organizations, depends on having the right number of people with the right mix of knowledge and skills. In our past work, we have found that strategic human capital management is essential to the success of any organization. Strategic human capital management focuses on two principles that are critical in a modern, results-oriented management environment: People are assets whose value can be enhanced through investment. An organization’s human capital approaches must be aligned to support the mission, vision for the future, core values, goals and objectives, and strategies by which the organization has defined its direction. In our model of strategic human capital management and our report on principles for strategic workforce planning, we lay out principles for managing human capital. Strategic workforce planning involves determining the critical skills and competencies needed to achieve current and future program results (these should be linked to long-term goals), analyzing the gaps between current skills and future needs, and developing strategies for filling gaps. Figure 2 shows the process of planning for workforce needs and the need for ongoing gap analyses based on program goals. Inventory of existing workforce capabilities FDA is not yet strategically managing its IT workforce, although it is taking some steps to address its IT human capital limitations. (As described in table 4, previously identified limitations include insufficient IT workforce and lack of investment in staff development.) For example, officials told us they have substantially increased the training budget this year for IT staff, although they could not provide actual dollar figures. Further, because the centers’ IT staffs have been centralized into the new Office of Information Management, IT human capital planning can be done centrally by the CIO. However, FDA has not yet inventoried the IT skills of its current IT workforce, determined present or future skills needs, or analyzed gaps. (A senior official said these activities were not undertaken because the centralization was too recent.) The CIO said that the agency is drafting a work order for an IT skills gap analysis, and agreed that the IT function is still understaffed. Even in the absence of an inventory, FDA officials were able to cite some skills areas as currently in short supply, such as project managers and network engineers. Finally, as mentioned earlier, the agency does not yet have an IT strategic plan; having a plan that describes future activities would improve the agency’s ability to accurately project its future staff and skill needs. Until it begins managing IT human capital strategically, FDA cannot be assured that it will have the workforce it needs to carry out its modernization projects. FDA is undertaking a variety of activities to address IT limitations that have hampered its mission, many of which the agency describes as urgent and some (such as PDUFA investments) as a result of federal laws and guidance. To help ensure that these important efforts are successful, the agency would be assisted by the kind of strategic view of its modernization initiatives provided by an appropriately comprehensive IT strategic plan. However, FDA does not have such a plan guiding its modernization efforts. FDA’s current agencywide plans lack many of the elements associated with a comprehensive IT strategic plan, such as strategies for managing the interdependencies among projects. In its modernization initiatives, FDA is taking steps to improve IT management. That is, it has begun implementing an enterprisewide approach to IT management, and it has put into place a foundation for investment management. However, FDA has weaknesses in certain IT management capabilities, including enterprise architecture, human capital, and security. Unless it further develops its enterprise architecture, the agency increases the risk that projects will not fully meet its strategic mission requirements, will be duplicative, and will not be integrated. In addition, the lack of a developed IT human capital management process increases the risk that projects will fail and that activities will continue to be hampered by a shortage of appropriately skilled staff. Finally, to address information security risks, the agency will need to ensure that it responds appropriately to the recommendations made by the HHS Inspector General. To help ensure the success of FDA’s modernization efforts, we recommend that the Commissioner of FDA require the CIO to take expeditious actions to set milestones and a completion date for developing a comprehensive IT strategic plan, including results-oriented goals, strategies, milestones, performance measures, and an analysis of interdependencies among projects and activities, and use this plan to guide and coordinate its modernization projects and activities; develop a documented enterprise architecture program management plan that includes a detailed work breakdown of the tasks, activities, and time frames associated with developing the architecture, as well as the funding and staff resources needed; complete the criteria for setting priorities for the segment architecture and accelerate development of the segment and enterprise architecture, including “as is,” “to be,” and transition plans, and in the meantime develop plans to manage the increased risk to modernization projects of proceeding without an architecture to guide and constrain their development; and develop a skills inventory, needs assessment, and gap analysis, and develop initiatives to address skills gaps as part of a strategic approach to IT human capital planning. The Acting Commissioner of Food and Drugs provided written comments on a draft of this report (the comments are reproduced in app. II). In the comments, FDA generally agreed with our recommendations and identified actions initiated or planned to address them. On developing a comprehensive IT strategic plan, for example, the agency stated that its efforts included performing a high-level analysis of FDA’s most immediate needs and priorities, and taking a longer-range view of the functionalities and capabilities it will need in the coming years. The agency added that it intends to complete a draft plan by the end of fiscal year 2009. In addition, with regard to its enterprise architecture, the agency stated that it was currently documenting a program management plan. It also indicated that it will use its ITIM processes to identify risks to its projects and programs and help ensure that they adhere to the agency’s “to be” architecture. Further, on developing a strategic approach to IT human capital planning, FDA stated that it plans to assess workforce needs, develop hiring plans based on the needs, and survey staff to identify their concerns with the organizational environment. The agency’s completion of the activities described, as well as other necessary actions to implement our recommendations, should increase the likelihood that FDA’s modernization projects and activities will accomplish their intended goals. In addition, the agency provided technical comments to clarify our discussion of its IT budget, which we have incorporated as appropriate. We are sending copies of this report to the Commissioner of the Food and Drug Administration, appropriate congressional committees, and other interested parties. In addition, the report is available at no charge on the GAO Web site at http://www.gao.gov. Should you or your staffs have questions on matters discussed in this report, please contact me at (202) 512-6304 or melvinv@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix V. Our objectives were to (1) evaluate the Food and Drug Administration’s (FDA) overall plans for modernizing its systems, including the extent to which the plans address identified limitations or inadequacies in the agency’s information technology (IT) capabilities, and (2) assess to what extent the agency has put in place key IT management policies and processes to guide the implementation of its modernization projects. To evaluate FDA’s overall plans for modernizing its IT systems, we examined criteria for strategic plans in guidance from the Office of Management and Budget (OMB), legislation (the Clinger-Cohen Act), and our previous reports. We analyzed studies of FDA’s IT conducted in the last several years to identify core limitations. We requested and received documentation from FDA on its agencywide modernization projects, including descriptions of their purpose and project summary status reports showing their expected completion dates and other milestones. We then analyzed these documents to determine which IT limitations these projects were intended to address. We analyzed the agency’s two main high-level planning documents that address IT, the agency’s Strategic Action Plan and the Prescription Drug User Fee Act (PDUFA) IV IT Plan, to determine whether they included elements of an IT strategic plan. We also assessed whether these plans were addressing IT limitations by analyzing whether they included strategies to address each limitation, and whether the plan included one or more projects intended to address each limitation. However, we did not assess the degree to which each limitation was addressed by FDA’s activities. Finally, we attended information sessions given by a contractor and an FDA inspector on one of the agency’s major initiatives—the Predictive Risk-based Evaluation for Dynamic Import Compliance Targeting (PREDICT) system—to gain understanding of the methodology and plans for implementing the system. To assess the IT management guiding the implementation and management of FDA’s modernization projects, we focused on key areas— investment management (including project management), information security, enterprise architecture development, and human capital management. We looked at whether policies or processes were in place for IT investment management, enterprise architecture, and human capital. We based our analysis on three frameworks: our Information Technology Investment Management (ITIM) framework, our Enterprise Architecture Management Maturity Framework, and our framework for strategic human capital management. The ITIM framework is a maturity model composed of five progressive stages of maturity that an agency can achieve in its IT investment management capabilities. Each stage specifies critical processes as well as specific key practices within each process. Stage 2 critical processes lay the foundation for sound IT investment management. We examined FDA’s implementation of three critical stage 2 processes (Instituting the Investment Board, Selecting an Investment, and Providing Investment Oversight). Within each process, we looked for the existence of policies, procedures, and organizational entities that would enable effective investment management and oversight. We did not do a complete ITIM assessment or audit specific IT projects to analyze how well the policies and procedures were implemented. Our Enterprise Architecture Maturity Framework (EAMMF) describes stages of maturity in managing enterprise architecture. Each stage includes core elements—descriptions of a practice or condition that is needed for effective enterprise architecture management. We evaluated FDA’s implementation of four core elements from stage 2 (Building the Enterprise Architecture Management Foundation). We did not do a complete EAMMF assessment, and we did not audit specific IT projects to analyze how well the policies and procedures were implemented. To supplement the EAMMF criteria, we used criteria from the Federal Enterprise Architecture Practice Guide issued by OMB and compared FDA’s progress on its architecture with these criteria. Our framework for strategic human capital management lays out principles for managing human capital. We evaluated FDA’s policies and procedures against this framework. To assess the agency’s management of information security, we analyzed the HHS Inspector General’s fiscal year 2009 FISMA report, which assessed FDA’s compliance with FISMA information security provisions. We did not do an independent review of the agency’s information security. In addition, we interviewed FDA officials, including the Chief Operating Officer, the Chief Information Officer (CIO), and officials from the new Office of Information Management and its five subdivisions. We also interviewed officials from the Office of Budget Presentation and Formulation, the Center for Biologics Evaluation and Research, and the Center for Drug Evaluation and Research. Further, we interviewed officials outside FDA, including a member of the Science Board study and a former FDA regulatory official to obtain additional perspectives on IT issues and proposed solutions at FDA. Finally, we obtained the perspectives of the Acting Commissioner regarding the IT issues identified in our review. We conducted this performance audit at FDA headquarters in Rockville, Maryland, from May 2008 through June 2009 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objective. According to FDA’s CIO, the agency defines mission-critical systems as those that support its centers and offices in accomplishing their mission. According to FDA, there are currently about 47 of these mission-critical systems. FDA’s CIO stated that the number of mission-critical systems is subject to change as legacy systems are retired and modernization projects create new systems to take their place. Mission-critical systems can be grouped by the key mission areas that they support: reviewing and evaluating applications for new products, overseeing manufacturing and production supply chains, and monitoring the safety of products on the market. In tables 5 to 7, we provide examples of systems that are currently in use and support a variety of internal users from each of FDA’s main centers and the Office of Regulatory Affairs (ORA). Regulatory tracking systems are currently used by each center for the day- to-day business activities supporting FDA’s regulatory review processes. These systems are used in the receipt and storage of externally generated applications, submissions, or other information for FDA’s regulatory review processes. Compliance systems are used to process or assess data used by FDA when overseeing conformance to regulatory requirements of an external entity or marketed product. These systems are generally used in the inspection of an FDA-regulated product or its manufacturing facilities. Adverse event reporting and analysis systems are used to process and/or assess data related to adverse reactions to FDA-regulated products. An adverse event could be illness due to food, injury caused by a device, or negative reaction to a drug or vaccine. FDA has defined its mission-critical infrastructure as IT equipment that must be available full time (24 hours a day, 7 days a week) in order for the agency to accomplish its mission. FDA identified the following infrastructure components as mission critical: Network components, which consist of Internet connectivity, domain name servers, active directory, e-mail, single sign on, and the routing infrastructure. Critical servers to run systems needed for operations that must run full time, such as the Prior Notice Center, which must be available full time for FDA to receive prior notice before food is imported into the United States. Other examples are servers to support Mission Accomplishments and Regulatory Compliance Services, Operational and Administrative System for Import Support, and Electronic Submission Gateway. Security components, such as the firewalls that protect the network from unauthorized users. Secure Remote Access infrastructure, which provides the ability for authorized users to securely access FDA computing resources from a non- FDA remote location. In addition to its mission-critical infrastructure, FDA provides other infrastructure services that support its mission, including telecommunications and help desk services. Reason study performed Main IT-related findings Undertaken to examine the effectiveness of the process used to develop requirements for a replacement for the agency’s dysfunctional AERS I system. FDA’s management of requirements development did not follow proper IT methodology; the Office of IT had poor procedures in the areas of procurement and communication with end users. Endorsed by FDA Management Council to ensure that FDA’s mission- critical IT activities are driven by proper business planning procedures. According to a survey of participants from FDA’s business centers done to understand the state of FDA business processes for use in FDA’s business process strategies, FDA’s IT capability to support processes needed significant improvement. Requested by members of the Congress to determine FDA’s ability to manage postmarket drug safety issues and assess the steps FDA is taking in this area. FDA databases cannot perform some actions needed to make postmarket drug safety decisions, and different types of data are not available to FDA. Requested by FDA to assess whether the agency’s science and technology can support current and future regulatory needs; to identify the broad categories of scientific and technologic capacities that FDA needs to fully support its core regulatory functions and decision making. FDA’s resources have not increased in proportion to the scientific demands on the agency, resulting in demand that far exceeds its capacity to respond. FDA cannot fulfill many of its core regulatory functions because its IT infrastructure is obsolete, unstable, and inefficient. Contracted by FDA to identify IT applications performing premarket processes, as defined by the Business Process Framework, with potential for agencywide use; also to find which applications were redundant, to retire them. Significant overlap exists among the IT applications assessed— opportunities exist to streamline these applications; 16 of 54 premarket applications had high enterprise potential for functionality, 25 were rated medium, and 13 were rated low. Reason study performed Main IT-related findings Requested by the Congress to investigate concerns regarding FDA’s foreign drug inspection program and make recommendations. FDA’s databases do not provide an accurate count of foreign establishments subject to inspection and do provide widely divergent counts. Because FDA does not know the number of establishments subject to inspection, the percentage of those inspected also cannot be calculated with certainty. Inconsistencies in its databases such as these have prevented FDA from ensuring compliance with corrective items from inspections that highlighted serious deficiencies. Required by OMB to determine FDA’s compliance with the Federal Information Security Management Act of 2002 (FISMA) in accordance with the OMB’s guidance; to determine if the FDA’s security program encompasses a risk-based life cycle approach to improving information security. Among other things, FDA did not fully implement a security program infrastructure to support its overall security program, and FDA did not conduct all required system development life cycle activities. Undertaken to allow FDA to better meet increased demand for information, and to make decisions more quickly and easily. Among other things, recommendations included development of information standards at an agency level, and use of these standards within a common enterprise information model within 7 to 10 years. In addition to the contact person named above, key contributors to this report were Cynthia Scott, Assistant Director; Shaun Byrnes; Barbara Collier; Neil Doherty; Rebecca Eyler; Anh Le; Glenn Spiegel; Shawn Ward; and Daniel Wexler. | The Food and Drug Administration (FDA) relies heavily on information technology (IT) to carry out its responsibility for ensuring the safety and effectiveness of certain consumer products. Recognizing limitations in its IT capabilities that had been previously identified in studies by FDA and others, the agency has begun various initiatives to modernize its IT systems. GAO was asked to (1) evaluate the agency's overall plans for modernizing its IT systems, including the extent to which the plans address identified limitations or inadequacies in the agency's capabilities, and (2) assess to what extent the agency has put in place key IT management policies and processes to guide the implementation of its modernization projects. GAO analyzed FDA's plans to determine whether they followed best practices and addressed capability limitations, reviewed key management policies and processes, and interviewed agency officials. In response to federal law and guidance and urgent mission needs, FDA is pursuing numerous modernization projects (including 16 enterprisewide initiatives), many of which are in early stages. However, FDA does not have a comprehensive IT strategic plan to coordinate and manage these initiatives and projects. Such a plan would describe what the agency seeks to accomplish, identify the strategies it will use to achieve desired results, and provide results-oriented goals and performance measures that permit it to determine whether it is succeeding. FDA has developed two high-level planning documents that include some of these elements, but not all: (1) The agency's Strategic Action Plan provides high-level goals and objectives related to modernization of infrastructure and systems, but it does not provide details on IT initiatives, such as milestones and performance measures. (2) An IT plan for FDA's user fee program for drugs and biological products focuses on selected projects in greater detail, but these projects are only a subset of the agency's modernization initiatives. As reflected by its projects and high-level plans, FDA intends to address most of the limitations in its IT systems and infrastructure that had been previously identified. However, successfully overcoming these limitations depends in part on the agency's developing and implementing appropriately detailed plans. A comprehensive IT strategic plan, including results-oriented goals and performance measures, is vital for guiding and coordinating the agency's numerous ongoing modernization projects and activities. Until it develops such a plan, the risk is increased that the agency's IT modernization may not adequately meet the agency's urgent mission needs. FDA has made mixed progress in establishing important IT management capabilities that are essential in helping ensure a successful modernization. These capabilities include investment management, information security, enterprise architecture development, and human capital management. For example, as part of a move to an enterprisewide approach to IT management, FDA has put policies in place for investment management and project management, and it is making progress in addressing information security. However, significant work remains with regard to enterprise architecture (that is, establishing modernization blueprints describing the organization's operation in terms of business and technology), particularly its "to be" architecture--a blueprint of where it wants to go in the future. Further, the agency is not strategically managing IT human capital--it has not determined its IT skills needs or analyzed gaps between skills on hand and future needs. In both these areas (enterprise architecture and human capital management), the agency's vision for the future, as captured in an IT strategic plan, would be an important asset. Without an effective enterprise architecture and strategic human capital management, FDA has less assurance that it will be able to modernize effectively and will have the appropriate IT staff to effectively implement and support its modernization efforts. |
VA operates the largest integrated health care system in the United States, providing care to nearly 5 million veterans per year. The VA health care system consists of hospitals, ambulatory clinics, nursing homes, residential rehabilitation treatment programs, and readjustment counseling centers. In addition to providing medical care, VA is the largest educator of health care professionals, training more than 28,000 medical residents annually, as well as other types of trainees. State licenses are issued by state licensing boards, which generally establish state licensing requirements governing their licensed practitioners. Current and unrestricted licenses are licenses that are in good standing in the states that issued them, and licensed practitioners may hold licenses from more than one state. To keep a license current, practitioners must renew their licenses before they expire and meet renewal requirements established by state licensing boards, such as continuing education. Renewal procedures and requirements vary by state and occupation. When licensing boards discover violations of licensing practices, such as the abuse of prescription drugs or the provision of poor quality of care that results in adverse health effects, they may place restrictions on licenses or revoke them. Restrictions from a state licensing board can limit or prohibit a practitioner from practicing in that particular state. Some, but not all, issued state licenses are marked in such a way as to indicate that the licenses have had restrictions placed on them. Generally, state licensing boards maintain a database of information on restrictions, which employers can often obtain at no cost either by accessing the information on a board’s Web site or by contacting the board directly. National certificates are issued by national certifying organizations, which are separate and independent from state licensing boards. These organizations establish professional standards that are national in scope for certain occupations, such as respiratory and occupational therapists. Practitioners who are required to have a national certificate to practice in VA may renew these credentials periodically by paying a fee and verifying that they obtained required educational credit hours. National certifying organizations can place restrictions on a certification or revoke certification for violations of the organization’s professional standards. Like state licensing boards, national certifying organizations maintain a database of information on disciplinary actions taken against practitioners with national certificates and many can be accessed at no cost. We identified key VA screening requirements that are intended to ensure that VA facilities employ health care practitioners who have valid professional credentials and personal backgrounds appropriate to deliver safe health care to veterans. Officials at VA facilities are required to verify whether credentialsstate licenses and national certificatesheld by applicants and employees are current and unrestricted. VA also requires its facilities to check the names of all applicants VA intends to hire against a federal list of individuals who have been excluded from participation in any federal health care programs and to compare applicants’ educational institutions against lists of fraudulent institutions. Additionally, VA requires that individuals in certain positions undergo a background investigation, which includes checking their fingerprints against a fingerprint-based criminal history database. VA policy requires officials at its facilities to screen applicants to determine whether they possess at least one current and unrestricted state license or an appropriate national certificate, whichever is applicable for the position they seek. We classified VA’s practitioners into three groups, depending upon the credentials and the verification process VA requires for employment. Figure 1 illustrates VA’s process for credentials verification with state licensing boards and national certifying organizations for these groups for applicants VA intends to hire and for employed practitioners, whose credentials are checked periodically. Groups A and B represent practitioners who must be licensed to work in VA. However, the requirements and process VA uses to verify professional credentials is different for each of these groups. Group C represents practitioners who must have a national certificate to work in VA and may also have a state license. The process used to screen applicants in all three groups has two stages. First, applicants are required to disclose, if applicable, their state licenses and national certificates. Second, VA facility officials are required to verify whether applicants’ state licenses or national certificates are current and unrestricted. To verify applicants’ credentials, VA officials are required to either contact state licensing boards or to physically inspect an applicant’s national certificate. Officials are also required to document that they verified the status of the professional credentials. VA also has requirements for verifying the credentials of its employed practitioners. Like the verification process for applicants, this process involves employed practitioners’ disclosures and VA verification of that information. VA employed practitioners in group A are required to disclose all of their current licenses, while those in group B must disclose only one license. For employed practitioners in group A, facility officials are also required to determine if any expired licenses disclosed as current and unrestricted at the beginning of employment had restrictions placed on them prior to their expiration. VA depends on its employed practitioners in group B to inform facility officials of any change in the status of their license, including any that have expired. Employed practitioners in group C must disclose a national certificate. VA officials must confirm that the disclosed licenses and certificates are current and unrestricted. For group A practitioners, VA facility officials contact the appropriate state licensing boards directly; for groups B and C they physically inspect the state license or national certificate. VA officials verify these credentials periodically depending on the occupation and the requirements of the state or national organization that issued the license or certificate. For example, a registered nurse with a state license from Virginia must renew the license every 3 years, while a respiratory therapist must renew the national certificate every 5 years. VA officials are required to document these verifications. If VA’s verification process identifies that a state licensing board or national certifying organization took disciplinary action against a practitioner, facility officials are required to determine the circumstances of the disciplinary action. Licensing boards and certifying organizations have various options for disciplining practitioners. For example, a nurse who is abusing drugs and voluntarily enters a drug abuse program may retain a license to practice with supervision when administering drugs. In contrast, a physician whose treatment results in the death or the permanent disability of a patient may have a license revoked. On the basis of a review of the action taken by the state licensing board or national certifying organization, VA officials are to determine whether an applicant should be hired or an employed practitioner should be retained or terminated. To supplement its checks with state licensing boards, VA has requirements for searching for disciplinary actions taken against licensed practitioners that might not have been disclosed by physician and dentist applicants and employed practitioners. VA requires its facilities to check national databases for information on disciplinary actions taken against these practitioners. Specifically, VA requires that facility officials query the Federation of State Medical Boards (FSMB) database, which includes records of disciplinary actions taken against physicians by all state licensing boards. Similarly, VA requires facility officials to query the National Practitioner Data Bank (NPDB), which contains information including disciplinary actions taken against physicians and dentists. Facility officials must document the results of these queries. VA policy requires its facilities to check the names of all applicants VA intends to hire against a federal list of individuals who have been excluded from participation in any federal health care program. The list, referred to as the List of Excluded Individuals and Entities (LEIE), is maintained by the Department of Health and Human Services’ Office of Inspector General. Since March 1999, VA facilities are to electronically query the LEIE Web site on all applicants prior to employment. VA also requires its facilities to make sure that an applicant’s educational degrees are authentic. VA requires that applicants for some positions, such as social workers, have degrees from accredited institutions. To prevent the use of fraudulent degrees to obtain employment, VA requires that its facilities compare the educational institutions listed by an applicant against existing lists of “diploma mills” that sell fictitious college degrees and other professional credentials. VA’s employed practitioners are required to undergo a background investigation that verifies their personal histories. A background investigation verifies, for instance, an individual’s history of employment, education, and residence. It also includes a fingerprint check that searches for evidence of criminal activity by comparing fingerprints against a database of criminal records. The Office of Personnel Management (OPM) conducts the investigations for VA and reports its results back to the facility that requested the investigation. In conjunction with the background investigation, VA employed practitioners are required to disclose information about their professional and personal backgrounds by filling out the Declaration for Federal Employment form—also known as form 306. Employed practitioners are asked to disclose, among other things, information about criminal convictions, employment terminations, military court-martials, and delinquencies on federal loans. Failure to disclose information requested on form 306 is grounds for dismissal. Facility officials compare the information obtained from form 306 with the results obtained through the background investigation to determine whether employed practitioners have been forthcoming in their disclosures. If the background investigation results include questionable issues, such as discrepancies in work or criminal histories, the facility has 90 days to take action. Gaps in key VA screening requirements result in vulnerabilities when screening certain health care practitioners. Although the screening requirements for some occupations, such as physicians, are adequate because they require verifying all licenses by contacting state licensing boards, screening requirements for other occupations are less stringent. These less stringent requirements do not require checking all licenses, and they require physical inspection of one license only rather than contacting the state licensing board. Similarly, VA does not require contacting national certifying organizations to verify national certificatesthe credentials held by health care practitioners, such as respiratory therapists. Physical inspection of credentials alone can be misleading; not all state licenses and national certificates indicate whether they are restricted, and licenses and certificates can be forged. While VA requires checking a national database for physicians and dentists, it does not require that facility officials query a national database that contains reports of disciplinary actions and criminal convictions involving all licensed practitioners. In addition, VA does not require that all practitioners undergo background investigations, including fingerprinting, to check for criminal records. These gaps create vulnerabilities because VA may remain unaware of health care practitioners who could place patients at risk. VA’s requirements for verifying the professional credentials of applicants it intends to hire and employed practitioners in group A, such as physicians and dentists, are complete and thorough. This is also the case for applicants VA intends to hire in group B, such as nurses and pharmacists. VA requires facility officials to verify all state licenses by contacting the appropriate state licensing boards. To supplement these requirements for physicians and dentists, VA officials also must query FSMB and NPDB to identify reports of any disciplinary actions involving these practitioners. In contrast to all practitioners in group A, the process for verification of licenses for group B practitioners has gaps, as illustrated in figure 2. VA’s verification process for group B practitioners that it intends to hire is as stringent as the process used for group A practitioners. However, the process used to verify licenses for continued employment of group B practitioners is less stringent, because facility officials are required to check only one state license, which is selected by the practitioner. Furthermore, officials are not required to contact the state licensing board directly, but instead may simply physically inspect the one state license to check its status. Employed practitioners in group B with multiple state licenses select the one state license under which they will continue to practice in VA. The license selected does not have to be from the state where the VA facility is located. VA officials check only that single license. As a result, these employed practitioners could have a restricted license in one state, or several restricted state licenses, but offer VA officials an unrestricted license from another state for verification. Moreover, the method required to periodically verify the status of licenses for continued employment of practitioners in group B is not thorough. VA facility officials are only required to physically inspect the license— instead of contacting the state licensing board. VA facility officials we interviewed were unaware of the inherent vulnerabilities in relying on a physical inspection. According to licensing board officials, one cannot determine with certainty that a license is valid and unrestricted unless the state licensing board is contacted directly. These officials explained that state licensing boards do not always exchange information. Furthermore, physical inspection of licenses alone can be misleading because not all state licensing boards mark a license to indicate that it is restricted, and licenses can be forged, even though licensing boards have taken steps to minimize this problem. Licensing board officials also pointed out that many state boards do not charge a fee to verify licenses. Unlike the national database queries of FSMB for physicians and NPDB for physicians and dentists, VA does not require facility officials to query the Healthcare Integrity and Protection Data Bank (HIPDB), a national database that contains information on disciplinary actions and criminal convictions involving all licensed practitioners. All government agencies, including state licensing boards, are required to report to HIPDB. VA accesses HIPDB when it queries the NPDB for physicians and dentists because the databases share information. However, VA does not require its facilities to query HIPDB for all licensed practitioners even though VA is authorized by statute to query this database at no charge. VA’s requirements for verifying the professional credentials of both applicants and employed practitioners in group C also have gaps, as illustrated in figure 3. For both applicants and employed practitioners in group C, which include respiratory therapists and dietitians for example, facility officials are only required to physically inspect the national certificate to check its status. The physical inspection is required when these practitioners apply for employment and periodically for continued employment. Additionally, VA requires applicants in group C to disclose all of the state licenses they have ever held, but does not require facility officials to verify any of these state licenses. However, according to officials from national certifying organizations, the authenticity and status of a national certificate can only be assured by contacting the national certifying organizations directly. For example, an official from the National Board for Respiratory Care told us that practitioners that were certified prior to July 2002 are not required to renew their certificates—they can voluntarily choose to recertify. Thus, physical inspection of a certificate will not ensure that there has been no disciplinary action taken by the board since the certificate was issued. VA has not implemented consistent background screening requirements, which include fingerprint checks, for all practitioners. Although VA requires background investigations for newly hired employed practitioners, it does not require background investigations for certain contract health care practitioners, practitioners who work without compensation from VA, medical consultants, and medical residents. VA, with prior approval from OPM, has the authority to determine which positions in VA require a background investigation. VA requested and received permission from OPM to exempt certain categories of health care practitioners from background investigations, based on VA’s assessment that these types of practitioners do not need a background investigation. Table 1 lists the types of practitioners that VA exempts from background investigations. VA requested and received permission from OPM, in 2001 and 2003, to perform fingerprint-only checks for contract health care practitioners, who work in a facility for 6 months or less and are currently exempt from background investigations, and for all volunteers who have access to patients, patient information, or pharmaceuticals. OPM began to offer a fingerprint-only checka new screening optionfor use by federal agencies in 2001. Compared to background investigations, which typically take several months to complete, fingerprint-only checks can be obtained within 3 weeks or less and cost less than $25, about a quarter of the cost of a background investigation. In commenting on a draft of this report, VA said that it planned to implement fingerprint-only checks for all contract health care practitioners, medical residents, medical consultants, and practitioners that work without direct compensation from VA, as well as certain volunteers. However, VA has not issued guidance to its facilities instructing them to implement fingerprint-only checks on all these practitioners. VA did issue guidance to its facilities to implement fingerprint-only checks for volunteers who have access to patients, patient information, or pharmaceuticals. Implementing fingerprint-only checks for practitioners who are currently exempt from background investigations would detect practitioners with a criminal history. According to the lead VA Office of Inspector General investigator in the Dr. Swango case, if Dr. Swango had undergone a fingerprint check at the VA facility where he trained, VA facility officials would have identified his criminal history and could have taken appropriate action. Additionally, one of the facilities we visited had implemented fingerprint-only checks of medical residents training in the facility and contract health care practitioners. An official at this facility stated that at a minimum, fingerprint-only checks of medical residents and contract practitioners were necessary to help ensure the safety of veterans in the facility. FSMB in 1996 recommended that states perform background investigations, including criminal history checks, on medical residents in order to better protect patients because residents have varying levels of unsupervised patient care. This recommendation, in part, resulted from reports that over a 4-year period more than 500 residents had performance, behavioral, or criminal problems during their training. In the four facilities we visited, we found mixed compliance with the existing key VA screening requirements which are intended to ensure that applicants and employed practitioners at VA facilities have valid professional credentials and personal backgrounds to deliver safe health care to veterans. None of the four VA facilities complied with all of the key requirements. Moreover, VA does not conduct oversight of its facilities to determine if they comply with these key screening requirements. In order to show the variability in the level of compliance among the four VA facilities we visited, we measured their performance against a compliance rate of at least 90 percent for each of five of the six screening requirements, even though VA allows no deviation from these requirements. Table 2 summarizes the rate of compliance among the four VA facilities we visited. For detailed information about our analysis and each facility’s compliance with a particular requirement, see appendixes I and II. For the sixth requirementmatching the educational institutions listed by a practitioner against lists of diploma millswe asked facility officials if they did this check and then asked them to produce the lists of diploma mills they use. All four facilities generally complied with VA’s existing policies for verifying the professional credentials of employed practitioners, either by contacting the state licensing board for practitioners, such as physicians, or physically inspecting the license or national certificate for practitioners, such as nurses and respiratory therapists. They also generally ensured that applicants VA intended to hire had completed the Declaration for Federal Employment form, which requires the applicants to disclose, among other things, information about criminal convictions, employment terminations, and delinquencies on federal loans. However, three of the four facilities did not follow VA’s policies for verifying all of the professional credentials of applicants and three facilities did not compare applicants’ names to LEIE prior to hiring them. Two of the four facilities conducted background investigations on their employed practitioners at least 90 percent of the time, but the other two facilities did not. We also asked officials whether their facilities checked the educational institutions listed by an applicant against a list of diploma mills to verify that the applicant’s degree was not obtained from a fraudulent institution. An official at one of the four facilities told us his staff consistently performed this check. Officials at the other three facilities stated they did not perform the check because they did not have a list of diploma mills. In addition to assessing the rate of compliance with the key screening requirements, we found that VA facilities varied in how quickly they took action to deal with background investigations that returned questionable results, such as discrepancies in work or criminal histories. OPM gives a VA facility up to 90 days to take action after the facility receives investigation results with questionable findings. We reviewed the timeliness of actions taken by facility officials from August 1, 2002, to August 23, 2003, at the four facilities we visited and six additional facilities geographically spread across the VA health care system. We found that officials at 5 of the 10 facilities took action within the 90-day time frame, with the number of days ranging on average from 13 to 68. Officials at 3 facilities exceeded the 90-day time frame on average by 36 to 290 days. One facility took action on its cases prior to OPM closing the investigation, and another facility did not have the information available to report. For additional information on the average number of days it took each facility to report its actions, see appendix II. One of the cases that exceeded the 90-day time frame involved a nursing assistant who was hired to work in a VA nursing home in June 2002. In August 2002, OPM sent the results of its background investigation to the VA facility, reporting that the nursing assistant had been fired from a non- VA nursing home for patient abuse. During our review, we found this case among stacks of OPM results of background investigations that were stored on a cart and in piles on the desk and on other work surfaces of a clerk’s office. After we brought this case to the attention of facility officials in December 2003, they reviewed the report and then terminated the employee for not disclosing this information on the Declaration for Federal Employment form 306. The employee had worked at the VA facility for more than 1 year. Another case at the same facility that exceeded the 90-day time frame involved an employee who had been convicted for possession of illegal drugs prior to being hired by VA. He had been hired at the facility in August 2002 and was to complete a background investigation form at that time. In June 2003, almost 1 year after being hired, a facility official realized the employee had not completed and returned this form and gave the employee the form to complete. The employee returned the completed form in the same month and it was sent to OPM, which returned the results of its investigation to the facility in July 2003, before the employee’s probationary period of 1 year was completed. The OPM report revealed numerous arrests for possession of illegal drugs. During our December 2003 review and about 120 days after the investigation results were returned from OPM, we found this report and brought it to the attention of the facility director. Later, a facility official told us that VA’s regional counsel stated that since the employee’s 1-year probationary period had ended and the employee had disclosed this information on the Declaration for Federal Employment form 306, the facility could not take action to terminate the employee. VA has not conducted oversight of its facilities’ compliance with the key screening requirements. Instead, VA has relied on OPM to do limited reviews of whether facilities were meeting certain human resources requirements, such as completion of background investigations. These reviews did not include determining whether the facilities were verifying professional credentials. Although VA established the Office of Human Resources Oversight and Effectiveness in January 2003 to conduct such oversight, the office has not conducted any facility compliance evaluations. There is no VA policy outlining the human resources program evaluations to be performed by this office, and the resources have not been provided to support the functions of this office. VA’s screening requirements are intended to ensure the safety of veterans by identifying applicants and employed practitioners with restricted or fraudulent credentials, criminal backgrounds, or questionable work histories. However, gaps in VA’s existing screening requirements allow some practitioners access to patients without a thorough screening of their professional credentials and personal backgrounds. For example, although the screening requirements for verifying professional credentials for some occupations, such as physicians, are adequate, VA does not apply the same screening requirements for all occupations with direct patient care access. Specifically, VA does not require that all licenses be verified, or that licenses and national certificates be verified by contacting state licensing boards or national certifying organizations. VA relies on two national databases to identify physicians and dentists who have had disciplinary actions taken against them. In addition, VA accesses a third national database, HIPDB, for physicians and dentists, because HIPDB is linked to one of the two national databases VA currently accesses. HIPDB is a national database that contains reports of disciplinary actions and criminal convictions involving all licensed practitioners, not just physicians and dentists. However VA does not require facility officials to query HIPDB for all licensed practitioners. As a result, practitioners such as nurses, pharmacists, and physical therapists do not have their state licenses checked against a national database. In addition, VA does not require all practitioners with direct patient care access, such as medical residents, to have their fingerprints checked against a criminal history database. These gaps create vulnerabilities that could allow incompetent practitioners or practitioners with the intent to harm patients into VA’s health care system. In addition to these gaps, compliance with the existing key screening requirements was mixed at the four facilities we visited. None of the four facilities complied with all of the key VA screening requirements. However, all four facilities generally complied with VA’s requirement to periodically verify the credentials of practitioners for their continued employment. Although VA created the Office of Human Resources Oversight and Effectiveness in January 2003 expressly to provide oversight of VA’s human resources practices at its facilities, it has not provided resources for this office to conduct oversight of VA facilities’ compliance with these requirements. Without such oversight, VA cannot provide reasonable assurance that its facilities comply with requirements intended to ensure the safety of veterans receiving health care in VA facilities. In light of the gaps we found and mixed compliance with the key screening requirements by VA facilities, we believe effective oversight could reduce the potential risks to the safety of veterans receiving health care in VA facilities. To better ensure the safety of veterans receiving health care at VA facilities, we recommend that the Secretary of Veterans Affairs direct the Under Secretary for Health to take the following four actions: expand the verification requirement that facility officials contact state licensing boards and national certifying organizations to include all state licenses and national certificates held by applicants and employed practitioners, expand the query of the Healthcare Integrity and Protection Data Bank to include all licensed practitioners that VA intends to hire and periodically query this database for continued employment, require fingerprint checks for all health care practitioners who were previously exempted from background investigations and who have direct patient care access, and conduct oversight to help ensure that facilities comply with all key screening requirements for applicants and current employees. In commenting on a draft of this report, VA generally agreed with our findings and conclusions. VA acknowledged that we identified gaps in its process for conducting background and credentialing checks and that we provided what appeared to be reasonable recommendations to close those gaps. VA stated that it would provide a detailed action plan to implement our recommendations when the final report was issued. VA said that our draft report inaccurately omitted VA’s querying HIPDB for practitioners who practice independently. We revised our report to clarify that NPDB queries performed by VA automatically check HIPDB for these practitioners because the databases are linked. However, VA does not perform queries of HIPDB for the majority of its licensed practitioners, which includes nurses, pharmacists, physical therapists, and dental hygienists. VA also incorrectly stated that the draft report did not include VA’s requirement to query FSMB for physicians. In addition, VA said that it planned to implement fingerprint-only checks for all contract health care practitioners, medical residents, medical consultants, and practitioners that work without direct compensation from VA, as well as certain volunteers. However, VA has not issued guidance to its facilities instructing them to implement fingerprint-only checks for all these practitioners. Further, VA stated that the title of the report implied that veterans are receiving inadequate care on a broad basis. We disagree. The title reflects vulnerabilities created by the gaps in the screening of practitioners that could place veterans at risk by allowing incompetent practitioners or those with the intent to harm patients into VA’s health care system. VA provided technical comments which we incorporated, as appropriate. VA’s written comments are reprinted in appendix III. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days after its date. We will then send copies of this report to the Secretary of Veterans Affairs and other interested parties. We also will make copies available to others upon request. In addition, the report will be available at no charge at the GAO Web site at http://www.gao.gov. If you or your staff have any questions about this report, please call me at (202) 512-7101. Another contact and key contributors are listed in appendix IV. We examined VA’s policies and practices that are intended to ensure that health care practitioners at its facilities have appropriate credentials and backgrounds to provide care to veterans. Specifically, we (1) identified key VA screening requirements for its health care practitioners, (2) determined the adequacy of these screening requirements, and (3) assessed the extent to which selected VA facilities complied with these screening requirements. To identify key VA screening requirements for its health care practitioners, we reviewed VA’s policies and VA Handbook 5005, which explains how to implement the screening policies. We limited our review to 43 occupations in VA that have direct patient care access or have an impact on patient care. See table 3 for a list of the occupations included in our review. To identify the 43 occupations, we consulted with VA human resource officials. We interviewed human resource officials at VA headquarters and at each facility we visited. We classified the practitioners who are required to have professional credentialsa state license or a national certificateinto three groups according to VA’s requirements for verifying these credentials. Groups A and B represent practitioners who must be licensed to work in VA. However, the requirements and process VA uses to verify professional credentials is different for each of these groups. Group C represents practitioners who must have a national certificate to work in VA and may also have a state license. Practitioners not included in the three groups are not required to have either a license or a national certificate to work in VA facilities. To determine the adequacy of the key VA screening requirements, we analyzed VA’s policies and procedures to identify whether there were inconsistencies in how the requirements were applied among various types of practitioners. We interviewed VA headquarters and facility officials and practitioners at the facilities we visited to determine how VA’s policies are implemented in facilities. In addition, we interviewed representatives from 13 state licensing boards and the District of Columbia Board of Nursing and 2 national certifying organizations to determine if VA’s requirements for verifying professional credentials are adequate for identifying practitioners without valid and unrestricted state licenses and national certificates. To assess the extent to which VA facilities we visited complied with the key screening requirements, we chose a judgmental sample of four VA facilities that varied in size, location, and medical school affiliations to assess the extent to which these selected facilities complied with these requirements. The four facilities are located in Big Spring, Texas; New Orleans, Louisiana; Seattle, Washington; and the District of Columbia. We chose these facilities based on geographic variation, affiliations with medical schools to train residents, and types of health care services provided. Of the four facilities we visited, three are large facilities located in major metropolitan areas and each are affiliated with at least one medical school. The remaining facility is small, providing mainly primary care and long-term care services to veterans and is located in a rural area. For each facility, VA provided, from its automated pay system, a list of current practitioners in the 43 occupations. As a result of using VA’s automated pay system, our sample does not include those practitioners providing care through a contract or training agreement or without direct compensation from VA. For each of the four facilities, we selected a random sample of 100 practitioners either hired or assigned to their current position no earlier than January 1, 1993. We chose to limit our review to approximately the last 10 years because VA changed its process for credentials verification in the early 1990s. For each of these practitioners, we reviewed their personnel files to check that the facility had complied with the following key VA screening requirements: verify state licenses and national certificates for applicants; verify state licenses and national certificates for employed practitioners; query the List of Excluded Individuals and Entities (LEIE) prior to hire; ensure completion of background investigations, including fingerprints; ensure completion of the Declaration for Federal Employment form, also known as form 306; and verify that the educational institutions listed by a practitioner VA intends to hire are checked against lists of diploma mills. In order to show the variability in the level of compliance among the four VA facilities we visited, we distinguished between facilities that had a compliance rate of at least 90 percent for each of five of the six screening requirements and those that did not. For each facility and key screening requirement, we compared the percentage of personnel files found in compliance to an acceptance level of 90 percent. In order to confirm that a requirement had a compliance rate less than 90 percent, we performed a one-sided significance test at the 95 percent confidence level. See appendix II for detailed information on the four VA facilities’ compliance with each of the key VA screening requirements. Our results from these four facilities cannot be generalized to other facilities. In order to determine compliance with the key screening requirement to verify that the educational institutions listed by a practitioner are not fraudulent, we asked facility human resources staff if they performed this screening and asked them to produce their lists of diploma mills. Additionally, we reviewed VA facilities’ response times to 214 background investigation results returned from OPM with questionable issues, from August 1, 2002, to August 23, 2003, at the four locations we visited and six other VA facilities selected based on geographic location. The six additional facilities were located in Boston, Massachusetts; East Orange, New Jersey; Indianapolis, Indiana; Palo Alto, California; Portland, Oregon; and San Diego, California. For these 10 facilities, we asked officials to provide the date they took action on cases returned from OPM, and from VA headquarters we obtained the dates when OPM returned the cases to the facility. We determined the average number of days it took each facility to take action after these cases were returned from OPM with questionable issues. Our results cannot be generalized to other VA facilities. See appendix II for detailed information on the results of our analysis. Tables 4 and 5 show the sample counts used to measure compliance and the results of our review for five of the requirements. Table 6 shows the average number of days it took each facility to take action after cases with questionable issues were returned from OPM. In addition to the contact named above, Jacquelyn T. Clinton, Jessica Cobert, Mary Ann Curran, Martha A. Fisher, Krister Friday, Lesia Mandzia, and Marie Stetser made key contributions to this report. | Cases of practitioners causing intentional harm to patients have raised concerns about the Department of Veterans Affairs' (VA) screening of practitioners' professional credentials and personal backgrounds. GAO was asked to (1) identify key VA screening requirements, (2) evaluate their adequacy, and (3) assess compliance with these screening requirements. GAO reviewed VA's policies and identified key VA screening requirements for 43 health care occupations; interviewed officials from VA, licensing boards, and certifying organizations; and randomly sampled about 100 practitioners' personnel files at each of four VA facilities we visited. GAO identified key screening requirements that VA uses to verify the professional credentials and personal backgrounds of its health care practitioners. These requirements include verifying professional credentials; completing background investigations for certain practitioners, including fingerprinting to check for criminal histories; and checking national databases that contain reports of practitioners who have been professionally disciplined or excluded from federal health care programs. GAO found adequate screening requirements for certain practitioners, such as physicians, for whom all licenses are verified by contacting state licensing boards. However, screening requirements for others, such as currently employed nurses and respiratory therapists, are less stringent because they do not require verification of all licenses and national certificates. Moreover, they require only physical inspection of the credential rather than contacting state licensing boards and national certifying organizations. Physical inspection alone can be misleading; not all credentials indicate whether they are restricted, and credentials can be forged. VA also does not require facility officials to query, for other than physicians and dentists, a national database that includes reports of disciplinary actions involving all licensed practitioners. In addition, many practitioners with direct patient care access, such as medical residents, are not required to undergo background investigations, including fingerprinting to check for criminal histories. VA has not conducted oversight of its facilities' compliance with the key screening requirements. This pattern of mixed compliance and the gaps in key VA screening requirements creates vulnerabilities to the extent that VA remains unaware of practitioners who could place patients at risk. |
According to IRS, during the mid to late 1990’s, abusive tax schemes reemerged across the country. The use of abusive tax shelters, anti- taxation arguments, abusive tax schemes, and frivolous returns last peaked in the 1980’s. IRS characterizes an abusive tax scheme as any plan or arrangement created and used to obtain tax benefits not allowable by law. Schemes run from simple to very complex, from clearly illegal to those carefully constructed to disguise the illegality of the scheme. Furthermore, users of schemes can range from those believing their position is correct to those who knowingly but willfully file incorrect tax returns. Schemes can be based on improper use of domestic and foreign trusts, inflated business expenses and deductions, falsely claimed tax credits and refunds, and various anti-tax arguments. Some schemes are created by tax professionals such as accountants, lawyers, and paid tax preparers, or by groups and individuals. Tax schemes are offered to taxpayers using various means, including conferences or seminars, publications, advertisements, and the Internet. Others are promoted by word-of-mouth. Abusive tax schemes that are generally used by individuals fall into four major categories. For the first two of these, frivolous returns and frivolous refunds, taxpayers submit a tax return that either states an argument that IRS can readily identify as frivolous, or a return with characteristics IRS has identified as reflecting a frivolous argument. For the other two types of schemes, abusive trusts and offshore compliance strategies, taxpayers’ returns are less likely to reveal use of a clearly abusive tax scheme. These schemes generally use any number of anti-tax arguments to incorrectly claim that income is exempt from taxation or that IRS otherwise lacks authority needed to tax income. These arguments have been well litigated in the courts and consistently ruled to be without merit. Examples include the following: Form 2555 Scheme: In this scheme, individuals file an IRS Form 2555, Foreign Earned Income, and claim that their income was not earned within the United States. This is also known as the “not a citizen” argument in which taxpayers file returns stating they are citizens of the “Republic of ” and not citizens of the United States, and thus, their income is not taxable. Section 861: Individuals using this scheme claim that under Internal Revenue Code section 861 income tax must only be paid on foreign income and, therefore, their income is not subject to tax or withholding. In these cases, taxpayers file a tax return and show a zero amount for wages. According to IRS, this argument has spread to some employers who are using it to avoid withholding and paying payroll- type taxes on their employees. According to IRS, credit and refund abusive tax schemes are designed to substantially reduce taxes or create a refund for the taxpayer, generally by claiming eligibility for a credit that does not exist or to which the taxpayer is not properly entitled. One such scheme that has received much attention is the Slavery Reparation Refund scheme. According to IRS, promoters circulate or publish information claiming African Americans are eligible for slavery reparations. Taxpayers claiming this credit generally enter a significant amount on their tax return as a credit that results in a taxpayer realizing a refund if not detected by IRS. A trust is a legitimate form of ownership, which completely separates asset responsibility and control from the benefits of ownership. As such, trusts are commonly used in matters such as estate planning. An abusive domestic trust scheme usually involves a taxpayer creating a trust that does not meet the Internal Revenue Code requirements that the assets and income of the trust not be subject to the control of the taxpayer. Once such an improper trust is established and the taxpayer has transferred business or personal assets to it, the scheme may involve further abuses, such as offsetting income of the trust by overstating its business expenses or including the taxpayer’s personal expenses—like a home mortgage—as an expense of the trust. The taxpayer will often use multiple entities such as partnerships, limited liability companies, or secondary level trusts that can be tiered or layered to mask the taxpayer’s continued ownership or control of the trust’s income or assets. Abuses that involve foreign locations can take a wide array of forms and attempt to use a number of techniques to improperly avoid paying taxes. One common technique is simply to use foreign locations to add another level of complexity in obscuring the true ownership of assets or income and thus obfuscating whether taxes are owed and by whom. Use of foreign locations, for instance, can be combined with use of trusts to make unraveling the true ownership of assets and income more difficult for IRS. According to IRS, criminals long have used offshore schemes to disguise the true nature of their enterprise and the resulting income. Promoters of abusive tax schemes have, according to IRS, increasingly devised schemes that in some fashion involve transferring income or title to assets to foreign locations. Often foreign locations are selected because they are tax havens with little or no taxation on income in their jurisdiction, have privacy rules that help schemers hide what they are doing, or have other characteristics favorable to carrying out the schemes. According to IRS, once such transfers are established, income is often repatriated back to the U.S. owners through loans, credit cards, or debit cards. By using complex transactions and multiple entities, the individuals using these schemes attempt to hide their income and avoid potential tax liabilities. According to IRS’s fiscal year 2003-2004 Small Business and Self-Employed (SB/SE) Division Strategic Assessment Report, abusive tax schemes represent a rapidly growing risk to the tax base. IRS estimates the potential revenue loss from these schemes to be in the tens of billions of dollars annually. According to an IRS official, to make accurate estimates in this area of noncompliance is difficult. For one reason, the nuances and types of schemes are constantly changing and evolving, particularly in the areas of abusive trusts and offshore compliance. Also, IRS’s detection of schemes is challenging. Trust schemes, in particular, often involve multiple entities that are vertically layered or tiered in an attempt to disguise the true ownership of the income or assets of the trust. The difficulty in determining the significance of offshore compliance is also exacerbated because these types of schemes generally use tax haven countries to disguise the transactions and prevent IRS from routinely collecting tax-related information on transactions. Despite the difficulties in accurately estimating the significance of abusive tax schemes, IRS provided us with estimates in four major scheme areas— Frivolous Returns, Frivolous Refunds, Abusive Domestic Trusts, and Offshore Schemes. According to IRS, its estimates were made in February of 2002 and were derived from information gathered during tax return processing and examination activities and from the work of IRS’s Criminal Investigation, the law enforcement arm of IRS. According to an IRS official, these estimates were derived from tax year 2000 information, the last full year for which data were available. IRS’s estimates are as follows: Frivolous returns: about 62,000 taxpayers with associated tax amounts approximating $1.8 billion. Frivolous refunds: about 105,000 taxpayers with associated tax amounts approximating $3.1 billion. Abusive domestic trusts: about 65,000 taxpayers with tax losses approximating $2.9 billion. Offshore schemes: about 505,000 taxpayers with tax losses ranging from $20 billion to $40 billion. IRS’s estimates for the numbers of taxpayers and taxes in connection with frivolous returns and refunds, although not precise, likely have less uncertainty than its estimates of the numbers of taxpayers and taxes at risk in connection with abusive domestic trusts and offshore schemes. IRS’s estimates for frivolous returns and refunds are based in large part on returns and refund claims that IRS has identified while processing tax returns and has addressed by pulling the associated returns and notifying the taxpayers that their returns contained errors that need to be corrected. Thus, in these cases, IRS has a fairly direct basis for counting the number of taxpayers involved and the amount of tax involved. Furthermore, because IRS has pulled these returns from processing, in general, improper refund claims have not been paid out, and IRS is pursuing collection of the proper amount of tax when taxpayers have failed to pay the full amount owed. In contrast, although taxpayers using domestic trusts and offshore schemes may file tax returns, those returns alone seldom provide enough information for IRS to determine whether an abusive scheme was used. Therefore, IRS’s estimates of the numbers of taxpayers and the taxes at risk for the domestic trust and offshore scheme categories generally rely on limited numbers of cases that have been examined or investigated, on intelligence obtained in the course of normal tax administration and Criminal Investigation activities, and on IRS officials’ professional judgments. Recognizing that offshore transactions are a significant factor in offshore schemes, IRS has been taking steps concerning the use of credit/debit cards issued by offshore banks to U.S. taxpayers. Although having an offshore credit card is not illegal, IRS believes that some U.S. taxpayers are using such cards to evade U.S. taxes. In October 2000, a federal judge authorized IRS to serve “John Doe” summonses on American Express and MasterCard to obtain limited information on U.S. taxpayers holding credit cards issued by banks in several tax haven countries. On the basis of information received from MasterCard, IRS identified about 235,000 accounts issued through 28 banks located in 3 countries. IRS’s ongoing analysis of these data leads it to estimate that between 60,000 and 130,000 U.S. customers are associated with these 235,000 accounts. In part because MasterCard is estimated to have about 30 percent of this market, IRS estimates that there could be 1 to 2 million U.S. citizens with credit/debit cards issued by offshore banks. However, this is a very preliminary estimate. IRS officials believe this estimate may be reduced because, among other things, a portion of these accounts may not be associated with abusive tax schemes. By comparison, only about 117,000 individual taxpayers indicated that they had offshore bank accounts in tax year 1999. On March 25, 2002, IRS petitioned for permission to serve a summons on VISA International, seeking records on transactions using cards issued by banks in over 20 tax-haven countries. The estimates of the number of individuals and dollar consequences associated with offshore credit/debit card schemes are very uncertain at this time. Nevertheless, IRS’s February 2002 estimate of $20 billion to $40 billion in tax dollars at risk from offshore schemes may grow as IRS learns more about the extent of the problem. No one individual or office could provide an agencywide perspective on IRS’s strategy, goals, objectives, performance measures, or program results, for its efforts to address abusive tax schemes. Consequently, a clear and consistent picture of IRS’s efforts was difficult to obtain. Available information indicates that IRS began increasing its efforts to combat abusive schemes over the past 2 or 3 years, continued to do so in 2001, and plans further future efforts. Limited data also suggest that these enhanced efforts have helped IRS convict more promoters and users of abusive schemes over the past 3 years, which IRS has publicized through enhanced communication strategies. Organizationally, IRS identifies and deals with schemes in two primary ways—during its processing and examination of tax returns (compliance and enforcement) and through the work of Criminal Investigation (CI). However, most of IRS’s programs to address abusive schemes are the responsibility of SB/SE and CI. IRS also works with various federal agencies in its efforts to identify and deal with abusive tax schemes. IRS has taken a number of steps to enhance its compliance and enforcement efforts—its audit and other civil enforcement activities—that focus on abusive tax schemes. In the past year, for example, IRS has increased staff years devoted to examining abusive tax scheme promoters, decided to assign about 50 more agents to promoter examinations and train them, and laid plans for assigning 200 or more additional staff to reviewing abusive tax schemes and offshore compliance schemes. Furthermore, IRS has created an organization that initially will focus on developing leads and cases related to abusive scheme promoters and that will monitor promoter web sites. IRS identifies many abusive tax schemes during its normal tax return processing and examination activities. For example, when tax returns initially are processed either manually or by computers, processes are in place to detect apparent frivolous returns or returns reflecting improper refunds. In these cases, the returns are pulled from processing to be forwarded elsewhere for follow-up action. Both the Wage and Investment (W&I) and SB/SE divisions in IRS process taxpayers’ tax returns and both have responsibilities for identifying tax returns that may involve abusive tax schemes. Three principal SB/SE efforts focusing on or related to abusive tax schemes are the Frivolous Return Program, the Office of Flow-Through Entities and Abusive Tax Schemes, and the National Fraud Program. The Frivolous Return Program identifies the tax returns of individuals who assert unfounded legal or constitutional arguments and refuse to pay their taxes or to file a proper tax return. The program also identifies returns claiming frivolous refunds, such as those involving slavery reparations. Generally, IRS provides guidance to those who process tax returns to identify the characteristics of returns claiming such frivolous arguments or refunds. IRS also has programmed its computers to do so. The Treasury Inspector General for Tax Administration helped IRS develop software programs to identify slavery reparation schemes. Since both W&I and SB/SE staff process tax returns, both divisions are involved in identifying such returns. Once identified, the returns are pulled out of the tax return processing stream and forwarded to the Frivolous Return Program unit where they are to be resolved with the taxpayer. The program was consolidated in January 2001, at the Ogden, Utah, Compliance Services Center. The compliance center staff enters information about each case into a database and assigns 1 of 31 different codes identifying the frivolous argument or refund being claimed by the taxpayer. Then, a notice requesting taxpayers to file a proper tax return is to be sent advising them that IRS has judged their tax return to include an argument that is without legal merit or a credit or tax refund to which they are not entitled. IRS officials indicate that the number of staff assigned to the Frivolous Return Program unit in Ogden grew from 18 employees in September 2000 to 45 employees in September 2001. Some of this increase may not reflect a net IRS-wide increase in full-time equivalents (FTE) for frivolous returns since the increase has, in part, been due to centralizing efforts in Ogden from other IRS locations. IRS officials expect to assign more employees to this program in fiscal year 2003. The Office of Flow-Through Entities and Abusive Tax Schemes became operational in January 2000. The office was created to organize IRS’s efforts in addressing abusive tax schemes, particularly trusts, and to identify their promoters and sellers. The unit’s goals are (1) to catalogue and profile schemes and trends, (2) direct compliance resources to examine schemes and promoters or refer tax scheme promoters and participants for criminal prosecution, (3) increase employee knowledge and skills related to abusive tax scheme issues, and (4) enhance coordination within IRS on issues related to abusive tax schemes. IRS expects to assign and train about 50 revenue agents this fiscal year to focus mainly on promoters of abusive tax schemes. The agents are to undergo training during the summer of 2002 and to begin examining cases by the fall of 2002. According to IRS, the number of abusive promoter leads increased from 25 in March 2001 to 155 in February 2002. In addition, the number of abusive promoter cases approved for further examinations has increased from 17 cases to 94 cases during the same period. The time spent on these cases is also increasing. IRS also reports that time spent on promoter examinations for fiscal year 2002 is expected to be 12.1 staff years, which is up from 4.4 and 1.2 staff years in fiscal 2001 and fiscal year 2000, respectively. Furthermore, IRS plans additional expansion of its abusive tax scheme compliance efforts. For example, IRS expects to develop units that will include 8 to 10 agents in each of 15 locations. These units will address abusive tax schemes and flow-through entities. In addition, given the growing significance of the offshore credit/debit card schemes, IRS plans to create four special enforcement groups. Each group will be staffed by approximately 8 agents and will concentrate on these offshore schemes. This growth in staffing reflects IRS’s increased priority for these schemes. IRS officials expect that the agents assigned to these units will be redirected largely from other compliance areas. Schedule K-1 Transcription and Matching. In the spring of 2001, the transcription of Schedule K-1 information became a major responsibility of the Office of Flow-Through Entities and Abusive Tax Schemes. According to IRS, information provided on Schedule K-1s is important for determining whether recipients of flow-through income have properly reported that income on their tax returns. IRS can use transcribed data for information-matching to determine whether proper reporting of income occurred. IRS believes that flow-through entities such as trusts and partnerships are increasingly being used in abusive tax schemes. IRS can also use these K-1 data in its return examination and tax collection activities to help identify abusive tax schemes. Tax year 1995 marked the last year that Schedule K-1 information was transcribed by IRS. From 1990 through 1995, IRS transcribed approximately 5 percent to 12 percent of the Schedule K-1s received. After 1995, IRS did not transcribe Schedule K-1 information submitted with paper returns nor did it match the income information contained on the schedules with the information presented on individual beneficiaries’ or partners’ tax returns. IRS again started to transcribe tax year 2000 K-1 information during the spring of 2001 and completed the process in December 2001. IRS officials told us that the matching of the K-1 information against individual tax returns was to begin in March 2002. IRS cites several reasons for reinstating its transcription and matching of Schedule K-1s. First, IRS has observed a significant increase in flow- through entities. The number of tax returns filed by Trusts, Partnerships, and S-Corporations has increased by 12 percent, 33 percent, and 35 percent, respectively, over the 6-year period from fiscal years 1995 through 2000. IRS also estimates an overall increase of nearly 2 million such returns by 2009. Second, based on a small study, in January 2002, IRS estimated that between 6 percent and 15 percent of total flow-through income would not be reported on tax year 2001 returns. Although data available to us at the time of this testimony were not clear, IRS estimates that income of about $1 trillion was distributed to taxpayers from flow- through entities for tax year 2000. Third, IRS expects its Schedule K-1 matching program not only to identify underreporting or nonreporting of income but also to improve taxpayer compliance. Transcription and matching of Schedule K-1 data are expected to increase accurate reporting of trust income on future tax returns just as matching of wage, interest, and other types of income has increased the accuracy of taxpayers’ tax returns. As a result, the Schedule K-1 program places taxpayers who receive flow-through income on a more equal footing with taxpayers who are wage earners. Lead Development Center. IRS has adopted a strategy of identifying promoters of tax schemes as a key to halting their promotion and identifying those who have taken advantage of the scheme and thus likely owe taxes. By early April 2002, SB/SE is to initiate a Lead Development Center. The center’s primary functions are to develop case leads and assemble case information for distribution to compliance field offices for further investigation. Initially the center will focus on abusive tax scheme promoters, and over time, it will expand to perform similar functions for fraud and anti-money laundering cases. Also, the center will operate a computer laboratory that, among other things, is expected to monitor possible abusive promoter sites on the Internet. In addition, the center is to serve as a coordinating link among various IRS groups that deal with abusive tax scheme issues and with outside stakeholders such as the Department of Justice, the Federal Trade Commission (FTC), and others. The National Fraud Program, which operates at IRS’s campuses and field offices, coordinates efforts and provides oversight to IRS’s compliance efforts to identify potential tax fraud. In addition, the program helps identify trends and disseminates the information within IRS and acts as a liaison on fraud cases involving bankruptcy and employment and excise taxes among other types of tax fraud. A National Fraud Program manager sets overall policy and program direction. Fraud managers are located in five area offices, and they oversee the activities of about 65 fraud referral specialists. These specialists assist other IRS revenue compliance staff in identifying cases with fraud potential, determining when indications of fraud are present, and developing potential cases. They also review fraud cases for technical accuracy and adequacy of supporting documentation to ensure appropriate and consistent application of fraud program guidelines and requirements. In cases where there is evidence of criminal activity, those cases are to be referred to criminal investigation within IRS. IRS’s Criminal Investigation investigates and pursues promoters and sellers of abusive schemes and the individuals using such schemes. CI’s role is the enforcement of the tax laws for individuals who willfully fail to comply with their obligation to file and pay taxes and who ignore IRS’s collection and compliance efforts. The most flagrant cases are recommended for criminal prosecution. Criminal Investigation also administers the Questionable Refund Program that focuses on stopping the payment of various false tax refunds and, if warranted, on prosecuting the taxpayers involved. Furthermore, CI develops education and publicity activities warning taxpayers about abusive tax schemes and placed public information officers (PIO) in the field to specifically generate publicity regarding IRS’s law enforcement efforts. CI’s enforcement strategy as it relates to fraudulent tax schemes is to focus primarily on the promoters of these schemes and on taxpayers who willfully use these schemes to evade taxes. For example, during a tax scheme investigation, CI generally attempts to gain access to a fraudulent promoter’s list of clients to whom the promoter sold the scheme. In addition to pursuing the promoter, CI can then use the list of clients to determine who may have used the abusive scheme. CI determines which users of the abusive scheme merit investigation for possible prosecution and which users merit referral to IRS operating divisions for possible compliance and civil enforcement action. Although CI has data on enforcement activity related to several types of tax scams (e.g., related to employment tax, refunds, return preparers, nonfilers, and domestic and foreign trusts), CI only separately tracked its promoter efforts for domestic and foreign trusts. (See table 1.) CI officials said that the number of full-time equivalent staff working on domestic and foreign trusts increased from 55 in fiscal year 1999 to 69 in fiscal year 2001. Although no consistent pattern exists across all of the categories in table 1, CI has had increases in the number of convictions obtained over the 3- year period. Furthermore, looking only at promoter-related cases, indictments, convictions, and active investigations increased over the period while the number of prosecutions recommended declined. For purposes of deterring individuals from engaging in abusive trusts, the pattern of increasing convictions has provided IRS an opportunity to publicize more cases in which individuals have been found guilty. Further, the increases in indictments and convictions of promoters may help deter promoter activity in particular. Because the investigative and legal processes can span several years, data like those in the table do not show whether the cases investigated lead to prosecutions, convictions, and indictments in that same year. Further, the data do not account for differences in the importance of cases, such as whether major fraudulent efforts are being successfully investigated and closed. IRS data do show that the average length of sentence for the abusive domestic and foreign trust program rose substantially from 35 months in 1999 to 64 months in 2001. To the extent that average length of sentence relates to the severity of the crime, IRS may be making headway in pursuing key abusive trust cases. The Questionable Refund Program (QRP), administered by CI, was established in 1977. The QRP was designed to identify false returns, stop the payment of false refunds, and prosecute scheme perpetrators. Various false refund schemes are pursued under this program, including ones involving the earned income tax credit, the fuel tax credit, social security refund schemes, and slavery reparations. Tax returns and return information are subject to manual processing or computerized information matching. IRS’s compliance staff identifies those returns claiming a possible false refund generally during these various return examination processes and referred to Questionable Refund Detection Teams (QRDT). The QRDT staff within CI determines which returns should be pursued within CI or civilly. Schemes with criminal potential are referred to CI field offices for investigation while schemes lacking criminal potential are referred to the appropriate IRS compliance or collection group. CI’s efforts to inform and educate the public about abusive tax schemes and to publicize the results of its enforcement activities related to such schemes take many forms and involve several types of media. CI has been particularly active in trying to disseminate information to the public to make them aware of IRS’s activities and accomplishments in combating abusive tax schemes. In addition, CI has PIOs located across the country who work with local media to publicize IRS’s efforts and results. CI Education and Publicity Activities. CI’s education and publicity activities focus on warning taxpayers about fraudulent tax schemes so that they will not be tempted to use such schemes. CI hopes that increasing media coverage of successful tax scheme prosecutions will deter the public from participating in tax schemes because the perceived risk of detection, prosecution, and resulting penalties and sanctions will be too high. In addition, CI officials believe that publicizing the prosecutions of promoters and users of tax schemes helps assure the public that people are paying their fair share of taxes. CI posted its web page (www.ustreas.gov/irs/ci) on the Internet in September 1997. According to CI officials, over the past 2 years the Internet site has evolved into an important tool for educating and alerting the public about tax schemes and about CI’s efforts to detect and deal with those who promote and use tax schemes. The Internet site provides fraud alerts warning the public of schemes where promoters are targeting unsuspecting taxpayers; information on topics including tax filing responsibilities, nonfilers, and abusive tax return preparers; summaries of cases and successful prosecutions of promoters and users of fraudulent schemes; and press releases and other IRS publications to generate a wide public distribution. Tax practitioners are also targets of CI’s publicity strategy. According to CI officials, some tax practitioners are using IRS’s materials directly from the Internet site to inform those clients who may believe that a given tax scheme is legal. For example, clients may ask the tax practitioner to set up a fraudulent trust to reduce their taxes, and the tax practitioner can simply print the brochure about “Too Good to be True? – Trusts” from CI’s Internet site to discourage the taxpayers from using such a trust. In conjunction with using the Internet site as an informational tool to educate and warn the public of frivolous schemes, CI has taken steps to increase IRS’s visibility and presence on the Internet. According to CI, it has recently intensified its efforts to improve the ranking of IRS’s web page through the use of “metatags” or keyword tags. By doing so, IRS seeks to have Internet users who enter various terms in available Internet search engines find IRS’s web page listed near the top of displayed search results. For example, CI is planning to add tags such as “pay no tax,” and “form 1040” so that entering these terms will result in CI’s Internet site being listed in the displayed search results. CI is pursuing other possible strategies to ensure that CI’s site rises to the top of Internet search responses. For example, CI staff has occasionally visited known promoter Internet sites to gather information on keywords used by those sites. IRS plans to incorporate those keyword tags into its Internet site. As a result, IRS expects to increase the odds that the CI Internet site would be included alongside Internet sites that promote questionable tax avoidance strategies. In addition, CI is working to create a web content manager position with responsibilities that include designing a strategy to maximize the potential of CI’s Internet site. The manager would be responsible for helping to integrate CI data into the pages in IRS’s Internet site that provide information to specific types of taxpayers. CI Public Information Officers. In October 2000, CI established PIOs in each of IRS’s 35 field offices. The PIOs serve as points of contact for all internal and external CI communications initiatives, including the issuing of press releases and the coordination of important law enforcement media events. Although IRS has other media relations specialists located in its field offices, their duties tend to focus on publicizing tax filing season information, including the benefits of electronic filing. CI PIOs generate publicity regarding IRS’s law enforcement activities including the detection and prosecution of abusive tax schemes. Primary functions of the PIOs include establishing contacts with editors, reporters, and news directors to educate them on tax issues and provide information about IRS and CI to enable them to write in-depth articles. encouraging media to include more stories on the detection and prosecution of abusive tax schemes. getting articles included in trade and professional journals and magazines that are read frequently by professionals such as doctors, lawyers, and accountants to make them aware of abusive tax schemes. developing a local media strategy. Part of CI’s local strategy involves generating a “hook” to get the stories focused more on communities. In addition, CI has employed a strategy of “bundling” news stories. For example, CI has been working cases on fraud involved in the restaurant industry. Once several such cases have been put together, CI will bundle these stories together into a single news story for possible publication in magazines and journals read by people in the restaurant industry. giving speeches and participating in a wide variety of presentations, panel discussions, and conferences with professional organizations, including the American Bar Association, the American Institute of Certified Public Accountants, and the American Medical Association, to create public awareness of CI’s activities and to provide information about fraudulent tax schemes. IRS works with various federal agencies in its efforts to identify and deal with fraudulent tax schemes. These include the Federal Trade Commission (FTC),the Securities and Exchange Commission (SEC), the Financial Crimes Enforcement Network (FinCen), the Department of Justice (DOJ), the Federal Bureau of Investigation (FBI), and the United States Attorneys Offices (USAO). In some cases, IRS’s coordination is on an informal basis, as it is with the FTC and the SEC, and involves the sharing of certain information and detection techniques. In other cases, the relationship is more formal, as in the case with DOJ or USAOs, which prosecute fraud and other tax-related cases with the assistance of IRS staff. IRS officials participate in various federal agency working groups, including a multiagency task force to share information, skills, and procedures for combating fraud on the Internet; an IRS and DOJ working group created to examine the use of civil injunctions against abusive promoters currently under criminal investigation; and a money-laundering- experts working group. According to the officials we interviewed, these working groups are invaluable for developing networking relationships between agencies which facilitate information-sharing among staff. IRS staff also attends quarterly meetings with staff from the FTC, SEC, and DOJ to develop joint initiatives to combat Internet fraud. These meetings have spawned other activities for IRS staff, including FTC-sponsored training seminars and periodic visits to FTC’s Internet laboratory to keep current with FTC efforts to combat Internet fraud. IRS has tried to develop a better understanding of the potential breadth of the problem of abusive tax schemes involving individual taxpayers and the steps needed to coordinate and manage numerous efforts to combat abusive tax schemes. In some cases, these steps have been recently implemented, and in other cases, IRS is working to implement them. These expanded efforts have not been accompanied, however, by performance goals or measures that Congress and IRS can use to assess IRS’s progress. The increased scope of the abusive tax scheme problem, and perhaps especially the offshore compliance schemes, could strain IRS’s audit resources. IRS is now beginning to gather data that will better enable it to estimate the magnitude and nature of the offshore credit and debit card schemes. Improved data will help IRS identify how many and what types of resources it may need to address the schemes. However, the evasive nature of these schemes may necessitate face-to-face audits in a significant portion of cases to determine whether taxes are owed and the amount owed. Even if the number of individuals involved in these schemes is a fraction of the reported estimate of 1 to 2 million, IRS’s staff may be challenged to audit them and maintain its current audit coverage as well. IRS’s face-to-face audits have been declining, decreasing from nearly 400,000 in fiscal year 1999 to nearly 200,000 in fiscal year 2001. Accordingly, IRS has begun considering whether other techniques than audits could be used to resolve these cases. For example, IRS is considering options such as disclosure initiatives, settlement initiatives, and self-correction programs. These techniques will need to be tested and refined to determine which, if any, are effective. The increased scope of abusive tax schemes has also led IRS to develop an improved process for selecting the best cases to pursue among the many that it identifies, develop a new policy to govern simultaneous criminal and civil enforcement investigations of taxpayers, consider how to ensure that increased volumes of scheme-related tax assessments are followed up by IRS’s collection function when taxpayers are unable to pay in full, and use its internal research group and a contractor to develop better models for identifying indicators that taxpayers may be participating in abusive tax schemes. In addition, a significant organizational change has just been implemented in SB/SE that is intended to increase program oversight and coordinate programs and units dealing with abusive schemes and related tax fraud activities. To that end, in the past few weeks SB/SE has divided its Office of Flow-Through Entities and Abusive Tax Schemes. Now, its efforts to ensure accurate reporting of income connected to flow-through entities will fall under a director for reporting compliance. IRS separated the flow- through entity effort from other abusive tax scheme efforts because it judged that the flow-through effort is more related to its traditional information-matching and examination programs than to its abusive scheme efforts. The flow-through effort will, however, also provide useful information for IRS to use elsewhere in investigations of abusive schemes. The rest of SB/SE’s major programs and efforts that are more directly focused on abusive tax schemes—the National Fraud program, the Abusive Tax Schemes program, the Lead Development Center, and the Anti-Money Laundering program—have been placed under a single executive for reporting enforcement. Monitoring the Internet and other media outlets where abusive tax schemes often are advertised will also be part of this centralized effort. To date, however, IRS has not provided information on its staff year investments in combating abusive tax schemes and has not established performance goals and measures that IRS and Congress can use to gauge whether these efforts are achieving desired results. We testified recently that the IRS commissioner identified four major areas of systematic noncompliance. These areas not only focus heavily on abusive tax schemes involving individuals, but also include corporate tax shelter activity. The fiscal year 2003 budget request includes increased resources for compliance efforts, but, excluding the Earned Income Credit program, it is unclear from IRS’s congressional budget justification how many resources IRS intends to devote to major areas of noncompliance or what performance measures will be available to Congress and IRS to assess progress. IRS has long-standing programs and related efforts aimed at detecting and dealing with abusive tax schemes, particularly those related to frivolous tax returns and fraudulent tax refund claims. Recently, IRS has begun to take a more assertive and coordinated approach to detecting and dealing with an ever-changing array of schemes, including those involving the use of domestic and offshore trusts. In the past year, IRS has added more resources to these efforts, created new programs, and improved others, and it is reorganizing its operations. Furthermore, based on the limited data available, IRS appears to be realizing some increased success in convicting those involved in schemes, publicizing these results, and uncovering previously hidden major offshore compliance problems. Nevertheless, it is difficult to get a clear picture of all that is underway in IRS how much is new as opposed to reemphasized or reorganized, and how the pieces combine to form a planned, coordinated effort with specific, defined outcomes. One of the difficulties we encountered in gathering information was that no central office, group, or executive could provide us with an agencywide focus or perspective on IRS’s strategy, goals, objectives, performance measures, or program results. Responsibility for the efforts was spread across various functions and groups within IRS. To some extent this lack of clarity is not surprising given the fairly rapid and ongoing change in IRS’s efforts, the expanding scope of the problem, and the difficulty in determining the difference between what is legitimate, aggressive tax planning and an abusive tax scheme. | Estimating the extent of abusive tax schemes used by individual taxpayers is difficult because they are often hidden. Nevertheless, the Internal Revenue Service (IRS) believes that the number and dollar consequences of these schemes has grown recently. IRS estimates that 740,000 taxpayers used abusive schemes in tax year 2000. IRS caught $5 billion in improper tax avoidance or tax credit and refund claims, but estimates that another $20 to $40 billion went undetected. Recent developments suggest that the number of individuals using an abusive tax scheme involving offshore accounts may be greater than estimated and potential lost revenues may be higher than estimated. Because no one individual or office could provide an agencywide perspective on IRS's strategy, goals, objectives, performance measures, or program results, it is difficult to provide a clear picture of IRS's efforts to address abusive tax schemes. IRS has created new offices, reemphasized and reorganized earlier efforts, and plans to assign at least 200 additional staff to its efforts. Limited data suggest that IRS's enhanced efforts have helped to successfully convict those promoting and taking advantage of abusive schemes, publicize these results, and uncover previously hidden major offshore compliance problems. The number of possible abusive tax schemes, however, could outstrip IRS's audit resources. Furthermore, identifying and handling these cases will require better coordination on IRS's part. IRS has not yet developed a way to track the resources used to combat abusive schemes, nor has it developed goals and measures to assess its progress. |
An influenza pandemic can occur when an existing virus mutates into a novel strain that is highly transmissible among humans, leading to outbreaks worldwide. Such strains can be highly pathogenic because there is little or no pre-existing immunity in the population. Some of the issues associated with the preparation for and responses to an influenza pandemic are similar to those for any other type of disaster or hazard. However, a pandemic poses some unique challenges. Unlike incidents that are discretely bounded in space or time (e.g., most natural or man-made disasters), an influenza pandemic is an event likely to come in waves, each lasting weeks, months, or years, and pass through communities of all sizes across the nation and the world. While a pandemic will not directly damage physical infrastructure such as power lines or computer systems, it could threaten critical systems by potentially removing the essential personnel needed to operate them from the workplace for weeks or months. The World Health Organization (WHO) and the U.S. Centers for Disease Control and Prevention (CDC) have said that in a severe pandemic, the absences of those who are ill, taking care of ill family members, and fearing infection could reach a projected 40 percent during the peak weeks of a community outbreak, with lower rates of absence during the weeks before and after the peak. In addition, an influenza pandemic could result in 200,000 to 2 million deaths in the United States, depending on its severity. Although representing a novel strain of flu, the H1N1 outbreak, first detected in the United States around April 2009, has caused illness ranging from mild to severe. While most people who have been sick have recovered without needing medical treatment, hospitalizations and deaths from infection with this virus have occurred, and recent CDC news bulletins have indicated the second wave of the disease potentially could be more severe, especially for children and other at-risk groups. As with most disasters, the initial governmental response to a pandemic will be at the state and local level and will aim to decrease people’s exposure to the virus. Initial responses may include encouraging and facilitating good hand hygiene, requiring ill individuals to isolate themselves, educating people about conditions that put them at high risk for complications, encouraging early treatment, and encouraging creative solutions to increase the distance between people at school and work. Under conditions of increased severity of illness, government response could escalate to include more aggressive actions such as closing schools, shutting down public transportation, and prohibiting large public gatherings at venues such as sporting events. These measures are intended to create “social distance” between people to prevent large numbers of people coming into direct contact in an attempt to minimize transmission of the disease. Similarly, individual organizations are also advised to increase the distance between people in workplaces. At the federal level, the National Strategy for Pandemic Influenza Implementation Plan calls for the Secretary of HHS to lead the federal medical response to a pandemic, and the Secretary of DHS to lead the overall domestic incident management and federal coordination. Protecting the nation’s critical infrastructure against natural and manmade catastrophic events, including pandemic, has been a concern of the federal government for over a decade. Several federal policies address the importance of coordination between the government and the private sector in critical infrastructure protection. Homeland Security Presidential Directive 7 (HSPD-7), issued in December 2003, identifies various federal agencies, including DHS, as having responsibility for ensuring that steps are taken to protect specific critical infrastructure sectors of the United States. HSPD-7 makes DHS responsible for, among other things, coordinating national critical infrastructure protection efforts and establishing uniform policies, approaches, guidelines, and methodologies for integrating federal infrastructure protection and risk management activities within and across these sectors. In addition to other sectors, DHS is the lead federal agency for two critical infrastructure sectors—information technology (IT) and communications—that are important for the Internet. Specifically, the entities within DHS responsible for coordinating national efforts to promote critical infrastructure protection activities for those sectors are the National Cyber Security Division and the Office of the Manager of the National Communications System (NCS), respectively. Although the vast majority of Internet infrastructure is owned and operated by the private sector, federal policy recognizes the need to be prepared for the possibility of debilitating disruptions in cyberspace. With the exception of the Department of Defense and intelligence community networks, DHS is the central coordinator for cyberspace security efforts and has responsibility for developing an integrated public-private plan for Internet recovery. FCC, which was established under the Communications Act of 1934 to regulate interstate and international communications by radio, television, wire, satellite, and cable—also oversees the telecommunications infrastructure on which the Internet depends. Because the functioning of the financial markets is important for our nation’s economy, the financial sector is one of the infrastructure sectors that has been designated as critical. Finally, under HSPD-7, the Department of the Treasury (Treasury) is responsible for infrastructure protection activities specifically within the banking and finance sector. The public Internet infrastructure is owned and operated primarily by private companies such as telecommunications companies, cable companies, and other Internet service providers. It is a network of many networks used around the world to communicate and share computing resources, engage in commerce, do research, and provide entertainment. As shown in figure 1, the various networks that make up the Internet include the national backbone and regional networks, as well as the residential Internet access networks and the networks run by individual businesses, or “enterprise” networks. The national backbone providers transmit data over long distances using high-speed fiber-optic lines. Because these providers do not service all locations worldwide, regional network providers provide regional service to supplement the long-haul traffic. When a user wants to access a Web site or send an e-mail to someone who is connected to the Internet through a different service provider, the data must be transferred between networks. Data travels from a user’s home computer to the Internet through various means, including coaxial cable, digital subscriber line (DSL), satellite, fiber, or wirelessly to a provider’s facility where it is aggregated with other users’ traffic. Data cross between networks at Internet exchange points, which can be either hub points where multiple networks exchange data or private interconnection points. At these exchange points computer systems called routers determine the optimal path for the data to reach their destination. The data then continue through the national and regional networks and exchange points, as necessary, to reach the recipient’s Internet service provider and the recipient. A functioning Internet will be important during a pandemic because it could be one important way that governments and private entities share necessary information with the public. Using the Internet to allow people to communicate effectively without coming together physically would assist in creating “social distance” to reduce the potential for illness to further spread. In addition, many organizations, including DHS, have been advocating that businesses and other enterprises consider increased use of telework by their workforce as a way to continue operations while maintaining physical separation from other workers during a pandemic. Doing so would typically involve employees working from home and accessing their business’s networks over an Internet connection. Some entities have also advocated the use of the Internet as a means for reducing the social isolation that could arise when people are asked to avoid contact with others. For the U.S. securities markets to function, ensuring that companies can raise capital to carry on commerce and investors can obtain returns on their savings for spending on necessities or for retirement security, various organizations must be able to operate. Individual investors and institutions such as mutual funds send their orders to buy and sell stocks and options to broker-dealers that, in turn, route these orders to be executed at one of the many exchanges or electronic trading venues in the United States and abroad. After a securities trade is executed, it undergoes clearance and settlement to verify the accuracy of the transaction. Ownership of the securities is then transferred from the seller to the buyer, and the necessary payment between the two parties is exchanged. Separate organizations complete the clearance and settlement process for stocks and for options. In general, a clearing organization collects and compares trade information to ensure the accuracy of the trade and calculates the amounts that are to be exchanged between parties. A depository organization then transfers ownership and maintains the records of securities held by broker-dealers and investors. To facilitate these interactions, the large broker-dealers have accounts directly with the clearing organizations, while smaller and independent broker-dealers act as introducing firms by sending their customers’ orders to an intermediary broker-dealer, known as a clearing firm, that accepts and processes the trades and clears and settles these trades with the central clearing organization. The clearing firm’s systems also maintain the records of the cash and securities holdings of the introducing broker-dealers, and their investor customers. The monies transferred as part of securities transactions are handled by the banks that maintain accounts for broker-dealers and accept and make payments for these firms’ securities activities. Payment processing systems operated by the Federal Reserve or private firms process the payments that are exchanged between the clearing banks used by the clearing organizations, broker-dealers, and their customers. Virtually all of the information processed is transferred electronically between parties; clearance and settlement and payment transactions take place over proprietary networks that do not traverse the public Internet infrastructure. Figure 2 illustrates how these various organizations participate in a trade. Although thousands of entities are active in the U.S. securities markets, certain key organizations are more critical to the ability of the markets to function, usually because they offer unique products or perform vital services. For example, markets cannot function without the activities performed by clearing organizations and in some cases, only one clearing organization exists for particular products. In addition, other market participants are critical to overall market functioning because they consolidate and distribute price quotations or information on executed trades. The inability of any one broker-dealer firm to continue operations during an event would not likely affect the markets as a whole, but a small number of large broker-dealers generally account for sizeable portions of the daily trading volume on many exchanges. If several of these large firms were unable or unwilling to operate, the markets might not have sufficient trading volume to function in an orderly or fair way. U.S. securities markets have evolved in the last decade, with trading occurring at a larger number of venues, including existing exchanges, electronic markets, and alternative trading networks operated by broker-dealers or others. As a result, the criticality of some participants to the overall functioning of the markets likely has changed since we began reviewing these issues in 2001, but all continue to play significant roles in U.S. markets. Various regulators oversee securities market participants: SEC regulates the stock and options exchanges and the clearing organizations for those products. In addition, SEC issues rules and oversees the broker-dealers that trade on those markets and other participants, such as mutual funds, which are active investors. Self-regulatory organizations also oversee broker-dealers directly and are responsible for ensuring that their members comply with the securities laws and these organizations’ own rules. FINRA is the primary self- regulatory organization for securities firms conducting business in the United States. As part of its responsibilities, this regulator conducts examinations of its members to ensure compliance with its rules and federal securities laws. The clearing banks that maintain accounts on behalf of securities market participants are overseen primarily by two different regulators. The Federal Reserve oversees bank holding companies and state-chartered banks that are members of the Federal Reserve System. The Office of the Comptroller of the Currency examines nationally chartered banks. As we reported in a series of reports issued since the September 11, 2001 terrorist attacks, securities market organizations have made significant progress in addressing various threats with the potential to disrupt their operations. As we reported in 2007, the group of organizations that we considered critical to overall operations of the securities markets— including exchanges, clearing organizations, and payment processors— have acted to significantly reduce the likelihood of physical disasters disrupting the functioning of U.S. securities markets. For example, all these organizations had developed the capability to perform their critical functions at alternate sites geographically dispersed from their primary sites. They all also had improved their physical and information security measures. The broker-dealers and clearing services banks that account for significant trading volumes had also taken steps to increase the distances between their sites for primary and backup operations for clearance and settlement activities and established dispersed backup trading locations. Market participants have also worked with financial regulators and other organizations on other efforts to improve the overall resiliency of the financial sector; these include periodically conducting industry-wide connectivity testing from backup locations. Coordinated by the Securities Industry and Financial Markets Association and other groups, these tests verify the ability of market participants to operate through an emergency using backup sites, recovery facilities, and backup communications capabilities across the industry; and to provide participants with an opportunity to exercise and check the ability of their backup sites to successfully transmit and receive communications between the backup sites of other market participants. In the 2008 test, more than 250 organizations, including broker-dealers, markets, service bureaus, and industry utilities participated, with test participants representing more than 85 percent of normal market volume. Overall, almost 98 percent of test connections among participants were successful. Financial market organizations have also taken steps to be better prepared for physical or information security attacks. For example, DHS’s Office of Infrastructure Protection assisted some financial market organizations by conducting assessments of the physical security measures these organizations were taking to prevent damage by physical attacks, including reviewing these organizations’ facilities and their physical security measures such as surveillance, perimeter, and intrusion technologies. Officials from Treasury and representatives of selected financial markets also participated in exercises conducted by DHS that involved tabletop events that were intended to create lifelike scenarios of disasters or cyber attacks. These exercises were to help participants better understand the effect of cross-sector dependency (or interdependencies) during such events. To assist in infrastructure protection issues, representatives from a broad range of financial regulatory agencies formed the Financial and Banking Information Infrastructure Committee (FBIIC). This group meets regularly to communicate information and coordinate efforts among the financial regulators and enhance the resiliency of the financial sector. In addition, representatives of the financial trade associations and other entities share information relating to infrastructure protection among financial market participants through the Financial Services Sector Coordinating Council for Critical Infrastructure Protection and Homeland Security (FSSCC). Formed in 2002, FSSCC acts as the private sector council that assists Treasury in addressing critical infrastructure protection issues within the banking and finance sector. FSSCC works to help reinforce the financial services sector’s resilience against terrorist attacks and other threats to the nation’s financial infrastructure. FSSCC has published reports summarizing best practices and lessons learned for issues of common concern to the industry at large. Members of FSSCC also meet periodically with the financial regulators to share information about common concerns and challenges. Financial market organizations also have received consolidated information through other sources. For example, the Financial Services Information Sharing and Analysis Center (FS/ISAC) consolidates threat information for the sector. The financial sector has also taken steps to ensure that key officials from financial regulators and financial market organizations will be able to communicate during disasters. Under the Government Emergency Telecommunications Service (GETS) Program, participating staff receive a card that provides them with a code that can be dialed to increase the priority of telephone calls they place during crises. To better ensure that critical communication among financial market participants occurs, FBIIC issued an interim policy on the GETS Card Program in July 2002 that outlines how staff from financial institutions can obtain such cards. To qualify for GETS sponsorship, the FBIIC policy states that organizations must perform functions critical to the operation of key financial markets. This priority currently is only available for voice calls and not for data communications over the Internet. Another FBIIC telecommunications effort involves the FCC’s Telecommunications Service Priority (TSP) Program, which is used to identify and prioritize telecommunication services that support national security or emergency preparedness missions. Under TSP, private-sector organizations, through the sponsorship of a selected group of federal agencies, including SEC and the Federal Reserve, can have some of their key telecommunications circuits added to an inventory maintained by NCS that will provide increased priority for restoration of these key circuits in the event of a disruption. Increased use of the Internet by students, teleworkers, and others during a severe pandemic is expected to create congestion in Internet access networks that serve metropolitan and other residential neighborhoods. For example, localities may choose to close schools and these students, confined at home, will likely look to the Internet for entertainment, including downloading or “streaming” videos, playing online games, and engaging in potential activities that may consume large amounts of network capacity (bandwidth). Additionally, people who are ill or are caring for sick family members will be at home and could add to Internet traffic by accessing online sites for health, news, and other information. This increased and sustained recreational or other use by the general public during a pandemic outbreak will likely lead to a significant increase in traffic on residential networks. If theaters, sporting events, or other public gatherings are curtailed, use of the Internet for entertainment and information is likely to increase even more. Furthermore, the government has recommended teleworking as an option for businesses to keep operations running during a pandemic. Thus, many workers will be working from home, competing with recreational and other users for bandwidth. According to a DHS study and Internet providers, this additional pandemic-related traffic is likely to exceed the capacity of Internet providers’ network infrastructure in metropolitan residential Internet access networks. Residential Internet users typically connect their computers to their Internet service providers’ network through a modem or similar Internet access device. These Internet access devices route home users’ traffic to a network device that aggregates it with that of other users before forwarding it to the other parts of the provider’s network and its ultimate destination on the Internet. As shown in figure 3, the traffic aggregating device differs depending on the technology used for Internet access—DSL, a cable network, or other means. But all these technologies use network architectures that basically aggregate the traffic of multiple users on a single device that then routes it to other parts of the providers’ networks. For example, within a DSL network architecture, the user’s traffic travels on a dedicated pair of copper wires from a home computer to the provider’s location—usually known as a central office—which houses a device called the digital subscriber line access multiplexer (DSLAM). The DSLAM aggregates this traffic and that of other users of this provider from individual residential neighborhoods before sending it on to regional networks and eventually to the national Internet backbone. Traffic from home users who connect to the Internet through a cable provider moves from the home computer over coaxial cables and fiber optic cables then ultimately to a network device known as a cable modem termination system (CMTS). The CMTS also aggregates this traffic with that of other users from other individual residential neighborhoods and sends it to the regional networks and the national Internet backbone. During a pandemic, congestion is most likely to occur in the traffic to or from the aggregation devices that serve residential neighborhoods, interfering with teleworkers’ and others’ ability to use the Internet. Congestion affecting home users is likely to occur because the parts of providers’ DSL, cable, satellite, and other types of networks that provide access to the Internet from residential neighborhoods are not designed to carry all the potential traffic that users could generate in a particular neighborhood or that all connect to a particular aggregating device for efficiency and cost reasons. Providers do not build networks to handle 100 percent of the total traffic that could be generated because users are neither active on the network all at the same time, nor are they sending maximum traffic at all times. Instead, providers use statistical models based upon past users’ patterns and projected growth to estimate the likely peak load of traffic that could occur and then design and build networks based on the results of the statistical model to accommodate at least this level. According to one provider, this engineering method serves to optimize available capacity for all users. For example, under a cable architecture, 200 to 500 individual cable modems may be connected to a provider’s CMTS, depending on average usage in an area. Although each of these individual modems may be capable of receiving up to 7 or 8 megabits per second (Mbps) of incoming information, the CMTS can transmit a maximum of only about 38 Mbps. Providers’ staff told us that building the residential parts of networks to be capable of handling 100 percent of the traffic that all users could potentially generate would be prohibitively expensive. A 2007 DHS study that was conducted in cooperation with various government, communication sector, and financial sector entities used modeling of residential and other network configurations to confirm that the increased traffic generated in neighborhoods during a severe pandemic is likely to exceed the capacity of the providers’ aggregation devices in metropolitan residential neighborhoods. The study examined the technical feasibility of the pandemic telecommuting strategy advocated by the government. The study also focused on identifying action plans to better prepare the nation for telecommuting during an influenza pandemic. As part of the study, a model was developed using data and assumptions from a large U.S. metropolitan area to represent a typical Internet provider’s network configuration, including devices and network capacities. For cost reasons, the study used DSL network architecture for the purposes of the congestion modeling, but the preparers acknowledged that other means of accessing the Internet had similar architectures and thus the impact of a pandemic would be similar. The contractors that prepared the study simulated Internet traffic in amounts that corresponded to the level of Internet use in a residential neighborhood under three scenarios of pandemic severity—20, 40, and 90 percent absenteeism from the workplace. The study’s model predicted that at the 40 percent absenteeism level—the level that health organizations have indicated is likely under a relatively serious pandemic—the highest point of congestion across the entire Internet infrastructure could occur within residential Internet access networks. Specifically, at the 40 percent absenteeism level, the study predicted that most users within residential neighborhoods would likely experience congestion when attempting to use the Internet. Based on our assessment of the study, we concluded that the methodology applied and the likely congestion points identified were reasonable. Furthermore, communication sector representatives we interviewed confirmed the likelihood of Internet congestion between a user’s home and the point at which that traffic combines with other users at the providers’ aggregation devices. Although this study assessed the impact on a large city, the severity of congestion could vary across neighborhoods or nationally depending on the capacities of residential neighborhood Internet access networks, with cities or areas with larger populations and higher incomes generally having large broadband capacities and less-populated rural or poorer areas possibly having less broadband capacity. However, the study used typical telecommunications network configurations for a large U.S. city and found that congestion was likely. As a result, we believe that its findings mean that most other locations in the United States could experience similar problems. Although predicting that the most severe congestion would occur within residential access networks, the study overseen by DHS also noted that pandemic-related congestion was possible in other parts of the networks that comprise the Internet. For example, users could experience congestion at the point at which traffic is transferred between service providers because of potential differences in transmission capacity. Additionally, teleworkers connecting to their companies’ networks (the “enterprise” networks) could overload various components of these networks, such as the devices that provide security—firewalls—or servers that provide access to various applications because some businesses’ networks may not have scaled these devices to accommodate the anticipated increase in telecommuting traffic during a pandemic. The steps being taken by financial organizations to ensure their enterprise networks are prepared for pandemic levels of use are discussed later in this report. Providers’ options for addressing expected pandemic-related Internet congestion include providing extra capacity, using network management controls, installing direct lines to organizations, temporarily reducing the maximum transmission rate, and shutting down some Internet sites. Each of these methods is limited either by technical difficulties or questions of authority. In the normal course of business, providers attempt to address congestion in particular neighborhoods by building out additional infrastructure—for example, by adding new or expanding lines and cables. Internet provider staff told us that providers determine how much to invest in expanding network infrastructure based on business expectations. If they determine that a demand for increased capacity exists that can profitably be met, they may choose to invest to increase network capacity in large increments using a variety of methods such as replacing old equipment and increasing the number of devices serving particular neighborhoods. Providers will not attempt to increase network capacity to meet the increased demand resulting from a pandemic, as no one knows when a pandemic outbreak is likely to occur or which neighborhoods would experience congestion. Staff at Internet providers whom we interviewed said they monitor capacity usage constantly and try to run their networks between 40 and 80 percent capacity at peak hours. They added that in the normal course of business, their companies begin the process to expand capacity when a certain utilization threshold is reached, generally 70 to 80 percent of full capacity over a sustained period of time at peak hours. However, during a pandemic, providers are not likely to be able to address congestion by physically expanding capacity in residential neighborhoods for several reasons. First, building out infrastructure can be very costly and takes time to complete. For example, one provider we spoke with said that it had spent billions of dollars building out infrastructure across the nation over time, and adding capacity to large areas quickly is likely not possible. Second, another provider told us that increasing network capacity requires the physical presence of technicians and advance planning, including preordering the necessary equipment from suppliers or manufacturers. The process can take anywhere from 6 to 8 weeks from the time the order is placed to actual installation. According to this provider, a major constraint to increasing capacity is the number of technicians the firm has available to install the equipment. In addition to the cost and time associated with expanding capacity, during a pandemic outbreak providers may also experience high absenteeism due to staff illnesses, and thus might not have enough staff to upgrade network capacities. Providers said they would, out of necessity, refrain from provisioning new residential services if their staff were reduced significantly during a pandemic. Instead, they would focus on ensuring services for the federal government priority communication programs and performing network management techniques to re-route traffic around congested areas in regional networks or the national backbone. However, these activities would likely not relieve congestion in the residential Internet access networks. Providing critical employees direct connections that bypass residential congestion may be another option for facilitating telework during a pandemic, but this option can be cost prohibitive to employers and is not widely used. Specifically, some providers offer network solutions such as private lines to businesses and governments. Private line services allow businesses to run their corporate networks and applications separately from public Internet traffic and could provide a point-to-point dedicated path between teleworkers’ homes and offices, bypassing the residential neighborhood congestion points. However, according to provider staff we spoke with, installing private lines in a residence requires advance planning and is expensive. One provider noted that a direct connection is not a solution that can be invoked when the pandemic strikes. In the current network environment, providers’ capability to address pandemic-related Internet congestion by prioritizing certain users’ traffic, including that of financial sector teleworkers, is limited. Specifically, provider systems are not designed to identify and provide priority to individual users when traffic is routed over the Internet and multiple networks are used for the connection. Furthermore, Internet providers’ networks also are not currently designed to identify particular types of customers connected to the Internet. For example, the networks cannot distinguish between critical employees teleworking and recreational users. Providers identified one technically feasible alternative that has the potential to reduce Internet congestion during a pandemic, but raised concerns that it could violate customer service agreements and thus would require a directive from the government to implement. Although providers cannot identify users at the computer level to manage traffic from that point, two providers stated that if the residential Internet access network in a particular neighborhood was experiencing congestion, a provider could attempt to reduce congestion by reducing the amount of traffic that each user could send to and receive from his or her network. Such a reduction would require adjusting the configuration file within each customer’s modem to temporarily reduce the maximum transmission speed that that modem was capable of performing—for example, by reducing its incoming capability from 7 Mbps to 1 Mbps. However, according to providers we spoke with, such reductions could violate the agreed-upon levels of services for which customers have paid. Therefore, under current agreements, two providers indicated they would need a directive from the government to take such actions. Shutting down specific Internet sites would also reduce congestion, although many we spoke with expressed concerns about the feasibility of such an approach. Overall Internet congestion could be reduced if Web sites that accounted for significant amounts of traffic—such as those with video streaming—were shut down during a pandemic. According to one recently issued study, the number of adults who watch videos on video- sharing sites has nearly doubled since 2006, far outpacing the growth of many other Internet activities. However, most providers’ staff told us that blocking users from accessing such sites, while technically possible, would be very difficult and, in their view, would not address the congestion problem and would require a directive from the government. One provider indicated that such blocking would be difficult because determining which sites should be blocked would be a very subjective process. Additionally, this provider noted that technologically savvy site operators could change their Internet protocol addresses, allowing users to access the site regardless. Another provider told us that some of these large bandwidth sites stream critical news information. Furthermore, some state, local, and federal government offices and agencies, including DHS, currently use or have plans to increase their use of social media Web sites and to use video streaming as a means to communicate with the public. Shutting down such sites without affecting pertinent information would be a challenge for providers and could create more Internet congestion as users would repeatedly try to access these sites. According to one provider, two added complications are the potential liability resulting from lawsuits filed by businesses that lose revenue when their sites are shutdown or restricted and potential claims of anticompetitive practices, denial of free speech, or both. Some providers said that the operators of specific Internet sites could shut down their respective sites with less disruption and more effectively than Internet providers, and suggested that a better course of action would be for the government to work directly with the site operators. Providers could help reduce the potential for a pandemic to cause Internet congestion by ongoing expansions of their networks’ capacities. Some providers are upgrading their networks by moving to higher capacity modems or fiber-to-the-home systems. For example, some cable providers are introducing a network specification that will increase the download capacity of residential networks from the 38 Mbps to about 152 to 155 Mbps. In addition to cable network upgrades, at least one telecommunications provider is offering fiber-to-the home, which is a broadband service operating over a fiber-optic communications network. Specifically, fiber-to-the-home Internet service is designed to provide Internet access with connection speeds ranging from 10 Mbps to 50 Mbps. Although not generally feasible in the current environment, the ability to prioritize individual user’s traffic is envisioned to be technically possible in future upgrades of the infrastructure of the Internet and telecommunications networks, but such capabilities are estimated to be years away. As we recently reported, DHS is working with international standards bodies to help develop standards that could allow greater flexibility to prioritize data communications in the future; this effort is a part of what is referred to as the Next Generation Networks. However, these capabilities are not expected to be ready for several years due to the complexity of the systems and the need to develop standards that work across varying providers’ infrastructures, including internationally. In addition, we reported DHS had difficulty getting its full budgets approved, which may have contributed to the delay in developing standards. As a result, the expanded features of this newer network architecture are not expected to be a viable solution for addressing pandemic-related Internet congestion in the near future. Although responsible for coordinating protection of the communications- critical infrastructure sector, which includes the networks that comprise the Internet, DHS has not yet developed a strategy to address pandemic- related Internet congestion, coordinated with federal partners, determined if sufficient authority exists to take necessary actions, or assessed the need for a public communications campaign to minimize congestion that is expected to occur during a pandemic. Under HSPD-7 and the National Strategy to Secure Cyberspace, DHS is the lead agency for coordinating the protection of critical assets in the communications sector from attacks. Also under these authorities, DHS is responsible for facilitating a public-private response to the recovery from major Internet disruption. In addition to being a focal point to the cyber-critical infrastructure protection effort, DHS has been designated as one of two federal agencies responsible for coordinating the United States’ pandemic response. As specified in the Implementation Plan for the National Strategy for Pandemic Influenza, DHS is to coordinate the nation’s response in conjunction with HHS. DHS has undertaken several pandemic planning activities. As discussed earlier in this report, DHS and representatives from the government, communications sector, and financial sector conducted a study to assess specifically the technical feasibility of the pandemic telecommuting strategy and identify ways for the nation to better prepare to support the strategy. In coordination with interagency partners and the critical infrastructure sector coordinating councils, DHS has completed individual sector-specific pandemic guidelines and provided Webinars to sector partners on their respective plans. These guidelines are intended to assist the sectors and businesses with the sectors’ plan for a severe influenza pandemic, and include some consideration of potential Internet congestion. For example, the guidelines for the information technology and communications sectors recommend that entities in these sectors consider advising employees to limit household use of streaming video or other bandwidth-intensive Internet activities. The guidelines also recommend consideration of obtaining multiple means of accessing the Internet. The guidelines have been provided to the sector coordinating councils via a secure DHS information portal, as well as to the members of the National Governor’s Association. DHS officials told us that some of the sectors have made the guidelines available to the public. More recently, DHS completed the DHS 2009-H1N1 Implementation Plan, which provides planning guidance for DHS and identifies specific roles and responsibilities for the DHS components such as the Office of Policy or the Transportation Security Administration. According to DHS officials, the plan also directs all DHS components to develop plans that address key preparation and response actions, performance of mission essential functions, workforce protection, continuity of operations, and communications with key stakeholders during the H1N1 influenza pandemic. However, while these planning activities are designed to help government and private sector partners prepare for a pandemic, they are limited in addressing the anticipated Internet congestion. Although serving as the coordinating agency for Internet recovery and pandemic response, DHS staff told us that their agency does not have a strategy to address Internet congestion. According to DHS staff, their agency has not begun developing such a strategy because since the September 11 terrorist attacks, they have had other crises such as Hurricane Katrina to address. A senior official at a financial markets regulator told us that leadership by the government had been lacking in addressing this potential risk to the financial sector. Without action by DHS to address this potential congestion, employees in critical sectors of the nation’s economy, including those in financial services, might not be able to effectively telework or otherwise communicate or transmit data over the Internet. In addition, although various federal and private sector organizations would likely be required to coordinate an effective Internet congestion response strategy, DHS has neither reached out nor coordinated with other partners, such as other federal or state agencies with telecommunications oversight authorities, to prepare such a strategy. As we previously reported, the experience of Hurricane Katrina showed the need to improve leadership at all levels of government in order to better respond to a catastrophic disaster. As part of this, the legal authorities, roles and responsibilities, and lines of authority at all levels of government must be clearly defined, effectively communicated, and well understood in order to facilitate rapid and effective decision making. In order to respond effectively to pandemic-related Internet congestion, DHS will need to effectively plan and work with other parts of the federal government and possibly state and local governments and the private sector in its efforts. Other organizations that could be relevant include FCC, which, as previously noted, is charged with regulating interstate and international communications by radio, television, wire, satellite, and cable. DHS staff representing the Office of Policy acknowledged that such coordination would be necessary to address Internet congestion effectively and ensure that the various parts of the federal government are not conducting conflicting activities. For example, the staff told us the Department of Education was hoping to have schools use the Internet during a pandemic to allow students to access remote learning courses if schools were closed. The staff acknowledged that, as a result, DHS would have to coordinate with the other relevant agencies to ensure that their various actions are appropriately taken into account in developing a congestion plan. According to DHS staff, DHS has engaged in dialogues with other agencies about pandemic-related issues on a regular basis. Agency staff once again cited time constraints and the need to focus on other crises as reasons for not having discussed the development of a coordinated strategy for addressing Internet congestion. However, unless DHS starts coordinating with other federal, state, and even private sector parties on possible Internet congestion solutions, there may not be sufficient time to develop a coordinated strategy to address a rapidly emerging severe pandemic. Further, although an effective congestion response strategy could require directing the private sector entities that operate the Internet’s infrastructure today to take actions that could negatively affect users, DHS has not determined whether it or other agencies have the necessary authorities to require providers to take such actions. We previously reported that the authorities of federal government agencies regarding the Internet were unclear. Given the importance of the Internet infrastructure to our nation’s communications and commerce, we suggested that Congress consider clarifying the legal framework guiding Internet recovery. Although DHS staff identified a list of potential authorities that may or may not apply, they told us they were not able to specify whether their agency had clear or specific authority to require telecommunications providers to take actions to address congestion, such as reducing customer transmission speeds or blocking entertainment Web sites. Instead, DHS’s approach would be to assess the authorities as part o the development of any such strategy. While this approach could help DH determine at some point if it or some other relevant federal agency had adequate authority to address potential Internet congestion, it would increase the risk that the federal government will not be able to respond rapidly or effectively if a pandemic quickly emerges. Other federal government agencies might have authority to direct providers to take certain actions during a pandemic, but whether these are adequate is uncertain. Under the Communications Act of 1934, as amended (the Act), FCC has authority to regulate the telecommunications providers specifically and has authority generally with respect to interstate and foreign communication by wire and radio. According to FCC staff, there may be actions the FCC could take regarding the Internet to address threats to national security or public safety. However, in commenting on a draft of this report, FCC officials noted that there is an ongoing court challenge to FCC’s authority regarding the Internet. In addition, FCC staff were not sure whether FCC would have sufficient authority to require private sector organizations to take all actions that may be deemed necessary in an emergency situation to relieve congestion and facilitate commerce, including teleworking by financial sector employees. As part of preparing a national broadband access plan, FCC has recently sought public comments on options for prioritizing Internet traffic in a pandemic. According to FCC staff, very few comment letters addressed the prioritization issue. Based on our review, some service providers expressed interest in the government considering including a prioritization scheme in the plan. Additionally, one provider suggested the plan should give providers flexibility to actively manage networks during a pandemic. Finally, one financial sector organization noted that the plan should include a prioritization scheme to prioritize Internet traffic based on how critical it is to national and economic security. Some observers have suggested that an authority granted to the President in the Communications Act of 1934 could conceivably be used to take actions to address Internet congestion during a pandemic. In their view, the President may have, under certain limited circumstances involving a state or threat of war, the power to authorize government control of the telecommunications systems and, if properly invoked and delegated, this might broadly provide authority for the government to require private sector entities to take actions intended to address congestion. However, according to FCC staff we spoke with, while the authority under the Act may grant the President powers over telecommunication systems during wartime, they did not know whether such powers could be exercised in a pandemic. However, until DHS, as the lead agency responsible for coordinating protection of telecommunications, including the Internet, takes action to work with other agencies to assess whether sufficient authorities exist to direct necessary actions by the private sector, the potential for a timely and effective federal response to congestion is reduced. Although its own study identified voluntary public reduction of Internet use as an effective means of reducing pandemic congestion, DHS has not begun steps to assess the feasibility and effectiveness of obtaining such public cooperation. According to the DHS study and to providers and others we spoke with, voluntary actions taken by the general public could have significant potential to reduce the surges in traffic loads that residential users may experience during a pandemic. For example, the general public could be asked to limit video streaming, gaming, and peer- to-peer and other bandwidth-intensive applications during daytime work hours. They could also be encouraged to use broadcast news sources in place of online news. A similar campaign developed by another agency— HHS—to publicize pandemic awareness strategies showed that such public education efforts can require months to prepare and cost millions of dollars to test and implement. For example, as part of creating various radio and television messages to provide information to the public about how to prepare for a pandemic, HHS conducted market research using various techniques, including focus groups, to gauge the public’s opinion about a pandemic. In 2005-2006, when they began this effort, HHS staff stated that it took the agency about 6 months to develop the public service announcements (PSA). In 2006-2007 HHS staff spent about 4 months planning and producing PSAs. The cost of running radio PSAs in 137 cities over an 11-month period in 2007 was about $1.5 million dollars. DHS staff acknowledged that such a campaign would also require cooperation and coordination among multiple federal and other agencies to be effective and avoid conflicting goals and activities. For example, agencies would need to work together to ensure that some were not planning to recommend increased use of the Internet to provide information, education, or for other purposes during a pandemic. For example, HHS may advocate using the Internet to maintain social ties during a pandemic, which would make the goal of easing congestion by staying off-line more challenging. However, DHS staff told us they had not begun efforts to evaluate the feasibility or effectiveness of such a campaign or taken steps to begin developing such an effort because other activities supporting its operational mission have taken priority. Until DHS takes such action, its ability to implement what its own study predicted would be an effective tool for reducing potential Internet congestion in a timely fashion is reduced. We reviewed seven organizations whose operations are critical to the overall functioning of U.S. securities markets and found that all have developed formal plans that address key elements of pandemic preparedness. But some have limitations that could increase the risk that aspects of their operations would be disrupted by a pandemic. In response to our last report, SEC and the banking regulators issued guidance to key financial market participants stipulating that an institution’s pandemic plan, at a minimum, must include the following five key elements: 1. a process for monitoring the pandemic’s progress and a series of escalating response steps as various pandemic phases are reached; 2. a preventive program to minimize, to the extent possible, illness among employees, including social distancing of employees by curtailing meetings; 3. a documented strategy of facilities or procedures designed to allow the organization to continue its critical operations in the event that large numbers of its staff are unavailable for prolonged periods; 4. a testing program to better ensure that the practices and capabilities that an organization implements to address a pandemic will be effective and allow it to continue its critical operations; and 5. an oversight program to ensure ongoing review and updates to the pandemic plan. All seven of the critical financial market organizations we reviewed have developed formal pandemic plans that call for them to monitor a pandemic’s progress and take escalating steps as the phases of a pandemic outbreak progress. Health authorities, including WHO and CDC, have issued phased timelines that track the progress of a pandemic from earliest detection to widespread global illness. Because being able to operate effectively at the height of a pandemic could require an organization to have taken steps in advance, an effective pandemic plan should contain more and stronger measures that would be taken as the phases of the pandemic progress. Such a strategy provides sufficient time to take steps that require more planning or lead time, such as purchasing needed supplies or conducting training in advance of the actual pandemic. Gradually implementing responses as the pandemic progresses also could prevent organizations from generating undue expenses if what appears to be a pandemic early on does not turn out to be one that significantly disrupts operations. Our analysis of the seven critical organizations’ pandemic plans showed that each included activities that escalated as the pandemic progressed. All the organizations are currently monitoring the information regarding the potential spread of viruses that could lead to a pandemic through the CDC or WHO Web sites and communicate closely with local authorities, such as the New York City Office of Emergency Management. In the early stages of a pandemic these organizations would take preventive actions, such as monitoring the world pandemic situation and creating awareness of wellness practices before widespread outbreaks begin (i.e., WHO Phases 1 through 3). But as the pandemic levels advance, the organizations’ plans generally call for them to implement more extensive responses, such as relocating staff to increase social distancing or sending some staff home to telework. For example, one organization’s pandemic plan describes efforts to impose business travel restrictions; prepare additional communications to employees, customers, and regulatory bodies; and stock up on additional critical supplies during WHO Phase 4 in case a pandemic disrupts supply chains. As the alert level rises to Phase 5, the plan escalates the actions to initiate daily absenteeism tracking, expand the deployment of hand-sanitizing gel, and do additional facility cleaning. When WHO declares a pandemic (i.e., WHO Phase 6), the organizations will take steps to implement social distancing, such as sending a number of employees to the backup facility and designating people to work from home. All of the plans follow this general design, and during the H1N1 outbreak, all the organizations began implementing some of these steps. In particular, as the alert level escalated from WHO Phase 4 to Phase 5 in April of 2009, several organizations communicated to staff on additional measures they were taking, which included placing more hand sanitizers in the workplace and cleaning facilities more often. As WHO raised its pandemic phase further to the highest level (i.e., WHO Phase 6), indicating that a broad outbreak of an influenza epidemic was believed to be imminent, organizations, according to SEC staff, were prepared to take further steps that correspond with an outbreak—such as performing medical screenings of staff reporting to work—although such measures ultimately were not necessary due to the milder nature of the H1N1 outbreak here in the United States. As a result of their experiences with the recent H1N1 flu outbreak, some market organizations and financial regulators told us they were considering developing modified trigger points in the plans that might not follow the WHO designations exactly. Officials from these organization said they had made this decision because of their experience with the relatively benign nature of the H1N1 virus in the United States. The health authorities’ pandemic phases were designed for a disease that causes high levels of severe illness, and even deaths, like some of the previous flu pandemics have caused. However, even though the United States continues to report the largest number of novel H1N1 cases of any country worldwide, most people who have become ill in 2009 have recovered without requiring medical treatment. As a result, staff from several of the critical market organizations did not need to fully implement their plans at that time because their employees were not seriously ill, if at all, and the plans could be modified to adapt to such a scenario. Our analysis indicated that all seven critical organizations also had fully addressed another key element of pandemic planning by instituting preventive programs intended to reduce the impact of a pandemic on their organizations. Because an organization has a much greater chance of continuing operations during a pandemic if fewer of its employees are ill, an effective pandemic plan should include a preventive program to reduce the likelihood of employees becoming sick. The steps the organizations took included providing information and educational campaigns to keep employees informed of pandemic news and developments. For example, during the recent H1N1 outbreak, staff at these seven organizations developed memos to employees on the status of the outbreak and steps the organizations were taking based on news and briefings from the federal, state, and local authorities. Further, all the organizations have developed internal Web sites to educate employees on general information on preventing spread of disease, including hand-washing techniques and coughing etiquette and provided personal hygiene items such as hand sanitizers and masks. In addition, three of the organizations prepared extensive education outreach campaigns (e.g., hand-washing awareness week) shortly after the financial regulators’ pandemic planning requirements were issued, in mid-2007. Most of the organizations have also developed policies regarding restricting travel as a way to reduce illness among their employees. For example, the organizations’ plans typically called for curtailing international travel at WHO Phase 4, and some required staff returning from abroad to quarantine themselves for a period, such as 7 days, to lower the chance of spreading illness. All seven critical securities market organizations we reviewed have developed plans with procedures intended to allow them to continue the functions critical to their operations despite high levels of absenteeism, but not all have fully analyzed or thoroughly documented their staffing levels or developed formal alternatives if teleworking proves unfeasible due to Internet congestion. Although congestion during a pandemic could interfere with individuals’ ability, including teleworkers and others, to access the Internet, the primary communications of the critical markets organizations would not be affected because these organizations and their participants communicate via high-capacity, proprietary networks that do not traverse the public Internet infrastructure. According to the health authorities, one of the most significant challenges of a pandemic will be staffing shortages due to absenteeism caused by employees either too ill to work, taking care of ill family members, or afraid to come to work because of the chance of infection. Unfortunately, organizations could also permanently lose critical staff if the pandemic causes significant levels of deaths. Therefore, a responsive pandemic plan should include procedures for ensuring that an organization can continue performing its critical functions even with as much as a 40 percent reduction in its workforce for a prolonged period—the level that the federal government has advised should be used for planning for a severe pandemic. In general, the seven critical organizations that we reviewed all intend to use existing geographically dispersed facilities to increase the distance among staff who perform critical functions. Staff from all seven critical organizations are spread among facilities located across the United States, including data centers, which are monitored by computer operators, and office or business centers with key staff that assist customers. Each of these organizations has created duplicate sites with redundant staffed data centers and locations or space for other critical staff. For example, officials from one organization told us that their three facilities are considerably distant from each other (i.e., hundreds of miles) in order to mitigate the effect of natural disasters, power and telecom outages, and other wide-scale regional disruptions, including a pandemic. The organizations plan to use these geographically dispersed sites to maximize social distancing and increase their ability to continue operating during a pandemic. Having sites with staff that perform critical functions in more than one location also provides these organizations with pools of cross- trained employees that they can draw on during a pandemic. For example, one organization’s pandemic plan relies on staff performing critical activities that are evenly divided across two geographically distant facilities in different regions of the country. This organization also has an alternate facility in the same metropolitan area as its primary location. Under its plan, during the final stage of a pandemic, when the United States is experiencing sustained transmission of the disease, some staff from its primary site are to report to the nearby alternate facility to do their critical activities, thus allowing the organization to increase the physical distance between the individual members of its critical staff. Although each organization has developed plans for continuing operations during a pandemic, our analysis indicated that three of the seven have not fully analyzed or documented the number of staff able to perform critical functions who would be available during a pandemic. With the federal government indicating that organizations should plan for absenteeism of 40 percent at the peak of a severe pandemic, under such circumstances approximately one in every three of an organization’s employees could be ill or caring for ill family members. Although regulators’ guidance does not specify the extent of cross-training required, we believe that, at a minimum, an organization would need two staff capable of performing each critical activity to allow for one to be absent while the other continues working. Organizations should probably have three staff capable of performing or cross-trained to take over these tasks to provide additional assurance that enough staff would be available. For example, the federal guidance on continuity of operations planning recommends that organizations should probably have three staff capable for key positions. Because these organizations have multiple operating sites with staff located in each that are capable of performing many of their critical activities, they have some assurance that they likely have enough employees to continuing operating during a pandemic. But, not all organizations have fully analyzed or documented the number of staff that could be available across all critical positions and tasks. All the organizations have identified their critical functions and all have lists of at least some of the essential staff for each of the departments performing those functions. Four critical organizations have developed lists that show the current staff for each critical function, backup staff, and sufficient numbers of staff who are cross-trained or already know these functions who could serve as additional backup support. One of these organizations rotates the performance of its critical functions through three geographically distant operation sites on an ongoing basis, ensuring a large group of cross-trained staff. For example, this organization has a list of 36 staff for one of the critical departments, all of whom are trained to perform functions normally requiring 8 staff. Thus the organization has 8 backup staff as well as 20 additional trained staff that it can draw upon. Another of the four organizations identified seven essential services that its organization needs to perform and prepared listings for each of these departments that identify the primary staff performing the functions, the backups for these staff, and additional staff that are knowledgeable or cross-trained to perform these duties. For example, one of the essential departments has a list of 19 staff that are trained to perform one set of critical functions that normally require only 5 employees—a surplus of 14. In addition, this organization cross-trained an additional 7 staff to serve as further backup support. Henceforth, these organizations identified additional staff beyond the primary and backup employees for each critical function—producing more than two staff capable of performing each critical activity—to have greater assurance of being able to perform their critical functions. The importance of sufficiently analyzing and documenting the adequacy of critical staffing was demonstrated by one of the critical organizations that has comprehensively identified its staff and backups. This organization participated in an industry-wide pandemic exercise that revealed it needed to identify even larger numbers of trained staff for some departments. The exercise simulated the impact of a pandemic by declaring that all staff with last names beginning with certain letters would be unavailable for work. Although at one point in the exercise the scenario called for 40 to 50 percent absenteeism, this organization found that in one of its critical departments, as many as 78 percent of its staff were projected to be unavailable. As a result, this organization has re-examined its staffing arrangement to identify staff that currently perform other activities that could be used to perform critical functions if needed. The results from the exercise demonstrated the need to determine, in advance of an outbreak, sufficient numbers of staff capable of performing critical functions. In contrast, three of the seven critical organizations have not fully developed lists of staff capable of performing critical functions. For example, at one organization each critical department listed essential staff, but only at a managerial level (e.g., vice president of a department, and one backup) but did not identify staff that perform the department’s functions on a day-to-day basis. The other two organizations created lists of essential staff by department, but the lists were completed only during the recent H1N1 outbreak rather than in advance. None of these three organizations listed primary, backup, or other staff for the critical functions. Officials at one of these organizations told us they have staff at several geographic locations and that business continuity tests for one of their critical departments demonstrated they can operate their organization’s critical information systems. As a result, they said that the geographic distance among locations and testing efforts provided them with a group of cross-trained staff that would be sufficient to continue operations even if 40 percent were absent. While this provides some assurance that this organization may be able to withstand a pandemic, as one organization learned, undergoing more extensive analysis and documentation allows organizations to identify gaps in staffing levels that would be unique to a pandemic, when large numbers of staff could be unavailable for prolonged periods. In addition, such analyses identify all critical tasks and those staff capable of performing them—primary, backup, and additional cross-trained staff—providing these organizations with greater assurance that adequate numbers of staff exist for each task within its critical departments. Until these organizations fully document their staffing analyses to ensure they have sufficient depth of staff capable of performing critical functions, some aspects of these organizations’ operations may be affected during a pandemic. In addition to better analyzing and documenting their staffing plans, some of the organizations that intend to use teleworking as part of their strategy for continuing operations during a pandemic need to address limitations in their teleworking plans. As noted previously, the critical market organizations included in our review generally rely on proprietary communications networks that will not likely be affected by any pandemic-related congestion. However, five of the seven critical organizations plan to have some of their critical staff telework during a pandemic, and the readiness of these organizations to successfully have employees telework varies. Based on our reviews, only one of the five organizations fully developed suitable alternatives to teleworking in case of Internet congestion. This organization identified hotels with increased broadband Internet access capability in the employees’ residential neighborhoods that staff could report to in order to improve their ability to telework. Another of the five organizations developed a plan for some of the critical staff that would be teleworking to come into one of its facilities that is currently prepared as a backup site. This facility is currently ready for operations and has ample space to provide adequate social distance for employees that find they cannot successfully telework due to congestion. However, the organization has not made adequate preparations for some critical staff in another geographic area to telework during a pandemic. If these employees are not able to telework, the organization plans to have them report to its office there and work in an unused part of the facility. But it has not outfitted this area with additional workstations that would allow its staff to work there effectively. Furthermore, three of the critical organizations whose plans include possibly having some of their critical employees telework have not fully developed plans for alternatives to teleworking should congestion arise. Our review of their plans show that the three organizations have not designated the necessary positions or employees who would telework. Determining the total number of teleworkers in advance of an outbreak would allow the organizations to confirm that their network systems can fully support that number, which would likely be higher than it might be in the course of a normal work day, and that these employees have full access to all the applications or systems they need in order to perform their critical duties effectively from home. These organizations have also not developed and assessed the feasibility of alternatives to teleworking in their plans. For example, one of these organizations told us that, if congestion occurs, they would bring staff back into their facilities and have them conduct their work wirelessly. However, they have not documented this in their planning or tested the feasibility of this approach for all potential critical activities. The other two organizations have not determined in their plans what steps they would take to respond to congestion problems experienced by their teleworking employees. Until all the critical organizations develop additional measures to ensure they have viable alternative strategies if teleworking proves difficult, they might be at greater risk of having some aspects of their operations disrupted during a pandemic. Our analysis shows that while all seven of the critical organizations we reviewed participated in an industry-wide pandemic scenario test, some have not conducted similar tests internally. All of the organizations reported that they participated in a 3-week industry-wide pandemic exercise, sponsored jointly by FBIIC, FSSCC, Treasury, and the Securities Industry and Financial Markets Association, which began in September 2007. The exercise simulated a pandemic occurring in three waves and reaching an absenteeism rate as high as 49 percent. As previously mentioned, each scenario update included an absenteeism distribution specified by first letters of employees’ last names, as a way to approximate the scenario’s target absenteeism rate. The scenario updates were provided to participants 1 week in advance so that each organization had adequate time to review its human resources records, identify the absent individuals, and determine the distribution of the absent employees among their various departments and units as appropriate. This method provided a probable picture of the range of absent employees, which could be from the lowest levels to the top of an organization. Organizations that did not want to carry out such a review of their records were allowed to simply use the provided absenteeism rate (25 percent, 49 percent, and 35 percent) for each scenario update. Officials from the critical organizations indicated that the exercise was useful in planning for a possible pandemic. As noted previously, one organization participating in this exercise experienced as much as 78 percent absenteeism in some of its departments—higher than the expected 49 percent projection—and has taken steps to identify additional staff capable of performing its critical functions. Officials from another organization said the exercise highlighted variation in human resource policies—for example, in the distribution of antiviral medication and the use of hazard pay across regional offices. As a result, the organization convened relevant staff to discuss consistent policy issues and infrastructure resiliency across regions. In addition to the industry-wide effort, three of the organizations have conducted or plan to conduct additional internal pandemic tests to ensure readiness. One of the organizations has conducted pandemic exercises for managers and staff at each facility, using a set of questionnaires corresponding to various scenarios. Another organization told us that it planned to conduct a full- day pandemic response test at all of its facilities in 2009. However, four of the organizations have not run internal pandemic scenario tests. As discussed earlier, the results from the industry-wide test demonstrated the need for the critical organizations to assess their staffing, backup, and cross-training levels to ensure they are sufficient to meet the organization’s needs during a pandemic. Internal pandemic scenario tests would give organizations just such an opportunity. In addition to pandemic scenario testing, all seven organizations have tested their abilities to run critical applications and functions at their alternate backup facilities as part of their business continuity testing. These efforts will provide further assurance that these sites will be viable for use during a pandemic. Some of the organizations rotate operations between the primary and the backup facility on a regular basis, while others operate certain processes simultaneously at the primary and the secondary facilities. For example, one of the organizations conducts six remote-site recovery tests per year, simulating failure of applications. Meanwhile, another organization has begun recovery testing by alternating its full production cycle between the two key facilities. Further, all but one of the five firms that intend to employ teleworking as part of their pandemic plan have assessed their work-from-home capabilities—an essential part of planning for extensive teleworking to ensure that the organizations’ telecommunications systems can support the large amounts of traffic that would be generated. One organization in particular tested work-from-home infrastructure to ensure continuity of daily production as early as mid-2007 and continues to test connectivity as its telecommunications infrastructure is upgraded. Another organization told us it conducted several work-from-home tests in 2008, including server stress test and tests featuring full-volume transaction levels. This kind of testing is critical to pandemic planning, especially for those organizations that intend to have some of their critical staff work from home. Our analysis also indicates that six of the seven of the organizations have procedures in place to ensure their pandemic plans are being reviewed and updated. Because pandemic plans should be sufficiently flexible to effectively address a wide range of possible effects that could result from a pandemic, ongoing review and updates will ensure a plan has up-to-date policies, standards, and procedures. Officials from the six organizations told us that the pandemic plans are reviewed on a regular basis, either at the business-department level or in some cases by the audit committee or the Board of Directors. For example, at one organization the audit committee reviews the pandemic plan and reports its assessment and findings to the Board of Directors on an annual basis. At another organization, the departmental plan is prepared by the department manager and is approved by the director. However, at one organization executives have seen the pandemic plan, but it has not been formally approved. This organization told us it recently instituted a pandemic flu committee that will formally review and approve the pandemic plan. Regular review and approval by senior management helps to ensure that adequate resources are dedicated to implementing the plan. Furthermore, with the changes occurring across financial organizations due to the recent market turmoil, regular review helps an organization confirm that its plan is still aligned effectively to its organizational structure. As the regulator that oversees stock and options exchanges, clearing organizations, and broker-dealers, SEC has taken various actions to ensure that market organizations are preparing for a pandemic, including issuing guidance and conducting examinations of market participants’ preparations, but could take additional steps to better oversee firms’ readiness. To ensure the readiness of the participants in the securities markets, SEC has issued various communications that provided guidance outlining its expectations for these entities’ pandemic preparation efforts. For example, in April 2006, SEC sent a letter to securities exchanges and clearing organizations advising them to plan for a pandemic and make preparations intended to keep the markets operating. SEC’s letter noted that the organizations’ existing business continuity programs were usually designed to address a discrete event and therefore could prove inadequate to address the potentially long-lasting impact of a pandemic. SEC staff also spoke at conferences, meetings with market participants, and other forums, such as those sponsored by industry trade associations, to share information about pandemic issues. Although securities regulators had taken various actions to help the financial markets prepare, our 2007 report indicated that additional actions could further improve the financial market’s readiness to withstand an influenza pandemic. In response to our recommendation, SEC provided more specific guidance between July and October 2007 to the securities exchanges, clearing organizations, and broker-dealers that indicated that these organizations’ pandemic plans include various key elements, such as procedures for continuing operations during even severe pandemics, and that the plans be in place by the end of 2007. To ensure that securities market organizations are taking adequate steps to be ready for a pandemic, SEC has been conducting examinations of various market participants’ preparations that cover, among other things, pandemic preparedness plans. To assess the extent to which securities exchanges, electronic markets, and clearing organizations are adequately managing risks to their operations, staff from SEC’s Division of Trading and Markets regularly conduct examinations through its Automation Review Program (ARP). Since beginning this program in the late 1980s, SEC has issued guidance and conducted examinations that address operations risk issues at these organizations, including reviewing physical and information security and business continuity planning. As of September 2009, 22 securities exchanges, electronic markets, and clearing organizations were subject to ARP’s guidance and examinations, including five of the organizations whose operations we consider critical to the securities markets. As part of the ARP examinations, SEC staff have been addressing these organizations’ pandemic preparedness during their reviews of business continuity issues. During these examinations, SEC staff were using an examination module, adapted from the Federal Reserve, to assess whether these organizations have developed plans that adequately address the five key elements of a pandemic plan, including whether the organizations identified their critical staff, had procedures for reducing the likelihood of their staff becoming ill, and tested their plans. From January 2007 to June 2009, SEC’s Division of Trading and Markets staff had conducted nine examinations addressing business continuity planning and pandemic preparedness at the critical organizations included in our review. Although examiners generally found in the 2007 examinations that organizations were in various stages of pandemic preparations, and in some cases had not addressed all the required elements of a pandemic plan, our review of the examination reports that SEC conducted in 2008 and 2009 indicate that these organizations improved their plans to better address the key elements of pandemic preparedness. However, after examining one of the organizations in October 2008, SEC staff made various recommendations to direct this organization to improve its pandemic planning. For example, SEC recommended that the organization’s plan better address the impact of staff reductions on its operations and that it test its pandemic procedures. Although SEC has conducted inspections to ensure that critical organizations are preparing plans that address all the key pandemic areas, SEC’s examination reports did not always cite as deficiencies the limited analysis or documentation of these organizations’ staffing levels. The pandemic assessment questions used by SEC staff addressed issues related to staff dependencies, including whether the organizations identified their key functions and staff for these functions and conducted cross-training of staff to ensure that sufficient staff would be available during a pandemic. However, as noted earlier, our reviews of the critical market organizations indicate that three of the five critical securities market organizations have not adequately documented the number of staff who could perform critical functions if many of the staff that currently perform those functions are unavailable during a pandemic. Our reviews of SEC’s examination reports show that SEC staff identified weaknesses in the staffing analysis at one of these critical organizations but not at the other two. SEC staff acknowledged that ensuring adequate numbers of critical staff is important. But they said they had not expected the organizations to document the adequacy of their staffing for all their positions because staff in critical departments were likely to be interchangeable and thus could fill in for each other. Moreover, in their opinion, specific staffing lists could quickly become out of date given the higher rate of staff turnover at these organizations during this current financial crisis. Although we agree that the critical organizations may have staff throughout their organizations that could step in for ill employees during a pandemic, until such staffing depth is better assessed and documented, these organizations cannot be fully assured of their ability to operate during such an event. As we noted previously, even organizations that created listings of the staff capable of performing critical functions found during testing that what they thought was sufficient depth in staffing was actually inadequate in some departments. In addition, this current period of increased staff turnover among financial organizations likely further increases the risk that an organization could have thinner staffing for some key positions that might not be identified until a pandemic is occurring. As a result, until SEC staff take steps to ensure that these organizations better document the adequacy of the depth of their critical function staffing, some aspects of these organizations’ operations could be disrupted during a severe pandemic. Although SEC’s ARP staff’s reviews address the extent to which the critical organizations plan to have employees telework during a pandemic, their examinations thus far have not included checking for viable alternate strategies if Internet congestion occurs. SEC staff told us that as part of their pandemic examinations at securities market organizations, they were reviewing various teleworking issues by addressing the relevant questions in their examination module regarding whether the organizations had remote access arrangements and whether the organizations had assessed the capacities of their communications links. The SEC module also asks whether an organization analyzed the locations of its staff’s homes to see if there were large numbers of staff that may be trying to connect from a single area and thus be more vulnerable if congestion or disruption occurs in that area. However, neither the SEC staff’s examination module nor the examination reports we reviewed address whether these organizations developed formal plans for what to do with their teleworking staff if congestion prevents that strategy from being viable. As noted earlier, our reviews of the critical organizations indicate that not all have developed adequate alternative strategies in the event that staff are unable to telework effectively. Until SEC staff take steps to ensure that all organizations develop such strategies, the risk exists that a pandemic could disrupt some areas of these organizations’ operations. In addition to taking steps to assess the readiness of securities exchanges, electronic markets, and clearing organizations to continue operating during a pandemic, SEC staff have been reviewing the preparations of large broker-dealers whose activities are important to overall market functioning. In the aftermath of the September 11, 2001 terrorist attacks, SEC and the banking regulators made coordinated efforts to ensure the resiliency of the U.S. securities markets with respect to clearance and settlement activities. As the attacks showed, the inability of individual securities market participants to promptly clear and settle transactions can pose significant financial risks to other participants. In response, SEC, the Federal Reserve, and the Office of the Comptroller of the Currency jointly issued the Interagency Paper on Sound Practices to Strengthen the Resilience of the U.S. Financial System (Sound Practices) in April 2003. The Sound Practices paper establishes business continuity expectations for the clearance and settlement activities of organizations that support critical financial markets. These organizations include the core clearing and settlement entities that process securities transactions (core organizations) and firms that play a significant role in critical financial markets (significant firms)—generally defined as those firms whose participation in the markets results in their consistently clearing or settling at least 5 percent of the value of the transactions in any of the product markets specified in the paper. Since issuing the paper, these regulators have been conducting examinations of the clearing organizations, significant broker-dealers, and clearing banks that are subject to these practices to ensure they have in place business continuity arrangements sufficient to meet various recovery goals for their clearance and settlement activities. All of the recent examinations that SEC staff conducted under the Sound Practices effort also addressed the pandemic preparations for the significant number of broker-dealers whose role in the critical financial market activities were deemed significant for selected securities and other product markets. In early 2008, staff in SEC’s Office of Compliance Inspections and Examinations, which is responsible for conducting examinations of broker-dealers, mutual funds, and investment advisers, conducted reviews of the then-largest existing broker-dealers. Because of these entities’ high trading volumes in various securities or other products, the markets could be significantly affected if they were unable to clear and settle their transactions. As part of these reviews, SEC staff obtained documentation on how these broker-dealers were addressing the key elements of pandemic planning. Based on these assessments, SEC staff found that the largest broker-dealers appeared to be implementing pandemic plans that generally addressed the key elements. However, as part of conducting some operations risk examinations of a broader group of broker-dealers during 2008, SEC staff also examined the extent to which four midsized firms that cleared trades for other broker-dealers had begun preparations for a pandemic. During these reviews, SEC staff found that, unlike the larger firms, three of these four clearing broker-dealers had no formal pandemic plans in place. In addition to the broker-dealers overseen by SEC, we also reviewed FINRA, the self-regulatory organization that oversees most broker-dealers in the United States. FINRA oversees broker-dealers, including “introducing” firms that accept customer orders and “clearing firms” that process introducing firms’ orders. Prior to H1N1 and our inquiries, FINRA had not fully assessed the pandemic readiness of broker-dealers, including clearing firms. However, since then, FINRA administered a voluntary survey of significant firms, in which a majority of the firms reported they are engaged in some level of pandemic planning. The results of the survey will be used to identify areas for improvement moving forward, including a new examination module that addresses pandemic readiness. For further information on FINRA’s activities, see appendix II. The increased demand on the Internet resulting from the number of students, workers, and other family members at home during the day during a severe pandemic is expected to create congestion by exceeding the current capacity of Internet providers’ network infrastructure in residential neighborhoods. Telecommunications providers will have limited options to expand network infrastructure during an outbreak, and possible network management techniques would likely require government action in order for providers to avoid violating existing customer service agreements. DHS is the federal agency responsible for working with the private sector to ensure that the critical communications sector, which includes the networks that comprise the Internet, is protected from attacks and other disasters. Although DHS has taken some actions relating to pandemic and possible Internet congestion, it has not taken the necessary steps to develop a strategy for addressing such congestion. In addition, developing an effective Internet congestion response plan will likely require coordination with various other federal agencies, including the Department of Education, HHS, and FCC. As the experience of Hurricane Katrina showed, working in advance of a crisis to understand the proper roles and responsibilities of various federal and other entities is important for ensuring an effective response, but DHS has not taken extensive actions to coordinate with other relevant federal and private sector entities about actions that could potentially reduce Internet congestion and how best to respond. In addition, an important step for ensuring the federal government is prepared to address pandemic-related Internet congestion will be identifying whether any federal entity currently has the needed authority to take any actions determined to be necessary. However, whether DHS, FCC, or others have sufficient existing authorities to direct private sector Internet providers to take the actions necessary to relieve congestion is not clear. Similarly, although its own study showed that obtaining public cooperation in reducing nonessential use of the Internet could greatly resolve the potential pandemic-related congestion, DHS has not taken steps to assess the effectiveness and feasibility of mounting such a campaign to begin developing one. Until DHS develops an effective response strategy, coordinates with federal and other partners on actions to take, determines whether sufficient authorities to act exist or are sought, and evaluates the need for a public campaign, employees in critical sectors of the nation’s economy, including those in financial services, might not be able to effectively telework or otherwise communicate or transmit data over the Internet during a pandemic. Seven critical securities market organizations that we reviewed have taken significant steps to better ensure they would be able to continue operating during a pandemic, including by developing plans that address the key elements of pandemic planning. However, some of these organizations could better document the adequacy of their staffing levels and ensure they have prepared viable alternatives in the event that their teleworkers experience Internet congestion. SEC has taken various steps, including issuing guidance and conducting examinations, to ensure that financial market organizations, including those critical to the overall functioning of the markets, are prepared to continue operating during a pandemic. However, taking additional steps during their examinations to ensure that these organizations have fully documented the adequacy of their staffing analyses, developed formal alternatives to teleworking, and tested these would provide greater assurance that the financial markets’ full range of operations will not be disrupted by a pandemic. To better ensure that securities market participants as well as organizations in other critical sectors of the economy will continue to have access to the Internet during a pandemic, we recommend that the Secretary of Homeland Security take the following four actions: develop a strategy outlining actions that could be taken to address coordinate with other relevant federal and private sector entities about actions that could potentially reduce Internet congestion, work with other federal partners to determine if sufficient authority exists for one or more relevant agencies to take any contemplated actions to address Internet congestion, and assess the effectiveness and feasibility, and undertake if warranted, a public education campaign to reduce such congestion. To better ensure that important securities market participants are making adequate preparations for pandemic, we recommend that the Chairman, SEC, ensure that SEC staff take steps to ensure that critical financial market organizations are fully documenting the adequacy of their staffing levels to withstand high absenteeism and have formally developed alternative strategies in the event that congestion limits teleworking effectiveness. We provided a draft of this report to the Secretary of Homeland Security, the Secretary of Health and Human Services, the Secretary of the Treasury, the Chairman of the Board of Governors of the Federal Reserve System, the Chairman of the Financial Industry Regulatory Authority, the Comptroller of the Currency, the Chairman of the Securities and Exchange Commission, and the Chairman of the Federal Communications Commission for their review and comment. In her letter, SEC’s Chairman noted that she shares our concern that Internet congestion could impair certain aspects of the securities markets during a pandemic (see app. IV). She noted that she also agrees that critical market organizations can take steps to improve their existing pandemic plans. Accordingly, the Chairman indicated that SEC will issue letters to these organizations recommending that they further document their staff cross-training arrangements and their plans to maintain operations if Internet congestion impairs their ability to rely on telework for support functions. Further, SEC staff will review compliance with this recommendation in future examinations of these organizations. The Chairman also noted that SEC is prepared to assist other agencies to help address the problem of potential Internet congestion. In a written response to a draft of this report, the Director of DHS’s Departmental GAO/OIG Liaison Office concurs in part with our recommendations that DHS should, among other things, develop a strategy outlining actions that could be taken to address potential Internet congestion. The Director’s letter states that the agency agrees to take these steps to mitigate the impact of any pandemic-related congestion on the systems that the federal government uses to communicate critical national security/emergency preparedness (NS/EP) information, but that addressing Internet congestion for other communications, as a general matter, does not fall within DHS’s responsibilities, and that DHS does not have the responsibility for developing an Internet congestion strategy separate and apart from assuring NS/EP communications. While we agree that DHS should ensure that NS/EP communications are maintained, DHS has been broadly tasked with leading efforts to prevent disruptions to the nation’s overall telecommunications infrastructure and is the agency best positioned to do so. As discussed in this report, federal policies and plans assign DHS lead responsibility for facilitating a public/private response to and recovery from major Internet disruptions. DHS was designated under HSPD-7 as the lead agency for coordinating the protection of the communications sector—a role it plays for several of the other sectors that have been identified as the nation’s critical infrastructures and key resources. As lead agency for this sector, DHS is to conduct vulnerability assessments and encourage risk management strategies to protect and mitigate against attacks. HSPD-7 also notes that agencies are responsible for working with their sectors to reduce the consequences of catastrophic failures not cased by terrorism. Similarly, the 2009 National Infrastructure Protection Plan notes that risk in the 21st century results from a complex mix of man-made and naturally occurring threats and hazards, including terrorist attacks, accidents, natural disasters, and other emergencies. Under this plan’s risk analysis and management framework, sector-specific agencies are to combine consequence, vulnerability, and threat information to produce assessments of risks to a sector and enhance protection by setting goals and objectives, establishing priorities for mitigating risks, and implementing protective programs and resiliency strategies. Based on the study that DHS itself led, congestion resulting from a pandemic appears to be one of the threats for which DHS is tasked with ensuring an adequate governmental response. Furthermore, for example, The National Strategy to Secure Cyberspace notes that the Internet is at the core of the information infrastructure upon which we depend, connecting millions of other computer networks and making most of the nation’s essential services and infrastructures work. According to this strategy, DHS has important responsibilities to develop plans to secure these key resources and infrastructures and provide assistance to the private sector and other government entities with respect to recovery plans for failures in critical information systems. DHS has already been working to address threats to the Internet, for example, by establishing an Internet Disruption Working Group to work with the private sector to establish priorities and develop action plans to prevent major disruptions of the Internet and to identify recovery measures in the event of a major disruption. DHS also has an ongoing relationship with the communications sector coordinating council, which consists of various private sector telecommunications providers, that could assist in assessing and developing solutions to this issue. As a result of these responsibilities and its existing capabilities, we believe that DHS is the appropriate agency to take the lead in developing a strategy to address potential pandemic-related Internet congestion and to coordinate with other relevant federal and private sector entities about actions that could reduce such congestion. DHS also commented that congestion that affects the Internet outside of NS/EP communications falls within the operational and administrative interests of other federal agencies. While we agree that other agencies, such as FCC, should play a role in addressing the potential negative impact on our nation’s commerce and economy from pandemic-related Internet congestion, under the existing governmental policies, DHS is the agency that is specifically tasked with addressing threats that have the potential to disrupt the critical communications sector. Furthermore, this report notes the uncertainty that exists over whether FCC has the authority to act to address Internet-related congestion problems. The uncertainty of roles and authorities regarding this issue is the reason we recommended that DHS work with other federal partners to determine if sufficient authority exists for one or more relevant agencies to take any actions necessary to address Internet congestion that may occur during and because of a severe pandemic crisis. While other agencies could play critical roles in addressing this issue, we believe that DHS, as the communications sector lead agency, should provide this leadership and coordinate a response. The Director’s letter also includes some additional technical comments that we address as appropriate in appendix V. We also received technical comments from FCC and HHS, which are incorporated as appropriate in the report. We are sending copies of this report to the Secretary of Homeland Security, the Chairman of the Securities and Exchange Commission, and other interested parties. The report will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff have any questions regarding this report, please contact Mathew Scire at (202) 512-8678 or sciremj@gao.gov; David Powner at (202) 512-9286 or pownerd@gao.gov; or Nabajyoti Barkakati at (202) 512-4499 or barkakatin@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix IV. Our objectives were to determine (1) the potential impact of a severe pandemic on the Internet and the actions telecommunications providers and government agencies are taking to address possible congestion, (2) the adequacy of the actions that securities market organizations are taking to prepare pandemic plans, and (3) steps that securities and other regulators are taking to assess the readiness of securities market organizations to continue operating during a pandemic. To describe the potential impact of a pandemic on the Internet and the actions that communications providers and relevant government agencies are taking to address possible congestion, we interviewed staff from two communications providers and two cable providers that are among the largest providers of Internet access service in the United States, as well as two industry associations representing such providers. In addition, we interviewed relevant officials at the Department of Homeland Security (DHS), Federal Communications Commission (FCC), and the Department of the Treasury to discuss their efforts and authorities to address potential Internet congestion. We also interviewed representatives from telecommunications and Internet providers that are members of the U.S. Communications Sector Coordinating Council that provides input to DHS regarding critical infrastructure protection issues. We also interviewed staff at the Department of Health and Human Services (HHS)—including staff from the Centers for Disease Control and Prevention—to learn about their efforts to educate the public about pandemic strategies. To assess the potential Internet congestion that could occur during a pandemic, we conducted a literature search and reviewed relevant studies and reports. Specifically, we reviewed a study conducted by DHS in cooperation with various government, communication sector, and financial sector representatives. The study evaluated the technical feasibility of the pandemic strategy advocated by the government and identified action plans to better prepare the nation for telecommuting during a pandemic influenza. Our review of the study included an evaluation of the study’s methodology, and interviews with the DHS staff who oversaw the research on this study, including the Director and Chief of Staff of the Office of Cyber Security and Communications. To confirm the accuracy of the study’s findings, we interviewed communication sector representatives who participated in the study. We also reviewed after action reports from two pandemic exercises—one sponsored by the Financial Services Sector Coordinating Council, Financial and Banking Information Infrastructure Committee, and the Securities Industry and Financial Markets Association, and another conducted by the United Kingdom financial sector to test the financial sectors’ resilience to pandemic influenza. To assess the actions that critical securities market organizations and key market participants are taking to prepare pandemic plans, we reviewed the actions of seven organizations—including exchanges, clearing organizations, and payment processors—whose ability to operate is critical to the overall functioning of the financial markets. To maintain the security and the confidentiality of their proprietary information, we agreed with these organizations that our report would not discuss their efforts to address pandemic readiness and ensure business continuity in a way that could identify them. To assess how these organizations ensure they can continue operations in the face of a pandemic outbreak, we discussed their business continuity and pandemic preparedness plans with their staff and visited their facilities. We reviewed and analyzed their pandemic plans and supporting business continuity documents and compared the plans to the key elements that banking and securities regulators have issued as guidance to financial organizations regarding pandemic planning. In evaluating these organizations’ pandemic readiness, we attempted to determine whether these organizations’ pandemic plans adequately address the five elements required by the regulators, including: (1) a process for monitoring the pandemic’s progress and a plan that escalates response steps as various pandemic phases are reached; (2) a preventive program to minimize, to the extent possible, illness among employees, including social distancing of employees by curtailing meetings; (3) a documented strategy of facilities or procedures designed to allow the organization to continue its critical operations in the event that large numbers of its staff are unavailable for prolonged periods, including an analysis of staffing levels needed for critical functions and, as applicable, an alternative to teleworking; (4) a testing program to ensure that the practices and capabilities will be effective and allow it to continue its critical operations; and (5) an oversight program to ensure ongoing review and updates to the pandemic plan. To assess financial regulators’ efforts to assess the readiness of securities market organizations to continue operating during a pandemic, we reviewed relevant regulations and guidance and interviewed officials at the Securities and Exchange Commission (SEC), the Board of Governors of the Federal Reserve System (Federal Reserve), and the New York Federal Reserve Bank, the Office of Comptroller of the Currency, and the Financial Industry Regulatory Authority (FINRA). We also collected and reviewed data and reports from SEC, FINRA, and the Federal Reserve on the examinations they conducted of exchanges, clearing organizations, and broker-dealers. Furthermore, we reviewed a random sample of exams conducted by FINRA of business continuity practices at clearing firms that provide order routing and post-trade clearance and settlement processing for other broker-dealers (introducing firms) from 2006 through 2008. We randomly selected 9 firms of varying sizes from a total population of 56. To assess whether the level of preparations varied by firm size, we reviewed examinations for 3 large firms (that provided clearing for 100 or more other broker-dealer firms), 3 medium-sized firms (those that cleared for between 20 and 99 firms), and 3 small firms (those clearing for 19 or fewer firms). We also interviewed officials at one of the larger clearing firms. We conducted this performance audit from June 2008 to October 2009 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Although the Financial Industry Regulatory Authority (FINRA)—the self- regulatory organization that oversees most broker-dealers in the United States—undertook some actions to improve broker-dealers’ awareness of the potential impact of a pandemic, it has only recently begun to take steps to more fully ensure such firms are making adequate preparations. In addition to oversight by the Securities and Exchange Commission (SEC), FINRA oversees broker-dealers conducting business domestically in the United States. The broker-dealers that FINRA oversees include, but are not limited to, two different types: Introducing broker-dealer firms whose staff open customer accounts and accept orders to buy and sell securities, but whose firms are not usually members of the exchanges or clearing organizations. Clearing firms that maintain accounts at the central securities clearing organization and process trades on behalf of their own customers as well as those for the customers of the introducing brokers that use them for trade execution or clearing processing. Clearing firms also maintain the cash and securities holdings for their introducing firms’ customers. According to data from FINRA, as of June 2009, 56 firms that clear for other broker-dealers (clearing firms) were operating in the U.S. markets, with some clearing for hundreds of firms but many clearing for less than 20 firms. Although most broker-dealers are not required to recommence operations after disasters, FINRA expects its member firms to have business continuity plans that, among other things, assess how pandemic risks could affect the firm. Unlike the core exchanges and clearing organizations and critical broker-dealers covered by the Interagency Paper on Sound Practices to Strengthen the Resilience of the U.S. Financial System, which SEC requires to be able to resume operations on the same business day on which a wide-scale disruption occurs, broker- dealers have the option of recommencing their operations or shutting down if they are unable to continue. Since 2004, FINRA has had rules that require broker-dealers to have a business continuity plan in place that describes how the firm will: maintain appropriate backup and recovery functions for critical data; alternate communications between the member and the employees; and maintain all mission critical systems, such as those that process taking orders, and clearing and settling securities trades. As a result, at a minimum the FINRA business continuity rule requires all of its member broker-dealers to have adequate plans for ensuring customers have prompt access to their funds and securities in the event that the broker-dealer discontinues business operations. Although FINRA’s business continuity rules were issued before a pandemic was widely recognized as a potential threat to the financial markets, the organization issued guidance in 2006 that encourages broker-dealers to ensure that they assess whether or not their business continuity plans would be suitable for prolonged, widespread public health emergencies, such as a pandemic outbreak. Also in 2006, FINRA requested comment on potential regulatory relief granted in response to a pandemic. FINRA officials told us they have also emphasized the importance of addressing pandemic as part of business continuity planning to the broker-dealer staff that attend industry conferences and workshops. However, prior to June 2009, FINRA had not begun to actively assess the readiness of broker-dealers, including clearing firms. FINRA examines firms on a rotational basis—depending on the risk level and complexity of firms’ operations—every 1, 2, or 4 years for compliance with a broad range of regulatory issues, including business continuity planning. According to data submitted to us by FINRA, across the 56 firms that clear for other broker-dealers, their staff conducted 40 examinations for compliance with the business continuity rules in 2006, 39 in 2007, 46 in 2008, with 33 completed or scheduled for 2009. When FINRA conducts the business continuity examinations, the inspectors use 1 or more of 13 business continuity planning examination modules to guide the inspection. However, the initial set of business continuity examination modules that FINRA staff have been using in their examinations to assess firms’ compliance with the business continuity rule did not include questions related to pandemic preparedness. In addition, our own review of FINRA-conducted inspections found that FINRA officials have not been addressing pandemic issues to a great extent in business continuity examinations conducted through June 2009. We reviewed FINRA business continuity exams from 2006 to 2008 for a randomly selected sample of 9 of the 56 clearing firms that clear for other firms to assess the extent to which pandemic issues were being addressed. To assess whether the level of preparations varied by firm size, we reviewed examinations for 3 large firms (that provided clearing for 100 or more introducing brokers), 3 medium-sized firms (those that cleared for between 20 and 99 firms), and 3 small firms (those clearing for 19 or fewer firms). Our review found that the inspections for 8 of the 9 firms showed evidence the FINRA examiner reviewed the firm’s plan for compliance with the 10 business continuity elements required to be addressed by FINRA’s business continuity rule. However, we found limited evidence that the examiners reviewed pandemic readiness at the firms. For three of the firms, the examination documentation included some general discussions about these firms’ pandemic planning, and in three cases we saw evidence that pandemic plans were included in the documents reviewed by the FINRA examiners. Although the full extent to which clearing firms are ready to continue operating during a pandemic has not been assessed, some evidence raised concerns that not all are making adequate preparations. We did not attempt to systematically determine clearing firms’ pandemic readiness, but we did interview staff at one of the largest clearing firms. This firm’s staff described a pandemic plan and procedures that appeared reasonably likely to be able to continue operations even in the face of significant absenteeism. However, as noted earlier, a limited review by SEC staff conducted in 2008 found that three of four midsized clearing firms have not developed plans for continuing operations during a pandemic. If clearing firms such as these are not able to continue operating, customers of the introducing broker-dealers that use the clearing firms experiencing such problems potentially could find access to their funds and securities curtailed for significant periods of time. For example, FINRA staff told us transferring the customer accounts of broker-dealers that cease operations can take several days or weeks, depending on the circumstances. In response to the recent H1N1 outbreak and our inquiries in relation to this review, FINRA staff told us they have begun various efforts to more broadly assess the readiness of broker-dealers, including clearing firms, for a pandemic. Beginning in June 2009, FINRA conducted a voluntary survey of broker-dealer firms to determine preparedness for a pandemic. The survey included questions asking, among other things, whether the firm has conducted a review of the potential impact of a pandemic, and whether the firm has a business continuity plan specifically addressing a pandemic, and if so, how that plan is being tested. The survey results show that almost all respondents report having conducted a review of the potential impact of a pandemic, and have business continuity plans that specifically address pandemic preparedness. FINRA is using the results of the survey to develop additional guidance on pandemic preparedness practices for the industry. In addition, FINRA staff told us they have developed a new examination module that addresses pandemic preparedness. This module requires their examiners to determine whether the firm’s business continuity arrangements for resuming business operations appear reasonable given the conditions likely to prevail during a pandemic. For example, the module directs the examiner to review the firm’s business continuity planning to determine if the procedures address risks associated with pandemic, such as taking steps to limit the spread of influenza among its staff, and assessing the firm’s operational capabilities using teleworking and the impact of requiring employees to work remotely. The new module was piloted by FINRA examiners during the summer of 2009, and then, once revised as needed, will be used in upcoming exams. FINRA officials told us they will conduct a pandemic preparedness review at all the firms that clear for other broker-dealers by the end of 2011. Banking regulators for the key clearing banks have taken actions to assess pandemic readiness among banks, including those that clear transactions for the securities markets. The Federal Reserve and the Office of Comptroller of the Currency issued guidance in 2006 that call for all banks under their supervision to include the unique impacts of a pandemic in their business continuity planning. Similar to securities regulators, the bank regulators had taken actions to help banks and thrifts address pandemic efforts in our last review. For example, in a joint notice from the regulators that oversee banks and thrifts, the agencies indicated that their institutions should review the U.S. government’s national pandemic strategy to consider what actions may be appropriate for their particular situations, and whether such actions should be included in their event response and contingency strategies. Furthermore, banking regulators had also begun to review pandemic planning in the context of their ongoing supervisory activities. However, in response to the recommendation we made in our 2007 report, the Federal Reserve and the Office of the Comptroller of the Currency subsequently notified institutions that play systemically important roles in securities and other markets that these entities should have plans that address even severe pandemics. In addition, the Federal Financial Institutions Examination Council issued an updated examination manual regarding information technology and business continuity issues that includes steps that banks should be taking related to pandemic planning. Banking regulators have also been conducting reviews to ensure that banks are preparing for possible pandemics, and through these efforts confirmed that the critical market institutions under their supervision met the 2007 deadline to have a pandemic plan in place, and that those plans include the required elements. For example, the Federal Reserve began a series of reviews—using a set of questionnaires to collect information on the planning elements established in the guidance—in January 2008 to assess the progress made by the top 15 banking organizations in the country and concluded that considerable progress has been made among its member banks in pandemic planning. The review objectives were to provide a broad perspective of the state of pandemic preparedness at systemic institutions, identifying trends within the pandemic preparedness planning process, and to provide peer benchmarking attributes to the participating institutions. Office of the Comptroller of the Currency officials told us they continue to monitor progress on pandemic planning in national banks through ongoing supervision rather than targeted exams, and they have been evaluating the banks’ efforts using the newly issued business continuity planning guidance that includes the requirements for pandemic plans. The following are GAO’s comments on the Department of Homeland Security’s letter dated October 14, 2009. 1. The likely usefulness of teleworking as a way for government agencies and businesses to continue operations during a pandemic is one of the reasons we believe that DHS should take the lead in addressing potential Internet congestion that could arise during a severe pandemic, including working with private sector providers to encourage them to take proper steps to be prepared not only to ensure that NS/EP communications are not affected, but that any adverse impact on all other communications is also mitigated. 2. Although not citing the FCC opinion and order by number, our report does discuss some of the network management techniques noted by those documents that providers might be able to use to relieve pandemic-related congestion. However, as our report notes, these techniques may have limitations in resolving the type of congestion envisioned to occur in residential neighborhoods. In addition, providers told us that they would require government direction to implement such techniques to reduce congestion, which is why we recommend that DHS begin taking steps to determine what strategies, actions, and authorities are needed to address this issue so that if it appears that private sector providers must be asked to take steps, such direction can come from the appropriate government source. Furthermore, as the report notes, providers told us their remote network management tools may be a way for them to continue their operations with reduced workforces resulting from pandemic-related absenteeism and that these tools could be used to re-route traffic around congested areas in regional networks or the national backbone, but not to relieve congestion in the residential neighborhoods. 3. As our report states, the DHS study of the impact of pandemic on Internet access notes that obtaining the cooperation of the general public in limiting bandwidth-intensive Internet activities was shown by the study’s modeling to be an effective way to relieve congestion. Uncertainty over whether such cooperation could be obtained is the reason that we recommend that DHS assess the effectiveness and feasibility of implementing a public information campaign, and if warranted, begin developing one. Regarding DHS’s suggested addition of the techniques noted in the FCC order, as we noted above, we discussed these techniques with providers and learned they may have limitations in addressing the type of congestion envisioned to arise in a pandemic and providers would likely require government direction to take such actions. 4. This comment was sent to us earlier as a part of DHS’s technical comments and we have revised the text to note that some of this information has been made available publicly. The best practices that DHS cites in response would likely improve telecommunication providers’ readiness for a pandemic, but likely would not be sufficient to relieve the congestion in residential neighborhoods. 5. This statement was intended to serve as an example of the types of comments FCC received regarding the prioritization issue. We did not assess whether this suggestion was feasible or comports with other FCC practices. 6. As noted above, the best practices DHS cites could assist providers in being better prepared for a pandemic. However, they are not likely sufficient to address residential neighborhood congestion, which is why DHS’s own study also proposed best practices for enterprises, teleworkers, and the public. Providers did not provide us information on any steps they were taking to advise the public about practices that could relieve congestion during a pandemic. In fact, one provider told us a good approach to manage Internet congestion effectively would be for the government to work with providers to publicize appropriate best practices and issue related guidance. As a result, we recommend that DHS assess the effectiveness and feasibility of such practices and implement such a campaign if warranted. 7. We changed the language in this report to note that DHS has not taken action related to evaluating a public education campaign because other activities supporting its operational mission have taken priority. Nevertheless, we believe that such activities should be undertaken to address potential pandemic-related congestion. 8. As this report discusses, much of the securities market’s critical communication would not be affected by congestion of the public Internet infrastructure because it travels over dedicated proprietary networks. However, financial sector organizations are planning to use teleworking to varying degrees as part of their plans to continue operations during a pandemic. As a result, these staff, as a well as the staff of other U.S. federal, state, or local governments and private businesses that plan to use teleworking from home during a pandemic would be affected by the congestion that is envisioned to affect residential neighborhoods. As a result, we recommend that DHS to take actions to address this issue. Furthermore, our report discusses securities market organizations’ activities to prepare themselves to effectively telework during a pandemic and describes the limitations we found in these efforts. As a result, we made recommendations to SEC to further improve its oversight, which it has agreed to implement. In addition to the contacts named above, Cody Goebel and Michael Gilmore, Assistant Directors; Chir-Jen Huang; Yola Lewis; Kristeen McLain; Marc Molino; Carl Ramirez; Linda Rego; and Hai Tran made major contributions to this report. | Concerns exist that a more severe pandemic outbreak than 2009's could cause large numbers of people staying home to increase their Internet use and overwhelm Internet providers' network capacities. Such network congestion could prevent staff from broker-dealers and other securities market participants from teleworking during a pandemic. The Department of Homeland Security (DHS) is responsible for ensuring that critical telecommunications infrastructure is protected. GAO was asked to examine a pandemic's impact on Internet congestion and what actions can be and are being taken to address it, the adequacy of securities market organizations' pandemic plans, and the Securities and Exchange Commission's (SEC) oversight of these efforts. GAO reviewed relevant studies, regulatory guidance and examinations, interviewed telecommunications providers and financial market participants, and analyzed pandemic plans for seven critical market organizations. Increased demand during a severe pandemic could exceed the capacities of Internet providers' access networks for residential users and interfere with teleworkers in the securities market and other sectors, according to a DHS study and providers. Private Internet providers have limited ability to prioritize traffic or take other actions that could assist critical teleworkers. Some actions, such as reducing customers' transmission speeds or blocking popular Web sites, could negatively impact e-commerce and require government authorization. However, DHS has not developed a strategy to address potential Internet congestion or worked with federal partners to ensure that sufficient authorities to act exist. It also has not assessed the feasibility of conducting a campaign to obtain public cooperation to reduce nonessential Internet use to relieve congestion. DHS also has not begun coordinating with other federal and private sector entities to assess other actions that could be taken or determine what authorities may be needed to act. Because the key securities exchanges and clearing organizations generally use proprietary networks that bypass the public Internet, their ability to execute and process trades should not be affected by any congestion. In analyzing seven critical market organizations, GAO found they had prepared pandemic plans that addressed key regulatory elements, including hygiene programs to minimize staff illness and continuing operations by spreading staff across geographic areas. However, not all had completed or documented analyses of whether they would have sufficient staff capable of carrying out critical activities if many of their employees were ill. Also, not all had developed alternatives to teleworking if congestion arises. SEC staff have been regularly examining market organizations' readiness, but could further reduce risk of disruptions by ensuring that these organizations prepare complete staffing analyses and teleworking alternatives. |
The electricity industry, as shown in figure 1, is composed of four distinct functions: generation, transmission, distribution, and system operations. Once electricity is generated—whether by burning fossil fuels; through nuclear fission; or by harnessing wind, solar, geothermal, or hydro energy—it is generally sent through high-voltage, high-capacity transmission lines to local electricity distributors. Once there, electricity is transformed into a lower voltage and sent through local distribution lines for consumption by industrial plants, businesses, and residential consumers. Because electric energy is generated and consumed almost instantaneously, the operation of an electric power system requires that a system operator constantly balance the generation and consumption of power. Utilities own and operate electricity assets, which may include generation plants, transmission lines, distribution lines, and substations—structures often seen in residential and commercial areas that contain technical equipment such as switches and transformers to ensure smooth, safe flow of current and regulate voltage. Utilities may be owned by investors, municipalities, and individuals (as in cooperative utilities). System operators—sometimes affiliated with a particular utility or sometimes independent and responsible for multiple utility areas—manage the electricity flows. These system operators manage and control the generation, transmission, and distribution of electric power using control systems—IT- and network-based systems that monitor and control sensitive processes and physical functions, including opening and closing circuit breakers. As we have previously reported, the effective functioning of the electricity industry is highly dependent on these control systems. However, for many years, aspects of the electricity network lacked (1) adequate technologies—such as sensors—to allow system operators to monitor how much electricity was flowing on distribution lines, (2) communications networks to further integrate parts of the electricity grid with control centers, and (3) computerized control devices to automate system management and recovery. As the electricity industry has matured and technology has advanced, utilities have begun taking steps to update the electricity grid—the transmission and distribution systems—by integrating new technologies and additional IT systems and networks. Though utilities have regularly taken such steps in the past, industry and government stakeholders have begun to articulate a broader, more integrated vision for transforming the electricity grid into one that is more reliable and efficient; facilitates alternative forms of generation, including renewable energy; and gives consumers real-time information about fluctuating energy costs. This vision—the smart grid—would increase the use of IT systems and networks and two-way communication to automate actions that system operators formerly had to make manually. Smart grid modernization is an ongoing process, and initiatives have commonly involved installing advanced metering infrastructure (smart meters) on homes and commercial buildings that enable two-way communication between the utility and customer. Other initiatives include adding “smart” components to provide the system operator with more detailed data on the conditions of the transmission and distribution systems and better tools to observe the overall condition of the grid (referred to as “wide-area situational awareness”). These include advanced, smart switches on the distribution system that communicate with each other to reroute electricity around a troubled line and high-resolution, time-synchronized monitors—called phasor measurement units—on the transmission system. Figure 2 illustrates one possible smart grid configuration, though utilities making smart grid investments may opt for alternative configurations depending on cost, customer needs, and local conditions. According to the National Energy Technology Laboratory, a Department of Energy (DOE) national laboratory supporting smart grid efforts, smart grid systems fall into several different categories: Integrated communications, such as broadband over power line communication technologies or wireless communications technologies. Advanced components, such as smart switches, transformers, cables, and other devices; storage devices, such as plug-in hybrid electric vehicles and advanced batteries; and grid-friendly smart home appliances. Advanced control methods, including real-time monitoring and control of substation and distribution equipment. Sensing and measurement technologies, such as smart meters and phasor measurement units. Improved interfaces and decision support, which includes software tools to analyze the health of the electricity system and real-time digital simulators to study and test systems. The use of smart grid systems may have a number of benefits, including improved reliability from fewer and shorter outages, downward pressure on electricity rates resulting from the ability to shift peak demand, an improved ability to shift to alternative sources of energy, and an improved ability to detect and respond to potential attacks on the grid. Both the federal government and state governments have authority for overseeing the electricity industry. For example, the Federal Energy Regulatory Commission (FERC) regulates rates for wholesale electricity sales and transmission of electricity in interstate commerce. This includes approving whether to allow utilities to recover the costs of investments they make to the transmission system, such as smart grid investments. Meanwhile, local distribution and retail sales of electricity are generally subject to regulation by state public utility commissions. State and federal authorities also play key roles in overseeing the reliability of the electric grid. State regulators generally have authority to oversee the reliability of the local distribution system. The North American Electric Reliability Corporation (NERC) is the federally designated U.S. Electric Reliability Organization, and is overseen by FERC. NERC has responsibility for conducting reliability assessments and enforcing mandatory standards to ensure the reliability of the bulk power system— i.e., facilities and control systems necessary for operating the transmission network and certain generation facilities needed for reliability. NERC develops reliability standards collaboratively through a deliberative process involving utilities and others in the industry, which are then sent to FERC for approval. These standards include critical infrastructure protection standards for protecting electric utility-critical and cyber-critical assets. The Energy Independence and Security Act of 2007 (EISA) established federal support for the modernization of the electricity grid and required actions by a number of federal agencies, including the National Institute of Standards and Technology (NIST), FERC, and DOE. With regard to cybersecurity, the act called for NIST and FERC to take the following actions: NIST was to coordinate development of a framework that includes protocols and model standards for information management to achieve interoperability of smart grid devices and systems. As part of its efforts to accomplish this, NIST planned to identify cybersecurity standards for these systems and also identified the need to develop guidelines for organizations such as electric companies on how to securely implement smart grid systems. In January 2011, we reported that NIST had identified 11 standards involving cybersecurity that support smart grid interoperability and had issued a first version of a cybersecurity guideline. FERC was to adopt standards resulting from NIST’s efforts that it deemed necessary to ensure smart grid functionality and interoperability. The act also authorized DOE to establish two initiatives to facilitate the development of industry smart grid efforts. These were the Smart Grid Investment Grant Program and the Smart Grid Regional Demonstration Initiative. DOE made $3.5 billion and $685 million of American Recovery and Reinvestment Act (“Recovery Act”) funds available for these two initiatives, respectively. The Smart Grid Investment Grant Program provided grant awards to utilities in multiple states to stimulate the rapid deployment and integration of smart grid technologies, while the Smart Grid Regional Demonstration Initiative was to fund regional demonstrations to verify technology viability, quantify costs and benefits, and validate new business models for the smart grid at a scale that can be readily adopted around the country. The federal government has also undertaken various other smart-grid-related initiatives, including funding technical research and development, data collection, and coordination activities. In January 2012, the DOE Inspector General reported that cybersecurity plans submitted by Smart Grid Investment Grant Program recipients were not always complete or they did not describe intended security controls in sufficient detail. The report also stated that DOE officials approved cybersecurity plans for smart grid projects even though some of the plans contained shortcomings that could result in poorly implemented controls. The report recommended, among other things, that DOE ensure that grantees’ cybersecurity plans were complete, including thorough descriptions of potential security risks and related mitigation through necessary controls. The responsible DOE office stated that it will continue to ensure that the security plans are complete and are implemented properly. Threats to systems supporting critical infrastructure—which includes the electricity industry and its transmission and distribution systems—are evolving and growing. In February 2011, the Director of National Intelligence testified that, in the past year, there had been a dramatic increase in malicious cyber activity targeting U.S. computers and networks, including a more than tripling of the volume of malicious software since 2009.sources may adversely affect computers, software, networks, organizations, entire industries, or the Internet. Cyber threats can be unintentional or intentional. Unintentional threats can be caused by software upgrades or maintenance procedures that inadvertently disrupt Different types of cyber threats from numerous systems. Intentional threats include both targeted and untargeted attacks from a variety of sources, including criminal groups, hackers, disgruntled employees, foreign nations engaged in espionage and information warfare, and terrorists. Moreover, these groups have a wide array of cyber exploits at their disposal. Table 1 provides descriptions of common types of cyber exploits. The potential impact of these threats is amplified by the connectivity between information systems, the Internet, and other infrastructures, creating opportunities for attackers to disrupt critical services, including electrical power. For example, in May 2008, we reported that the corporate network of the Tennessee Valley Authority (TVA)—the nation’s largest public power company, which generates and distributes power in an area of about 80,000 square miles in the southeastern United States— contained security weaknesses that could lead to the disruption of control systems networks and devices connected to that network. We made 19 recommendations to improve the implementation of information security program activities for the control systems governing TVA’s critical infrastructures and 73 recommendations to address specific weaknesses in security controls. TVA concurred with the recommendations and has taken steps to implement them. As government, private sector, and personal activities continue to move to networked operations, the threat will continue to grow. GAO-07-1036 and GAO-12-92. grid. When several key transmission lines in northern Ohio tripped due to contact with trees, they initiated a cascading failure of 508 generating units at 265 power plants across eight states and a Canadian province. Davis-Besse power plant. The Nuclear Regulatory Commission confirmed that in January 2003, the Microsoft SQL Server worm known as Slammer infected a private computer network at the idled Davis-Besse nuclear power plant in Oak Harbor, Ohio, disabling a safety monitoring system for nearly 5 hours. In addition, the plant’s process computer failed, and it took about 6 hours for it to become available again. While presenting significant potential benefits, the smart grid vision and its increased reliance on IT systems and networks also expose the electric grid to potential and known cybersecurity vulnerabilities, which could be exploited by a wide array of cyber threats. This creates an increased risk to the smooth and reliable operation of the grid. As we and others have reported, these vulnerabilities include an increased number of entry points and paths that can be exploited by potential adversaries and other unauthorized users; the introduction of new, unknown vulnerabilities due to an increased use of new system and network technologies; wider access to systems and networks due to increased connectivity; an increased amount of customer information being collected and transmitted, providing incentives for adversaries to attack these systems and potentially putting private information at risk of unauthorized disclosure and use. We and others have also reported that smart grid and related systems have known cyber vulnerabilities. For example, cybersecurity experts have demonstrated that certain smart meters can be successfully attacked, possibly resulting in disruption to the electricity grid. In addition, we have reported that control systems used in industrial settings such as electricity generation have vulnerabilities that could result in serious damages and disruption if exploited. Further, in 2009, the Department of Homeland Security, in cooperation with DOE, ran a test that demonstrated that a vulnerability commonly referred to as “Aurora” had the potential to allow unauthorized users to remotely control, misuse, and cause damage to a small commercial electric generator. Moreover, in 2008, the Central Intelligence Agency reported that malicious activities against IT systems and networks have caused disruption of electric power capabilities in multiple regions overseas, including a case that resulted in a multicity power outage. In our January 2011 report, we identified a number of key challenges that industry and government stakeholders faced in ensuring the cybersecurity of the systems and networks that support our nation’s electricity grid. Among others, these challenges included the following: Lack of a coordinated approach to monitor whether industry follows voluntary standards. As mentioned above, under EISA, FERC is responsible for adopting cybersecurity and other standards that it deems necessary to ensure smart grid functionality and interoperability. However, FERC had not developed an approach coordinated with other regulators to monitor, at a high level, the extent to which industry will follow the voluntary smart grid standards it adopts. There had been initial efforts by regulators to share views, through, for example, a collaborative dialogue between FERC and the National Association of Regulatory Utility Commissioners (NARUC), which had discussed the standards-setting process in general terms. Nevertheless, according to officials from FERC and NARUC, FERC and the state public utility commissions had not established a joint approach for monitoring how widely voluntary smart grid standards are followed in the electricity industry or developed strategies for addressing any gaps. Moreover, FERC had not coordinated in such a way with groups representing public power or cooperative utilities, which are not routinely subject to FERC’s or the states’ regulatory jurisdiction for rate setting. We noted that without a good understanding of whether utilities and manufacturers are following smart grid standards, it would be difficult for FERC and other regulators to know whether a voluntary approach to standards setting is effective or if changes are needed. Lack of security features being built into certain smart grid systems. Security features had not been consistently built into smart grid devices. For example, according to experts from a panel convened by GAO, currently available smart meters had not been designed with a strong security architecture and lacked important security features, such as event logging and forensics capabilities, which are needed to detect and analyze attacks. In addition, these experts stated that smart grid home area networks—used for managing the electricity usage of appliances and other devices in the home—did not have adequate security built in, thus increasing their vulnerability to attack. Without securely designed smart grid systems, utilities may not be able to detect and analyze attacks, increasing the risk that attacks would succeed and utilities would be unable to prevent them from recurring. Lack of an effective mechanism for sharing cybersecurity information within the electricity industry. The electricity industry lacked an effective mechanism to disclose information about smart grid cybersecurity vulnerabilities, incidents, threats, lessons learned, and best practices in the industry. For example, experts stated that while the industry has an information-sharing center, it had not fully addressed these information needs. According to these experts, information regarding incidents such as both successful and unsuccessful attacks must be able to be shared in a safe and secure way; this is crucial to avoid publicly revealing the reported organization and penalizing entities actively engaged in corrective action. Such information sharing across the industry could provide important information regarding the level of attempted attacks and their methods, which could help grid operators better defend against them. In developing an approach to cybersecurity information sharing, the industry could draw upon the practices and approaches of other industries. Without quality processes for information sharing, utilities may not have the information needed to adequately protect their assets against attackers. Lack of industry metrics for evaluating cybersecurity. The electricity industry was also challenged by a lack of cybersecurity metrics, making it difficult to measure the extent to which investments in cybersecurity improve the security of smart grid systems. Experts noted that while such metrics are difficult to develop, they could help in comparing the effectiveness of competing solutions and determining what mix of solutions best secure systems. Further, our panel of experts noted that having metrics would help utilities develop a business case for cybersecurity by helping to show the return on a particular investment. Until such metrics are developed, increased risk exists that utilities will not invest in security in a cost-effective manner or be able to have the information needed to make informed decisions about their cybersecurity investments. Accordingly, in our January 2011 report, we made multiple recommendations to FERC, including that it develop an approach to coordinating with state regulators to evaluate the extent to which utilities and manufacturers are following voluntary smart grid standards and develop strategies for addressing any gaps in compliance with standards that are identified as a result. We further recommended that FERC, working with NERC as appropriate, assess whether commission efforts should address any of the cybersecurity challenges identified in our report. FERC agreed with our recommendations and described steps the commission intended to take to address them. We are currently working with FERC officials to determine the status of their efforts to address these recommendations. In summary, the electricity industry is in the midst of a major transformation as a result of smart grid initiatives and this has led to significant investments by many entities, including utilities, private companies, and the federal government. While these initiatives hold the promise of significant benefits, including a more resilient electric grid, lower energy costs, and the ability to tap into alternative sources of power, the prevalence of cyber threats aimed at the nation’s critical infrastructure and the cyber vulnerabilities arising from the use of new technologies highlight the importance of securing smart grid systems. In particular, it will be important for federal regulators and other stakeholders to work closely with the private sector to address key cybersecurity challenges posed by the transition to smart grid technology. While no system can be made 100 percent secure, proven security strategies could help reduce risk to an acceptable level. Chairman Stearns, Ranking Member DeGette, and Members of the Subcommittee, this completes our statement. We would be happy to answer any questions you have at this time. If you have any questions regarding this statement, please contact Gregory C. Wilshusen at (202) 512-6244 or wilshuseng@gao.gov or David C. Trimble at (202) 512-3841 or trimbled@gao.gov. Other key contributors to this statement include Michael Gilmore (Assistant Director), Jon R. Ludwigson (Assistant Director), Paige Gilbreath, Barbarol J. James, and Lee A. McCracken. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The electric power industry is increasingly incorporating information technology (IT) systems and networks into its existing infrastructure as part of nationwide effortscommonly referred to as the smart gridaimed at improving reliability and efficiency and facilitating the use of alternative energy sources such as wind and solar. Smart grid technologies include metering infrastructure (smart meters) that enable two-way communication between customers and electricity utilities, smart components that provide system operators with detailed data on the conditions of transmission and distribution systems, and advanced methods for controlling equipment. The use of these systems can bring a number of benefits, such as fewer and shorter outages, lower electricity rates, and an improved ability to respond to attacks on the electric grid. However, this increased reliance on IT systems and networks also exposes the grid to cybersecurity vulnerabilities, which can be exploited by attackers. Moreover, for nearly a decade, GAO has identified the protection of systems supporting our nations critical infrastructurewhich include the electric gridas a governmentwide high-risk area. GAO is providing a statement describing (1) cyber threats facing cyber-reliant critical infrastructures and (2) key challenges to securing smart grid systems and networks. In preparing this statement, GAO relied on its previously published work in this area. The threats to systems supporting critical infrastructures are evolving and growing. In a February 2011 testimony, the Director of National Intelligence noted that there had been a dramatic increase in cyber activity targeting U.S. computers and systems in the previous year, including a more than tripling of the volume of malicious software since 2009. Varying types of threats from numerous sources can adversely affect computers, software, networks, organizations, entire industries, and the Internet itself. These include both unintentional and intentional threats, and may come in the form of targeted or untargeted attacks from criminal groups, hackers, disgruntled employees, hostile nations, or terrorists. The interconnectivity between information systems, the Internet, and other infrastructures can amplify the impact of these threats, potentially affecting the operations of critical infrastructures, the security of sensitive information, and the flow of commerce. Moreover, the smart grids reliance on IT systems and networks exposes the electric grid to potential and known cybersecurity vulnerabilities, which could be exploited by attackers. As GAO reported in January 2011, securing smart grid systems and networks presented a number of key challenges that required attention by government and industry. These included: A lack of a coordinated approach to monitor industry compliance with voluntary standards. The Federal Energy Regulatory Commission (FERC) is responsible for regulating aspects of the electric power industry, which includes adopting cybersecurity and other standards it deems necessary to ensure smart grid functionality and interoperability. However, FERC had not, in coordination with other regulators, developed an approach to monitor the extent to which industry will follow the voluntary smart grid standards it adopts. As a result, it would be difficult for FERC and other regulators to know whether a voluntary approach to standards setting is effective. A lack of security features built into smart grid devices. According to a panel of experts convened by GAO, smart meters had not been designed with a strong security architecture and lacked important security features. Without securely designed systems, utilities would be at risk of attacks occurring undetected. A lack of an effective information-sharing mechanism within the electricity industry. While the industry has an information-sharing center, it had not fully addressed the need for sharing cybersecurity information in a safe and secure way. Without quality processes for sharing information, utilities may lack information needed to protect their assets against attackers. A lack of metrics for evaluating cybersecurity. The industry lacked metrics for measuring the effectiveness of cybersecurity controls, making it difficult to measure the extent to which investments in cybersecurity improve the security of smart grid systems. Until such metrics are developed, utilities may not invest in security in a cost-effective manner or be able to make informed decisions about cybersecurity investments. GAO made several recommendations to FERC aimed at addressing these challenges. The commission agreed with these recommendations and described steps it is taking to implement them. |
Terrorism and drug trafficking exact a tremendous cost from society. According to the Federal Aviation Administration (FAA), the estimated cost of one bombed aircraft is about $1 billion, including the price of litigation for the loss of human lives and property loss. This estimate does not include the cost to national security in terms of U.S. military and law enforcement response or terrorism’s psychological effect on society—neither of which has been measured. FAA is expected to spend an estimated $281 million on aviation security during fiscal year 1997 for research and development, the purchase of detection technology equipment, regulatory enforcement, and policy- and rule-making. The annual social cost of narcotics, according to the Office of National Drug Control Policy (ONDCP), is estimated to be about $67 billion, mostly from the consequences of drug-related crime. This cost does not include what Americans spend to purchase illegal drugs, estimated at $49 billion for 1993, the last year for which data is available. Federal agencies are expected to spend about $15 billion during fiscal year 1997 on drug control activities, including research and development, law enforcement, demand reduction, interdiction, and international programs. Four organizations—FAA, the National Security Council (NSC), the Office of Management and Budget (OMB), and the Department of Transportation—are responsible for overseeing or developing explosives detection technologies. FAA has the primary responsibility for the development of explosives detection technologies used to protect commercial aircraft. From fiscal year 1992 to 1996, FAA provided about $131 million, or an average of $26.2 million per year, for detection technology development. NSC established the Technical Support Working Group (TSWG) in 1986 to oversee and coordinate counterterrorism research and development, including explosives detection technology. TSWG funding for explosives detection efforts totaled about $14.3 million during fiscal years 1992-96. OMB and the Department of Transportation play more limited roles in overseeing the FAA budget dealing with explosives detection technologies. OMB officials explained that OMB’s role is limited because of the small size of FAA’s explosives detection technology development program. The Department of Transportation has played a somewhat more active role in FAA and interagency working groups that assess the capabilities of the technologies to detect explosives. In the aftermath of the TWA 800 explosion in July 1996, the President established the White House Commission on Aviation Safety and Security. The Commission recommended, among other things, that Customs assume an enhanced role in screening outbound international air cargo for explosives. In September 1996, Congress provided the Secretary of the Treasury the authority to establish scientific certification standards for explosives detection canines and to provide for the certification of canines used for such purposes at U.S. airports. The Bureau of Alcohol, Tobacco, and Firearms (ATF) has assumed responsibility for this effort. In February 1997, the Commission recommended that ATF continue to work to develop governmentwide standards for canine teams. Senior FAA officials have stressed that delays in deploying advanced explosives detection technology are, in part, a function of the history of their technology planning and development efforts. FAA was criticized in 1990 when it announced plans to mandate the deployment of a specific technology for screening checked baggage on international flights following the December 1988 crash of Pan Am 103 over Lockerbie, Scotland. At the time, the technology could not detect the amount of explosives that blew up Pan Am 103 without an unacceptably high rate of false alarms. The airline industry objected to the technology’s high cost, large size, slow speed in processing baggage, and high rate of false alarms. The Aviation Security Improvement Act (P.L. 101-604 of Nov. 1990) provided a framework for FAA’s technology planning. The act prohibited FAA from mandating a particular technology until it was certified as capable of detecting various types and quantities of explosives, using certification procedures developed in conjunction with the scientific community. In addition, the act required that FAA establish a scientific advisory panel to review its counterterrorism research and development program and recommend future program areas, including the need for long-range research to prevent catastrophic damage to commercial aircraft by the next generation of terrorist weapons. FAA’s scientific advisory panel recently recommended, among other things, a reallocation of 1997 research and development funds to provide an immediate increase in resources for long-term research to identify and counter emerging terrorist threats. In response, FAA increased its request for fiscal year 1997 funding for aircraft hardening and chemical weapons detection. In its final report dated February 12, 1997, the White House Commission on Aviation Safety and Security addressed the question of whether FAA is the appropriate government agency to regulate aviation security. The Commission concluded that because of its extensive interactions with airlines and airports, FAA is the appropriate agency. However, the Commission also stressed that the intelligence and law enforcement agencies’ roles in supporting FAA must be clearly defined and coordinated. NSC provides a number of forums for coordinating explosives detection technology issues. As the primary agency responsible for aviation security, FAA sought interagency support within one of NSC’s forums in early 1996 for a proposal to improve aviation security. Another forum, TSWG, has been involved in developing detection technology for countering the threat from terrorist use of explosives for several years. In January 1996, FAA briefed the NSC’s Coordinating Sub-Group on Terrorism on threats to civil aviation and the need for a high-level national policy review on ways of increasing domestic aviation security. FAA used this forum because it believed that the threat of terrorism in the United States was not limited to aviation and responsibilities for countering terrorism crossed federal agency lines. Although FAA discussed the possible use of a presidential commission to obtain consensus and a legislative mandate on increasing aviation security domestically, it was agreed instead to establish a working group within FAA to review the threat against aviation and recommend options for increasing security in the United States. On July 17, 1996, FAA’s Aviation Security Advisory Committee formed a Baseline Working Group to examine everyday security measures at U.S. airports and recommend specific initiatives to strengthen those measures. On December 12, 1996, the group recommended several immediate and long-term improvements, including expansion of FAA’s research and development efforts for explosives detection. TSWG has an Explosives Detection Technology Subgroup, chaired by an FAA representative, to ensure compatibility between TSWG and FAA research and development programs in the explosives detection technology arena. TSWG funds explosives detection technology projects near the $2.9 million level annually. NSC uses TSWG to develop coordinated views regarding the development of explosives detection technologies. For example, in August 1996, the NSC Coordinating Sub-Group on Terrorism requested the State Department’s Coordinator for Counterterrorism to review research in explosives detection equipment and to determine whether additional funds should be invested in such research. The Coordinator directed TSWG to undertake this task. In October 1996, TSWG recommended (1) accelerating the development of methods that reduce or eliminate the human element from the initial threat detection process, (2) increasing the emphasis on and funding for explosive detection research and development, and (3) improving the interagency exchange of information. According to an NSC official, the first two recommendations have been implemented through increased funding. Regarding the third, he pointed out that improved information exchange is the constant goal of all agencies. The White House Commission on Aviation Safety and Security recommended new roles for Customs in screening outbound international air cargo for explosives, including updating and acquiring technologies to do that screening. Customs had previously not been involved in developing explosives detection technologies, although it had developed technologies to screen cargo for various types of contraband. Consequently, it had not worked closely with FAA, the airlines, or TSWG on specifically developing explosives detection technologies. In response to the Commission’s recommendations, Customs is using $16 million to develop a system to identify high-risk cargo for closer inspection and $34 million to purchase detection technologies. Customs is now determining how to develop an automated targeting system to process outbound cargo information. In addition, Customs may develop a new X-ray technology for examining pallets or improve other technologies before acquisition. Customs’ new role presents challenges in coordinating its efforts with FAA and the aviation industry. For example, the Customs’ targeting system may be adapted to enable FAA to screen domestic cargo shipments transported within the United States. In addition, Customs may be required to ensure that its narcotics detection technologies can meet FAA standards for screening cargo for explosives. To date, Customs and FAA have held informal discussions on technical issues but have not prepared a memorandum of understanding setting out their respective roles to help meet these challenges. As a part of its new role, Customs must also enter into agreements with the airline industry for the joint use of the detection technologies. In a letter to the Commission dated January 13, 1997, Customs stated that a memorandum of understanding is being established with FAA to coordinate the identification and deployment of the “joint use” screening equipment. Customs has decided that such an agreement will be limited to sharing information and to possibly developing joint research and development projects. FAA strongly believes that such a memorandum of understanding should include standards for the use of explosives detection systems, development of a joint-use strategy, the resolution of liability concerns, and the development of profiling and targeting systems to identify potentially threatening passengers and cargo. Although FAA has used canines for explosives detection at airports since the 1970s, in September 1996 Congress authorized the Secretary of the Treasury to develop governmentwide explosives detection certification standards for canines and to certify such canines for use at airports. ATF has assumed responsibility for this effort, and an interagency working group has been established to develop uniform standards. FAA believes that a memorandum of understanding is needed with ATF to address the deployment of canine teams at airports. Both FAA and ATF have canine programs. As of February 25, 1997, FAA’s program had 81 certified explosives detection canine teams deployed to 31 airports. FAA requires intensive training in aviation environments on aircraft, in terminals, and around baggage, airport vehicles, and cargo. In fiscal year 1997, FAA received $8.9 million for certifying an additional 114 canines. ATF has certified 115 explosives and firearms detection canine teams for use by 7 foreign countries in support of the Department of State’s Antiterrorism Assistance Program. According to an ATF official, these ATF-certified canines are trained to perform preblast detection duties in various overseas environments, including airports. In fiscal year 1997, ATF received $7.5 million, of which $3.5 million was specifically earmarked for construction and expansion of a canine training facility. Congress also authorized the Secretary of the Treasury to establish scientific certification standards for canines and to certify, on a reimbursable basis, canines employed by federal agencies at airports in the United States. In 1996, the House Appropriations Committee expressed concern about multiple and possibly duplicative or wasteful programs for training dogs to detect explosives. The Committee directed that ATF establish a pilot canine explosives detection program with FAA to foster cooperation, coordination, and consistency between their two programs. The two agencies are working out the details for the pilot program. In August 1996, the Coordinating Sub-Group on Terrorism requested a study on the use of canines for counterterrorism purposes. As a result, a joint effort was begun by FAA and ATF, which agreed to rely principally on a group comprised of various agencies’ chemists and canine trainers to make recommendations to them. Since 1992, TSWG has used its own funds, as well as funds provided by DOD, FAA, and ONDCP, for canine research projects. The White House Commission of Aviation Safety and Security recommended that FAA establish federally mandated standards for security enhancements, including the deployment of explosive detection canine teams. FAA believes that a memorandum of understanding is needed with ATF to address standards for deploying canine teams at airports because ATF has assumed responsibility for establishing governmentwide certification standards for explosives detecting canines. OMB officials said that they play a limited role in overseeing FAA’s explosives detection technology development program because of the small amount of funding for that program relative to funding for all of FAA. They also told us that the extent of their oversight has traditionally been to ensure that the FAA budget meets presidential priorities and is adequately justified. However, OMB became more active and participated in FAA’s Baseline Working Group because of the increased threat of terrorism. Several FAA officials stated that OMB participation was important because the cost of improving security was being estimated at billions of dollars and consideration was being given to shifting the responsibility of funding from the airlines to the government. An OMB official expressed the view that the government might need to be more concerned about research and development efforts if it has to pay for equipment resulting from such efforts. In addition, OMB prepared the President’s fiscal year 1997 antiterrorism proposal, including incorporating the recommendations of the White House Commission. As such, OMB worked with FAA on such issues as pricing explosives detection technologies that FAA would purchase with the additional funding. Four agencies—ONDCP, Customs, DOD, and OMB—are primarily responsible for coordinating or developing narcotics detection technologies. The congressionally established Counterdrug Technology Assessment Center (CTAC) within ONDCP is responsible, among other things, for coordinating federal counterdrug technology efforts and assessing and recommending narcotics detection technologies. Customs, because of its mission to interdict drugs at U.S. ports of entry, is ultimately responsible for deciding on the types of technologies to be developed and used. As congressionally directed, DOD has been primarily responsible for funding and developing most of the innovative narcotics detection technologies for Customs. Recently, OMB became involved in overseeing Customs’ plans for developing and deploying narcotics detection technologies. Agencies have not always agreed on the most appropriate technologies to detect narcotics at U.S. ports of entry. Two technologies funded at about $30 million have been developed but not deployed. More recently, differing views between ONDCP and Customs regarding the type of systems needed along the southwest border led to varying directions from congressional committees. These differing views between ONDCP and Customs stem, in part, from recommendations presented in a congressionally mandated study on costs and benefits of specific technologies. These differences may be resolved as Customs, in coordination with ONDCP, develops a methodology and a 5-year plan for transitioning technologies from development to deployment. CTAC coordinates the counterdrug technology research and development efforts of 21 federal agencies. In addition, CTAC funds its own development projects to address gaps in technologies that provide the greatest support to the various counterdrug activities of federal, state, and local agencies. During fiscal years 1992-96, CTAC funding for detection technologies amounted to about $8.4 million, or an average of about $1.7 million per year. In coordinating the counterdrug research and development program, CTAC attempts to prevent duplication of effort and to ensure that, whenever possible, those efforts provide capabilities that transcend the needs of any single agency. CTAC relies on its interagency Science and Technology Committee to help prioritize projects supported with CTAC funds. The projects are generally managed by a member agency. In addition, a Contraband Detection Working Group was established under this Committee to provide an interagency forum to focus other agencies’ research activities on technology areas that support the contraband detection requirements of law enforcement agencies. In August 1996, the Director, ONDCP, committed to revitalizing the Science and Technology Committee and its working groups. Among other things, the Director proposed that the Committee act as a steering body with membership at a level senior enough to make commitments to research and development policy decisions. An ONDCP official informed us that the Committee is currently focusing on developing a 5-year technology plan. While Customs has the operational need for detection technologies, Congress tasked DOD to develop most of these technologies because DOD was already developing technologies that could be adapted for narcotics detection. During fiscal years 1992-96, DOD funded detection technologies for about $73 million, or an average of about $14.6 million per year. Over the same period, Customs funded detection technologies amounting to about $3.1 million, or an average of about $620,000 per year. In 1990, the House Appropriations Committee tasked DOD, in coordination with Customs, to develop a comprehensive plan for developing drug detection technology for use in inspecting cargo containers. The Committee cited cargo containers as a major threat for the import of illegal drugs into the United States and identified specific technologies that should be pursued. In April 1994, DOD began testing a high-energy X-ray system capable of penetrating fully loaded containers, at a specially constructed port in Tacoma, Washington. DOD and CTAC viewed the system as a key step toward the development of effective, nonintrusive cargo inspection technologies. The tests showed that high-energy X-ray technology could be an effective tool in detecting drugs in a broad range of vehicles and in containers carrying varying types of cargo. DOD spent about $15 million for facility construction and system testing. However, ONDCP, Customs, and DOD agreed in December 1994 to dismantle the site because Customs did not believe that the system was affordable, safe, or operationally suitable for its needs. Based on experiences with the Tacoma high-energy system, Customs and DOD entered into a restructured development program to ensure that DOD would develop only those technologies that would be transitioned by Customs into an operational environment. Based on this understanding, DOD also discontinued work on a Pulsed Fast Neutron Analysis projectafter spending about $15 million because Customs was likewise concerned about its affordability, safety, and operational suitability. For fiscal years 1996 and 1997, OMB questioned Customs’ funding requests for truck X-ray systems to be placed at U.S. ports of entry along the southwest border. These systems use a low-energy X-ray source capable of penetrating empty trucks and other conveyances. OMB limited Customs’ use of the funds until certain conditions were met, citing its concern that a low-energy system had limited capabilities for inspecting fully loaded containers. OMB requested a comprehensive border technology plan that would focus effective inspection technologies in the areas of greatest need. In response, Customs prepared a plan favoring the use of fixed-site truck X-ray systems as well as mobile or relocatable systems. Customs stated that the large number of empty trucks crossing the southwest border presents a very high threat because they sometimes carry drugs. As a result, Customs wanted a system to inspect for drugs concealed within the structure of the truck. Customs stated that the low-energy X-ray system has been effective in detecting drugs concealed in these empty trucks, is safe, and fits into available space. In addition, acquisition costs are estimated at $3 million, operating expenses are low, and training requirements are minimal compared to the high-energy X-ray system built at Tacoma and the Pulsed Fast Neutron Analysis system. OMB continues to believe that Customs needs a range of technologies for the southwest border. Thus, OMB plans to stay informed on issues dealing with the development of those technologies and has started attending ONDCP meetings on developing narcotics detection technologies so that it can become aware of emerging issues. Congressional committees have provided differing direction regarding the development and acquisition of narcotics detection technologies. One committee, supporting Customs needs, recommended funding for a certain technology, while another committee, responding to ONDCP concerns, directed a moratorium on the purchase of such technology. The differences stem, in part, from recommendations presented in a congressionally mandated study on costs and benefits of specific technologies. In September 1994, Congress mandated a study on the cost and benefit tradeoffs in different nonintrusive inspection systems. The study, released in September 1996, concluded that Customs should accelerate the development and implementation of an automated system for screening documents to target cargo for further inspection. For land ports, the study recommended that only the automated targeting system be deployed. Conferees on the National Defense Appropriations Act for 1997 provided DOD with $6 million for DOD’s purchase of low-energy truck X-ray systems to be used by Customs. Conferees to the 1997 Treasury, Postal Service, and General Appropriations Act stated that they were aware of the tradeoff study’s conclusion that deployment of advanced technology at land sites and seaports can make a significant improvement to drug interdiction efforts. The conferees directed a moratorium on the purchase of the low-energy systems until Customs reevaluated its plans regarding the automated targeting system and to both low- and high-energy systems. They further directed that Customs present Congress with an integrated plan responding to the recommendations in the tradeoff study. Customs issued a response February 6, 1997, which stated that empty trucks crossing the southwest border are a very high threat. As a result, Customs wanted a system to examine trucks returning empty to the United States. Customs also stated that it would work with DOD and ONDCP to identify and evaluate new inspection technologies that would complement the capabilities of the low-energy system. According to ONDCP, a promising technology currently under development may be as effective. This system, which will be mobile, is expected to cost about one fifth the estimated $3 million cost of the low-energy system. Over the next few months, Customs and DOD will evaluate this new technology to inspect empty trucks. Development of the current generation of narcotics detection technologies is nearing completion, but Customs does not have a detailed methodology for determining which technologies should be acquired. Nonetheless, Custom’s future development efforts are expected to be a part of the Director of ONDCP’s recent proposal for a 5-year technology plan for developing narcotics supply and demand reduction technologies. The congressionally mandated tradeoff study recommended that Customs adopt a methodology similar to the one it used for assessing procurement options. The study also pointed out that the variation among the ports require a port-by-port analysis to assess the need for specific technologies at each port. Customs has acknowledged that a methodology was needed but noted that the methodology presented in the study was only one of several possible approaches and did not realistically consider personnel and funding constraints. ONDCP and other federal agencies are creating a 5-year technology plan. As part of this plan, the agencies will prepare a road map for developing nonintrusive inspection technologies and upgrading existing systems. For example, Customs and DOD are expected to set out their plans for developing mobile or relocatable high-energy systems for drug interdiction. Both Customs and DOD plan to evaluate the capabilities of the high-energy X-ray system for its ability to detect narcotics concealed in cargo containers. ONDCP plans to review the results of this evaluation. We reported earlier that various technologies, with modifications, can be used to detect both explosives and narcotics. During work on this report, we found that formal coordination between developers of explosives and narcotics detection technologies was not a two-way street. We did find, however, that results of research and testing are shared among the technology developers and overseers through personal contacts or through symposiums. In addition, Customs and FAA have done joint work on systems such as TNA and trace detectors. Canines provide a special opportunity for coordination because they can be trained to respond in specific ways to smells of explosives and narcotics. The developers of explosives detection technologies are active participants on committees that oversee the development of narcotics detection technologies. FAA has participated in ONDCP’s Science and Technology Committee and its Contraband Detection Working Group since their inception to provide a linkage between explosives and narcotics detection technology development. However, the developers of narcotics detection technologies have generally not been included in committees that oversee the development of explosives detection technologies. Customs has not been a member of the scientific advisory panel that reviews FAA’s research and development program and recommends ways to improve the program. Based on our inquiries, an FAA official said that including Customs on the panel may add some additional insight from the developers of narcotics detection technologies. FAA included Customs as a member of the panel effective February 13, 1997. Although Customs is a member of TSWG, it has not participated in the explosives detection subgroup. Officials agreed that Customs would benefit from participating in this subgroup because of its interagency coordination activities. Customs says that it plans to begin participating in the subgroup. The relocatable Pulsed Fast Neutron Analysis system is an example of a technology development that may benefit from closer coordination. In fiscal year 1996, Congress provided TSWG with $6.2 million to evaluate the capabilities of a relocatable Pulsed Fast Neutron Analysis system to detect explosives hidden in cargo. This evaluation will cover a 30-month period and eventually cost about $19 million. As noted earlier, this technology was developed to detect narcotics concealed in large containers but was not adopted for use by Customs because it did not believe that the system was affordable, safe, or operationally suitable for its needs. Customs advised TSWG that it wants to participate in the development of the system. A Customs official said that should the system meet concerns about safety and other operational issues, they would support its installation at a seaport where fully load containers are of concern and its performance could be assessed for both counterdrug and counterterrorism applications. Our work identified efforts underway that if successfully completed could significantly strengthen development of explosives and narcotics technologies. For example, in explosives detection technology development, FAA is working closely with Customs and ATF, both of which have new roles to play. In narcotics detection technology development, Customs is working with ONDCP on a 5-year technology plan and with TSWG on an explosives detection system that may have application to narcotics detection. However, these agencies have not yet established formal understandings on how to develop standards for aviation security enhancements and numerous related issues. Moreover, comprehensive reports on the U.S. government’s efforts to develop explosives and narcotics detection technology are not periodically provided to key decisionmakers. Regarding explosives detection technology development, we found that: FAA and Customs are preparing a memorandum of understanding setting out how they will share information and possibly conduct joint research and development projects regarding detection technologies of mutual interest. ATF has assumed a new role to develop governmentwide standards for explosives detection canines and has begun a joint effort with FAA by cochairing a policy group. They agreed to rely principally on a group comprised of various agencies’ chemists and canine trainers, including a representative from TSWG, to make recommendations to the policy group. FAA strongly believes that memorandums of understanding are needed with Customs and ATF for developing standards for aviation security enhancements, including the use of explosives detection systems, development of a joint-use strategy, resolution of liability concerns, development of profiling and targeting systems to identify potentially threatening passengers and cargo, and deployment of canine teams at airports. However, to date, little or no progress has been made in achieving such understandings, and the involved agencies have not developed a coordinated approach for handling such issues. Regarding narcotics detection technology development, we found the following: ONDCP and Customs disagree on the appropriate methodology for deciding which technologies to transition from development to deployment. According to ONDCP, the methodology should require a port-by-port analysis to assess the need for specific technologies at each port. On the other hand, Customs prefers a methodology that does not add to its or industry’s data-reporting requirements. Nevertheless, both agencies are working on a 5-year technology plan to develop new detection technologies, and Customs told us that it intends to develop a methodology that is acceptable to ONDCP. Customs advised the NSC’s TSWG that it would participate in the development of a system that may have counterdrug application. In addition, a Customs official has been informally monitoring the system’s development. However, as now being developed, the system will not include requirements unique to a narcotics detection application. ONDCP believes that Customs’ involvement with the system will be a worthwhile effort. Our review indicated that no one in the executive branch has aggregated into a single report information on the totality of what is being done on the development of explosives and narcotics detection technology, the nature and extent of resources that the various agencies are applying, the informal coordination and integration efforts, and the types of emerging issues that must be addressed. Currently, no reports are periodically provided to key decisionmakers in the executive branch or Congress. We generally endorse the actions being undertaken by the agencies as the initial steps to strengthening the coordination of explosives and narcotics detection technology development. However, FAA, Customs, and ATF need to work closer as a team to solve complex technological issues. Establishing memorandums of understanding among the agencies could help define the agencies’ roles and enhance cooperation in resolving the numerous issues associated with the development of standards for aviation security enhancements. Further, the resolution of differences in views between ONDCP and Customs on needed technology should help serve as a springboard to acting jointly on the broader problems. In addition, joint development of technology may prove beneficial for both explosives and narcotics detection. Periodic reports to oversight authorities can help keep focus on the efforts being taken to develop and deploy technologies at ports of entry, including airports. In line with the White House Commission of Aviation Safety and Security’s call for more clearly defining and coordinating the roles of law enforcement agencies in supporting the FAA, we recommend that the Secretaries of Transportation and the Treasury establish a memorandum of understanding on how FAA, Customs, ATF, and other agencies are to work together in establishing standards, including the use of explosives detection systems, development of a joint-use strategy, resolution of liability concerns, development of profiling and targeting systems to identify potentially threatening passengers and cargo, and deployment of canine teams at airports. Because no single agency in the executive branch has aggregated into a single report information on what is being done on the development of explosives and narcotics detection technology, Congress may wish to direct the Secretaries of Transportation and the Treasury and the Director, ONDCP, to jointly provide to appropriate congressional oversight committees an annual report on all of the government’s efforts to develop and field explosives and narcotics detection technology. NSC, the Departments of State and Transportation, FAA, ATF, ONDCP, Customs, DOD, and OMB reviewed a draft of this report and provided oral or written comments. They generally agreed with the facts presented, and their suggested technical corrections have been incorporated where appropriate. The written comments of State, FAA, Customs, ATF, and DOD are presented in appendixes I, II, III, IV, and V, respectively. In responding to a draft of this report, FAA, Customs, and ATF have taken varying positions on how to develop standards for aviation security enhancements and address numerous related issues. We have therefore modified the report to recommend that the department secretaries establish a memorandum of understanding for FAA, Customs, ATF, and other agencies to work together on these issues. Based on previous work, we initially focused on five agencies that play the largest roles in developing detection technologies. During the course of our work, we identified other agencies that are beginning to play larger roles in technology development. For our work on agencies involved with developing explosives detection technologies or coordinating their development, we contacted officials of the Departments of Transportation, Defense, and State; FAA; NSC; OMB; Customs; and ATF. We interviewed officials to identify processes and mechanisms to resolve conflicts when establishing policy, setting priorities, selecting projects, and requesting funding. We also obtained and reviewed key documents, such as FAA’s research and development plan, and identified circumstances surrounding cases in which agencies disagreed on technology development. For our work on agencies involved with developing narcotics detection technologies or coordinating their development, we contacted officials of ONDCP, Customs, DOD, and OMB. We again interviewed officials to identify processes and mechanisms to resolve conflicts when establishing policy, setting priorities, selecting projects, and requesting funding. We also obtained and reviewed key documents, such as the ONDCP’s counterdrug research and development plan, and identified circumstances surrounding cases in which agencies disagreed on technology development. To identify mechanisms for coordinating joint development, we interviewed officials and gathered information from the NSC’s TSWG and FAA on the committees that oversee explosives detection technology development efforts. In addition, we interviewed officials and gathered information from ONDCP on similar committees that oversee ONDCP’s narcotics detection technology development efforts. We analyzed the membership of these committees to see if there is representation from both the explosives and narcotics detection technology development communities. We also examined the minutes of the committees’ meetings to verify that member agencies from both communities participated in these meetings. In addition, we gathered information on a particular technology to show the benefits of coordination between the two communities. Finally, we asked about attendance at various symposiums or other professional forums. Based on our objectives, we identified efforts being initiated to strengthen coordination of detection technology development and opportunities to enhance that development. OMB did not provide us with all the information we requested. OMB officials met with us but did not provide documentation on its interactions with other federal agencies responsible for developing explosives and narcotics detection technologies. As a result, we relied on other agencies’ records to document OMB’s role. In addition, NSC officials declined to meet with us to clarify its interaction with the other agencies. We performed this phase of work between October 1996 and February 1997 in accordance with generally accepted government auditing standards. We are sending copies of this report to the Chairmen and Ranking Minority Members of other appropriate congressional committees; the Secretaries of the Treasury, State, Defense, and Transportation; the Directors, OMB, ONDCP, and ATF; the Administrator, FAA; and the Commissioner, U.S. Customs Service. If you have any questions regarding this report, please call me on (202) 512-4841. Major contributors to this report are listed in appendix VI. The following are GAO’s comments on the Department of State’s letter dated March 20, 1997. 1. We have not shown the Departments of Defense (DOD), State, and Energy as agencies responsible for overseeing or developing explosives detection technologies. Instead of showing these agencies separately, we have grouped them under the National Security Council’s Technical Support Working Group (TSWG). Specifically, we state on page 4 that the Department of State provides overall policy guidance to and oversees the operations of TSWG and that DOD and the Department of Energy cochair TSWG. We also state that all three agencies fund the TSWG program, with DOD providing most of the funding. We also have not shown the Bureau of Alcohol, Tobacco and Firearms (ATF) as an agency responsible for overseeing or developing explosives detection technologies. In the introduction to the explosives section, we say that ATF and Customs have assumed new roles. We believe that the reference to ATF at this point is sufficient. 2. We have modified the report to reflect this comment. The following are GAO’s comments on the Federal Aviation Administration’s (FAA) letter dated March 28, 1997. 1. We have incorporated FAA’s technical comments in the text where appropriate. 2. We have modified the report to reflect these comments. The following are GAO’s comments on the U.S. Customs Service letter dated March 20, 1997. 1. In a letter dated January 13, 1997, to the White House Commission on Aviation Safety and Security, Customs stated a memorandum of understanding (MOU) is being established with the FAA to coordinate the identification and deployment of “joint use” screening equipment. The letter further stated that a strategy for the joint-use resources is being developed, with a target date of January 30, 1997, for completion. Based on Customs’ response to a draft of this report, we have concluded that Customs has changed its position on establishing an MOU on joint-use. In its response to our draft report, FAA supports the establishment of such an MOU covering a number of issues. As a result, there appears to be disagreement between Customs and FAA as to how they should address these important issues. We have therefore modified the report to recommend that the department secretaries establish a MOU for FAA, Customs, ATF, and other agencies to work together on these issues and have also suggested that Congress may wish to require the involved agencies to periodically report on these efforts. 2. We have modified the report to state that Customs and FAA are developing an MOU for sharing information and possibly conducting joint-use research and development projects. 3. We have modified the report to reflect this comment. 4. FAA has informed us that it may adapt the Customs’ targeting system for screening domestic cargo shipments transported within the United States. FAA pointed out that the White House Commission on Aviation Safety and Security’s report, dated February 12, 1997, states that Customs and FAA are working with an FAA contractor to study technical issues associated with converting Customs’ targeting system, which was originally designed for sea cargo analysis, to air cargo analysis. The following is GAO’s comment on the Bureau of Alcohol, Tobacco, and Firearm’s letter dated March 25, 1997. 1. We have incorporated ATF’s technical comments in the text where appropriate. The following is GAO’s comment on DOD’s letter dated March 24, 1997. 1. We have incorporated DOD’s technical comments in the text where appropriate. David E. Cooper Ernest A. Döring Charles D. Groves John K. Harper John P.K. Ting The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 6015 Gaithersburg, MD 20884-6015 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (301) 258-4066, or TDD (301) 413-0006. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO reviewed how the U.S. government is organized to develop technologies for detecting explosives and narcotics, focusing on: (1) the roles, responsibilities, and authority of agencies that establish policy, provide funds or oversee funding requests, and develop explosives and narcotics detection technologies; (2) mechanisms used to coordinate the joint development of technologies; and (3) efforts to strengthen detection technology development. GAO noted that: (1) numerous federal organizations are involved in developing technologies for detecting explosives and narcotics; (2) the Federal Aviation Administration (FAA) is the key agency responsible for developing explosives detection technologies for civil aviation security; (3) in response to the explosion of TWA Flight 800, the President established the White House Commission on Aviation Security and Safety to recommend ways of improving security against terrorism; (4) the Commission's recommendations included assigning a new role to the U.S. Customs Service in screening outbound, international cargo for explosives; (5) in September 1996, Congress gave the Secretary of Treasury authority to develop governmentwide standards for canine teams; (6) regarding narcotics detection, the Office of National Drug Control Policy (ONDCP) is responsible for coordinating federal counterdrug technology efforts and assessing and recommending detection technologies; (7) Customs, with technology development support and funding from the Department of Defense, ultimately decides which technologies will be developed and deployed at U.S. ports of entry; (8) Customs has not deployed some technologies because it did not believe that they were affordable, safe, or operationally suitable for its needs; (9) in addition, Customs and ONDCP have differing views regarding the types of detection technologies needed along the Southwest border; (10) joint technology development is important because the types of technologies used to detect explosives and narcotics are similar; (11) the developers of narcotics detection technologies have not always participated in committees that oversee the development of explosives detection technologies; (12) in the future, Customs plans to participate in these committees; (13) at the direction of Congress, an interagency working group on counterterrorism plans to spend $19 million to develop a system for detecting explosives that Customs may possibly use in a seaport environment to detect drugs; (14) despite efforts to strengthen development of explosives and narcotics technologies, GAO found that the cognizant agencies have not yet agreed to formal understandings on how to establish standards for explosives detection systems, profiling and targeting systems, and deploying canine teams at airports; (15) in addition, they have not agreed on how to resolve issues related to a joint-use strategy and liability; and (16) furthermore, key decisionmakers are not receiving periodic comprehensive reports on the aggregated efforts of the various government entities to develop and field explosives and narcotics detection technologies. |
Title XIX of the Social Security Act authorizes federal funding to states for Medicaid, which finances health care services including acute and long- term care for certain low-income, aged, or disabled individuals. States have considerable flexibility in designing and operating their Medicaid programs. Within broad federal requirements, each state determines which services to cover and to what extent, establishes its own eligibility requirements, sets provider payment rates, and develops its own administrative structure. In addition to groups for which federal law requires coverage—such as children and pregnant women at specified income levels and certain persons with disabilities—states may choose to expand eligibility or add benefits that the statute defines as optional. Medicaid is an open-ended entitlement: states are generally obligated to pay for covered services provided to eligible individuals, and the federal government is obligated to pay its share of a state’s expenditures under a CMS-approved state Medicaid plan. The federal share of each state’s Medicaid expenditures is based on a statutory formula linked to a state’s per capita income in relation to national per capita income. In 2002, the specified federal share of each state’s expenditures ranged from 50 percent to 76 percent; in the aggregate, the federal share of total Medicaid expenditures was 57 percent. The Social Security Act provides that up to 60 percent of the state share of Medicaid spending can come from local- government revenues and sources. Some states design their Medicaid programs to require local governments to contribute to the programs’ costs. For more than a decade, some states have used various financing schemes, some involving IGTs, to create the illusion of a valid state Medicaid expenditure to a health care provider. This payment has enabled states to claim federal matching funds regardless of whether the program services paid for had actually been provided. As various schemes have come to light, Congress and CMS have taken actions to curtail them (see table 1). Many of these schemes involve payment arrangements between the state and government-owned or government-operated providers, such as local- government-operated nursing homes. A variant of these creative financing arrangements involves states’ exploitation of Medicaid’s upper payment limit (UPL) provisions. These schemes share certain characteristics, including IGTs, with other financing schemes from prior years (see table 1). In particular, these arrangements create the illusion that a state has made a large Medicaid payment— separate from and in addition to Medicaid expenditures that providers have already received for covered services—which enables the state to obtain a federal matching payment. In reality, the large payment is temporary, since the funds essentially make a round-trip from the state to the Medicaid providers and back to the state. As a result of such round-trip arrangements, states obtain excessive federal Medicaid matching funds while their own state expenditures remain unchanged or even decrease. Figure 1, which is based on our earlier work, illustrates how this mechanism operated in one state (Michigan). As shown in figure 1, the state made Medicaid payments totaling $277 million to certain county health facilities; the total included $155 million in federal funds and $122 million in state funds (step 1). On the same day that the county health facilities received the funds, they transferred all but $6 million back to the state, which retained $271 million (steps 2 and 3). From this transaction, the state realized a net gain of $149 million over the state’s original outlay of $122 million. In cases like this, local-government facilities can use IGTs to easily return the excessive Medicaid payments to the state via electronic wire transfers. We have found that these round-trip transfers can be accomplished in less than 1 hour. The IGT is critical, because if the payment does not go back to the state, the state gains no financial benefit and actually loses from the arrangement because it has simply paid the provider more than its standard Medicaid payment rate for the services. In a variant of this practice, some states require a few counties to initiate the transaction, by taking out bank loans for the total amount the states determined they can pay under the UPL. The counties wire the funds to the states, which then send most or all of the funds back to the counties as Medicaid payments. The counties use these “Medicaid payments” to repay the bank loans. Meanwhile, the states claim federal matching funds on the total amount. Consistent with past actions, Congress and CMS have taken steps to curtail UPL financing schemes when they have come to light. At the direction of Congress, the agency—then called the Health Care Financing Administration (HCFA)—finalized a regulation in 2001 that significantly narrowed the UPL loophole by limiting the amount of excessive funds states could claim. HCFA estimated that its 2001 regulation would reduce the federal government’s financial liability due to inappropriate UPL arrangements by $55 billion over 10 years; a related 2002 regulation was estimated to yield an additional $9 billion over 5 years. CMS recognized that some states had developed a long-standing reliance on these excessive UPL funds, and the law and regulation authorized transition periods of up to 8 years for states to come into compliance with the new requirements. As we recently reported, however, even under the new regulations, states can still aggregate payments to all local-government nursing homes under one UPL to generate excessive federal matching payments beyond their standard Medicaid claims. For example, CMS information about states complying with the new regulation indicates that, through UPL arrangements with public nursing homes and other public facilities, states can still claim about $2.2 billion annually in federal matching funds exceeding their standard Medicaid claims. States’ use of these creative financing mechanisms undermines the federal-state Medicaid partnership as well as the program’s fiscal integrity in at least three ways. First, state financing schemes effectively increase the federal matching rate established under federal law by increasing federal expenditures while state contributions remain unchanged or even decrease. For example, for one state we analyzed (Wisconsin), we estimated that by obtaining excessive federal matching payments and using these funds as the state share of other Medicaid expenditures, the state effectively increased the federal matching share of its total Medicaid expenditures from 59 percent to 68 percent in state fiscal year 2001. The state did so by generating nearly $400 million in excessive federal matching funds via round-trip arrangements with three counties. Similarly, the HHS Office of the Inspector General found that a comparably structured arrangement in Pennsylvania effectively increased that state’s statutorily determined matching rate from 54 percent to about 65 percent. Second, CMS has no assurance that these increased federal matching payments are used for Medicaid services. Federal Medicaid matching funds are intended for Medicaid-covered services for the Medicaid-eligible individuals on whose behalf payments are made. Under state financing schemes, however, states can use funds returned to them at their own discretion. We recently examined how six states with large UPL financing schemes involving nursing homes used the federal funds they generated. As in the past, some states in our review deposited excessive funds from UPL arrangements into their general funds, which the states may or may not use for Medicaid purposes. For example, one state (Oregon) has used funds generated by its UPL arrangement to help finance education programs. Table 2 provides further information on how states used their UPL funds in recent years, as reported by the six states we reviewed. Third, these state financing schemes undermine the fiscal integrity of the Medicaid program because they enable states to make to providers payments that significantly exceed their costs. In our view, this practice is inconsistent with the statutory requirement that states ensure that Medicaid payments are economical and efficient. Under UPL financing arrangements, some states pay a few public providers excessive amounts, well beyond the cost of services provided. We found, for example, that Virginia’s proposed arrangement would allow the state to pay six local- government nursing homes, on average, $670 in federal funds per Medicaid nursing home resident per day—more than 12 times the $53 daily federal payment these nursing homes normally received, on average, per Medicaid resident. Although CMS and the Congress have often acted to curtail states’ financing schemes, problems persist. Improved CMS oversight and additional congressional action could help address continuing concerns with UPL financing schemes and other inappropriate arrangements. We recently reported that CMS has taken several actions to improve its oversight of state UPL arrangements, including forming a team to coordinate its review of states’ proposed and continuing arrangements, drafting internal guidelines for reviewing state methods for calculating UPL amounts, and conducting financial reviews that have identified hundreds of millions of dollars in improper claims. Starting in August 2003, when considering states’ proposals to change how they would pay nursing homes or other institutions, CMS also began to ask states to provide previously unrequested information. The information includes sources of state matching funds for supplemental payments to Medicaid providers, the extent to which total payments would exceed providers’ costs, how a state would use the additional funds, and whether a state required providers to return payments (and, if so, how the state planned to spend such funds). As of October 2003, CMS indicated that it had asked 30 states with proposed state Medicaid plan amendments to provide additional information, and the agency was in the process of receiving and reviewing states’ initial responses. We also reported, however, that CMS’s efforts do not go far enough to ensure that states’ UPL claims are for Medicaid-covered services provided to eligible beneficiaries. Moreover, we remain concerned that in carrying out its oversight responsibilities, CMS at times takes actions inconsistent with its stated goals for limiting states’ use of these arrangements. For example, we previously reported that while the agency was attempting to narrow the glaring UPL loophole in 2001, it was allowing additional states to engage in the very schemes it was trying to shut down, at a substantial cost to the federal government. More recently, we reported that CMS’s granting two states the longest available transition period of 8 years, for phasing out excessive claims under their UPL arrangements, was not consistent with the agency’s stated goals. We estimated that, as a result of these decisions, these two states can claim about $633 million more in federal matching funds under their 8-year transition periods than they could have claimed under shorter transition periods consistent with CMS’s stated policies and goals. In our view, additional congressional action also could help address continuing concerns about Medicaid financing schemes. Although Congress and CMS have taken significant steps to help curb inappropriate UPL arrangements and other financing schemes, states can still claim federal matching funds for more than a public provider’s actual costs of providing Medicaid-covered services. As long as states are allowed to make payments exceeding a facility’s actual costs, the loophole remains. A recommendation open from one of our earlier reports would, if implemented, close the existing loophole and thus mitigate these continuing concerns. We previously recommended that Congress consider prohibiting Medicaid payments that exceed actual costs for any government-owned facility. If this recommendation were implemented, a facility’s payment would be limited to the reasonable costs of covered services it actually provides to eligible beneficiaries, thus eliminating the possibility of the exorbitant payments that are now passed through individual facilities to states. The Administration appears to support such legislative action; the President’s budget for fiscal year 2005 sets forth a legislative proposal to cap Medicaid payments to government providers (such as public hospitals or county-owned nursing homes) to the actual cost of providing services to Medicaid beneficiaries. The term “IGTs” has come to be closely associated—if not synonymous— with the abusive financing schemes undertaken by some states in connection with illusory payments for Medicaid services to claim excessive federal matching funds. IGTs are a legitimate state budget tool and not problematic in themselves. But when they are used to carry out questionable financial transactions that inappropriately shift state Medicaid costs to the federal government, they become problematic. We believe the problem goes beyond IGTs. An observation we made in our first report on this issue in 1994 is as valid today as it was then: in our view, the Medicaid program should not allow states to benefit from arrangements where federal funds purported to benefit providers are given to providers with one hand, only to be taken back with the other. State financing schemes, variants of which have been applied for a decade or longer, circumvent the federal and state funding balance set under law. They have also resulted in the diversion of federal funds intended to pay for covered services for Medicaid-eligible individuals to whatever purpose a state chooses. Although Congress and CMS have often acted to address Medicaid financing schemes once they become apparent, new variations continue to emerge. Experience shows that some states are likely to continue looking for creative means to supplant state financing, making a compelling case for the Congress and CMS to sustain vigilance over federal Medicaid payments. Understandably, states that have relied on federal funding as a staple for their own share of Medicaid spending are feeling the budgetary pressure from the actual or potential loss of these funds. The continuing challenge remains to find the proper balance between states’ flexibility to administer their Medicaid programs and the shared federal-state fiduciary responsibility to manage program finances efficiently and economically in a way that ensures the program’s fiscal integrity. Mr. Chairman, this concludes my prepared statement. I will be happy to answer any questions that you or Members of the Subcommittee may have. For future contacts regarding this testimony, please call Kathryn G. Allen at (202) 512-7118. Katherine Iritani, Tim Bushfield, Ellen W. Chu, Helen Desaulniers, Behn Miller Kelly, and Terry Saiki also made key contributions to this testimony. Medicaid: Improved Federal Oversight of State Financing Schemes Is Needed. GAO-04-228. Washington, D.C.: February 13, 2004. Major Management Challenges and Program Risks: Department of Health and Human Services. GAO-03-101. Washington, D.C.: January 2003. Medicaid: HCFA Reversed Its Position and Approved Additional State Financing Schemes. GAO-02-147. Washington, D.C.: October 30, 2001. Medicaid: State Financing Schemes Again Drive Up Federal Payments. GAO/T-HEHS-00-193. Washington, D.C.: September 6, 2000. State Medicaid Financing Practices. GAO/HEHS-96-76R. Washington, D.C.: January 23, 1996. Michigan Financing Arrangements. GAO/HEHS-95-146R. Washington, D.C.: May 5, 1995. Medicaid: States Use Illusory Approaches to Shift Program Costs to Federal Government. GAO/HEHS-94-133. Washington, D.C.: August 1, 1994. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Medicaid, the federal-state health financing program for many of the nation's most vulnerable populations, finances health care for an estimated 53 million lowincome Americans, at a cost of $244 billion in 2002. Congress structured Medicaid as a shared fiduciary responsibility of the federal government and the states, with the federal share of each state's Medicaid payments determined by a formula specified by law. In 2002, the federal share of each state's expenditures ranged from 50 to 76 percent under this formula; in the aggregate, the federal share of total Medicaid expenditures was 57 percent. Some states have used a number of creative financing schemes that take advantage of statutory and regulatory loopholes to claim excessive federal matching payments. GAO was asked to summarize prior work on how some of these schemes operated, including the role of intergovernmental transfers (IGT), which enable government entities--such as the state and local-government facilities like county nursing homes--to transfer funds among themselves. GAO was also asked to discuss these schemes' effects on the federalstate Medicaid partnership and to discuss what can be done to curtail them. For many years states have used varied financing schemes, sometimes involving IGTs, to inappropriately increase federal Medicaid matching payments. Some states, for example, receive federal matching funds on the basis of large Medicaid payments to certain providers, such as nursing homes operated by local governments, which greatly exceed established Medicaid rates. In reality, the large payments are often temporary, since states can require the local-government providers to return all or most of the money to the states. States can use these funds--which essentially make a round-trip from the states to providers and back to the states--at their own discretion. States' financing schemes undermine the federal-state Medicaid partnership, as well as the program's fiscal integrity, in at least three ways. The schemes effectively increase the federal matching rate established under federal law by increasing federal expenditures while state contributions remain unchanged or even decrease. GAO estimated that one state effectively increased the federal matching share of its total Medicaid expenditures from 59 percent to 68 percent in state fiscal year 2001, by obtaining excessive federal funds and using these as the state's share of other Medicaid expenditures. There is no assurance that these increased federal matching payments are used for Medicaid services, since states use funds returned to them via these schemes at their own discretion. In examining how six states with large schemes used the federal funds they generated, GAO found that one state used the funds to help finance its education programs, and others deposited the funds into state general funds or other special state accounts that could be used for non-Medicaid purposes or to supplant the states' share of other Medicaid expenditures. The schemes enable states to pay a few public providers amounts that well exceed the costs of services provided, which is inconsistent with the statutory requirement that states ensure economical and efficient Medicaid payments. In one state, GAO found that the state's proposed scheme increased the daily federal payment per Medicaid resident from $53 to $670 in six local-government-operated nursing homes. Although Congress and the Centers for Medicare & Medicaid Services have acted to curtail financing schemes when detected, problems persist. States can still claim excessive federal matching funds for payments exceeding public facilities' actual costs. GAO suggests that Congress consider a recommendation open from prior work, that is, to prohibit Medicaid payments that exceed actual costs for any government-owned facility. |
The FBI primarily grants security clearances to state and local law enforcement officials who participate in FBI task forces and to state and local chiefs of police and sheriffs. Since September 11, the FBI has expanded its JTTF initiative from 35 JTTFs to 84 JTTFs. This increase in JTTFs created more opportunities for information sharing and participation by state and local officials, thus increasing the number of security clearance applications submitted to the FBI by state and local officials. After September 11, an increasing number of state and local officials who were not assigned to JTTFs began requesting security clearances to obtain terrorism-related information that might affect their jurisdictions. In some cases, state and local officials assigned to JTTFs were not able to share classified information with their state and local superiors because the superiors lacked security clearances. In addition to forming the JTTFs, the FBI launched the LEO initiative to focus on granting security clearances to officials in an executive, decision-making role who were not posted to a JTTF. According to the FBI, the launch of this initiative also increased the number of state and local security clearance applications. Presidential Executive Order 12968, Access to Classified Information, dated August 1995, established federal executive branch criteria for granting access to classified information. As implemented by the FBI, the primary criterion for granting access to classified information is an individual’s “need to know,” which is defined as the determination made by an authorized holder of classified information that a prospective recipient requires access to specific classified information in order to perform or assist in a lawful and authorized governmental function. In addition to possessing a need to know, individuals must have a security clearance based upon a favorable adjudication of an appropriate background investigation, been briefed on the responsibilities for protecting classified information, signed a nondisclosure agreement acknowledging those responsibilities, and agreed to abide by all appropriate security requirements. An amendment to Presidential Executive Order 12958, Executive Order 13292, issued in 2003, also allows federal agencies, including the FBI, to share classified information in an emergency with individuals who lack a prior security clearance when necessary to respond to an imminent threat to life or in defense of the homeland. Prior to this executive order, during high-priority cases when there was a threat to life, the FBI would provide pertinent information to those with a need to know by granting them interim clearances. Federal government-wide policies, as implemented by the FBI, apply to state and local officials seeking FBI-issued security clearances. The security clearance process for state and local officials involves six broad steps that consist of the (1) FBI field office officials’ determining the applicant’s need to know classified national security information and level of clearance required, (2) applicant’s submission of application materials to an FBI field office, (3) applicant fingerprinting and interview conducted by FBI field office officials, (4) FBI field officials’ routing of application materials to FBI headquarters and the FBI investigators’ completion of a background investigation, (5) FBI headquarters’ adjudication of clearance applications based on federal government adjudication standards, and (6) notification of an adjudication. The FBI’s policies for granting access to classified information requires state and local officials to undergo the same background investigation and adjudication procedures as do individuals who have an employment relationship with the FBI or other federal government agencies and require access to classified national security information. The FBI received its authority to grant security clearances from the Department of Justice in 1993 (see fig. 1). Presidential Executive Order 12968, Access to Classified Information, dated August 1995, established federal executive branch criteria for granting access to classified information. The FBI’s Manual of Investigative Operations and Guidelines outlines the bureau’s policies and procedures for investigating and adjudicating various categories of security clearance cases, including FBI employees, contractors, and state and local officials. The FBI cannot grant a security clearance to any individual based simply on the individual’s rank or position. Police chiefs, for example, are not automatically granted security clearances and must undergo the same procedures as all other individuals. State and local officials do not have a direct employment relationship with the FBI, but sometimes they require access to FBI workspaces and classified information. The procedures for conducting background investigations and granting access to classified information are identical to those for all FBI cases. For example, the FBI requires a background investigation of the last 10 years of a person’s life for all individuals in need of a top secret security clearance. This also applies to state and local officials. Generally, state and local officials’ contact with the FBI is primarily through the local FBI field office, and field office management is personally held accountable by FBI headquarters for all security clearance applications processed through their offices. Nevertheless, a number of stakeholders, including state and local officials themselves, are involved in various parts of the application process which, as figure 2 depicts, can be summarized in six steps. To initiate the security clearance application process, state and local officials first identify the individuals in their departments who require access to classified information, taking into consideration JTTF assignments, internal staffing needs, and FBI field office guidance. Once individuals are identified, the FBI field office specify the need to know for each individual, which in turn determines the appropriate level of the security clearance: top secret or secret. Next, FBI field officials distribute application materials to the selected state and local officials. The instructions for completing the application materials (see fig. 3) require the applicant to provide information reaching back 10 years, and must include all previous employment, residences, foreign travel and contacts, and references. The applicant also signs a statement authorizing access to his or her credit records. Upon submission of the application materials, an FBI field office official reviews the materials, conducts a face-to-face interview, and fingerprints the applicant (see fig. 4). In addition, the FBI is moving toward requiring all security clearance applicants to undergo a polygraph examination to be completed during the security interview. FBI field office officials summarize the information collected for submission to the FBI headquarters. According to FBI headquarters guidance, the FBI field office also must compile a portion of the investigative information for submission to FBI headquarters within 10 days of receipt of the application. However, the FBI was not able to estimate the actual processing time for these preliminary tasks. This procedure opens the investigation and corresponds to the beginning of FBI headquarters’ involvement in the security clearance application process. Following the opening of the case, FBI headquarters conducts government-wide required security clearance National Agency Checks. Such inquiries include checks of the National Crime Information Center, fingerprint checks, and Office of Personnel Management and Defense Clearance and Investigations Index inquiries . FBI headquarters then refers the case to the FBI's Background Investigation Contract Service (BICS) Unit. BICS conducts background investigations for FBI security clearance applications, including applications from state and local officials. Typical background investigations include verification of citizenship, credit, and criminal history checks. For a top secret security clearance, the background investigation includes additional checks such as the verification of education, employment, and residences within the past 10 years. Interviews of friends, coworkers, supervisors, and neighbors also are conducted. The background investigation may be expanded if an applicant has resided abroad or has a history of mental health disorders, drug or alcohol abuse. Once all required background information has been submitted to FBI headquarters, the adjudicator assigned to the case reviews the information. In some cases, the adjudicator must request additional leads to follow up on information uncovered in the investigation that might increase the risk of granting the individual access to classified information. To confirm or mitigate information collected, the adjudicator may require a second interview with the applicant. For example, if a negative credit history is uncovered, the applicant may be asked how he or she is attempting to remedy that situation. The adjudicator summarizes this information in a report, which includes a recommendation as to whether to grant a security clearance based on 13 federal government-wide adjudicative standards. These standards include an assessment of the applicant’s allegiance to the United States, personal conduct, mental health, and associations with undesirable persons or foreign nationals, among other things. A supervisory official reviews this recommendation and grants final approval. In cases where a denial of clearance is recommended, the adjudicator’s unit chief also must review the report. According to the FBI, the same day the adjudication is made regarding whether to grant access to classified information, the FBI field office is notified. In cases of a favorable decision, according to FBI policies, the FBI field office then has 10 days to set up a security briefing with the applicant. The security clearance does not take effect until the security briefing has been completed and the applicant has signed a non-disclosure agreement. Documentation of the interview and the signed agreement are sent to FBI headquarters to record that the clearance has taken effect. The FBI’s goal for processing top secret security clearance applications is 6 to 9 months, and its goal for processing secret security clearance applications is 45 to 60 days, though actual completion times can vary from case to case. The majority of the applications the FBI received for top secret security clearances since September 11, were processed within the FBI’s stated time frame goals. In contrast, the majority of secret applications received since September 11, were not processed within the FBI’s time frame goals. However, during the last half of 2003 the FBI nearly doubled its success rate for completing secret security clearance applications within its time frame goals. The FBI’s goal is to process top secret security clearance applications in 6 to 9 months, though actual completion times can vary from case to case. The FBI has received 1,211 applications for top secret security clearances since September 11, 2001. Of the top secret applications received, about 835 were granted and 276 were pending at the time our review. The FBI denied clearances to 7 applicants. The FBI often assigns greater priority to processing applications for state and local JTTF members, who are required to have top secret clearances. Consequently, as figure 5 shows, about 92 percent of the 835 applications for top secret security clearances granted were processed within the FBI’s time frame goal of 6 to 9 months since September 11. Of the 276 top secret security clearance applications that were still pending completion at the time of our review, over 90 percent were processed within the FBI’s time frame goal of 6 to 9 months. As noted earlier, the FBI’s average actual time frame for processing top secret security clearances for state and local officials has declined since September 11. As shown in figure 6, the average number of days for completing a top secret security clearance application declined from 244 days in the last quarter of 2001 to 70 days in the third quarter of 2003. The FBI’s time frame goal for granting secret security clearances is 45 to 60 days, though actual completion times can vary from case to case. The FBI received 2,363 applications for secret security clearances from state and local officials since September 11. Of the secret security clearance applications received, 2,021 had been granted and 267 were pending at the time of our review. The FBI did not deny any secret security clearance applications. The FBI processed about 26 percent of secret applications within its time frame goals, as shown in figure 7. The remainder, about 72 percent, were granted in more than 60 days. An analysis of the last 6 months of data collected during our review shows that the FBI has begun to process secret applications more quickly. The FBI received about 483 secret security clearance applications between June 2003 and December 2003. About 46 percent of those applications were processed within the FBI’s time frame goals (see fig. 8), nearly doubling the success rate for completing secret security clearance applications within the designated time frame goals. The FBI has also improved its overall actual completion time frames for secret security clearance applications since the second quarter of 2003, as shown in figure 9. The average number of days for processing a secret security clearance application was about 90 in the last quarter of 2001. By the fourth quarter of 2002, the average number of days for processing a secret security clearance application had risen to about 115, and in the third quarter of 2003, the average number of days declined to about 60. FBI officials stated that the bureau’s time frame goals are approximate averages and actual completion times can vary depending on a number of factors. For example, numerous past residences or foreign travel can extend the time it takes to conduct a background investigation. In addition, background investigations can take additional time when information from the applicant’s application does not match the information collected during investigation interviews. FBI guidance to state and local officials states that the processing time for each application will vary depending on its complexity. Also, according to a DOJ official, the number of applications in the FBI’s queue, as well as the FBI’s staffing resources, affect the time needed to process a security clearance application. According to FBI officials, the volume of security clearance applications increased substantially after September 11. Although the FBI’s security clearance process begins when a need for a clearance is determined, as shown in figure 2, the FBI does not begin to track the processing of security clearance applications until after the FBI field office completes its preliminary tasks and forwards the application package to FBI headquarters. These preliminary tasks can include checks with the internal affairs unit of the department where the official is employed, administering a polygraph examination, and verifying citizenship with the U.S. Citizenship and Immigration Services. The official must also undergo a face-to-face personnel security interview. Though guidance to FBI field offices requires that these tasks be performed and the application forwarded to headquarters within 10 days of receipt of the security clearance application, the FBI was not able to estimate the actual processing time for these preliminary tasks. According to the FBI, the time required to perform the initial steps of the process varies by field office and is dependent on the unique circumstance of each candidate. For example, state and local officials who are located in a law enforcement agency that is distant from the FBI field office may require additional time to schedule tasks that take place at the field office. According to FBI officials, the time frames for completion of these preliminary tasks may also depend on the laws, regulations, policies, and union agreements that may affect the local police agency. The FBI has undertaken various steps to enhance its process for granting security clearances to state and local officials and to facilitate information sharing with state and local law enforcement agencies. Since September 11, the FBI has consulted with state and local officials to collect their views and recommendations regarding information sharing and improving the security clearance process. The FBI identified state and local officials’ unfamiliarity with the requirements for processing security clearance applications as one of the main impediments to timely processing of applications. To improve understanding of its policies for granting security clearances, the FBI published an informational brochure for state and local officials and continued to meet with state and local law enforcement organizations. The FBI also developed policy guidance and a checklist of procedures for FBI field office officials, added staff to the headquarters unit responsible for processing state and local security clearance applications, and developed database resources at headquarters to track applications. To promote information sharing between the FBI and state and local law enforcement agencies, in 2002 and 2003, the FBI increased the number of JTTFs and encouraged state and local officials to participate in this and other information-sharing initiatives. In addition, the FBI distributed terrorism-related bulletins to state and local agencies and has made terrorist threat-related information available via various FBI electronic networks. In response to an increased interest in information sharing between the FBI and state and local law enforcement agencies following the terrorist attacks of September 11, high-level FBI officials met with state and local law enforcement leaders to discuss ways to prevent or respond to terrorist attacks. These discussions included ways to improve information sharing between the FBI and state and local law enforcement agencies, as well as the FBI’s requirements and process for granting security clearances. According to FBI officials, representatives of major state and local law enforcement groups continue to meet periodically with the FBI Director and other FBI officials. In addition, the FBI created the Office of Law Enforcement Coordination (OLEC), which is headed by a former city police chief, to address state and local officials’ concerns. OLEC’s general responsibilities include serving as the FBI’s primary liaison to national law enforcement associations and communicating the perspectives of state and local law enforcement agencies to the FBI. In addition to participating in state and local law enforcement organizations’ meetings with the FBI Director, OLEC staff regularly attend conferences of state and local law enforcement organizations and respond to inquiries from state and local officials. In addition to the FBI’s direct consultation with state and local officials, after September 11, some state and local law enforcement organizations expressed their members’ concerns regarding information sharing with the FBI and the FBI’s security clearance application process. These organizations shared their members’ views with the FBI via white papers, and through meetings and conversations with FBI officials, including OLEC representatives. Officials from the FBI’s Security Division, which houses the units responsible for processing security clearance applications and granting access, also participated in many of these activities. According to FBI and state and local officials, initially following September 11, there was little understanding of, and some confusion regarding the security clearance process among both FBI field office and state and local officials. According to an FBI official, some FBI field office and state and local officials lacked guidance for identifying individuals in need of security clearances. According to an FBI official, some police chiefs or officers equated receiving a security clearance with status or importance, or believed they should be granted a clearance based solely on their status. According to an FBI official, failure to understand or follow FBI guidelines for identifying individuals in need of a security clearance initially resulted in an unnecessary increase in the number of applications submitted. For example, one FBI official stated that a police agency had asked for security clearances for most of its officers, rather than for the one or two officials who needed a clearance. In addition, according to representatives of some law enforcement professional organizations, some state and local officials said they did not have adequate guidance for filling out and submitting the appropriate application forms. According to FBI officials, in some cases, state and local officials were reluctant or refused to provide the information required to conduct the background investigation. For example, police chiefs sometimes omitted negative information because they feared it might affect their standing in the community (e.g., past drug use or poor credit). If an incomplete application was submitted, it had to be sent back to the individual for completion, extending the time required to investigate and process an application. State and local officials expressed discontent with the time frames for processing a security clearance application, as well as dissatisfaction with their inability to check the status of their security clearance applications. With the assistance of the newly created OLEC, in November 2002, the FBI distributed an informational brochure to state and local law enforcement agencies to help improve these officials’ understanding of the FBI’s security clearance requirements and process. The FBI also made this brochure available on its Web site. According to FBI guidance, this brochure, shown in figure 10, is included in the packet of application materials given to state and local security clearance applicants. In addition to distributing this brochure, FBI officials have attended conferences and meetings of national law enforcement organizations. OLEC officials have fielded requests for information, as well as complaints, from state and local officials. For example, state and local officials who wanted to know the status of their applications sometimes contacted OLEC officials. FBI officials acknowledged that it is extremely difficult to give status updates when a background investigation is under way; the FBI can tell an applicant simply whether the investigation is complete or is still being processed. This may not always satisfy the interests of state and local officials requesting updates on the status of the application. Representatives of some law enforcement professional organizations we interviewed stated that the FBI’s guidance and consultation with these organizations has helped improve state and local officials’ understanding of the security clearance application process. An OLEC official also said these efforts had reduced the number of calls it received from state and local officials. To clarify FBI field office officials’ role in processing state and local security clearance applications, FBI headquarters developed policy guidance and a checklist of procedures for FBI field office officials. In an electronic communication sent in December 2002, FBI headquarters spelled out the necessary justification for granting security clearances to state and local officials and laid out processing procedures for various FBI officials. In addition, according to FBI officials, the FBI holds quarterly conference calls with FBI field office security officials to address any issues with the security clearance process. FBI officials cited staffing shortages as an impediment to the security clearance process. According to an FBI official, the unit responsible for state and local law enforcement security clearances initially was understaffed, despite experiencing an increase in workload after September 11. To address staffing needs, the FBI created a new unit within its Security Division Personnel Security Section specifically to handle state and local law enforcement security clearance applications. The FBI also requested increased funding for the division responsible for processing security clearances. FBI officials stated that additional staff had been added to this unit, and caseloads presently are not as heavy as in the period immediately following September 11. Following September 11, and the subsequent increase in state and local security clearance applications submitted to the FBI, the FBI Security Division designed databases to track the submission of applications and application completion dates, among other things. FBI officials said the databases have served as management tools for tracking state and local security clearance applications and monitoring application trends and percentages. These databases are to be integrated into an FBI-wide security clearance tracking database currently under development and set for rollout in the latter part of fiscal year 2004. The FBI credits its JTTFs with enhancing its ability to coordinate its counterterrorism efforts with state and local law enforcement agencies, as well as with other government organizations. Since 1996, the FBI has continued to increase the number of JTTFs in operation across the country. By the end of fiscal year 2001, 35 JTTFs were in operation at FBI field office locations. In response to the terrorist attacks of September 11, and with additional congressional support, the FBI expanded this number to 66 by the end of fiscal year 2002, and to 84 JTTFs by the end of 2003. According to an FBI report on the JTTF program, the FBI anticipates that the number of JTTFs will increase in coming years, depending on the availability of funding. According to one FBI official, the FBI received 30 requests for additional JTTF annexes in fiscal year 2003 and was able to grant 18 of these. As of fiscal year 2003, more than 800 state and local officials served full- time on these forces, along with FBI and other federal government officials. The FBI encourages state and local officials in cities with established or newly created JTTFs to join these forces. FBI field office executives work with their state and local law enforcement counterparts to make JTTF assignments according to staff and other resource availability. In addition, the FBI also established the National Joint Terrorism Task Force (National JTTF) in July 2002, to coordinate the FBI’s local JTTFs. In addition to federal agency officials, state and local officials also serve on the National JTTF, which plans to establish a fellowship program for state and local law enforcement officers. In addition to the JTTFs and the LEO initiative, which was created to provide classified information to state and local officials, the FBI utilizes several means of disseminating terrorism-related information to agencies and individuals outside the FBI, including state and local officials. On a weekly basis, the FBI distributes Intelligence Bulletins, which are tailored specifically for state and local officials. According to a DOJ inspector general’s report, the intent of sending these bulletins is to raise general awareness of terrorism-related issues, though some information may help state and local officials detect or uncover criminal activity related to international terrorism. The FBI also distributes Quarterly Terrorist Threat Assessments to state and local officials. These reports provide a general overview of the terrorist threat and provide summaries of events in different regions in the world that might have an impact on this threat. In addition to distributing these bulletins, the FBI grants state and local officials access to various unclassified networks and databases. State and local officials generally make use of the Law Enforcement Online database and network, as well as the National Law Enforcement Telecommunications System, over which the FBI sends e-mail messages. The FBI also posts names from its Terrorist Watch List on the National Crime Information Center database and state and local law enforcement personnel can access the National Instant Criminal Background Check System. Other efforts, though less widespread, include regular meetings of some FBI field office supervisory agents in charge with state and local officials to discuss terrorism-related investigations taking place in their jurisdictions. The Department of Justice and the FBI provided comments on a draft of this report. The Department of Justice and the FBI generally concurred with our findings, and the FBI provided technical comments, which were incorporated as appropriate. We are providing copies of this report to the Ranking Minority Member of the Senate Judiciary Subcommittee on Administrative Oversight and the Courts. We will also make copies available to others on request. In addition, the report will be made available at no charge on GAO’s Web site at http:/www.gao.gov. If you or your staff have any questions, please contact me at (202) 512-8777 or by e-mail at ekstrandl@gao.gov or Charles Michael Johnson, Assistant Director, at (202) 512-7331. Other key contributors to this report are Todd M. Anderson, Leo Barbour, Nettie Richards, and Jerry Sandau. To describe the FBI’s processes for granting security clearances to state and local officials, we reviewed presidential executive orders, Security Council guidelines, and the Code of Federal Regulations for federal government-wide criteria for granting access to classified national security information (NSI). We reviewed Department of Justice (DOJ) guidance delegating the authority to grant access to classified NSI to the FBI. We also reviewed copies of the relevant sections of the FBI’s Manual of Investigative Operations and Guidelines, as well as copies of guidance sent to FBI field office officials. We reviewed a copy of the brochure the FBI distributed to state and local law enforcement organizations and agencies. We also interviewed officials from DOJ, Justice Management Division. For background on classified information and granting access, we interviewed the Director of the Information Security Oversight Office at the National Archives and Records Administration, as well as staff from our Security and Safety office. We gathered additional information from interviews with officials from the FBI’s Security Division Personnel Security Section, which oversees the unit responsible for granting security clearances to state and local officials. To determine the extent to which the FBI is meeting its timeliness goals for processing security clearances for state and local officials, we analyzed data from the FBI Security Division’s state and local law enforcement security clearance databases collected between September 2001 and December 2003. These data generally included case numbers, beginning and ending dates of background investigations, adjudication resolution, and level of clearance granted, among other things. We determined that the data were suitable and sufficiently accurate for the purpose of our review. We obtained additional context for understanding the FBI’s data through interviews with FBI Security Division officials. To describe the efforts undertaken by the FBI to enhance its state and local law enforcement security clearance and information-sharing processes, we reviewed a variety of reports and studies, including literature published by state and local law enforcement organizations. We also interviewed representatives of state and local law enforcement organizations, including the International Association of Chiefs of Police (IACP), Major Cities Chiefs Association, Police Executive Research Forum, National Association of Chiefs of Police, and the National Sheriffs Association. We also attended an IACP annual conference panel session on the FBI’s information-sharing initiatives. We conducted interviews with FBI Security Division officials, as well as a representative of FBI’s Office of Law Enforcement Coordination. We also obtained and reviewed planning, organization, budget, and staffing documentation from the FBI. We performed our work from July 2003 to March 2004 in Washington, D.C., in accordance with generally accepted government auditing standards. We requested comments on a draft of this report from the Department of Justice and the FBI. | The free flow of information among federal, state, and local law enforcement agencies could prove vital to fighting the war on terrorism. State and local law enforcement officials are key stakeholders in the United States' efforts to combat terrorism, and as such, they may require access to classified national security information to help prevent or respond to terrorist attacks. In order to gain access to such information, state and local law enforcement officials generally need federal security clearances. The Federal Bureau of Investigation (FBI) grants security clearances and shares classified information with state and local law enforcement officials. Immediately following September 11, 2001, some state and local law enforcement officials expressed frustration with the complexity of the process for obtaining security clearances. Others expressed frustration with the length of time it took to obtain a security clearance. These frustrations exacerbated the general concern among law enforcement stakeholders that the lack of security clearances could impede the flow of critical information from the FBI to the state and local level, from the state and local level to the FBI, and laterally from one state or local agency to another. In turn, this potential lack of access to critical terrorism-related information might place local law enforcement officials at a disadvantage in their efforts to respond to or combat a terrorist threat. The Ranking Minority Member, Subcommittee on Administrative Oversight and the Courts, Senate Committee on the Judiciary asked us to examine several issues regarding the FBI's process for granting security clearances to state and local law enforcement officials. This report provides information on: (1) the FBI's process for granting security clearances to state and local law enforcement officials, (2) the extent to which the FBI has met its time frame goals for processing security clearance applications for state and local law enforcement officials and factors that could affect the timely processing of security clearance applications, and (3) efforts undertaken by the FBI to enhance its security clearance and information-sharing processes with state and local law enforcement officials. The FBI's process for granting access to classified information requires state and local law enforcement officials to undergo the same background investigation and adjudication procedures as do individuals who have an employment relationship with the federal government and require access to classified national security information. The FBI's goal is to complete the processing for secret security clearances within 45 to 60 days and top secret security clearances within 6 to 9 months, beginning with the FBI headquarters' receipt of the application from the FBI field office. Since September 11, about 92 percent of applications for top secret security clearances were processed within the FBI's time frame goals. During this same period, about 26 percent of secret security clearance applications were processed within the FBI's time frame goals, although substantial improvements have been made in the most recent quarters for which we have data. The FBI was more successful with processing top secret security clearances within its stated time frame goals than secret security clearances, in part because the FBI often assigns greater priority to processing applications for state and local Joint Terrorism Task Force (JTTF) members, who are required to have top secret clearances. For either secret or top secret security clearance applications, processing timeframes can vary depending on the complexity of individual cases. The FBI has taken a number of steps to enhance its process for granting security clearances to, and sharing information with, state and local law enforcement officials. One of the impediments highlighted was the state and local officials' and the FBI field office staff's lack of a clear understanding of the FBI's security clearance granting process. In response to this impediment, the FBI headquarters widely distributed step-by- step guidance to state and local law enforcement officials and reeducated the FBI field staff on the FBI security clearance process and goals. In addition, the FBI added staff to its headquarters unit responsible for adjudicating state and local security clearance applications and created databases to track state and local security clearance applications. Efforts undertaken by the FBI to enhance information sharing with state and local officials include increasing the number of JTTFs from 35 to 84 and increasing state and local law enforcement officials' participation on these forces. Serving on JTTFs provides state and local law enforcement officials the opportunity to interact with the FBI on a daily basis. The FBI also circulates declassified intelligence through a weekly bulletin and provides threat information to state and local law enforcement officials via various database networks. |
The Middle Class Tax Relief and Job Creation Act of 2012 (2012 act) established numerous responsibilities for FirstNet, most of which relate directly to developing the nationwide public-safety broadband network (hereafter, public safety network). It is intended that this network will be a new, nationwide, broadband telecommunications network. Public safety users, and potentially other “secondary” users, may be assessed fees to use the network, much as they may currently pay for the use of commercial wireless networks. In establishing this network, FirstNet must issue open, transparent, and competitive Requests for Proposals (RFP) to private sector entities for the purpose of building, operating, and maintaining the public safety network; develop RFPs that include appropriate timetables for construction, and network coverage areas and service levels, among other things; enter into agreements to use, to the maximum extent economically desirable, existing commercial, federal, state, local, and tribal infrastructure; promote competition in the public-safety equipment marketplace by requiring that equipment for the public safety network be built to open, non-proprietary standards; promote integration of the public safety network with public-safety address special considerations for areas with unique homeland or require deployment phases that include substantial rural coverage milestones; and develop the technical and operational requirements for the public safety network, as well as the practices and procedures for managing and operating it. In establishing the infrastructure for the public safety network, the 2012 act requires FirstNet to include the following network components, as shown in figure 1: core network of data centers and other elements, all based on commercial standards; connectivity between the radio access network (RAN) and the public Internet or the Public Switched Telephone Network, or both; and network cell-site equipment, antennas, and “backhaul” equipment, based on commercial standards, to support wireless devices operating on frequencies designated for public safety broadband. In developing the public safety network, FirstNet must work with a variety of stakeholders. The 2012 act required FirstNet to be headed by a 15- member board with 3 permanent members (the Secretary of Homeland Security, the Attorney General, and the Director of the Office of Management and Budget) and 12 individuals appointed by the Secretary of Commerce. The appointed members are required to have public safety experience or technical, network, or financial expertise. The 2012 act also required FirstNet to establish a standing public-safety advisory committee to assist it in carrying out its responsibilities and consult with federal, regional, state, local, and tribal jurisdictions on developing the network. For state, local, and tribal planning consultations, FirstNet is required to work with the state Single Points of Contact (SPOC) who have been designated by each state, territory, and the District of Columbia (hereafter, states). The 2012 act requires FirstNet to notify the states when it has completed its RFP process for building, operating, and maintaining the public safety network. Once a state receives the details of FirstNet’s plans, it has 90 days either to agree to allow FirstNet to construct a RAN in that state or notify FirstNet, NTIA, and FCC of its intent to deploy its own RAN. A state that “opts out” of FirstNet’s plan to build that state’s RAN has an additional 180 days from that notification to develop and complete RFPs for the construction, maintenance, and operation of the RAN within the state; the state must send FCC an alternative plan for constructing, operating, and maintaining its RAN. The plan must demonstrate that the state’s proposed RAN would comply with certain minimum technical requirements and be interoperable with FirstNet’s network. FCC shall either approve or disapprove the plan. In addition, a state that opts out is required to apply to NTIA for an agreement to use FirstNet’s spectrum capacity. Various federal agencies will provide assistance and support to FirstNet. For example, The National Institute of Standards and Technology (NIST) within the Department of Commerce, in consultation with FCC, DHS, and the Department of Justice’s National Institute of Justice, is required to conduct research and assist with the development of standards, technologies, and applications to advance wireless public-safety communications. The Public Safety Communications Research (PSCR) program, a joint NTIA and NIST effort, performs research on behalf of FirstNet to advance public-safety communications interoperability. DHS, through both the Office of Emergency Communications and the Office for Interoperability and Compatibility, supports the establishment of the public safety network and collaborates with public safety and government officials at the federal, state, local, and tribal levels to help ensure the network meets the needs and technical requirements of users in the public safety community. The 2012 act also required FCC, the entity responsible for managing and licensing commercial and non-federal spectrum use—including spectrum allocated to public safety—to take certain steps to support FirstNet. It also authorized FCC to provide FirstNet with technical assistance and to take any action necessary to assist FirstNet in effectuating its statutory duties and responsibilities. Radio frequency spectrum is an essential resource for wireless communications, including the planned public-safety network. The energy in electronic telecommunications transmissions converts airwaves into signals to deliver voice, text, and images. These signal frequencies are allocated for specific purposes, such as television broadcasting or Wi-Fi, and assigned to specific users through licenses. Allocating sufficient spectrum for wireless emergency communications has long been a concern for Congress. The 2012 act required FCC to reallocate the “D Block,” a previously commercial spectrum block located in the upper 700 megahertz (MHz) band, to public safety and to grant a license to FirstNet for the use of both the existing public-safety broadband spectrum in the upper 700 MHz band and the D Block. FirstNet now has a license to operate the nationwide public-safety broadband network on spectrum in the upper 700 MHz band, specifically 758-769 MHz and 788- 799 MHz (see fig. 2 below). In October 2013, FCC adopted consolidated technical service rules governing the 700 MHz band of spectrum licensed to FirstNet and was prepared to accept and process applications for equipment certification in that spectrum band consistent with the newly consolidated rules. FCC has conducted specified spectrum auctions, as required by the 2012 act, so that auction proceeds could be used to fund FirstNet. The 2012 act provides $7 billion from the spectrum auction proceeds to FirstNet for “buildout” of the public safety network, reduced by the amount needed to establish FirstNet. The 2012 act requires FirstNet to be self-funding beyond this initial $7 billion. Efforts to establish local and regional public-safety networks are also ongoing, and predate the 2012 act. In 2009, public safety entities began requesting waivers from FCC’s rules to allow early deployment of interoperable public-safety wireless broadband networks in the upper 700 MHz spectrum band authorized for public safety use. Through 2011, FCC granted waivers to 22 jurisdictions for early deployment of such networks.altered the regulatory landscape for the 700 MHz band” and sought However, FCC recognized that the 2012 act “fundamentally comment on the disposition of waiver jurisdiction deployments. In May 2012, NTIA partially suspended seven grant projects for the waiver jurisdictions that received funding from the Broadband Technology Opportunities Program (BTOP), a federal grant program to promote the expansion of broadband infrastructure. NTIA suspended the projects in part to ensure the projects supported the development of FirstNet’s public safety network. In July 2012, FCC decided to hold all waiver authorizations ineffective as of September 2, 2012, and said it would allow such jurisdictions to continue to deploy public-safety broadband services in the existing public-safety broadband spectrum (later extended to include the D Block) under certain limited circumstances consistent with FirstNet’s future use of the spectrum. FCC granted FirstNet the 700 MHz public-safety broadband-spectrum license in November 2012, and in February 2013, FirstNet announced it would begin negotiations with the original waiver jurisdictions on SMLAs. With FirstNet in control of the 700 MHz public-safety broadband-spectrum license pursuant to its FCC license, the jurisdictions had to secure an SMLA with FirstNet to lift their partial funding suspension and resume work. By August 2014, FirstNet secured SMLAs with five original waiver jurisdictions; in this report, we refer to these jurisdictions as early builder projects. Figure 3 shows timelines for the early builder projects and the amount of federal funding they received to deploy their networks. As part of their SMLAs with FirstNet, the early builder projects agreed to report to FirstNet on their use of the spectrum, their progress, and on key-learning conditions, among other things. FirstNet has not yet determined if or how the early builder project networks will be incorporated into its nationwide network. Three other original waiver jurisdictions were awarded BTOP funds to deploy local and regional public-safety broadband networks, including the State of Mississippi, the San Francisco Bay Area in California, and the City of Charlotte in North Carolina. However, for various reasons, these jurisdictions were unable to reach SMLAs with FirstNet and the projects were canceled. Even with the establishment of the public safety network, first responders will continue to rely on their current LMR systems for mission-critical voice communications. We previously reported that a major limitation of a public-safety broadband network is that it will not provide first responders with “mission-critical” voice capabilities—that is, voice capabilities that meet a high standard for reliability, redundancy, capacity, and flexibility necessary during emergencies—for several years. Long Term Evolution (LTE), the technical standard for the public-safety broadband network, is a wireless broadband standard that is not currently designed to support mission-critical voice communications.are currently providing voice over LTE capabilities, but these capabilities do not currently meet public safety’s mission-critical voice requirements because key elements needed for mission-critical voice—such as group calls—are not part of the LTE standard. Therefore, public safety agencies will continue to rely on their LMR systems for mission-critical voice for the foreseeable future, and the public safety network will supplement, rather than replace, current LMR systems. FirstNet has made progress carrying out its statutory responsibilities in three areas—(1) establishing its organizational structure, (2) planning the public safety network, and (3) consulting with stakeholders—but could face challenges in each of these areas. Figure 4 provides a timeline of select events in FirstNet’s progress meeting its statutory responsibilities since the 2012 act’s passage. As a newly created entity within the federal government, FirstNet has taken a number of steps to establish its organizational structure and hire staff. As required by the 2012 act, the Secretary of Commerce appointed FirstNet’s inaugural board members in August 2012. FirstNet’s Board established Governance and Personnel, Finance, Technology, and Outreach Committees to review, approve, and oversee FirstNet’s activities. As required by the 2012 act, FirstNet’s Board also established the Public Safety Advisory Committee (PSAC) in February 2013—one year after the 2012 act’s passage—and adopted the PSAC’s charter in June 2014. The PSAC’s membership represents a broad cross section of public safety disciplines and state, local, and tribal governments. In addition to the board and the PSAC, FirstNet has established organizational units and hired key personnel to lead and perform its work. In April 2013—over a year after the 2012 act’s passage—the FirstNet Board selected an Executive Director to lead its day-to-day operations. Since then, FirstNet has hired, and continues to hire, other senior management personnel to lead its organizational units (such as a Chief Counsel and Chief Administrative, Financial, and Information Officers), and Directors and organizational chiefs to further lead and perform its work. FirstNet has also hired, and continues to hire, general staff. As of February 2015, FirstNet had over 120 employees, including full-time equivalent federal employees, contractors, and personnel on detail from other agencies. FirstNet confirmed in February 2015 that it may hire up to 42 additional full-time equivalent federal employees in fiscal year 2015 and that it will evaluate and adjust needed staffing levels on an ongoing basis. See figure 5 for an overview of FirstNet’s organizational structure. Additionally, in June 2014, FirstNet opened its “corporate” headquarters in Reston, Virginia, and, in March 2014, a technical center to serve as its technical, engineering, and network design headquarters in Boulder, Colorado. Stakeholders we spoke with and surveyed expressed concern that organizational issues have slowed FirstNet’s progress and could continue to do so. In particular, in response to our survey, numerous SPOCs noted either that FirstNet’s placement within NTIA could create “bureaucratic” obstacles or that FirstNet should be more independent from NTIA. One survey respondent and another subject matter expert we interviewed also noted that this placement creates conflicting expectations—that is, stakeholders expect FirstNet to behave like both a commercial wireless carrier and a government entity, and these expectations can sometimes be in conflict. However, FirstNet officials told us that while FirstNet has leveraged its relationship with NTIA in administrative and legal matters, it exercises strong independence in decisions that are directly program- related. Numerous stakeholders we surveyed and interviewed were also concerned about the pace of FirstNet’s hiring, noting that the federal hiring process is too slow, has not allowed FirstNet to hire staff quickly enough, and has delayed FirstNet’s progress. For example, FirstNet hired its management team for state plans, consultation, and outreach in late 2013 to early 2014, then hired State and Local, Tribal, and Federal Outreach Leads in June and August 2014, almost 2 years after FirstNet’s board members were appointed. Additionally, FirstNet is still in the process of hiring staff for key positions. For example, as of February 2015 FirstNet was re-filling the Chief User Advocacy Officer and Director of Communications positions, was planning for the hiring of a Chief Procurement Officer, and was in the process of hiring staff to lead regional consultation teams that it plans to establish, among other positions. As of February 2015, it was also still hiring key technical positions—including Director of Devices, Director of Standards, and Director of Core Network—and re-filling the Chief Technology Officer position. According to FirstNet officials, FirstNet faces challenges hiring as quickly as it would like, in part due to government hiring procedures, but is seeking direct hire authority from the Office of Personnel Management and is exploring other authorities it could use to expedite hiring. To plan the public safety network and help ensure that its approach is open and transparent and meets the 2012 act’s requirements and that interested parties have a formal way to comment on FirstNet’s strategy and decisions, FirstNet intends to follow the Federal Acquisition Regulation (FAR) process for its comprehensive network-services procurement. This process will culminate in one or multiple RFPs for “network solution(s)”—that is, proposals for the building, deployment, operation, and maintenance of the public safety network. To help draft the RFP(s), FirstNet has sought comments through a Notice of Inquiry (NOI), RFIs, and Public Notices. Specifically, in October 2012, NTIA, on behalf of FirstNet, issued an NOI to seek input on the network’s potential architecture.RFIs on specific technical aspects of the network and on devices and From April to November 2013, FirstNet issued 12 detailed applications for public safety. NTIA and FirstNet received over 400 comments to the NOI and RFIs, and FirstNet has used these as a source of market research in assessing industry capabilities and developing the technical design of the network. Building off these efforts, in September 2014 FirstNet issued an RFI that sought further comment on a number of issues, such as network coverage, pricing, deployment strategies, security, prioritization among network users, and customer service, among other things. According to this RFI, FirstNet’s current approach is to follow a performance-based procurement strategy, whereby FirstNet does not dictate the specific network solution that bidders responding to the RFP(s) must provide, but, rather, outlines objectives and encourages bidders to develop proposals that will meet those objectives. Additionally, although the 2012 act exempts FirstNet from portions of the Administrative Procedure Act (APA), FirstNet has chosen to follow an APA-like process in order to provide more opportunities for interested parties to comment on its interpretations of the 2012 act’s requirements. Therefore, in September 2014, FirstNet also issued a Public Notice seeking comment on a number of its preliminary interpretations, such as how to define its rural coverage milestones and eligible network users.FirstNet received 185 responses to the September 2014 RFI and Public Notice from a broad range of respondents, such as states, public safety entities, private companies, and associations. FirstNet issued a second Public Notice in March 2015 and expects to issue a draft RFP in spring 2015. In addition to its efforts seeking public comment, FirstNet has also received technical input from a variety of stakeholders, such as FCC, the PSAC, the National Public Safety Telecommunications Council (NPSTC), and the PSCR program. As required by the 2012 act, the FCC’s Technical Advisory Board for First Responder Interoperability issued minimum technical requirements for the public safety network in May 2012. The report provided recommendations on LTE standards; network user equipment, device management, and evolution; and quality of service, among other things. FirstNet has also asked the PSAC to provide recommendations on various topics, including a framework for establishing the priority of public safety entities on the network. In May 2014, NPSTC provided FirstNet with guidance on what makes communication systems “public-safety grade.” Additionally, FirstNet has provided the PSCR—which is based in Boulder, Colorado, where FirstNet’s technical center is also located—with funding to research interoperability standards, test and evaluate potential network features, and model and simulate network data traffic. Specifically, SPOCs responded this 33 percent of the time. For question wording, see appendix II question 5 (b-g). network, and the practices, procedures, and standards for managing and operating the network. Still, even in these instances, SPOCs’ level of satisfaction or dissatisfaction was mostly “moderate.” While many stakeholders we surveyed and interviewed noted that FirstNet’s progress has been too slow, some also noted that they were satisfied with FirstNet’s progress given the complex nature of FirstNet’s tasks and that it is a government entity subject to federal rules and regulations. Some also said that FirstNet’s progress has improved recently, especially as FirstNet has hired more staff. FirstNet has initiated a process to consult with the public safety community in each state through the SPOC. FirstNet first worked to establish informal mechanisms for coordinating with the SPOCs. For example, FirstNet began holding regional workshops in May 2013, quarterly SPOC webinars in January 2014, and monthly calls with SPOCs in February 2014. FirstNet began its formal state consultation process by delivering an initial consultation package to each SPOC in April 2014; the package contained a checklist for the SPOCs to complete in preparation for an initial consultation meeting. FirstNet is using these meetings to gather information on each state’s unique challenges, needs, and processes to inform its development of the public safety network and, in particular, state RAN plans. Initial state consultation meetings began in July 2014—when FirstNet conducted its first consultation with Maryland— and FirstNet expects these initial meetings to continue through 2015, with additional rounds to follow. As of April 16, 2015, FirstNet had conducted initial consultation meetings with 18 states, Puerto Rico, and the District of Columbia. In addition to its consultation with SPOCs, FirstNet officials have conducted outreach to other stakeholders. For example, from October 2013 to February 2015, FirstNet officials visited 39 states and territories while participating in 187 events, such as state town halls and public safety, industry, and government (including federal, state, local, and tribal) conferences. To engage with tribal entities, FirstNet staff visited eight tribal nations in 2014, and sought the advice of the PSAC on tribal outreach, education, and inclusive consultation strategies. To engage with federal entities, FirstNet has hired a director of federal outreach, designated DHS’s Emergency Communications Preparedness Center (ECPC) as the primary body to FirstNet for federal consultation, and participated in meetings with ECPC and other agencies. To engage with vendors interested in doing business with FirstNet, FirstNet has held vendor meetings and appointed an industry liaison to coordinate vendor outreach. In addition to this targeted outreach, FirstNet also launched a public website in March 2014, where it regularly posts updates, presentations, board-meeting minutes, a list of upcoming speaking engagements, and other information. The 2012 act also created a State and Local Implementation Grant Program (SLIGP), administered by NTIA, to provide states with funds to plan for the public safety network and to consult with FirstNet. Starting in July 2013, NTIA awarded $116 million in SLIGP grants. We believe FirstNet’s consultation and outreach activities generally align with core principles for effective stakeholder participation that we have developed and used in previous reports. For example, we found during the course of our review that FirstNet is using an open and clearly defined decision-making process, actively conducting outreach, involving stakeholders throughout the process, using formal and informal participation methods, and including all stakeholders. The majority of stakeholders we surveyed were generally satisfied with the level of FirstNet’s consultation and outreach, but others were dissatisfied and said that they would like more new information. In response to our survey, 54 percent of SPOCs said they were either “moderately” or “very” satisfied with FirstNet’s overall level of consultation, coordination, and communication, while 22 percent said they were either “moderately” or “very” dissatisfied. However, numerous stakeholders we surveyed and interviewed said that they would like more new and detailed information and that they would like FirstNet to focus more on certain aspects of the public safety network during outreach. For example, in response to our survey, SPOCs most frequently indicated that there should be a “large increase” in how much FirstNet focuses on various technical aspects of developing the network in its state outreach.SPOCs also said that the lack of new information from FirstNet, such as details about the network’s design, hampers their ability to conduct local outreach. However, in response to our survey many SPOCs also acknowledged that the level of new and detailed information exchanged will likely increase once they hold their initial state consultation meeting with FirstNet. Additionally, officials from one federal agency with public safety responsibilities said they would have liked to see more in-depth outreach to federal entities, given that federal agencies will have a large pool of potential network users. In January 2015, FirstNet formally initiated its federal consultation process. As part of this process, FirstNet will conduct meetings with federal entities in 2015 similar to the initial consultation meetings it is conducting with states. Internal controls are the plans, methods, policies, and procedures that an entity uses to fulfill its mission, strategic plan, goals, and objectives. An effective internal control system increases the likelihood that an entity will achieve its objectives. We assessed FirstNet’s policies and practices against two components of an effective federal internal control system: risk assessment and control environment.two components because risk assessment provides the basis for developing appropriate risk responses and control activities and the control environment is the foundation for an internal control system. For those two components, we found that—although FirstNet has begun establishing policies and practices that are consistent with federal standards—FirstNet lacked certain elements that contribute to the proper implementation of effective internal control systems. While FirstNet has stated that it is relying on the Department of Commerce’s and NTIA’s internal controls where it has not developed its own, it is also important for FirstNet to implement its own controls, as the Commerce Office of Inspector General (OIG) noted as early as February 2014 in a memo on the management challenges facing FirstNet. In an internal control system, according to federal internal control standards, management should assess risks facing the entity as it seeks to achieve its objectives. Specifically, entities should first clearly define their objectives then identify and analyze risks from both internal and external sources. Analyzing risks generally includes estimating the risk’s significance, assessing the likelihood of its occurrence, and deciding how to respond to it. Risk assessments inform an entity’s policies, planning, and priorities, and help entities develop responses to the risks they face, so that they can achieve their objectives. Control activities respond to these risks. FirstNet has set objectives and taken some steps to assess risks. Specifically, FirstNet has set three key objectives: 1. Provide FirstNet services that are critical to public safety users and differentiate FirstNet services from commercial broadband services, such as through reliability, resiliency, coverage, functionality, interoperability, quality of service, priority access, pre-emption, and applications. 2. Reduce costs for public safety entities by leveraging the value of excess network capacity with partners. 3. Provide mechanisms for public safety entities (directly or indirectly through the states) to benefit from the economies of scale created by FirstNet in terms of purchasing, partnering, and information/data. FirstNet has further delineated how it will accomplish these objectives in a “roadmap” that identifies additional long-term and short-term objectives and milestones. FirstNet has established a Program Management Office to, according to FirstNet officials, help set internal timelines and monitor the completion of tasks needed to achieve these objectives and reach these milestones. Additionally, the Department of Commerce and FirstNet have performed some risk assessment activities. In February 2014, the Commerce OIG issued a memorandum outlining the top management challenges facing FirstNet. To support development of the roadmap, FirstNet created a “risk register” that identifies some risks related to its financial sustainability as well as possible counter-measures. However, FirstNet has not yet fully assessed risks it may face in accomplishing its objectives. FirstNet officials told us in November 2014 that they have not yet done so because they are in the process of defining risk factors and, again in December 2014, because they are in the process of conducting a legal compliance risk assessment of certain key risk areas. In December 2014, FirstNet officials also said that they intend to perform periodic risk assessments in various areas to manage and mitigate risks on an iterative basis. However, as of February 2015, FirstNet has not yet completed these risk assessment activities. As a result, we were unable to evaluate the extent to which these activities align with the elements of risk assessment detailed in the federal internal control standards, and therefore, it remains unclear how effective FirstNet’s efforts will be in helping it to identify and respond to obstacles to fulfilling its responsibilities. As FirstNet completes these assessments, we believe that it is important that it incorporate all of the elements of risk assessment detailed in the federal internal control standards. Lacking complete risk assessments (that is, assessments that incorporate these elements), FirstNet’s control activities may not be designed to respond to the appropriate risks. In previous work, we found that when an agency worked quickly to establish a new program, it resulted in the agency’s hastily designing an internal control system that was not based on complete risk assessments and that the agency responded to risks in a reactive, rather than a proactive, manner. Thus, it was unclear whether the controls appropriately responded to risk or were the best use of the agency’s resources. Additionally, we are concerned that the complexity of FirstNet’s responsibilities and objectives, makeup of FirstNet’s Board of Directors, and challenges that FirstNet will face attracting users to its network and becoming self-funding illustrate the multitude of potential risks FirstNet faces in achieving its objectives. For example, as we point out later in this report, various factors could hinder whether public safety entities adopt the public safety network—and thus how much user fee revenue FirstNet can collect—which could pose risks to FirstNet’s ability to become self- funding. Given this, complete risk assessments could help FirstNet appropriately design its full internal control system and achieve its objectives while maximizing use of its available resources. The control environment is the foundation for an internal control system and provides the basic structure that helps an entity achieve its objectives. To help set this environment, according to federal internal control standards, an entity should establish an organizational structure and delegate authority establish appropriate human-capital practices for hiring, developing, evaluating, retaining, supervising, and disciplining personnel; demonstrate a commitment to competence, such as by establishing expectations of competence and holding personnel accountable by evaluating their performance; demonstrate a commitment to integrity and ethical values, such as by setting a positive “tone at the top,” providing and evaluating adherence to ethical and behavioral guidance, and removing temptations for unethical behavior; and have an oversight body that oversees the entity’s internal control system. FirstNet has taken a number of steps to begin establishing an effective control environment. As described above, FirstNet has established an organizational structure with clearly designated responsibilities and has explored hiring options that would allow it to recruit individuals more quickly. According to FirstNet officials, to develop staff, FirstNet has also leveraged its relationship with the Department of Commerce by using the Department’s training facilities as it works to create FirstNet-specific training. Per Department of Commerce policy, FirstNet’s Senior Executive Service employees are subject to applicable competency and evaluation plans and other employees are evaluated using a performance appraisal process. Additionally, FirstNet has taken steps to identify and manage potential conflicts of interest. For example, FirstNet has held ethics briefings, distributed ethics documents, held ethics counseling and training, and instituted a Board Member Vendor Interaction Policy to establish processes for board members interacting with vendors with a potential interest in FirstNet’s procurement efforts. FirstNet has also established and disseminated a variety of guidance documents, such as policies on employee timekeeping, expenses, travel, information technology rules of behavior, and telework. Finally, FirstNet’s Board Committees and FirstNet senior management’s Compliance Committee oversee FirstNet’s activities and the Commerce OIG and others perform additional oversight. However, FirstNet has not yet finished establishing its control environment. While FirstNet has a variety of separate guidelines and policies, it does not have a uniform and cohesive standards of conduct policy. Specifically, although FirstNet officials told us that they intend to develop a code/standards of conduct policy, which is an important form of ethical and behavioral guidance for personnel, they have not yet developed this item as of February 2015. According to FirstNet officials, FirstNet has not yet done so because, as a “start-up” entity, building up the organization while making progress meeting statutory responsibilities is a balancing act affected by FirstNet’s priorities and resources. Nonetheless, absent standards of conduct, we are concerned that FirstNet may not be able to address deviations in its personnel’s conduct and performance, and take corrective actions in a timely manner. Indeed, FirstNet itself established a special committee in May 2013 to review ethical concerns raised by one of its board members. Similarly, in a December 2014 report, the Commerce OIG identified concerns with FirstNet’s financial disclosure reporting and contracting practices, among other things. The report highlighted that the FirstNet Board, out of necessity, includes members with significant ties to the telecommunications industry that make strategic decisions regarding FirstNet’s operations and, thus, are at increased risk of encountering conflicts of interests. Although FirstNet has taken corrective actions since the Commerce OIG’s investigation, we believe that establishing and evaluating adherence to standards of conduct may help FirstNet ensure that all its personnel are held accountable for their actions. By establishing this item, FirstNet could also foster stakeholder trust in its ability to meet its statutory responsibilities and be a good steward of public funds. Various entities have estimated the cost to construct and operate a nationwide network for public safety from a low of $12 billion to a high of between $34 and $47 billion, over the first 10 years. As shown in table 1, a variety of entities have developed cost estimates for a public-safety broadband network, although they have used different assumptions about the network’s scope. Key assumptions influencing these estimates include whether the network is constructed, operated, or financed in partnership with commercial entities, and the number of sites needed to provide the network’s coverage. For instance, FCC’s 2010 estimate assumes costs would be decreased through a high level of cooperation with commercial carriers. These estimates also vary, for instance, on how much they expect ongoing maintenance and operation to cost. However, some differences among these estimates are difficult to identify since some of the estimates do not explicitly state all their assumptions. The actual costs per site for early builder projects vary but are generally less than the estimates above. As discussed more below, five early builders are constructing local or regional broadband networks for public safety. Some of these projects have begun construction and provided us with cost data. The new cell towers (and associated LTE equipment) for the regional network in the Los Angeles area average $196,000 per site. For equipment being attached to existing towers and sites, the average cost is $102,000. For the regional project in Adams County, Colorado, the cost per site is approximately $75,000. This cost reflects sites utilizing existing infrastructure that does not require strengthening. For budgetary purposes, the Adams County project has estimated the development of a new site at $500,000, an estimate that includes acquisition of land, tower construction, and utilities. GAO-09-3SP. associated documentation were deemed business sensitive. Therefore, we cannot say if the estimate is in line with the best practices associated with the credible and accurate characteristics of our Cost Estimating and Assessment Guide. Cost estimates notwithstanding, various factors will influence the cost of constructing and operating FirstNet’s public safety network, including (1) the business model used, especially the extent of commercial partnerships; (2) use of existing infrastructure; (3) efforts to ensure network reliability; and (4) network coverage. FirstNet’s business model, especially the extent to which it partners with commercial carriers or other private enterprises, will influence the cost to construct and operate the public safety network. The 2012 act gives FirstNet the authority to engage in a variety of commercial partnerships. Such partnerships could involve a private-sector partner that would contribute resources to the network (e.g., infrastructure) and accept some risk in the form of profits or losses. FirstNet would then contribute other resources to the partnership (e.g., spectrum) conditioned on the network satisfying social objectives (i.e., enhancing public safety communications). In a partnership, public safety and commercial users could share the public safety network’s infrastructure and spectrum, with public safety given priority to all network capacity during times of emergency. Regardless of the approach, some public safety stakeholders we spoke with maintained the need for FirstNet to work with commercial partners in building and operating the network for it to be financially sustainable. One study calculated that the value of serving both commercial and public safety users is greater than the costs of the additional capacity and signal reliability requirements placed on the network to serve both public safety and commercial users, demonstrating a strong business case for a public-private partnership. The use of existing infrastructure will influence the cost to construct and operate FirstNet’s public safety network. Under agreements to share existing wireless network infrastructure, FirstNet may be able to make use of, for example, cell towers, antennae, cabling, radio-processing equipment, backup power facilities, and the links between towers and the nearest communications hub, to the extent economically desirable to do so. According to FCC estimates, capital costs would be 2.5 times greater without this form of sharing. Given these potential financial savings, a few public safety stakeholders we spoke with maintained that FirstNet should use at least some existing infrastructure for the network. The use of existing infrastructure can have limitations though. For instance, using existing infrastructure can limit the design and coverage of the network, since existing towers and buildings can only facilitate certain network coverage given their physical location. Negotiating access to existing infrastructure can also be a time-consuming process, especially with government-owned or controlled facilities, and where contracts must by executed with multiple owners, ultimately slowing down network deployment. For example, when we spoke with Swedish officials about the public-safety communications network in their county, they said that they sought to use existing infrastructure, to save costs, when constructing their network, but faced problems in their largest cities convincing tower owners to allow the government to rent the towers. Furthermore, there is a risk when public safety entities rely on infrastructure owned by commercial operators, particularly if they have to rely on a single provider in any given location that can then charge high fees. FirstNet’s approach to ensure the public safety network is safe, secure, and resilient (that is, the overall reliability of the network) will also influence the cost to construct the network. FirstNet is required by the 2012 act to “ensure the safety, security, and resiliency of the network.” This interoperable network in Sweden is called the “Rakel” network and is primarily used for voice communications. Pub. L. No. 112-96, § 6206(b)(2)(A), 126 Stat. 156, 212. FirstNet is still determining how it will satisfy this requirement. As previously described, NPSTC published a report to provide guidance for FirstNet as it constructs and implements the public safety network. NPSTC concluded that a “public-safety grade” communications system should be designed to resist failures due to manmade or natural events as much as practical, and that the public safety network must be constructed to meet as many of these requirements as possible. If FirstNet implements all of NPSTC’s best practices, though, it will significantly add to the cost of building the network. For example, transmission sites, such as cell towers, should have backup power sources when used for public safety communications, according to NPSTC. Existing commercial sites, however, generally do not have such backup, primarily to reduce costs in extremely competitive markets. FCC, for instance, has reported that it could cost $35,000 per site to harden existing commercial LTE sites. The public safety network’s coverage will also influence the cost to construct and operate the network. The 2012 act requires FirstNet to establish a “nationwide” network, but does not define the level of coverage that constitutes “nationwide.” Generally speaking, increasing the area covered by the network, as well as the extent to which coverage penetrates buildings, increases the amount of infrastructure needed, and thus the cost of the network. It may be relatively affordable, for example, to cover large segments of the population concentrated in relatively small areas. For instance, one stakeholder we spoke with suggested that FirstNet could provide service to as much as 25 percent of all potential customers by covering just the 8 largest metropolitan areas. Providing coverage outside dense metropolitan areas can be particularly expensive. One study has shown that a nationwide public-safety broadband network would generally be profitable in urban areas and unprofitable in rural areas. This study demonstrated that a network built with a commercial partnership could cover 94 percent of the U.S. population and break even because urban areas could subsidize coverage in rural areas. Although FirstNet has various revenue options it is authorized to use to become self-funding, it is unclear how FirstNet will use those authorities. As the cost estimates discussed above illustrate, FirstNet’s network will likely cost tens of billions of dollars to construct and initially operate. As also noted above, FirstNet is required to be self-funding. To meet the costs of building and maintaining the network, FirstNet may generate revenue through user fees and commercial partnerships, the latter of which can involve the secondary use of the network for non-public safety services. However, FirstNet faces difficult decisions determining how to best utilize these revenue sources. Additionally, regardless of the effectiveness of FirstNet’s use of these revenue sources, the public safety network will likely have net negative income in the first few years of operation. More than 75 percent of survey respondents noted that the network would be “very useful” to emergency management, emergency medical services, fire services, and law enforcement. For question wording, see appendix II question 1. ability to have reliable access to broadband data throughout the state would improve interoperability,” as well as a first responder’s ability to deliver critical services in a timely manner. Despite the demand for the public safety network, a variety of challenges could hinder adoption and thus user fee revenue: Fee size: If FirstNet’s user fee is too high it could hinder public safety adoption, and if it is too low it could bring in too little revenue. Numerous stakeholders we spoke with noted that FirstNet’s cost would play a role in whether they adopt the public safety network and that user fees must be competitive with existing commercial services. According to a few public safety entities we spoke with and the SPOCs we surveyed, public safety entities currently pay $20-$100 per user or device, per month, for commercial services. While low user fees would be attractive to public safety entities and therefore may increase adoption, they would also bring in a relatively smaller amount of revenue per user. As a FirstNet Senior Program Manager reported in December 2014, there is a trade-off with low user fees between adoption and the network’s financial sustainability. Some stakeholders also noted that the cost of equipment and devices needed to access the public safety network could limit adoption, especially since public safety entities are continuing to invest in their LMR devices and equipment. User base: While a large user base can potentially bring in significant user fee revenue, it could be challenging to manage. The 2012 act established that FirstNet’s primary customers will be entities that provide “public safety services.” How FirstNet interprets the definitions of “public safety services” established in the 2012 act will expand or contract the potential sources of revenue. As one public safety official we spoke with noted, the network has more value to public safety entities when there are more users on the network, because entities will all be able to communicate with each other. Another public safety official we spoke with suggested that certain users could be required to adopt the public safety network. Government users in Sweden, for instance, are required to pay a user fee, regardless of their use of the network, to generate revenue that is necessary to support the national network’s maintenance and operations. However, a large user base can require priority and preemption rules, if certain users are to have privileged access to the network. According to some public safety officials we spoke with, such rules can be difficult to establish among public safety entities. Coverage: Widespread network coverage can attract more users, and thus user fee revenue, but is expensive to construct and maintain. As the FirstNet Senior Program Manager reported in December 2014, there is a trade-off with increased coverage between adoption and the network’s financial sustainability. Further, FirstNet does not have total control over the network’s coverage, since states may opt out and build their own RANs. Nevertheless, some of the public safety entities we spoke with said that the network’s coverage would play a role in whether they adopt the public safety network, noting in particular that the coverage should be at least as good as existing commercial services. One public safety entity we spoke with said that existing commercial coverage is inadequate, while two other entities said that the coverage is adequate normally, but the service becomes unusable during large events because of the number of users on the network. However, as noted above, providing extensive coverage, especially in rural areas, can be very costly. Indeed, a few SPOCs noted in survey responses that providing rural coverage in their states would be challenging, with one commenting that “it is inconceivable that FirstNet will be able to deploy a terrestrial network in the vast areas that are unpopulated or sparsely populated.” Reliability: Although FirstNet is required to construct a resilient network, practices to ensure this can be costly. Some public safety officials we spoke with said that the network’s reliability would play a role in whether they adopt the public safety network. A few officials specifically said that if the network did not reliably work when first utilized by public safety, adoption would suffer, since public safety has a low tolerance for unreliable technology. However, as noted above, ensuring reliability requires significant capital expenses. As the FirstNet Senior Program Manager reported in December 2014, there is also a trade-off with hardening the network between the extent of adoption and the network’s financial sustainability. Compounding these challenges are other factors that might hinder adoption. For instance, officials from some public safety entities told us that public safety tends to take a “wait and see” approach to adopting new technology. Public safety, according to an official we spoke with, can be reluctant to buy and use new technology and services because if those things do not work it can put lives in jeopardy. Furthermore, according to officials from two public safety entities we spoke with, some entities may not see a need for a nationwide public-safety network if they operate in areas with few large-scale emergencies. FirstNet can also generate revenue through commercial partnerships, but the extent of commercial interest in these partnerships, and therefore the value of this authority for FirstNet, is currently unknown. Under the 2012 act, FirstNet can receive payment for the use of the public safety network’s capacity by non-public safety users as well as use of the network’s infrastructure. The value of secondary access to the public safety network’s capacity depends in part on the availability of the spectrum, which itself will be determined in part by the capacity available given the network’s design. According to one major carrier we spoke with, no business is likely to enter into a partnership with FirstNet because its public-safety user base has not been defined, and thus the network’s capacity available to secondary (commercial) users is unknown. According to this carrier, the risk would be too high for a commercial entity to enter into an agreement without knowing exactly how they will be able to use FirstNet’s network. If public safety preempts all commercial traffic, then the commercial entity will struggle to generate income from this venture and may lose favor with its customers. However, another major carrier we spoke with maintained that FirstNet will have to partner with at least one commercial carrier to be financially sustainable, and given the significant investments in LTE infrastructure made by commercial carriers, FirstNet would do well to utilize some of this infrastructure through commercial partnerships. Although the historic Advanced Wireless Services spectrum auction that FCC concluded in late January 2015 could indicate demand for spectrum capacity among commercial carriers, the extent of carrier interest in partnering with FirstNet is not yet fully known. In particular, there may be some benefits to existing commercial carriers in partnering with FirstNet, but these companies may prefer to expand their businesses by directly competing with FirstNet and offering their own public safety products. Notably, when FCC presented the D Block spectrum for auction in 2008 with public safety encumbrances, it received no qualifying bids and thus the D Block was not licensed. The lack of commercial interest in the D Block was due in part to uncertainty about how the public-private partnership would work, which raises further questions about FirstNet’s ability to partner with commercial carriers. Officials from the five early builder projects, as well as the three canceled projects, told us that they have learned a number of lessons while developing their public-safety broadband networks that may be useful as FirstNet develops its public safety network. Specifically, the early builders identified lessons about (1) governance, (2) financing the network, (3) conducting outreach, and (4) planning for network deployment. Officials from the early builder projects cited governance lessons associated with developing a new network for public safety. As we reported in February 2012, governance is a key element for interoperable networks. By providing a framework for collaboration and decision making with the goal of achieving a common objective, governance structures can promote interoperability and help ensure public safety networks are secure and reliable. The 2012 act established FirstNet as the governing entity for a nationwide public-safety network, and as such, early builder project officials described governance challenges that FirstNet may face. For example, officials from one project told us some public safety entities may not have a clear understanding of FirstNet’s goals and plans. The officials told us localities are willing to participate in the public safety network, but FirstNet will face difficulty in establishing timely technical decisions and effective policies that keep pace with local enthusiasm to participate. The officials said FirstNet can address this challenge by setting expectations about what the network will provide, including the specific intent, purpose, and planned capabilities. A SPOC working with one of the projects described challenges that FirstNet will face in determining how to diplomatically work with tribal nations. For example, whereas each state will have a designated SPOC, tribes are sovereign nations within a state but will not have a designated SPOC, which could pose governance challenges. The SPOC told us it is unclear whether FirstNet has planned for how the “opt in” or “opt out” process will work for tribal nations, when the 2012 act requires state governors to make the state decision. The SPOC also told us that a tribe representative in his state met with him to share concerns about how public safety entities only have broadband coverage near the edge of their reservation’s limits. The SPOC noted that the tribe has become a partner of the early builder project and that this partnership underscores how their state has developed close relationships with tribes through state-level liaison efforts, federal grant programs, and its early builder project. According to the SPOC, FirstNet should work closely with states and leverage these relationships to work with tribal nations. Officials from the early builder projects also learned lessons related to financing a new public safety network that could be applicable to FirstNet. Although the 2012 act provides FirstNet with funding sources and options, as described above, utilizing such sources could be challenging according to early builder project stakeholders. Officials from a project told us they will face sustainability challenges due to the limited number of users that will be able to utilize their network. According to the officials, their project will not be able to charge their users enough to make the operations sustainable without pricing the users out of the services. The officials told us a possible way to address this challenge would be to expand the service to public safety entities in neighboring metropolitan areas and airport services areas that have established broadband infrastructure. Similarly, officials from one of the projects said their network will be in an “uneasy” financial position initially because the number of users subscribing to the network will likely be low. Further, the officials noted that a small user base could make it difficult to maintain the network because user fees are expected to pay for a significant portion of network operations and maintenance. According to a SPOC working with one of the projects, determining user fees to cover ongoing maintenance and administrative costs has been a financial challenge. To decide user fees, project officials are considering factors including the potential for public-private partnerships and how to define the scope of the user base. As we found in February 2012, obtaining adequate funds to build, operate, and maintain a public safety network could be challenging. Mobile “deployables” can also be referred to as “deployable networks.” A deployable network typically includes “deployable assets” such as “cells on wheels” that provide localized wireless network service to areas where coverage is minimal or compromised. These assets typically provide fully functional service via vehicles such as trailers, vans, and trucks. challenges in their state on the topic of network coverage. According to the SPOC, their state’s public safety community and tribal nations became frustrated with FirstNet’s inconsistent messaging about the network’s coverage and capability. The inconsistent messaging created challenges for the SPOC in convincing frustrated potential users to remain engaged in the state’s FirstNet planning efforts. We also spoke with a SPOC closely involved in one of the early builder projects who described strategies that his state’s project used in conducting outreach. For example, the project developed contact lists for each site deployment location and for primary and secondary stakeholders, and distributed materials in public safety locations including police stations. According to the SPOC, FirstNet will need to sustain a level of excitement in its outreach for the public safety network. Officials from the early builder projects as well as the canceled public- safety projects also described lessons they learned about planning their network’s deployment that could benefit FirstNet. Officials from one project told us they learned specific lessons about site selection, permits and site access agreements, and equipment choices. The officials also told us they selected the RFP response that was the most economical in its use of existing infrastructure. Officials from another project told us how they chose to deploy in densely populated areas with high-crime rates, where public safety coordination is typically challenging and in most need of improvement. The officials also said they provided guidance to local police departments that helped them avoid procuring communications devices that would not be compatible with the project or FirstNet’s public safety network. An official from one of the canceled projects said his team faced several challenges including local zoning conditions that affected project schedule and cost, a newly passed city code that required towers to withstand higher wind loads and that increased costs, and commercial competitors lowering their subscription rates to compete with the planned public-safety network. Officials from an ongoing project told us their project initially identified “buildout” sites but learned that environmental assessments would need to be completed for each site. Doing these assessments would threaten the project’s ability to follow its project schedule. To address the challenge, the project narrowed its buildout site pool to exclude marshlands and other areas with obstructive tree lines and to include publicly owned sites such as police and fire stations. With the publicly owned sites identified, project officials worked with their state’s legislature to pass an exemption to state environmental reviews. According to a SPOC we spoke with, the project’s efforts on this issue will reduce the project’s overall build time. The official from the project told us that an important lesson learned is to thoroughly understand all of the process steps and risks prior to plan execution. According to the project official, it will be important for FirstNet to be able to navigate similar issues and challenges. While FirstNet has taken steps to collect and evaluate lessons learned from the early builder projects, it could do more to ensure that the lessons are properly evaluated. We have previously found that a well-developed evaluation plan for projects like the early builder projects can help ensure that agencies obtain the information necessary to make effective program and policy decisions. A well-developed evaluation plan should include, at a minimum, several key features including the following: well-defined, clear, and measurable objectives; criteria or standards for determining project performance; a clear plan that details the type and source of data necessary to evaluate the project, methods for data collection, and timing and frequency of data collection; and a detailed data-analysis plan to track performance and evaluate the projects’ final results. procedures and experiences. According to a SPOC involved in one of the projects, FirstNet’s key-learning conditions cover a comprehensive set of issues and should provide valuable data for FirstNet. Under the SMLAs, the projects also agree to provide FirstNet with quarterly reporting on their project’s use of FirstNet’s spectrum, progress achieving project milestones, and in some cases, the experiences of their network users. In October 2014, FirstNet provided the projects with quarterly report templates, instructions, and timing for completing the reports. Additionally, FirstNet intends to gain knowledge from the projects through contractors who have been assigned to each project to provide information and collect formal and informal lessons.contractors are using an informal lessons log to track observed lessons and whether they have been incorporated into the technical documents used to guide FirstNet’s acquisition of a comprehensive network solution. FirstNet officials also told us that they hold weekly meetings to review early builder project status, progress on SMLA key-learning conditions, and informal key lessons. Finally, in April 2014 FirstNet authorized the PSAC to establish an Early Builder Working Group to provide advice on the strategies and lessons learned related to the early builder network development, outreach, and consultation. As of January 2015, the PSAC planned to submit the Early Builder Working Group’s first series of recommendations to FirstNet. Although FirstNet has taken these steps, we are concerned that it lacks a detailed data-analysis plan to track the performance and results of the early builder projects. For the early builder projects, their performance and results are captured in the observations and lessons learned reported to FirstNet and identified by consultants. Tracking the early builder projects’ observations and lessons against FirstNet technical documentation is necessary to ensure that the lessons have been addressed and also facilitates transparency and accountability for FirstNet’s decision-making. Even though FirstNet staff and contractors remain in close contact with the early builder projects, without a data- analysis plan to track those projects it is unclear how FirstNet intends to evaluate the projects’ observations and lessons and determine whether or how the lessons are addressed. As a result, we believe that FirstNet could miss opportunities to leverage key lessons related to governance, finance, outreach, and network deployment. Given that the early builder projects are doing, in part, on a regional and local level what FirstNet must eventually do on a national level, a complete evaluation plan that includes a detailed data-analysis plan could play a key role in FirstNet’s strategic planning and program management, providing feedback on both program design and execution. Furthermore, such a plan could provide FirstNet officials the opportunity to make informed midcourse changes as they plan for the public safety network, and help ensure that lessons from these projects are evaluated in ways that generate reliable information to inform future program-development decisions. The lack of interoperability in public safety communications has been a long-standing concern given the essential role these communications play in protecting lives, health, and property. The 2012 act provided FirstNet with the basic resources necessary—such as spectrum and initial capital—to establish an interoperable broadband network for all public safety entities. However, there is no guarantee that FirstNet will be able to successfully develop and operate this nationwide network. FirstNet is tasked with a complex and challenging mission to establish the network, which researchers have estimated could cost as much as $47 billion to construct and operate over its first 10 years. Furthermore, FirstNet faces a multitude of risks, significant challenges, and difficult decisions in meeting its statutory responsibilities, including determining how to become a self-funding entity. If FirstNet fails to generate enough revenue to operate the network over the long-term, for example, it could jeopardize the existence of this new public safety network. At this time, the extent to which FirstNet can generate revenue through partnerships with commercial carriers remains unknown, especially given that some commercial carriers could choose to compete with FirstNet. However, FirstNet is taking certain actions to help ensure that its public safety network is successful. For instance, to date it has effectively consulted with stakeholders and maintained a rigorous cost estimate. Although FirstNet has started to establish an internal control system to help it meet its statutory responsibilities, we found FirstNet could strengthen its internal controls. In particular, while FirstNet has begun taking some steps to assess risks, it has not fully assessed the risks it faces. Complete risk assessments would help FirstNet respond to risks in a proactive manner and make the best use of its resources by appropriately responding to the most pressing risks. FirstNet could also strengthen its internal controls by fully establishing its control environment, which is the foundation for an effective internal control system. Currently, FirstNet has not established a cohesive standards of conduct policy, which means FirstNet may not be able to address deviations in conduct and performance and take corrective actions in a timely manner. Establishing and evaluating adherence to standards of conduct would help FirstNet ensure that all its personnel are held accountable for their actions and foster stakeholder trust in FirstNet’s ability to meet its statutory responsibilities. Early builder projects have learned important lessons related to governance, finance, outreach, and network deployment that could be useful to FirstNet as it develops its plans to establish a nationwide network. However, FirstNet lacks a detailed data-analysis plan to track the projects’ observations and lessons learned. Without such a plan, it is unclear how FirstNet intends to evaluate the early builder projects and ensure that the lessons have been addressed and incorporated, if applicable, into FirstNet’s planning. As a result, FirstNet could miss opportunities to leverage the key lessons the projects learned. A complete evaluation plan for the early builder projects that includes a detailed data-analysis plan would increase transparency and help FirstNet’s strategic planning and program management, which are important given the complexity of FirstNet’s mission. Furthermore, such a plan would provide FirstNet officials the opportunity to make informed midcourse changes as they plan the nationwide network and help ensure that lessons from these projects are evaluated in ways that generate reliable information to inform future network-deployment decisions. To improve the accountability and transparency of FirstNet’s operations, and ensure that FirstNet is gaining as much knowledge from the early builder projects as possible, we recommend that FirstNet take the following two actions: strengthen FirstNet’s internal control system by fully assessing risks, developing standards of conduct, and evaluating performance against these standards, and develop an evaluation plan that includes a detailed data-analysis plan for the early builder projects’ performance and results, including how the observations and lessons learned reported to FirstNet and identified by consultants will be evaluated. We provided a draft of this report to the Departments of Commerce and Homeland Security, FCC, and FirstNet for their review and comment. The Department of Commerce and FirstNet provided written comments, reprinted in app. III and IV, respectively. DHS, FCC, FirstNet, and NTIA (within the Department of Commerce), provided technical comments that we incorporated as appropriate. In its written comments, the Department of Commerce stressed that it takes its oversight responsibilities with respect to FirstNet seriously, is committed to the success of the public-safety broadband network, and supports the response provided to us from FirstNet. In its written comments, FirstNet stated that it agreed with all of our recommendations and noted activities that it will undertake to implement them. Regarding strengthening internal controls, FirstNet stated that it is cognizant that, as a newly formed government entity, it must continue its efforts to establish comprehensive internal control policies and procedures. FirstNet reiterated that it has initiated a legal compliance risk assessment focusing on key legal risk areas and stated that it will also undertake a full risk assessment. FirstNet also stated that it plans to establish supplemental standards of conduct, which will operate in conjunction with applicable regulations and existing FirstNet and Department of Commerce policies. Regarding an evaluation plan for lessons learned from the early builder projects, FirstNet stated that it will develop an appropriate evaluation plan consistent with the principles we specified. FirstNet noted that it has already enacted a standard operating procedure that ensures information and lessons from the projects are appropriately distributed within FirstNet, and it will use this process to disseminate the findings gathered under the evaluation plan. We will send copies of this report to FirstNet as well as the Secretary of Commerce, Secretary of Homeland Security, Chairman of the Federal Communications Commission, and appropriate congressional committees. In addition, the report will be available at no charge on GAO’s website at http://www.gao.gov. If you or members of your staff have any questions about this report, please contact me at (202) 512-2834 or goldsteinm@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Major contributors to this report are listed in appendix V. This report examines the First Responder Network Authority (FirstNet) and its progress towards establishing a nationwide public-safety broadband network (hereafter, the public safety network). Specifically, we reviewed (1) the extent to which FirstNet is carrying out its responsibilities and establishing internal controls for developing the public safety network, (2) how much the public safety network is estimated to cost to construct and operate and how FirstNet plans to become a self-funding entity, and (3) what lessons can be learned from local and regional public-safety network projects. To determine the extent to which FirstNet is carrying out its responsibilities and establishing internal controls, we reviewed FirstNet documentation and conducted interviews. We reviewed all of FirstNet’s Requests for Information, Notices, and annual reports to Congress. We also reviewed relevant board-meeting materials and resolutions, presentations to stakeholders, press releases and blog posts, and other documentation. We interviewed and received written responses from FirstNet, the National Telecommunications and Information Administration, and Department of Commerce officials to obtain further information on their efforts. We compared FirstNet’s efforts and progress carrying out its responsibilities against requirements established in the Middle Class Tax Relief and Job Creation Act of 2012 (2012 act). However, we did not review FirstNet’s progress against every responsibility established for it in the 2012 act, because it is not possible for FirstNet to have made progress on some responsibilities. For instance, FirstNet cannot develop terms of service for the use of the public safety network without first developing the network, the design of which is itself dependent on consultation with a wide variety of stakeholders. We did compare FirstNet’s efforts establishing internal controls against criteria established in the federal Standards for Internal Control.policies and practices against the first two components of internal control: control environment and risk assessment. We chose these two components because the control environment is the foundation for an For the scope of this review, we chose to evaluate FirstNet’s internal control system and risk assessment provides the basis for developing appropriate risk responses and control activities. We also assessed FirstNet’s outreach efforts against core principles for effective stakeholder participation identified by our previous reports. Further, as described below, we interviewed a variety of public safety officials about their perspectives on FirstNet’s progress to date. GAO, GAO Cost Estimating and Assessment Guide: Best Practices for Developing and Managing Capital Program Costs, GAO-09-3SP (Washington, D.C.: Mar. 2, 2009). detailed understanding of the cost model, such as how it was prepared, the assumptions underlying it, and the documentation supporting it. We also interviewed FirstNet officials about how it plans to become self- funding. To assess the factors that will influence the cost of the public safety network and challenges FirstNet may face in becoming self-funding, we conducted a variety of interviews and reviewed documents. As described below, we interviewed officials involved in early builder projects, as well as state and local public-safety entities. We also interviewed subject matter experts who were interviewed for our previous report on emergency communications. Additionally, we interviewed two major commercial wireless carriers for their perspectives on building and operating a public safety network. We also reviewed FirstNet documentation, academic literature suggested to us by subject matter experts, and reports published by FCC and the National Public Safety Telecommunications Council (NPSTC). To identify lessons that can be learned from local and regional public- safety broadband network early builder projects, we interviewed project officials and reviewed documentation from FirstNet and the projects. Specifically, we conducted site visits and phone interviews with officials from—and involved in—the five current projects (Los Angeles, CA; Adams County, CO; New Jersey; New Mexico; and Harris County, TX), the three projects that were canceled (Charlotte, NC; Mississippi; and San Francisco, CA), and state and local public-safety entities in the project jurisdictions. We also reviewed the Spectrum Manager Lease Agreements that each of the five current projects established with FirstNet, and documentation related to how FirstNet plans to collect lessons learned from the projects, such as Key Learning Condition Plans, project quarterly reporting requirements, and other documentation provided by project officials. We also interviewed and obtained written responses from FirstNet officials to obtain more information about their plans to evaluate and utilize lessons from the projects. We assessed FirstNet’s plans to evaluate and utilize lessons from the projects against key features of a well-developed evaluation plan for pilot projects identified by our previous reports. The 5 U.S. territories we surveyed were American Samoa, Guam, the Northern Mariana Islands, Puerto Rico, and the U.S. Virgin Islands. service, three emergency management or communications, and two general public safety entities. In addition to the individual named above, Sally Moino (Assistant Director), Susan Baker, Kyle Browning, David Hooper, Kristen Kociolek, Abishek Krupanand, Jason Lee, Josh Ormond, Nalylee Padilla, Amy Rosewarne, Kelly Rubin, Grant Simmons, Andrew Stavisky, and Michael Sweet made key contributions to this report. | For communications during emergencies, public safety officials rely on thousands of separate systems, which often lack interoperability, or the ability to communicate across agencies and jurisdictions. The 2012 act created FirstNet within the Department of Commerce to establish, for public safety use, a nationwide, interoperable, wireless broadband network, which will initially support data transmissions. The 2012 act established numerous responsibilities for FirstNet, provided $7 billion for network construction, and required FirstNet to be self-funding beyond this initial allocation. As part of the effort, FirstNet is working with five “early builder projects” that are building local and regional public-safety broadband networks. GAO was asked to examine FirstNet's progress in establishing the network. GAO assessed (1) FirstNet's progress carrying out its responsibilities and establishing internal controls, (2) how much the network is estimated to cost and how FirstNet plans to become self-funding, and (3) what lessons can be learned from the early builder projects. GAO reviewed FirstNet documentation and public-safety network cost estimates, surveyed all state-designated FirstNet contacts, and interviewed FirstNet officials and public safety stakeholders selected for their telecommunications and public safety experience. The First Responder Network Authority (FirstNet) has made progress carrying out its responsibilities established in the Middle Class Tax Relief and Job Creation Act of 2012 (the 2012 act) but lacks certain elements of effective internal controls. FirstNet is charged with the complex and challenging task of establishing a new, nationwide, wireless broadband network for public safety entities, in consultation with federal, state, local, and tribal stakeholders. The network will initially support interoperable data communications, and later integrate mission-critical voice capabilities as public safety standards for voice communications are developed. FirstNet has made progress establishing an organizational structure, planning for the network, and consulting with stakeholders. FirstNet has also begun establishing policies and practices consistent with federal internal control standards. Officials told GAO that they plan to continue to do so. However, FirstNet has not fully assessed its risks or established standards of conduct—which is an important form of ethical guidance for its personnel . Given that FirstNet faces numerous risks to achieve its complex objectives, fully assessing risks could help FirstNet achieve its objectives and maximize use of its resources. Developing standards of conduct could also help FirstNet address any performance issues in a timely manner. A nationwide public-safety broadband network has been estimated by various entities to cost billions of dollars, and FirstNet faces difficult decisions determining how to fund the network's construction and ongoing operations. These estimates indicate the cost to construct and operate such a network could be from $12 to $47 billion over the first 10 years. The actual cost of FirstNet's network will be influenced by FirstNet's (1) business model, especially the extent of commercial partnerships; (2) use of existing infrastructure; (3) efforts to ensure network reliability; and (4) network coverage. For example, the cost of the network may be higher if FirstNet does not utilize partnerships and some existing infrastructure. To become self-funding, FirstNet is authorized to generate revenue through user fees and commercial partnerships. However, FirstNet faces difficult decisions in determining how to best utilize these revenue sources. For instance, widespread network coverage can attract more users and revenue, but is expensive to construct and maintain, especially in rural areas. FirstNet has taken steps to collect and evaluate information and lessons from the five “early builder projects” that are developing local and regional public-safety networks, but could do more to ensure that the lessons are properly evaluated. For example, FirstNet has asked the projects to report on the experiences of their networks' users and has assigned contractors to collect and log lessons. However, FirstNet does not have a plan that clearly articulates how it will evaluate those experiences and lessons. Although FirstNet told GAO that it remains in close contact with early builder projects, GAO has previously found that a well-developed evaluation plan for projects like these can help ensure that agencies obtain the information necessary to make effective program and policy decisions. Given that the early builder projects are doing on a local and regional level what FirstNet must eventually do nationally, an evaluation plan can play a key role in FirstNet's strategic planning and program management, providing feedback on both program design and execution and ensuring FirstNet has not missed opportunities to incorporate lessons the projects have identified. FirstNet should complete its risk assessment, develop standards of conduct, and develop an evaluation plan for early builder projects. FirstNet concurred with the recommendations. |
The U.S. Department of Commerce controls the export of U.S. dual-use commodities, software, and technology by requiring validated licenses prior to shipment unless a license exception applies. A license exception, formerly known as a general license, is a broad grant of authority to export certain goods and technology without prior government review. Dual-use telecommunications items are controlled under the Commerce Control List, which specifies the items that require validated licenses, generally for reasons of national security. During the Cold War, the United States sought international coordination of export controls through the Coordinating Committee for Multilateral Export Controls (COCOM), which was made up primarily of North Atlantic Treaty Organization member countries. With the end of the Cold War, COCOM members pushed for liberalization of many dual-use items, including telecommunications equipment, computers, and machine tools. However, before COCOM ended on March 31, 1994, participant countries agreed to continue controlling dual-use items, at each country’s discretion, after COCOM’s expiration. The new U.S. export control arrangement, established in 1996, focuses export controls on several potential aggressor countries, such as Iraq, Libya, and North Korea, and away from former communist countries. Broadband telecommunications equipment, such as ATM and SDH, have numerous civilian and military applications and are becoming increasingly available in China as it strives to improve its telecommunications networks to meet international standards. According to U.S. government officials, overall improvements to China’s telecommunications networks will likely benefit the Chinese military as well. Broadband networks employ ATM and SDH equipment to transfer data, voice, and video communications simultaneously at high rates of speed. ATM is a flexible switching technique used to transfer information over advanced telecommunications networks. SDH technology is used for high speed transmission of data, video, and voice traffic. Both ATM and SDH equipment were developed by commercial industry for civil applications, and they are now increasingly used worldwide on broadband telecommunications networks. Civil applications of ATM and SDH equipment generally allow work groups to collaborate quickly and efficiently over a computer network. These applications include holding a meeting simultaneously in several locations (video-conferencing), having people in different locations work simultaneously on the same document (virtual notebook), and transmitting X-rays and other medical records from one location to another (telemedicine). ATM and SDH equipment are necessary to have the speed of delivery, quick data retrieval, and clear video images that are needed for these applications to operate. Military applications for ATM and SDH equipment are similar to civilian applications. These applications include sharing of intelligence, imagery and video between several locations, command and control of military operations using video-conferencing, and medical support and telemedicine between the battlefield and remote hospitals. When used in a military application, both types of equipment requires encryption devices to protect communications from interception. The Department of Defense buys such equipment “off-the-shelf” and is still testing it for potential military uses. According to Defense officials, ATM and SDH equipment will be beneficial to military functions, such as command and control, and will form the basis for future Defense Department communications networks. Both ATM and SDH equipment are considered to be advanced technologies in the telecommunications industry and are now increasingly deployed worldwide. While ATM equipment has only recently become readily available, SDH has been commercially available for at least the last 5 years. In 1993, a U.S. company claimed that foreign producers of SDH equipment were marketing to China and urged the Commerce Department to reduce its controls on such equipment. SDH equipment was found to be foreign available by the Commerce Department in 1994. Former COCOM member countries tightly controlled the export of such equipment until 1994, when controls on telecommunications items were relaxed. Prior to the liberalization of exports of dual-use telecommunications equipment in April 1994, the export of ATM and SDH equipment to China would have required a validated license from the Commerce Department. According to company officials, since the removal of most export restrictions on telecommunications equipment, the market for such equipment in China has grown quickly and large quantities of SDH equipment have been sold to China to modernize its commercial long-distance telecommunications network. Advanced telecommunications equipment, particularly SDH, is increasingly used to be consistent with the emerging international communications standards, and it should vastly improve China’s outdated and underdeveloped telecommunications systems. According to industry officials, U.S. companies sold tens of millions of dollars worth of SDH equipment in China in 1994, and several joint ventures are currently manufacturing SDH equipment in China to build China’s telecommunications network. In contrast, there is little demand for ATM equipment in China and there is no in-country production of such equipment. Company officials noted the demand for ATM equipment in the United States is also limited. Officials at the Defense Department told us that the Chinese military would like to improve its telecommunications capabilities. According to these officials, the Chinese military is seeking to acquire ATM and SDH equipment, which may increase their operational readiness by the end of the next decade. Defense Department officials told us that broadband telecommunications equipment could be used to improve the Chinese military’s command and control communications networks. These officials also observed that the Chinese military will benefit from the improvements to China’s overall public telecommunications system, since they also make use of that system. The Chinese military may also benefit from the use of broadband equipment in its nonmilitary activities. For example, Chinese Army hospitals, which also serve civilians, are interested in acquiring high-speed telecommunications links using ATM switches to develop telemedicine capabilities and to extend health care to rural areas. The Chinese military operates at least one-third of the hospitals in China. The Commerce Department created GLX in 1994, allowing nearly all dual-use telecommunications items to be exported to Chinese civil end users without validated licenses. As a result, the ATM and SDH equipment exported to HuaMei did not require prior government approval before being shipped. According to U.S. government officials, the Commerce Department created GLX in April 1994 to ease export restrictions and reduce administrative burdens on U.S. exporters. In September 1993, the Trade Promotion Coordinating Committee, chaired by the Secretary of Commerce, recommended removing export controls on computers and telecommunications products to aid key U.S. industries. Commerce officials noted that there was an interagency effort to determine which items to decontrol before COCOM expired at the end of March 1994. The Commerce Department used the advisory notes from the Commerce Control List as a basis for what items were to be included under GLX. Advisory notes are included in the list to advise exporters of the items likely to be approved for export to certain locations. The Commerce Department removed the requirement for a validated license and prior government review on all items with advisory notes because it viewed the items as being less sensitive. According to a Commerce Department official, the advisory note items included under GLX did not necessarily include the entire category of equipment, nor did they apply to China. For example, one advisory note stated that licenses would likely be approved for the export of telecommunications transmission equipment, such as SDH, to China, provided that such equipment did not exceed certain capabilities. Under GLX, all SDH equipment is eligible for export to China, regardless of capability. Another advisory note stated that licenses would likely be approved for the export of all controlled dual-use telecommunications equipment, including ATM, for civil end use in Estonia, Latvia, and Lithuania. According to a Commerce Department official, since ATM equipment was included in this advisory note, it became eligible for export under GLX, even though the note did not apply to China. Our analysis of the telecommunications items included under GLX showed that each item category was covered by some advisory note on the Commerce Control List. Because most of the telecommunications items on the list were covered by advisory notes, they became eligible for export under GLX. The only telecommunications items covered by an advisory note that were not included under GLX were radio frequency hopping equipment and radio receivers with certain scanning and frequency switching capabilities. Items included under GLX were telecommunications transmission equipment, switching equipment, optical fiber communication cables, and phased array antennas. However, technology for the development and production of the telecommunications equipment may not be exported under GLX. The Commerce Department officially requested comments from other agencies on GLX on March 17, 1994, and the creation of GLX was announced in the Federal Register on April 4, 1994. According to government officials, the period of time given to other agencies to comment on the new license category was limited due to the expiration of COCOM at the end of March 1994. Senior Commerce and Defense Department officials stated that while the comment period provided to the agencies for approving GLX was limited, the items included under GLX were less sensitive and had been previously considered for liberalization under earlier COCOM-related negotiations. The Departments of Defense, Energy, and State all approved the new license category. At the time that AT&T exported U.S. telecommunications equipment to HuaMei in 1994, the Commerce Department had already created GLX, so AT&T did not need to apply for a license to ship ATM and SDH equipment to China. In 1993, SCM Brooks Telecommunications entered into a joint venture with Galaxy New Technology, a Chinese company controlled by the Commission of Science, Technology, and Industry for National Defense (COSTIND), an agency of the Chinese military, to form Guangzhou HuaMei Communications Limited. The purpose of the HuaMei project was to demonstrate the commercial capabilities of a broadband telecommunications network, using several hotels in Guangzhou. The demonstration included video-conferencing, virtual notebook, and teleradiology applications. HuaMei contracted with AT&T to provide the equipment necessary for the project. AT&T exported three ATM switches from the United States under a general license and four SDH transmission systems from the Netherlands under a Dutch validated license to Guangzhou in the fall of 1994. GLX allows the export of ATM and SDH equipment to China without a validated license, if the exports are going to “civil end-users for civil end-uses.” However, Commerce Department regulations do not define a “civil end-user” or offer any guidance on how an exporter is to determine who is a civil end user in China. U.S. company and government officials stated that determining end users in China is problematic because the Chinese military is often involved in commercial activities. The Chinese military invests in and owns numerous commercial entities in China to obtain profits and to help fund its military activities. For example, according to industry officials, the military has investments in such enterprises as the Palace Hotel in Beijing, various entertainment projects, and even restaurants. HuaMei, while a commercial enterprise, has as its principal Chinese partner, a company controlled by the Chinese military. As shown in figure 1, SCM Brooks Telecommunications and Galaxy New Technology each own 50 percent of HuaMei. However, the Chinese military is the primary shareholder of Galaxy New Technology, with two other Chinese government agencies each holding a minority interest in the company. Several members of the HuaMei board of directors are military officers or have direct ties to the Chinese military. Such a high degree of involvement in HuaMei could indicate a strong military interest in this company. U.S. company and government officials stated that HuaMei was a civil end user. In determining that, company officials considered the end use of the equipment, the location of the installed equipment, and the customers for the equipment. Since the equipment was being used for video-conferencing among several Chinese hotels to demonstrate the commercial applications of this technology, company officials were confident that the export of this equipment would satisfy the civil end-user requirement of GLX. According to Commerce Department officials, the agency does not have guidance for its staff to use in making civil end-user determinations for exports under GLX (or its successor, CIV), nor has it issued guidelines on end-user determination for exporters. For an export to China that requires a validated license, the Commerce Department normally conducts a review of the application and determines if the export is going to a military or civil end user on a case-by-case basis, using available government resources such as embassy personnel and intelligence reports. Exporters do not have such resources available to them when making a civil end-user determination. Currently, there is no guidance or criteria available to exporters on how much military involvement in a commercial entity is needed before it is considered a military end user. According to Commerce Department officials, an exporter is responsible for knowing its end user when exporting under GLX (or CIV). An exporter is encouraged to come to the Commerce Department for an end-user check if there are any abnormal circumstances in a transaction that indicate the export may be going to an inappropriate end user. According to Commerce Department guidance, these circumstances can include orders for items that are inconsistent with the purchaser’s needs or a customer declining installation and testing when included in the sale price. If there are no suspicious circumstances, the exporter is not required to verify the buyer’s representations of civil end use. AT&T officials stated that they did not ask the Commerce Department to determine if HuaMei was a civil end user, nor were they required to under GLX. Commerce officials stated that the civil end-user requirement in GLX was specifically included to allow Commerce to review exports going to the military. However, in the export of telecommunications equipment to HuaMei, the Commerce Department did not have an opportunity to review the end user because prior government review is not required under GLX. Consequently, the equipment was exported to HuaMei without Commerce review, even though the company was partially controlled by several high-level members of the Chinese military. In commenting on a draft of this report, the Departments of Commerce, Defense, and State generally agreed with the information we presented. Commerce noted that our suggestion that it assess the need to provide additional guidance to exporters on determining civil end users as it gains experience under CIV (formerly known as GLX) was helpful. Commerce also asserted that our report confirmed that its decision to make certain telecommunications equipment available to China without requiring a validated license was consistent with both U.S. national security and economic interests. However, it should be noted that determining consistency with U.S. interests was not within the scope of our work. We made minor technical corrections to the report where appropriate based on suggestions provided by the Departments of Commerce and Defense. Comments from the Departments of Commerce, Defense, and State are presented in appendixes I, II, and III, respectively. To obtain information about the HuaMei project—when it was created, its purpose, and the equipment it employed—we interviewed SCM Brooks and AT&T officials and examined company agreements and project descriptions. To determine civil and military applications of ATM and SDH equipment, availability of the equipment, and the importance of those applications to China’s military, we interviewed officials from AT&T and SCM Brooks and obtained technical descriptions of the products and potential applications. We also interviewed officials from the National Security Agency, the Defense Technology Security Administration, the Defense Intelligence Agency, and the Defense Information Security Agency, as well as the Commerce Department Bureau of Export Administration to get expert assessments of advanced telecommunications equipment and data on telecommunications equipment availability in China. In addition, we interviewed telecommunications industry and U.S. embassy officials in China to obtain information about the applications and availability of ATM and SDH equipment in China. To examine the process and rationale for liberalizing the export of broadband telecommunications equipment, we reviewed a chronology of the export of AT&T equipment. We also interviewed officials from the Departments of Defense, Commerce, and State, and the National Security Agency to obtain data on the rationale and purposes of creating the GLX license category, as well as agency positions on the inclusion of telecommunications equipment under GLX. We reviewed Defense and Commerce Department documentation on the development of GLX. We also compared the items on GLX with the Commerce Control List items covered by advisory notes to confirm the method used to develop GLX. We performed our review from March to October 1996 in accordance with generally accepted government auditing standards. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution of this report until 10 days after its issue date. At that time, we will send copies to other congressional committees and the Secretaries of Defense, Commerce, and State. We will also make copies available to other interested parties upon request. Please contact me at (202) 512-4383 if you or your staff have any questions concerning this report. Major contributors to this report were Karen Zuckerstein, David C. Trimble, and John Neumann. The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 6015 Gaithersburg, MD 20884-6015 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (301) 258-4066, or TDD (301) 413-0006. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO reviewed the transfer of broadband telecommunication equipment to HuaMei, a joint venture between SCM Brooks Telecommunications, a U.S. limited partnership, and Galaxy New Technology, a Chinese company primarily owned by an agency of the Chinese military, focusing on the: (1) civil and military applications of the exported telecommunications equipment, its availability, and the importance of these applications to China's military; and (2) process and rationale for liberalizing the export of telecommunications equipment shipped to China. GAO found that: (1) there are numerous civil and military applications for broadband telecommunications equipment, such as Asynchronous Transfer Mode (ATM) and Synchronous Digital Hierarchy (SDH) equipment, including video-conferencing, remote command and control, and telemedicine; (2) since the liberalization of exports of advanced telecommunications equipment, such equipment is now readily available in China; (3) SDH equipment, in particular, is being manufactured and used to upgrade China's telecommunications networks to international standards; (4) according to U.S. government officials, the Chinese military is seeking to acquire ATM and SDH equipment, which may benefit its command and control networks by the end of the next decade; (5) furthermore, these officials stated that as China's telecommunications infrastructure is modernized, the Chinese military will also benefit; (6) the creation of the new General License category, GLX, by the Commerce Department in April 1994, allowed the export of ATM and SDH equipment to HuaMei without a validated license having to be issued by the Commerce Department; (7) ATM and SDH equipment were two of a number of dual-use telecommunications items that were included under GLX; (8) GLX includes many items that would have been typically approved for export to civil end users in the licensing process; (9) according to U.S. government officials, GLX was created in response to the end of the Cold War and the expiration of the Coordinating Committee for Multilateral Export Controls, to ease export restrictions and reduce administrative burdens on U.S. exporters; (10) determining who is a civil end user under GLX is the responsibility of the exporting companies; (11) however, this is particularly difficult in China because of the Chinese military's significant involvement in various commercial ventures; (12) there is no information readily available to exporters on how much military involvement in a commercial entity constitutes a military end user; and (13) based on this one case, GAO is not making any recommendations; however, as the Commerce Department gains experience under GLX, it may want to assess the need to provide additional information or guidance to exporters to help them determine when they should request a government review of an end user. |
In the past, the ICC regulated almost all of the rates that railroads charged shippers. The Railroad Revitalization and Regulatory Reform Act of 1976 and the Staggers Rail Act of 1980 greatly increased reliance on competition to set rates in the railroad industry. Specifically, these acts allowed railroads and shippers to enter into confidential contracts that set rates and prohibited ICC from regulating rates where railroads had either effective competition or rates negotiated between the railroad and the shipper. Furthermore, the ICC Termination Act of 1995 abolished ICC and transferred its regulatory functions to STB. Taken together, these acts anchor the federal government’s role in the freight rail industry by establishing numerous goals for regulating the industry, including to allow, to the maximum extent possible, competition and demand for services to establish reasonable rates for transportation by rail; minimize the need for federal regulatory control over the rail transportation system and require fair and expeditious regulatory decisions when regulation is required; promote a safe and efficient rail transportation system by allowing rail carriers to earn adequate revenues, as determined by STB; ensure the development and continuation of a sound rail transportation system with effective competition among rail carriers and with other modes to meet the needs of the public and the national defense; foster sound economic conditions in transportation and ensure effective competition and coordination between rail carriers and other modes; maintain reasonable rates where there is an absence of effective competition and where rail rates provide revenues that exceed the amount necessary to maintain the rail system and attract capital; prohibit predatory pricing and practices to avoid undue concentrations of market power; and provide for the expeditious handling and resolution of all proceedings. While the Staggers Rail and ICC Termination Acts reduced regulation in the railroad industry, they maintained STB’s role as the economic regulator of the industry. The federal courts have upheld STB’s general powers to monitor the rail industry, including its ability to subpoena witnesses and records and to depose witnesses. In addition, STB can revisit its past decisions if it discovers a material error, or new evidence, or if circumstances have substantially changed. Two important components of the current regulatory structure for the railroad industry are the concepts of revenue adequacy and demand-based differential pricing. Congress established the concept of revenue adequacy as an indicator of the financial health of the industry. STB determines the revenue adequacy of a railroad by comparing the railroad’s return on investment with the industrywide cost of capital. For instance, if a railroad’s return on investment is greater than the industrywide cost of capital, STB determines that railroad to be revenue adequate. Historically, ICC and STB have rarely found railroads to be revenue adequate—a result that many observers relate to characteristics of the industry’s cost structure. Railroads incur large fixed costs to build and operate networks that jointly serve many different shippers. Some fixed costs can be attributed to serving particular shippers, and some costs vary with particular movements, but other costs are not attributable to particular shippers or movements. Nonetheless, a railroad must recover these costs if the railroad is to continue to provide service over the long run. To the extent that railroads have not been revenue adequate, they may not have been fully recovering these costs. The Staggers Rail Act recognized the need for railroads to use demand- based differential pricing to promote a healthy rail industry and enable it to raise sufficient revenues to operate, maintain and, if necessary, expand the system in a deregulated environment. Demand-based differential pricing, in theory, permits a railroad to recover its joint and common costs—those costs that exist no matter how many shipments are transported, such as the cost of maintaining track— across its entire traffic base by setting higher rates for traffic with fewer transportation alternatives than for traffic with more alternatives. Differential pricing recognizes that some customers may use rail if rates are low—and have other options if rail rates are too high or service is poor. Therefore, rail rates on these shipments generally cover the directly attributable (variable) costs, plus a relatively low contribution to fixed costs. In contrast, customers with little or no practical alternative to rail—”captive” shippers—generally pay a much larger portion of fixed costs. Moreover, even though a railroad might incur similar incremental costs while providing service to two different shippers that move similar volumes in similar car types traveling over similar distances, the railroad might charge the shippers different rates. Furthermore, if the railroad is able to offer lower rates to the shipper with more transportation alternatives, that shipper still pays some of the joint and common costs. By paying even a small part of total fixed cost, competitive traffic reduces the share of those costs that captive shippers would have to pay if the competitive traffic switched to truck or some other alternative. Consequently, while the shipper with fewer alternatives makes a greater contribution toward the railroad’s joint and common costs, the contribution is less than if the shipper with more alternatives did not ship via rail. The Staggers Rail Act further requires that the railroads’ need to obtain adequate revenues to be balanced with the rights of shippers to be free from, and to seek redress from, unreasonable rates. Railroads incur variable costs—that is, the costs of moving particular shipments—in providing service. The Staggers Rail Act stated that any rate that was found to be below 180 percent of a railroad’s variable cost for a particular shipment could not be challenged as unreasonable and authorized ICC, and later STB, to establish a rate relief process for shippers to challenge the reasonableness of a rate. STB may consider the reasonableness of a rate only if it finds that the carrier has market dominance over the traffic at issue—that is, if (1) the railroad’s revenue is equal to or above 180 percent of the railroad’s variable cost (R/VC) and (2) the railroad does not face effective competition from other rail carriers or other modes of transportation. The changes that have occurred in the railroad industry since the enactment of the Staggers Rail Act are widely viewed as positive. In addition, rail rates have generally declined since 1985, even though rates began to increase in 2001 and experienced a 9 percent annual increase between 2004 and 2005—the largest annual increase in 20 years. Likewise, rail rates have declined since 1985 for certain commodity groups and routes despite some increases since 2001, but rates have not declined uniformly. Railroads have also shifted other costs to shippers, such as the cost of rail car ownership, and have increased the revenue they report as miscellaneous more than 10-fold between 2000 and 2005. There is widespread consensus that the freight rail industry has benefited from the Staggers Rail Act. Various measures indicate an increasingly strong freight railroad industry. Freight railroads have also cut costs by streamlining their workforces; right-sizing their rail networks; and reducing track miles, equipment, and facilities to more closely match demand. Freight railroads have also expanded their business into new markets—such as the intermodal market—and implemented new technologies, including larger cars, and are currently developing new scheduling and train control systems. Rail rates across the freight railroad industry have generally declined since 1985 despite a recent increase. Rates began to rise in 2001 and experienced a 9 percent annual increase from 2004-2005, which represents the largest annual increase in rates during the 20-year period from 1985 through 2005. This increase also outpaced inflation—about 3 percent in 2005. However, despite these increases, rates for 2005 remain below their 1985 levels and below the rate of inflation. Because the set of rail rate indexes we used to examine trends in rail rates over time does not account for inflation we also included the price index for the gross domestic product (GDP) in figure 1. Rates for several commodities in 2005 remain lower than in 1985. Similar to overall industry trends, rates for individual commodities have increased from 2004-2005. In 2005, rates increased for all 13 commodities that we reviewed. Figure 2 depicts rate changes for coal, grain, miscellaneous mixed shipments, and motor vehicles from 1985 through 2005. Over 20 years, freight railroad companies have shifted other costs to shippers. Our analysis shows a 20 percent shift in railcar ownership (measured in tons carried) since 1987. In 1987, railcars owned by freight railroad companies moved 60 percent of tons carried. In 2005, they moved 40 percent of tons carried, meaning that freight railroad company railcars no longer carry the majority of tonnage (see fig. 3). In 2005 the amount of industry revenue reported as miscellaneous increased ten-fold over 2000 levels, rising from about $141 million to over $1.7 billion (see fig. 4). Miscellaneous revenue is a category in the Carload Waybill Sample for reporting revenue outside the standard rate structure. This miscellaneous revenue can include some fuel surcharges, as well as revenues such as those derived from congestion fees and railcar auctions (in which the highest bidder is guaranteed a number of railcars at a specified date). In 2004, miscellaneous revenue accounted for 1.5 percent of freight railroad revenue reported. In 2005, this percentage had risen to 3.7 percent. Also, in 2005, 20 percent of all tonnage moved in the United States generated miscellaneous revenue. In October 2006, we reported that captive shippers are difficult to identify and STB’s efforts to protect captive shippers have resulted in little effective relief for those shippers. We also reported that economists and shipper groups have proposed a number of alternatives to address remaining concerns about competition – however, each of these alternative approaches have costs and benefits and should be carefully considered to ensure the approach will achieve the important balance set out in the Staggers Act. It remains difficult to determine precisely how many shippers are “captive” to one railroad because the proxy measures that provide the best indication can overstate or understate captivity. One measure of potential captivity—traffic traveling at rates equal to or greater than 180 percent R/VC—is part of the statutory threshold for bringing a rate relief case before STB. STB regards traffic at or above this threshold as “potentially captive,” but, like other measures, R/VC levels can understate or overstate captivity. Since 1985, tonnage and revenue from traffic traveling at rates over 180 percent R/VC have generally declined. (see fig. 5). In 2005, industry revenue generated by traffic traveling at rates over 180 percent R/VC dropped by roughly half a percent. Tonnage traveling at rates over 180 percent R/VC dropped by a smaller percentage. While traffic traveling at rates over 180 percent R/VC has generally declined, traffic traveling at rates substantially over the threshold for rate relief has generally increased from 1985 to 2005 (see fig. 6). This traffic declined in 2003 and 2004, but rose in 2005. Some areas with access to one Class I railroad also have more than half of their traffic traveling at rates that exceed the statutory threshold for rate relief. For example, parts of New Mexico and Idaho with access to one Class I railroad had more than half of all traffic originating in those same areas traveling at rates over 180 percent R/VC. However, we also found instances in which an economic area may have access to two or more Class I railroads and still have more than 75 percent of its traffic traveling at rates over 180 percent R/VC, as well as other instances in which an economic area may have access to one Class I railroad and have less than 25 percent of its traffic traveling at rates over 180 percent R/VC. STB has taken a number of actions to provide relief for captive shippers. While the Staggers Rail and ICC Termination Acts encourage competition as the preferred way to protect shippers and to promote the financial health of the railroad industry, they also give STB the authority to adjudicate rate cases to resolve disputes between captive shippers and railroads upon receiving a complaint from a shipper; approve rail transactions, such as mergers, consolidations, acquisitions, and trackage rights; prescribe new regulations, such as rules for competitive access and merger approvals; and inquire into and report on rail industry practices, including obtaining information from railroads on its own initiative and holding hearings to inquire into areas of concern, such as competition. Under its adjudicatory authority, STB has developed standard rate case guidelines, under which captive shippers can challenge a rail rate and appeal to STB for rate relief. Under the standard rate relief process, STB assesses whether the railroad dominates the shipper’s transportation market and, if it finds market dominance, proceeds with further assessments to determine whether the actual rate the railroad charges the shipper is reasonable. STB requires that the shipper demonstrate how much an optimally efficient railroad would need to charge the shipper and construct a hypothetical, perfectly efficient railroad that would replace the shipper’s current carrier. As part of the rate relief process, both the railroad and the shipper have the opportunity to present their facts and views to STB, as well as to present new evidence. STB also created alternatives to the standard rate relief process, developing simplified guidelines, as Congress required, for cases in which the standard rate guidelines would be too costly or infeasible given the value of the cases. Under these simplified guidelines, captive shippers who believe that their rate is unreasonable can appeal to STB for rate relief, even if the value of the disputed traffic makes it too costly or infeasible to apply the standard guidelines. Despite STB’s efforts, we reported in 2006 that there was widespread agreement that STB’s standard rate relief process was inaccessible to most shippers and did not provide for expeditious handling and resolution of complaints. The process remained expensive, time consuming, and complex. Specifically, shippers we interviewed agreed that the process could cost approximately $3 million per litigant. In addition, shippers said that they do not use the process because it takes so long for STB to reach a decision. Lastly, shippers stated that the process is both time consuming and difficult because it calls for them to develop a hypothetical competing railroad to show what the rate should be and to demonstrate that the existing rate is unreasonable. We also reported that the simplified guidelines also had not effectively provided relief for captive shippers. Although these simplified guidelines had been in place since 1997, a rate case had not been decided under the process set out by the guidelines when we issued our report in 2006. STB had held public hearings in April 2003 and July 2004 to examine why shippers have not used the guidelines and to explore ways to improve them. At these hearings, numerous organizations provided comments to STB on measures that could clarify the simplified guidelines, but no action was taken. STB observed that parties urged changes to make the process more workable, but disagreed on what those changes should be. We reported that several shipper organizations told us that shippers were concerned about using the simplified guidelines because they believe the guidelines will be challenged in court, resulting in lengthy litigation. STB officials told us that they—not the shippers—would be responsible for defending the guidelines in court. STB officials also said that if a shipper won a small rate case, STB could order reparations to the shipper before the case was appealed to the courts. Since our report in October 2006, STB has taken steps to refine the rate relief process. Specifically, in October 2006, STB revised procedures for deciding large rate relief cases. By placing restraints on the evidence and arguments allowed in these cases, STB predicted that the expense and delay in resolving these rate disputes would be reduced substantially. In September 2007, STB altered its simplified guidelines for small shippers to enable shippers who are seeking up to $1 million in rate relief over a 5- year period to receive a STB decision within 8 months of filing a complaint. STB also created a new rate relief process for medium size shipments to allow shippers who are seeking up to $5 million in rate relief over a 5-year period to receive a STB decision within 17 months of filing a complaint. Additionally, STB also stated that all rail rate disputes would require nonbinding mediation. Shipper groups, economists, and other experts in the rail industry have suggested several alternative approaches as remedies that could provide more competitive options to shippers in areas of inadequate competition or excessive market power. These groups view these approaches as more effective than the rate relief process in promoting a greater reliance on competition to protect shippers against unreasonable rates. Some proposals would require legislative change, or a reopening of past STB decisions. These approaches each have potential costs and benefits. On the one hand, they could expand competitive options, reduce rail rates, and decrease the number of captive shippers as well as reduce the need for both federal regulation and a rate relief process. On the other hand, reductions in rail rates could affect railroad revenues and limit the railroads’ ability and potential willingness to invest in their infrastructure. In addition, some markets may not have the level of demand needed to support competition among railroads. It will be important for policymakers, in evaluating these alternative approaches, to carefully consider the impact of each approach on the balance set out in the Staggers Act. The targeted approaches frequently proposed by shipper groups and others include the following: Reciprocal switching: This approach would allow STB to require railroads serving shippers that are close to another railroad to transport cars of a competing railroad for a fee. The shippers would then have access to railroads that do not reach their facilities. This approach is similar to the mandatory interswitching in Canada, which enables a shipper to request a second railroad’s service if that second railroad is within approximately 18 miles. Some Class I railroads already interchange traffic using these agreements, but they oppose being required to do so. Under this approach, STB would oversee the pricing of switching agreements. This approach could also reduce the number of captive shippers by providing a competitive option to shippers with access to a proximate but previously inaccessible railroad and thereby reduce traffic eligible for the rate relief process (see fig. 7). Terminal agreements: This approach would require one railroad to grant access to its terminal facilities or tracks to another railroad, enabling both railroads to interchange traffic or gain access to traffic coming from shippers off the other railroad’s lines for a fee. Current regulation requires a shipper to demonstrate anticompetitive conduct by a railroad before STB will grant access to a terminal by a nonowning railroad unless there is an emergency or when a shipper can demonstrate poor service and a second railroad is willing and able to provide the service requested. This approach would require revisiting the current requirement that railroads or shippers demonstrate anticompetitive conduct in making a case to gain access to a railroad terminal in areas where there is inadequate competition. The approach would also make it easier for competing railroads to gain access to the terminal areas of other railroads and could increase competition between railroads. However, it could also reduce revenues to all railroads involved and adversely affect the financial condition of the rail industry. Also, shippers could benefit from increased competition but might see service decline (see fig. 8). to its tracks to another railroad, enabling railroads to interchange traffic beyond terminal facilities for a fee. In the past, STB has imposed conditions requiring that a merging railroad must grant another railroad trackage rights to preserve competition when a merger would reduce a shipper’s access to railroads from two to one. While this approach could potentially increase rail competition and decrease rail rates, it could also discourage owning railroads from maintaining the track or providing high- quality service, since the value of lost use of track may not be compensated by the user fee and may decrease return on investment (see fig. 9). “Bottleneck” rates: This approach would require a railroad to establish a rate, and thereby offer to provide service, for any two points on the railroad’s system where traffic originates, terminates, or can be interchanged. Some shippers have more than one railroad that serves them at their origin and/or destination points, but have at least one portion of a rail movement for which no alternative rail route is available. This portion is referred to as the “bottleneck segment.” STB’s decision that a railroad is not required to quote a rate for the bottleneck segment has been upheld in federal court. STB’s rationale was that statute and case law precluded it from requiring a railroad to provide service on a portion of its route when the railroad serves both the origin and destination points and provides a rate for such movement. STB requires a railroad to provide service for the bottleneck segment only if the shipper had prior arrangements or a contract for the remaining portion of the shipment route. On the one hand, requiring railroads to establish bottleneck rates would force short-distance routes on railroads when they served an entire route and could result in loss of business and potentially subject the bottleneck segment to a rate complaint. On the other hand, this approach would give shippers access to a second railroad, even if a single railroad was the only railroad that served the shipper at its origin and/or destination points, and could potentially reduce rates (see fig. 10). Paper barriers: This approach would prevent or, put a time limit on, paper barriers, which are contractual agreements that can occur when a Class I railroad either sells or leases long term some of its track to other railroads (typically a short-line railroad and/or regional railroad). These agreements stipulate that virtually all traffic that originates on that line must interchange with the Class I railroad that originally leased the tracks or pay a penalty. Since the 1980s, approximately 500 short lines have been created by Class I railroads selling a portion of their lines; however, the extent to which paper barriers are a standard practice is unknown because they are part of confidential contracts. When this type of agreement exists, it can inhibit smaller railroads that connect with or cross two or more Class I rail systems from providing rail customers access to competitive service. Eliminating paper barriers could affect the railroad industry’s overall capacity since Class I railroads may abandon lines instead of selling them to smaller railroads and thereby increase the cost of entering a market for a would-be competitor. In addition, an official from a railroad association told us that it is unclear if a federal agency could invalidate privately negotiated contracts (see fig. 11). STB has taken some actions to address our past recommendations, but it is too soon to determine the effect of these actions. In October 2006 we reported that the continued existence of pockets of potential captivity at a time when the railroads are, for the first time in decades, experiencing increasing economic health, raises the question whether rail rates in selected markets reflect justified and reasonable pricing practices, or an abuse of market power by the railroads. While our analysis provided an important first step, we noted that STB has the statutory authority and access to information to inquire into and report on railroad practices and to conduct a more rigorous analysis of competition in the freight rail industry. As a result, we recommended that the Board undertake a rigorous analysis of competitive markets to identify the state of competition nationwide and to determine in specific markets whether the inappropriate exercise of market power is occurring and, where appropriate, to consider the range of actions available to address such problems. STB initially disagreed with our recommendation because it believed the findings underlying the recommendation were inconclusive, their on-going efforts would address many of our concerns, and a rigorous analysis would divert resources from other efforts. However, in June 2007, STB stated that it intended to implement our recommendation using funding that was not available at the time of our October report to solicit proposals from analysts with no connection to the freight railroad industry or STB proceedings to conduct a rigorous analysis of competition in the freight railroad industry. On September 13, 2007, STB announced that it had awarded a contract for a comprehensive study on competition, capacity, and regulatory policy issues to be completed by the fall of 2008. We commend STB for taking this action. It will be important that these analysts have the ability that STB has through its statutory authority to inquire into railroad practices as well as sufficient access to information to determine whether rail rates in selected markets reflect justified and reasonable pricing practices, or an abuse of market power by the railroads. We also recommended that STB review its method of data collection to ensure that all freight railroads are consistently and accurately reporting all revenues collected from shippers, including fuel surcharges and other costs not explicitly captured in all railroad rate structures. In January 2007, STB finalized rules that require railroads to ensure that fuel surcharges are based on factors directly affecting the amount of fuel consumed. In August 2007, STB finalized rules that require railroads to report their fuel costs and revenue from fuel surcharges. While these are positive steps, these rules did not address how surcharges are reported in the Carload Waybill Sample. In addition, STB has not taken steps to address collection and reporting of other miscellaneous revenues— revenues deriving from sources other than fuel surcharges. As stated earlier, STB has also taken steps to refine the rate relief process since our 2006 report. STB has made changes to the rate relief process that it believes will reduce the expense and delay of obtaining rate relief. While these appear to be positive steps that could address longstanding concerns with the rate relief process, it is too soon to determine the effect of these changes to the process, and we have not evaluated the effect of these changes. Mr. Chairman, this concluded my prepared statement. I would be happy to respond to any questions you or other Members of the Committee may have at this time. For questions regarding this testimony, please contact JayEtta Z. Hecker on (202) 512-2834 or heckerj@gao.gov. Individuals making key contributions to this testimony include Steve Cohen (Assistant Director), Yumiko Jolly, and John W. Shumann. Freight Railroads: Industry Health Has Improved, but Concerns About Competition and Capacity Should Be Addressed. GAO-07-94. Washington, D.C.: Oct. 6, 2006). Freight Railroads: Updated Information on Rates and Other Industry Trends. GAO-07-291R. Washington, D.C.: Aug. 15, 2007. Freight Railroads: Preliminary Observations on Rates, Competition, and Capacity Issues. GAO-06-898T. Washington, D.C.: June 21, 2006. Freight Transportation: Short Sea Shipping Option Shows Importance of Systematic Approach to Public Investment Decisions. GAO-05-768. Washington, D.C.: July 29, 2005. Freight Transportation: Strategies Needed to Address Planning and Financing Limitations. GAO-04-165. Washington, D.C.: December 19, 2003. Railroad Regulation: Changes in Freight Railroad Rates from 1997 through 2000. GAO-02-524. Washington, D.C.: June 7, 2002. Freight Railroad Regulation: Surface Transportation Board’s Oversight Could Benefit from Evidence Better Identifying How Mergers Affect Rates. GAO-01-689. Washington, D.C.: July 5, 2001. Railroad Regulation: Current Issues Associated with the Rate Relief Process. GAO/RCED-99-46. Washington, D.C.: April 29, 1999. Railroad Regulation: Changes in Railroad Rates and Service Quality Since 1990. GAO/RCED-99-93. Washington, D.C.: April 6, 1999. Interstate Commerce Commission: Key Issues Need to Be Addressed in Determining Future of ICC’s Regulatory Functions. GAO-T-RCED-94-261 Washington, D.C.: July 12, 1994. Railroad Competitiveness: Federal Laws and Policies Affect Railroad Competitiveness. GAO/RCED-92-16. Washington, D.C.: November 5, 1991. Railroad Regulation: Economic and Financial Impacts of the Staggers Rail Act of 1980. GAO/RCED-90-80. Washington, D.C.: May 16, 1990. Railroad Regulation: Shipper Experiences and Current Issues in ICC Regulation of Rail Rates. GAO/RCED-87-119. Washington, D.C.: September 9, 1987. Railroad Regulation: Competitive Access and Its Effects on Selected Railroads and Shippers. GAO/RCED-87-109, Washington, D.C.: June 18, 1987. Railroad Revenues: Analysis of Alternative Methods to Measure Revenue Adequacy. GAO/RCED-87-15BR. Washington, D.C.: October 2, 1986. Shipper Rail Rates: Interstate Commerce Commission’s Handling of Complaints. GAO/RCED-86-54FS. Washington, D.C.: January 30, 1986. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The Staggers Rail Act of 1980 largely deregulated the freight railroad industry, giving the railroads freedom to price their services according to market conditions and encouraging greater reliance on competition to set rates. The act recognized the need for railroads to recover costs by setting higher rates for shippers with fewer transportation alternatives. The act also recognized that some shippers might not have access to competitive alternatives and might be subject to unreasonably high rates. It established a threshold for rate relief and granted the Interstate Commerce Commission and the Surface Transportation Board (STB) the authority to develop a rate relief process for those "captive" shippers. GAO's reported on rates, competition, and other industry trends in reports issued in October 2006 and August 2007. This statement is based on those reports and discusses (1) the changes that have occurred in the railroad industry since the enactment of the Staggers Rail Act, including changes in rail rates since 1985, (2) the extent of captivity in the industry and STB's efforts to protect captive shippers, and (3) STB's actions to address GAO's recent recommendations. The changes that have occurred in the railroad industry since the enactment of the Staggers Rail Act are widely viewed as positive, since the financial health of the industry has improved and most rates have declined since 1985. The freight railroad industry's financial health improved substantially as railroads cut costs through productivity improvements and new technologies. However, rates began to increase in 2001, and in 2005 rates jumped nearly 9 percent--the largest annual increase in twenty years--and rates increased for all 13 commodities that we reviewed. Revenues that railroads report as "miscellaneous revenue"--a category that includes some fuel surcharges--increased more than ten-fold from $141 million in 2000 to over $1.7 billion in 2005. It is difficult to precisely determine how many shippers are "captive" because available proxy measures can overstate or understate captivity. However some data indicate that potentially captive traffic appears to have decreased, while at the same time, data also indicates that traffic traveling at rates significantly above the threshold for rate relief has increased. In October 2006, we reported that STB's rate relief process to protect captive shippers have resulted in little effective relief for those shippers. We also reported that economists and shipper groups have proposed a number of alternatives to address remaining concerns about competition--however, each of these alternative approaches have costs and benefits and should be carefully considered. STB has taken some actions to address our past recommendations, but it is too soon to determine the effect of these actions. Our October 2006 report noted that the continued existence of pockets of potentially "captive shippers" raised questions as to whether rail rates in selected markets reflected reasonable pricing practices, or an abuse of market power. We recommended that the Board undertake a rigorous analysis of competitive markets to identify the state of competition. STB has awarded a contract to conduct this study; while this is an important step, it will be important that these analysts have STB's authority and access to information to determine whether rail rates in selected markets reflect reasonable pricing practices. We also recommended that STB ensure that freight railroads are consistently reporting all revenues, including miscellaneous revenues. While STB has revised its rules on fuel surcharges, these rules did not address how fuel surcharges are reported and STB has not yet taken steps to accurately collect data on other miscellaneous revenues. STB has also taken a number of steps to revise its rate relief process. While these appear to be promising steps, it is too soon to tell what effect these changes will have and we have not evaluated them. |
Information technology should enable government to better serve the American people. However, according to OMB, despite spending more than $600 billion on IT over the past decade, the federal government has achieved little of the productivity improvements that private industry has Too often, federal IT projects run over budget, behind realized from IT.schedule, or fail to deliver promised functionality. In combating this problem, proper oversight is critical. Both OMB and federal agencies have key roles and responsibilities for overseeing IT investment management. OMB is responsible for working with agencies to ensure investments are appropriately planned and justified. Additionally, each year, OMB and federal agencies work together to determine how much the government plans to spend on IT projects and how these funds are to be allocated. Congress enacted several laws to assist the federal government in better managing IT investments. The three key laws are the Paperwork Reduction Act of 1995, the Clinger-Cohen Act of 1996, and the E-Government Act of 2002: The Paperwork Reduction Act of 1995 specified OMB and agency responsibilities for managing information resources, including the management of IT. Among its provisions, this law established agency responsibility for assessing and managing the risks of major information systems initiatives.and oversee policies, principles, standards, and guidelines for federal agency IT functions, including periodic evaluations of major information systems. It also required that OMB develop The Clinger-Cohen Act of 1996 placed responsibility for managing investments with the heads of agencies and established chief information officers (CIO) to advise and assist agency heads in carrying out this responsibility. Additionally, this law required OMB to establish processes to analyze, track, and evaluate the risks and results of major capital investments in information systems made by federal agencies and report to Congress on the net program performance benefits achieved as a result of these investments. The E-Government Act of 2002 established a federal e-government initiative, which encouraged the use of web-based Internet applications to enhance the access to and delivery of government information and service to citizens, to business partners, to employees, and among agencies at all levels of government. The act also required OMB to report annually to Congress on the status of e- government initiatives. In these reports, OMB is to describe the administration’s use of e-government principles to improve government performance and the delivery of information and services to the public. OMB uses the following mechanisms to help it fulfill its required oversight responsibilities of federal IT spending during the annual budget formulation process. OMB requires 27 federal departments and agencies to provide information related to their IT investments, including agency IT investment portfolios (called exhibit 53s) and capital asset plans and business cases (called exhibit 300s). In June 2009, OMB publicly deployed the IT Dashboard, which is intended to display near real-time information on the cost, schedule, and performance of all major IT investments. For each major investment, the Dashboard provides performance ratings on cost and schedule, a CIO evaluation, and an overall rating. The CIO evaluation is based on his or her evaluation of the performance of each investment and takes into consideration multiple variables. This evaluation is to be updated when new information becomes available that would affect the assessment of a given investment. The CIO also has the ability to provide written comments regarding the status of each investment. The Dashboard replaced OMB’s Management Watch List and High-Risk List, which were previously used to highlight poorly planned or poorly performing investments on a quarterly basis. As of August 2011, the Dashboard displayed information on the cost, schedule, and performance of 797 major federal IT investments at 27 federal agencies. According to OMB, the public display of investment data on the IT Dashboard is intended to allow OMB, other oversight bodies, and the general public to hold government agencies accountable for results and progress. In addition, the Dashboard allows users to download exhibit 53 data, which provide details on the more than 7,200 federal IT investments (totaling $78.8 billion in planned spending for fiscal year 2011). Figure 1 shows the number of IT investments and planned spending by federal agency. As we have previously reported, while the IT Dashboard provides IT investment information for 27 federal agencies, it does not include any information about 61 other agencies’ investments. Specifically, it does not include information from 58 independent executive branch agencies (such as the Securities and Exchange Commission, the Central Intelligence Agency, and the Federal Communications Commission) and 3 other agencies (such as the Legal Services Corporation). It also does not include information from the legislative or judicial branch agencies. Accordingly, we recommended that OMB specify which executive branch agencies are included when discussing the annual federal IT investment portfolio. OMB disagreed with this recommendation, stating that the agencies included in the federal IT portfolio are already identified in OMB guidance and on the IT Dashboard. However, we maintained that the recommendation had not been fully addressed because OMB officials frequently refer to the federal IT portfolio without clarifying that it does not include all agencies. Despite required roles and responsibilities and OMB’s oversight mechanisms, the federal government spends billions of dollars on poorly performing IT investments, as the following examples illustrate: In April 2008, due to problems identified during testing and cost overruns and schedule slippages, the Secretary of Commerce announced a redesign of the 2010 Census, resulting in a $205 million increase in life-cycle costs. In February 2010, the Defense Integrated Military Human Resources System was canceled after 10 years of development and approximately $850 million spent, due, in part, to a lack of strategic alignment, governance, and requirements management, as well as the overall size and scope of the effort. In July 2010, OMB directed the National Archives and Records Administration (NARA) to halt development of its Electronic Records Archive system at the end of fiscal year 2011 (1 year earlier than planned). OMB cited concerns about the system’s cost, schedule, and performance and directed NARA to better define system functionality and improve strategic planning. Through fiscal year 2010, NARA had spent about $375 million on the system. In January 2011, the Secretary of Homeland Security ended the Secure Border Initiative Network program after spending about $1.5 billion because it did not meet cost-effectiveness and viability standards. In February 2011, the Office of Personnel Management canceled its Retirement Systems Modernization program, after several years of trying to improve the implementation of this investment.the Office of Personnel Management, it spent approximately $231 million on this investment. In March 2011, we reported that while DOD’s Navy Next Generation Enterprise Network investment’s first increment is estimated to cost $50 billion, the program was not well positioned to meet its cost and schedule estimates. As such, we recommended DOD limit further investment until it conducts an interim review to reconsider the selected acquisition approach and addresses its investment management issues. DOD stated that it did not concur with the recommendation to reconsider its acquisition approach, but we maintain that without doing so, DOD cannot be sure it is pursuing the most cost-effective approach. Additionally, as of August 2011, according to the IT Dashboard, 261 of the federal government’s approximately 800 major IT investments— totaling almost $18 billion—are in need of management attention (rated “yellow” to indicate the need for attention or “red” to indicate significant concerns). (See fig. 2.) In recognizing that wasteful spending continues to plague IT investment management, OMB has recently implemented additional efforts to address this problem. These efforts include the following: TechStat reviews. In January 2010, the Federal CIO began leading reviews—known as “TechStat” sessions—of selected IT investments involving OMB and agency leadership to increase accountability and transparency and improve performance. OMB officials stated that, as of December 2010, 58 sessions had been held and resulted in improvements to or termination of IT investments with performance problems. For example, the June 2010 TechStat session for NARA’s Electronic Records Archive investment (mentioned above) resulted in the halting of development funding pending the completion of a strategic plan. In addition, OMB has identified 26 additional high-priority IT projects and plans to develop corrective action plans with agencies at future TechStat sessions. According to the former Federal CIO, OMB’s efforts to improve management and oversight of IT investments have resulted in $3 billion in savings. IT reform. In December 2010, the Federal CIO issued a 25 Point Implementation Plan to Reform Federal Information Technology Management. This 18-month plan specified five major goals: strengthening program management, streamlining governance and improving accountability, increasing engagement with industry, aligning the acquisition and budget processes with the technology cycle, and applying “light technology” and shared solutions. As part of this plan, OMB outlined actions to, among other things, strengthen agencies’ investment review boards and consolidate federal data centers. The plan stated that OMB will work with Congress to consolidate commodity IT spending (e.g., e-mail, data centers, content management systems, and web infrastructure) under agency CIOs. Further, the plan called for the role of federal agency CIOs to focus more on IT portfolio management. In addition to these efforts to improve government spending on IT, avoiding unnecessary duplicative investments is critically important. In February 2002, OMB established the FEA initiative. According to OMB, the FEA is intended to facilitate governmentwide improvement through cross-agency analysis and identification of duplicative investments, gaps, and opportunities for collaboration, interoperability, and integration within and across agency programs. The FEA is composed of five “reference models” describing the federal government’s (1) business (or mission) processes and functions, independent of the agencies that perform them; (2) performance goals and outcome measures; (3) means of service delivery; (4) information and data definitions; and (5) technology standards. Since the fiscal year 2004 budget cycle, OMB has required agencies to categorize their IT investments in their annual exhibit 53s according to primary function and sub-function as identified in the FEA reference models. For fiscal year 2012 submissions, agencies chose from the primary functions listed in table 1. In their fiscal year 2011 submissions, agencies reported the greatest number of IT investments in Information and Technology Management (1,536 investments), followed by Supply Chain Management (777 investments), and Human Resource Management (622 investments). Similarly, planned expenditures on investments were greatest in Information and Technology Management, at about $35.5 billion. Figure 3 depicts, by primary function, the total number of investments within the 27 federal agencies that report to the IT Dashboard. Additionally, agencies were required to choose a sub-function for each investment related to the primary function. These sub-functions are to be selected from the business reference model. Table 2 provides examples of primary functions and their corresponding sub-functions. During the past several years, we have issued multiple reports and testimonies and made numerous recommendations to OMB and federal agencies to identify and reduce duplication within the federal government’s portfolio of IT investments. In March 2011, we reported an overview of federal programs and functional areas where unnecessary duplication, overlap, or fragmentation existed.agencies, offices, or initiatives had similar or overlapping objectives or provided similar services to the same populations, or where government missions were fragmented across multiple agencies or programs. These areas spanned a range of government missions: agriculture, defense, economic development, energy, general government, health, homeland security, international affairs, and social services. Within and across these missions, the report touched on hundreds of federal programs, including IT programs, affecting virtually all major federal departments and agencies. Specifically, we identified 34 areas where We reported that overlap and fragmentation among government programs or activities could be harbingers of unnecessary duplication. Thus, the reduction or elimination of duplication, overlap, or fragmentation could potentially save billions of tax dollars annually and help agencies provide more efficient and effective services. For example, we reported that, according to OMB, the number of federal data centers (defined as data processing and storage facilities) grew from 432 in 1998 to more than 2,000 in 2010. These data centers often house similar types of equipment and provide similar processing and storage capabilities. These factors have led to concerns associated with the provision of redundant capabilities, the underutilization of resources, and the significant consumption of energy. Operating such a large number of centers places costly demands on the government. In an effort to address these inefficiencies, in February 2010, OMB launched the Federal Data Center Consolidation Initiative to guide federal agencies in consolidating data centers. Specifically, OMB and agencies plan to close over 950 of the more than 2,100 federal data centers by 2015. As of November 2011, agencies reported that a total of 149 data centers have been closed across the federal government. For example, 16 DOD data centers, 3 DOE centers, and 7 DHS centers have been closed. In September 2011, we reported that limitations in OMB’s guidance hindered efforts to identify IT duplication. Specifically, OMB guidance stated that each IT investment needs to be mapped to a single functional category within the FEA to allow for the identification and analysis of potentially duplicative investments across agencies. We noted that this limits OMB’s ability to identify potentially duplicative investments both within and across agencies because similar investments may be organized under different functions. Accordingly, we recommended that OMB revise guidance to federal agencies on categorizing IT investments to ensure that the categorizations are clear and that it allow agencies to choose secondary categories, where applicable, which will aid in identifying potentially duplicative investments. OMB officials generally agreed with this recommendation and stated that they plan to update the FEA reference models in the fall of 2011 to provide additional clarity on how agencies should characterize investments in order to enhance the identification of potentially duplicative investments. We also reported that results of OMB initiatives to identify potentially duplicative investments were mixed and that several federal agencies did not routinely assess their entire IT portfolios to identify and remove or consolidate duplicative systems. Specifically, we said that most of OMB’s recent initiatives have not yet demonstrated results, and several agencies did not routinely assess legacy systems to determine if they are duplicative. As a result, we recommended that OMB require federal agencies to report the steps they take to ensure that their IT investments are not duplicative as part of their annual budget and IT investment submissions. OMB generally agreed with this recommendation. Although the Departments of Defense, Energy, and Homeland Security utilize various processes to prevent and reduce investment in duplicative programs and systems, potentially duplicative IT investments exist. Further complicating agencies’ ability to identify and address duplicative investments is miscategorization of investments within agencies. Each of the agencies has recently initiated plans to address many of these investments. DHS’s efforts have resulted in the identification and elimination of duplication, but DOD’s and DOE’s initiatives have not yet led to the elimination or consolidation of duplicative investments or functionality. Until DOD and DOE demonstrate progress on their efforts to identify and eliminate duplicative investments, and correctly categorize investments, it will remain unclear whether they are avoiding investment in unnecessary systems. Each of the agencies we reviewed has IT investment management processes in place that are, in part, intended to prevent, identify, and eliminate unnecessary duplicative investments. For example, DOD’s Information Technology Portfolio Management Implementation guide requires the evaluation of existing systems to identify duplication and determine whether to maintain, upgrade, delete, or replace identified systems. Similarly, DOE’s Guide to IT Capital Planning and Investment Control specifies that investment business case summaries should be reviewed for redundancies and opportunities for collaboration. Additionally, according to DHS’s Capital Planning and Investment Control Guide, proposed investments must be reviewed at the department level to determine if the proposed need is, among other things, being fulfilled by another DHS program, or already fulfilled by an existing capability. Even with such investment review processes, of the 810 investments we reviewed, we identified 37 potentially duplicative investments at DOD and DOE within three FEA categories (Human Resource Management, Information and Technology Management, and Supply Chain These investments account for about $1.2 billion in total Management).IT spending for fiscal years 2007 through 2012. Specifically, we identified 31 potentially duplicative investments totaling approximately $1.2 billion at DOD, and 6 potentially duplicative investments totaling approximately $8 million at DOE. The 37 investments comprise 12 groups of investments that appear to have duplicative purposes based on our analysis of each investment’s description, budget information, and other supporting documentation from agency officials (see table 3). For example, we identified three investments at DOE that were each responsible for managing the back- end infrastructure at three different locations. We also identified four DOD Navy personnel assignment investments—one system for officers, one for enlisted personnel, one for reservists, and a general assignment system—each of which is responsible for managing similar assignment functions. Additionally, the Air Force has five investments that are each responsible for contract management, and within the Navy there are another five contract management investments. Table 3 summarizes the 12 groups of potentially duplicative investments we identified by purpose and agency. (See app. II for details on each of the 37 potentially duplicative investments.) We did not identify any potentially duplicative investments at DHS within our sample; however, DHS has independently identified several duplicative investments and systems. Specifically, DHS officials have identified and, more importantly, reduced duplicative functionality in four investments by consolidating or eliminating certain systems within each of these investments. DHS officials have also identified 38 additional systems that they have determined to be duplicative. For example, officials identified multiple personnel action processing systems that could be consolidated. Officials from the three agencies reported that duplicative investments exist for a number of reasons, including decentralized governance within the departments and a lack of control over contractor facilities. For example, DOE investments for the management of back-end infrastructure are for facilities which DOE oversees but does not control. In addition, DOD officials indicated that a key reason for potential duplication at the Department of the Navy is that it had traditionally used a decentralized IT management approach, which allowed offices to develop systems independent of any other office’s IT needs or acquisitions. Further complicating the agencies’ ability to prevent investment in duplicative systems or programs is the miscategorization of investments. Among the 810 investments we reviewed, we identified 22 investments where the selected agencies assigned incorrect FEA primary functions or Specifically, we identified 13 miscategorized investments sub-functions.at DOD, 4 at DOE, and 5 at DHS. Examples are as follows: DOD’s Computer Aided Procurement System was initially categorized within the Information and Technology Management primary function, but DOD agreed that this investment should be classified within the Supply Chain Management primary function. DOE’s Environmental Management Headquarters Central Internet Database was initially categorized within the Information and Technology Management primary function, but DOE agreed that this investment could be assigned the Environmental Management primary function and the Environmental Monitoring and Forecasting sub-function. DHS’s Federal Emergency Management Agency—Minor Personnel/Training Systems investment was initially categorized within the Employee Performance Management sub-function, but DHS agreed that this investment should be assigned to the Human Resources Development sub-function. Agency officials agreed that they had inadvertently miscategorized 15 of the 22 investments we identified. However, proper categorization is necessary in order to analyze and identify duplicative investments, both within and across agencies. Each improper categorization represents a possible missed opportunity to identify and eliminate an unjustified duplicative investment. Until agencies correctly categorize their investments, they cannot be confident that their investments are not duplicative and are justified, and they may continue expending valuable resources developing and maintaining unnecessarily duplicative systems. DHS has taken action to improve its processes for identifying and eliminating duplicative investments, which has produced tangible results. Specifically, in 2010 and 2011, the DHS CIO conducted program and portfolio reviews of hundreds of IT investments and systems. DHS evaluated portfolios of investments within its components to avoid investing in systems that are duplicative or overlapping, and to identify and leverage investments across the department. Among other things, this effort contributed to the identification and consolidation of duplicative functionality within four investments. DHS also has plans to further consolidate systems within these investments by 2014, which is expected to produce approximately $41 million in cost savings. The portfolio reviews also contributed to the identification of 38 additional systems that are duplicative. Additionally, the DHS CIO and Chief Human Capital Officer are coordinating to streamline and consolidate the department’s human resources investments. A summary of the investments for which DHS eliminated duplicative functionality and systems is provided in table 4 below. DOD has begun taking action to address 29 of the 31 duplicative investments we identified. For example, according to DOD officials, four of the DOD Navy acquisition management investments—two for Naval Sea Systems Command and two for Space and Naval Warfare Systems Command—will be reviewed to determine whether these multiple support systems are necessary. In addition, DOD reported that the Air Force is in the process of developing a single contract writing system to replace the five potentially duplicative investments we have identified. Moreover, the Department of the Navy has implemented an executive oversight board that is chaired by the Navy CIO, and it is now the Navy’s single senior information management and technology policy and governance forum. The Department of the Navy also required all IT expenditures greater than $100,000 to be centrally reviewed and approved by the Navy CIO to ensure that they are not duplicative. Officials reported that these initiatives will include the review of Navy’s 22 potentially duplicative investments that we identified. Similarly, DOE has plans under way to address each of the 6 investments we identified as potentially duplicative. Specifically, DOE officials established working groups that are addressing the two groups of duplicative investments we identified. These working groups are to address records management and back-end infrastructure, and are looking across the department to minimize redundancy in each of these areas. In addition, the CIO stated that DOE has developed a departmental strategy for electronic records management whereby a small number of approved records management applications will be identified for departmentwide use. Moreover, in a broader effort to reduce duplication across the department, in September and October 2011, DOE held technical strategic reviews, known as “TechStrat” sessions, which are aimed at exploring opportunities to consolidate DOE’s commodity IT services, such as e-mail and help desk support, among the various DOE offices. The first two sessions provided opportunities for DOE bureaus to identify and share lessons learned, and established action items to improve DOE’s IT investment portfolio. While these efforts could eventually yield results, DOD’s and DOE’s initiatives have not yet led to the consolidation or elimination of duplication. For example, while DOD provided us with documented milestones—several of which have passed—for improving the Department of the Navy’s IT investment review processes, officials did not provide us with any examples of duplicative investments that they had consolidated or eliminated. Similarly, while DOE officials have documented time frames for consolidating DOE’s commodity IT services, electronic records management investments, and identity management investments, officials were unable to demonstrate that they have consolidated or eliminated unjustified duplicative investments. Additionally, DOD does not have plans under way to address the remaining 2 of the 31 potentially duplicative investments. DOD officials stated that they do not have plans to address these investments because they do not agree that they are potentially duplicative. However, agency officials were unable to demonstrate that investing in these systems and programs was justified. Table 5 provides more information on the unaddressed potentially duplicative investments at DOD. Table 6 summarizes the number of potentially duplicative investments for which Defense and Energy have actions under way, as well as the number of investments that remain unaddressed. Until DOD and DOE demonstrate, through existing transparency mechanisms such as OMB’s IT Dashboard, that they are making progress in identifying and eliminating duplicative investments, it will remain unclear whether they are avoiding investment in unnecessary systems. While agencies have various investment review processes in place that are partially designed to avoid investing in systems that are duplicative, we have identified 37 potentially duplicative investments at DOD and DOE. These investments account for about $1.2 billion in total IT spending for fiscal years 2007 through 2012. Given that our review covered 11 percent (810 investments) of the total number of IT investments that agencies report to OMB, it raises questions about how much more potential duplication exists. DHS’s recent efforts have resulted in the identification and consolidation of duplicative functionality in several investments and related systems. DOD and DOE have also recently initiated plans to address many investments that we identified, but these recent initiatives have not yet resulted in the consolidation or elimination of duplicative investments or functionality. Further complicating agencies’ ability to prevent, identify, and eliminate duplicative investments is miscategorization of investments within agencies. Without demonstrating the progress of efforts to identity and eliminate duplicative investments, DOD and DOE will be unable to provide assurance that they are avoiding investment in unnecessary systems. Similarly, until DOD, DOE, and DHS, correctly categorize their investments, they are limiting their ability to identify opportunities to consolidate or eliminate duplicative investments. To better ensure agencies avoid investing in duplicative investments, we recommend that the Secretary of Defense direct the CIO to take the following two actions: utilize existing transparency mechanisms, such as the IT Dashboard, to report on the results of the department’s efforts to identify and eliminate, where appropriate, each potentially duplicative investment we have identified, as well as any other duplicative investments; and correct the miscategorizations for the DOD investments we identified and ensure that investments are correctly categorized in agency submissions. We recommend that the Secretary of Energy direct the CIO to take the following two actions: utilize existing transparency mechanisms, such as the IT Dashboard, to report on the results of the department’s efforts to identify and eliminate, where appropriate, each potentially duplicative investment we have identified, as well as any other duplicative investments; and correct the miscategorizations for the DOE investments we identified and ensure that investments are correctly categorized in agency submissions. We recommend that the Secretary of Homeland Security direct the CIO to take the following action: correct the miscategorizations for the DHS investments we identified and ensure that investments are correctly categorized in agency submissions. We provided a draft of our report to the three departments selected for our review and to OMB. In commenting on the draft, DOD and DHS generally concurred with our recommendations. DOE generally agreed with our first recommendation and disagreed with parts of our second recommendation. In addition, OMB provided oral technical comments that we incorporated, where appropriate. Each department’s comments are discussed in more detail below, and the written comments are reprinted in appendixes IV, V, and VI. DOD’s Deputy CIO for Information Management, Integration, and Technology within the Office of the Assistant Secretary of Defense for Networks and Information Integration provided written comments, which stated that the department agreed with both of our recommendations. DOD also provided technical comments, which we incorporated, where appropriate. The Director of DHS’s Departmental GAO/Office of Inspector General Liaison Office provided written comments, which stated that the department agreed with our recommendation to correct the miscategorized investments and ensure that investments are correctly categorized. Additionally, DHS provided documentation showing that the department had recently corrected the miscategorizations in response to our recommendation. The department also provided technical comments, which we incorporated as appropriate. The DOE CIO provided written comments in which the department generally agreed with the first recommendation and disagreed with parts of the second recommendation. Regarding our first recommendation, to identify and eliminate potentially duplicative investments as appropriate, DOE generally agreed with the recommendation and stated that the Office of the CIO is committed to increasing its IT investment oversight. The department added that for the non-major investments that GAO identified as being potentially duplicative, it will update GAO on its progress through means other than the IT Dashboard, since non-major investments are not individually tracked on the Dashboard. However, DOE also indicated that it does not believe certain investments that we identified are potentially duplicative. Specifically, DOE did not agree that the two card issuance and maintenance, and three logical access control investments were potentially duplicative. Rather, it stated that the investments in these groups were listed individually on the exhibit 53 for reporting purposes, in order to show how the funding was being distributed at various locations. According to DOE, these costs were for the labor involved in deploying the technology, and could not be avoided given the separate geographical locations. We reviewed this additional information, and subsequently removed these five investments from our list of potentially duplicative investments. Regarding our second recommendation to correct miscategorizations and ensure that investments are correctly categorized, DOE disagreed with parts of this recommendation. Specifically, DOE agreed that two of the four investments could be recategorized. However, it disagreed that the two training center investments should be recategorized, and stated that they should continue to be categorized under the Employee Performance Management FEA sub-function because of how they are funded. However, OMB guidance defines Employee Performance Management as activities that enable managers to make distinctions in performance and link individual performance to agency goals and mission accomplishment. In other words, this sub-function involves enabling managers to assess the performance of personnel—and does not involve providing training to personnel. In contrast, the Human Resources Development sub-function—which OMB guidance defines as administering, delivering, and designing employee development programs—is a more appropriate category.position. Additionally, DOE stated that we identified only 4 miscategorized investments from its total population of 876 investments. However, this implies we reviewed all 876 investments. As stated in our report, we looked at 19 percent of DOE’s reported IT investment population, or 167 investments, and identified 4 miscategorized investments from that subset. Therefore, we maintain our In addition, DOE stated that in our September 2011 report we highlighted limitations in OMB’s guidance regarding proper categorization of investments and further stated that, while OMB agreed to make improvements to the guidance, agencies and OMB did not have time to implement the changes before our new audit began. In our September report, we noted that, under OMB’s guidance, agencies were unable to designate a secondary category, in addition to the primary category for each of the investments. However, in this report, our concern is with the accuracy of agencies’ selections of the primary categories for certain investments. These are two independent concerns with investment categorization—both of which need to be addressed and are not necessarily dependent on each other. In other words, regardless of whether agencies are able to designate a secondary category, in addition to a primary category, it is still critically important that the primary category is accurate. DOE made several additional comments that we address below: The department stated that it has implemented various investment review processes to help identify potentially duplicative investments and to manage these investments. We acknowledge in the report that DOE has such processes in place, and we provide examples of the department’s existing IT investment management processes that are, in part, intended to prevent, identify, and eliminate duplicative investments. DOE stated that our draft report mentions the Federal Data Center Consolidation Initiative but that we did not specifically discuss DOE’s accomplishment in this area. In response, we added the number of federal data centers that DOE reportedly closed. The department stated that prior to the GAO audit, DOE officials realized potential duplicative investments may exist in back-end infrastructure and that a working group has been meeting regularly to identify duplicative investments and investigate the possibility of consolidating. We agree with this statement, and we acknowledge the working group’s efforts in the report. However, as we report, this initiative has not yet resulted in the consolidation or elimination of duplicative investments or functionality. According to DOE, it had developed a departmental strategy for electronic records management whereby a small number of approved records management applications will be identified for department- wide use. It added that the three records management investments cited in our report will remain in place while the departmental strategy is being implemented. In response to this comment, we updated the report to acknowledge that the CIO stated that DOE has developed a departmental strategy, in addition to establishing an electronic records management working group. However, similar to the back-end infrastructure, these efforts have not yet resulted in the consolidation or elimination of duplicative investments or functionality, and thus, DOE may continue investing in unnecessary systems until such actions are taken. Lastly, DOE noted that in our report we discuss the Department’s TechStrat sessions related to commodity IT services but did not discuss the TechStrat sessions conducted by its Office of Environmental Management on its major investments. We did not add this activity to the report, because supporting documentation was not provided to indicate that this session was conducted to specifically reduce duplication, rather than to review major investments with performance problems. Finally, OMB’s Chief Architect provided comments regarding the office’s efforts to oversee IT investments, which we incorporated, as appropriate. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 11 days from the report date. At that time, we will send copies of this report to the appropriate congressional committees; the Secretaries of Defense, Energy, and Homeland Security; the Director of the Office of Management and Budget; and other interested parties. In addition, the report also will be available at no charge on GAO’s website at http://www.gao.gov. If you or your staff members have any questions on the matters discussed in this report, please contact me at (202) 512-9286 or pownerd@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix VII. Our objective was to identify potentially duplicative information technology (IT) investments at selected agencies and actions these agencies are taking to address them. To select agencies for review, we used the Office of Management and Budget’s (OMB) fiscal year 2011 exhibit 53. Specifically, we downloaded this data from OMB’s IT Dashboard and used it to identify the agencies and their number of IT investments as reported on the Dashboard. We used this analysis to select for review three of the agencies with the highest number of IT investments—the Departments of Defense (DOD), Energy (DOE), and Homeland Security (DHS). To identify potentially duplicative investments, we further narrowed our analysis of the exhibit 53 data to the largest Federal Enterprise Architecture (FEA) primary functions, by number of investments. Within each of the selected primary functions, we selected the two sub-functions with the most investments. Table 7 identifies the FEA primary functions and FEA sub-functions used to select the investments for review. This resulted in a nongeneralizable sample of 810 IT investments, which is 11 percent of the total number of IT investments that agencies report to OMB through the IT Dashboard (810 of 7,227).The investments we reviewed represent approximately 24 percent of DOD’s IT portfolio in terms of number of investments that it reports to the Dashboard, 19 percent of DOE’s, and 16 percent of DHS’s. To determine the reliability of the data on the IT Dashboard, we reviewed recent GAO reports that identified issues with the accuracy and reliability of agency data on the IT We determined that the data were sufficiently reliable for the Dashboard.purpose of this report, which was to identify selected investments to include in our review. We then reviewed the name and narrative description of each investment’s purpose to identify similarities among related investments within each agency (we did not review investments across agencies). This formed the basis of establishing groupings of similar investments. We discussed the groupings with each of the selected agencies, and we obtained further information from agency officials. We also reviewed and assessed agencies’ rationales for having multiple systems that perform similar functions. Additionally, when analyzing each investment’s description, we compared each investment’s designated FEA primary category and sub-category to OMB’s definitions for each FEA primary category and sub-category and determined whether the investment was placed in the correct FEA category. We obtained additional information from agency officials about these discrepancies. To identify the actions agencies have taken to address the potentially duplicative investments we identified, we reviewed agency documentation, such as agency memos and working group charters, and interviewed officials. We also reviewed documentation and interviewed agency officials to identify what investments were consolidated, eliminated, or modified to decrease duplication and the estimated cost savings (if available) associated with these actions. We conducted this performance audit from June 2011 to February 2012 in accordance with generally accepted government auditing standards. Those standards required that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objective. The tables in this appendix provide information on the 37 investments that we identified as potentially duplicative within the three selected FEA functions (Human Resource Management, Information and Technology Management, and Supply Chain Management). Specifically, we identified 31 potentially duplicative IT investments at DOD and 6 at DOE. Highlighted investments indicate the instances in which the agency does not currently have plans under way to address the potential duplication. The tables in this appendix provide information on the 22 investments that we identified as incorrectly categorized by the selected agencies according to OMB’s FEA. Specifically, we identified 13 miscategorized investments at DOD (2 within Air Force, 2 within Army, 3 within Navy, and 6 enterprisewide), 4 at DOE, and 5 at DHS. Highlighted investments indicate the seven instances in which the agency did not agree that the investments were miscategorized. In addition to the individual named above, the following staff made key contributions to this report: Shannin O’Neill, Assistant Director; Cortland Bradford; Javier Irizarry; Lee McCracken; and Kevin Walsh. | The federal government spends billions of dollars on information technology (IT) each year, with such investments accounting for at least $79 billion in fiscal year 2011. Given the size of these investments, it is important that federal agencies avoid duplicative investments when possible to ensure the most efficient use of resources. GAO has previously reported on initiatives under way to address potentially duplicative IT investmentsi.e., investments providing similar functions across the government. GAO was asked to review the extent to which potentially duplicative IT investments exist within three categories at selected agencies (the Departments of Defense (DOD), Energy (DOE), and Homeland Security (DHS)) and actions these agencies are taking to address them. To accomplish this, GAO analyzed budget data on agency IT investments, reviewed agency information related to efforts to address duplication, and interviewed agency officials. Although the Departments of Defense (DOD) and Energy (DOE) use various investment review processes to identify duplicative investments, GAO found that 37 of its sample of 810 investments were potentially duplicative. These investments account for about $1.2 billion in total information technology (IT) spending for fiscal years 2007 through 2012. For example, GAO identified four DOD Navy personnel assignment investmentsone system for officers, one for enlisted personnel, one for reservists, and a general assignment systemeach of which is responsible for managing similar functions. While GAO did not identify any potentially duplicative investments at the Department of Homeland Security (DHS) within its sample, DHS officials have independently identified several duplicative investments and systems. DOD and DOE officials offered a variety of reasons for the potential duplication, such as decentralized governance and a lack of control over certain facilities. Further complicating agencies ability to identify and eliminate duplicative investments is that investments are, in certain cases, misclassified by function. Until agencies correctly categorize their investments, they cannot be confident that their investments are not duplicative. DHS has taken action to improve its processes for identifying and eliminating duplicative investments. For example, through reviewing portfolios of IT investments, DHS has identified much, and eliminated some, duplicative functionality in certain investments. Additionally, DOD and DOE have recently initiated plans to address potential duplication in many of the investments GAO identified, which include consolidating or eliminating systems. While these efforts may eventually yield results, they have not yet led to the elimination of duplication. For example, while DOD and DOE have specific plans to improve their IT investment review processes, officials did not provide examples of duplicative investments that had been consolidated or eliminated. Until DOD and DOE demonstrate progress on these efforts, the agencies will be unable to provide assurance that they are avoiding investment in unnecessary systems. GAO recommends that DOD and DOE report on the progress of efforts to identify and eliminate duplication, where appropriate. GAO is also recommending that DOD, DOE, and DHS correct misclassifications of investments. DOD and DHS agreed with the recommendations. DOE generally agreed with the first recommendation, but disagreed with parts of the second recommendation regarding the number of misclassified investments. However, GAO believes the number is accurate. |
The Great Lakes Basin is a large area that extends well beyond the five lakes proper to include their watersheds, tributaries, connecting channels, and a portion of the St. Lawrence River. The basin encompasses nearly all of the state of Michigan and parts of Illinois, Indiana, Minnesota, New York, Ohio, Pennsylvania, Wisconsin, and the Canadian province of Ontario. The lakes form the largest freshwater system on earth, accounting for 20 percent of the world’s fresh surface water and over 95 percent of the U.S. fresh surface water supply for the contiguous 48 states. Millions of people in the United States and Canada rely on the five Great Lakes—Superior, Michigan, Erie, Huron, and Ontario—as a principal source of their drinking water, recreation, and economic livelihood. Over time, industrial, agricultural, and residential development on lands adjacent to the lakes has seriously degraded the lakes’ water quality, posing threats to human health and the environment, and forcing restrictions on activities such as swimming and fish consumption. To protect the Great Lakes Basin and to address water quality problems, the governments of the United States and Canada entered into the bilateral Great Lakes Water Quality Agreement in 1972. In the agreement, the United States and Canada agreed to restore and maintain the chemical, physical, and biological integrity of the Great Lakes Basin. A new agreement with the same name was reached in 1978 and amended in 1983 and 1987. The agreement prescribes prevention and cleanup measures to improve environmental conditions in the Great Lakes. The agreement obligates the International Joint Commission (IJC), an international body, to assist in and report on the implementation of the agreement. The Clean Water Act directs EPA to lead efforts to meet the goals of the Great Lakes Water Quality Agreement and establishes GLNPO within EPA, charging it with, among other things, cooperating with federal, state, tribal, and international agencies to develop action plans to carry out the responsibilities of the U.S. under the agreement. GLNPO is further responsible for coordinating the agency’s actions both in headquarters and in the regions to improve Great Lakes’ water quality. In addition to GLNPO, numerous federal, state, binational, and nonprofit organizations conduct activities that focus on improving the overall Great Lakes Basin environment or some specific environmental issue within the basin. About 200 programs—148 federal and 51 state—fund restoration activities within the Great Lakes Basin. Most of these programs, however, involve the localized application of national or state environmental initiatives and do not specifically focus on basin concerns. Officials from 11 federal agencies identified 115 of these broadly scoped federal programs, and officials from seven of the eight Great Lakes states identified 34 similar state programs. EPA administers the majority of the federal programs that provide a broad range of environmental activities involving research, cleanup, restoration, and pollution prevention. For example, EPA’s nationwide Superfund program funds cleanup activities at contaminated areas throughout the basin. While these broadly scoped federal and state programs contribute to basin restoration, program officials do not track or try to isolate the portion of funding directed toward specific areas, such as the basin, which makes it difficult to determine their contributions to total Great Lakes spending. However, basin-specific information was available on some of these programs. Specifically, basin-related expenditures for 53 of the 115 broadly scoped federal programs totaled about $1.8 billion in fiscal years 1992 through 2001. Expenditures for 14 broadly scoped state- funded programs totaled $461.3 million during approximately the same time period. Several federal and state programs were specifically designed to focus on environmental conditions across the Great Lakes Basin. Officials from seven federal agencies identified 33 Great Lakes specific programs that had expenditures of $387 million in fiscal years 1992 through 2001. Most of these programs funded a variety of activities, such as research, cleanup, or pollution prevention. An additional $358 million was expended for legislatively directed Corps of Engineers projects in the basin, such as a $93.8 million project to restore Chicago’s shoreline. Officials from seven states reported 17 Great Lakes specific programs that expended about $956 million in 1992 through 2001, with Michigan’s programs accounting for 96 percent of this amount. State programs focused on unique state needs, such as Ohio’s program to control shoreline erosion along Lake Erie and Michigan’s program to provide bond funding for environmental activities. Besides federal and state government agencies, other organizations, such as foundations, fund a variety of restoration activities in the Great Lakes Basin by approving grants to nonprofit and other organizations. Other governmental and nongovernmental organizations fund restoration activities. For example, individual municipalities, township governments, counties, and conservation districts are involved in various restoration activities. Restoration of the Great Lakes Basin is a major endeavor involving many environmental programs and organizations. The magnitude of the area comprising the basin and the numerous environmental programs operating within it require the development of one overarching strategy to address and manage the complexities of restoring the basin’s environmental health. The Great Lakes region cannot hope to successfully receive support as a national priority without a comprehensive plan for restoring the Great Lakes. In lieu of such a plan, organizations at the binational, federal, and state levels have developed their own strategies for the Great Lakes, which have inadvertently made the coordination of the various programs operating in the basin more challenging. The Great Lakes Basin needs a comprehensive strategy or plan similar to the plans developed for other large ecosystem restoration efforts, such as those for the South Florida ecosystem and the Chesapeake Bay. In South Florida, federal, state, local and tribal organizations joined forces to participate on a centralized task force formalized in the Water Resource Development Act of 1996. The strategic plan developed for the South Florida ecosystem by the task force made substantial progress in guiding the restoration activities. The plan identifies the resources needed to achieve restoration and assigns accountability for specific actions for the extensive restoration effort, estimated to cost $14.8 billion. The Chesapeake Bay watershed also has an overarching restoration strategy stemming from a 1983 agreement signed by Maryland, Virginia, and Pennsylvania; the District of Columbia; the Chesapeake Bay Commission; and EPA. The implementation of this strategy has resulted in improvements in habitat restoration and aquatic life, such as increases in bay grasses and in the shad population. Several organizations have developed strategies for the basin at the binational, federal, or state levels that address either the entire basin or the specific problems in the Great Lakes. EPA’s Great Lakes Strategy 2002, developed by a committee of federal and state officials, is the most recent of these strategies. While this strategy identified restoration objectives and planned actions by various federal and state agencies, it is largely a description of existing program activity relating to basin restoration. State officials told us that the states had already planned the actions described in it, but that these actions were contingent on funding for specific environmental programs. The strategy included a statement that it should not be construed as a commitment for additional funding or resources, and it did not provide a basis for prioritizing activities. In addition, we identified other strategies that addressed particular contaminants, the restoration of individual lakes, or the cleanup of contaminated areas. Ad hoc coordination takes place among federal agencies, states, and other environmental organizations in developing these strategies or when programmatic activity calls for coordination. Other Great Lakes strategies address unique environmental problems or specific geographical areas. For example, a strategy for each lake addresses the open lake waters through Lakewide Management Plans (LaMP), which EPA is responsible for developing. Toward this end, EPA formed working groups for each lake to identify and address restoration activities. For example, the LaMP for Lake Michigan, issued in 2002, includes a summary of the lake’s ecosystem status and addresses progress in achieving the goals described in the previous plan, with examples of significant activities completed and other relevant topics. However, EPA has not used the LaMPs to assess the overall health of the ecosystem. The Binational Executive Committee for the United States and Canada issued its Great Lakes Binational Toxics Strategy in 1997 that established a collaborative process by which EPA and Environment Canada, in consultation with other federal departments and agencies, states, tribes and the province of Ontario work toward the virtual elimination of persistent toxic substances in the Great Lakes. The strategy was designed to address particular substances that bioaccumulate in fish or animals and pose a human health risk. Michigan developed a strategy for environmental cleanup called the Clean Michigan Initiative. This initiative provides funding for a variety of environmental, parks, and redevelopment programs. It includes nine components, including Brownfields redevelopment and environmental cleanups, nonpoint source pollution control, clean water, cleanup of contaminated sediments, and pollution prevention. The initiative is funded by a $675 million general obligation bond and, as of early 2003, most of the funds had not been distributed. Although there are many strategies and coordination efforts ongoing, no one organization coordinates restoration efforts. We found that extensive strategizing, planning, and coordinating have not resulted in significant restoration. Thus, the ecosystem remains compromised and contaminated sediments in the lakes produce health problems, as reported by the IJC. In addition to the absence of a coordinating agency, federal and state officials cited a lack of funding commitments as a principal barrier that impedes restoration progress. Inadequate funding has also contributed to the failure to restore and protect the Great Lakes, according to the IJC biennial report on Great Lakes water quality issued in July 2000. The IJC restated this position in a 2002 report, concluding that any progress to restore the Great Lakes would continue at a slow incremental pace without increased funding. In its 1993 biennial report, the IJC concluded that remediation of contaminated areas could not be accomplished unless government officials came to grips with the magnitude of cleanup costs and started the process of securing the necessary resources. Despite this warning, however, as we reported in 2002, EPA reduced the funding available for ensuring the cleanup of contaminated areas under the assumption that the states would fill the funding void. States, however, did not increase their funding, and restoration progress slowed or stopped altogether. Officials for 24 of 33 federal programs and for 3 of 17 state programs reported insufficient funding for federal and state Great Lakes specific programs. The ultimate responsibility for coordinating Great Lakes restoration programs rests with GLNPO; however, GLNPO has not fully exercised this authority. Other organizations or committees have been formed to assume coordination and strategy development roles. The Clean Water Act provides GLNPO with the authority to fulfill the responsibilities of the U.S. under the GLWQA. Specifically, the act directs EPA to coordinate the actions of EPA’s headquarters and regional offices aimed at improving Great Lakes water quality. It also provides GLNPO authority to coordinate EPA’s actions with the actions of other federal agencies and state and local authorities for obtaining input in developing water quality strategies and obtaining support in achieving the objectives of the GLWQA. The act also provides that the EPA Administrator shall ensure that GLNPO enters into agreements with the various organizational elements of the agency engaged in Great Lakes activities and with appropriate state agencies. The agreements should specifically delineate the duties and responsibilities, time periods for carrying out duties, and resources committed to these duties. GLNPO officials stated that they do not enter into formal agreements with other EPA offices but rather fulfill their responsibilities under the act by having federal agencies and state officials agree to the restoration activities contained in the Great Lakes Strategy 2002. However, the strategy does not represent formal agreements to conduct specific duties and responsibilities with committed resources. EPA’s Office of Inspector General reported the absence of these agreements in September 1999. The report stated that GLNPO did not have agreements as required by the act and recommended that such agreements be made to improve working relationships and coordination. To improve coordination of Great Lakes activities and ensure that federal dollars are effectively spent, we recommended that the Administrator, EPA, ensure that GLNPO fulfills its responsibility for coordinating programs within the Great Lakes Basin; charge GLNPO with developing, in consultation with the governors of the Great Lakes states, federal agencies, and other organizations, an overarching strategy that clearly defines the roles and responsibilities for coordinating and prioritizing funding for projects; and submit a time-phased funding requirement proposal to the Congress necessary to implement the strategy. The Great Lakes Water Quality Agreement, as amended in 1987, calls for establishing a monitoring system to measure restoration progress and assess the degree to which the United States and Canada are complying with the goals and objectives of the agreement. However, implementation of this provision has not progressed to the point that overall restoration progress can be measured or determined based on quantitative information. Recent assessments of overall progress, which rely on a mix of quantitative data and subjective judgments, do not provide an adequate basis for making an overall assessment. The current assessment process has emerged from a series of biennial State of the Lakes Ecosystem Conferences (SOLEC) initiated in 1994 for developing indicators agreed upon by conference participants. Prior to the 1987 amendments to the GLWQA, the 1978 agreement between the two countries also contained a requirement for surveillance and monitoring and for the development of a Great Lakes International Surveillance Plan. The IJC Water Quality Board was involved in managing and developing the program until the 1987 amendments gave this responsibility to the United States and Canada. This change resulted in a significant reduction in the two countries’ support for surveillance and monitoring. In fact, the organizational structure to implement the surveillance plan was abandoned in 1990, leaving only one initiative in place—the International Atmospheric Deposition Network (IADN), which involved a network of 15 air-monitoring stations located throughout the basin. With the surveillance and monitoring efforts languishing, IJC established the Indicators for Evaluation Task Force in 1993 to identify the appropriate framework to evaluate progress in the Great Lakes. In 1996, the task force proposed that nine desired measurements and outcomes be used to develop indicators for measuring progress in the Great Lakes. Shortly before the task force began its work, the United States and Canada had agreed to hold conferences every 2 years to assess the environmental conditions in the Great Lakes in order to develop binational reports on environmental conditions to measure progress under the agreement. Besides assessing environmental conditions, the conferences were focused on achieving three other objectives, including providing a forum for communication and networking among stakeholders. Conference participants included U.S. and Canadian representatives from federal, state, provincial, and tribal agencies, as well as from other organizations with environmental restoration or pollution prevention interests in the Great Lakes Basin. The 1994 SOLEC conference culminated in a “State of the Great Lakes 1995” report, which provided an overview of the Great Lakes ecosystem at the end of 1994 and concluded that overall the aquatic community health was mixed or improving. This same assessment was echoed in the 1997 state of the lakes report. Meanwhile the IJC agreed that the nine desired outcome areas recommended by the task force would help assess overall progress. It recommended that SOLEC, during the conference in 2000, establish environmental indicators that would allow the IJC to evaluate what had been accomplished and what needed to be done for three of the nine indicators—the public’s ability to eat the fish, drink the water, and swim in the water without any restrictions. However, the indicators developed through the SOLEC process and the accomplishments reported by federal and state program managers do not provide an adequate basis for making an overall assessment for Great Lakes restoration progress. The SOLEC process is ongoing, and the indicators that are still being developed are not generally supported by sufficient underlying data for making progress assessments. The number of indicators considered during the SOLEC conferences has been pared down from more than 850 indicators in 1998 to 80 indicators in 2000, although data was available for only 33 of them. After the SOLEC 2000 conference, IJC staff assessed the indicators supported by data that measured the desired outcomes of swimmability, drinkability, and the edibility of fish in the Great Lakes. Overall, the IJC commended SOLEC’s quick response that brought together information regarding the outcomes and SOLEC’s ongoing efforts. The IJC, however, recognized that sufficient data were not being collected throughout the Great Lakes Basin and that the methods of collection, the data collection time frames, the lack of uniform protocols, and the incompatible nature of some data jeopardized their use as indicators. Specifically, for the desired outcome of swimmability, the IJC concurred that it was not always safe to swim at certain beaches but noted that progress for this desired outcome was limited because beaches were sampled by local jurisdictions without uniform sampling or reporting methods. At the 2002 SOLEC conference, the number of indicators assessed by conference participants increased from 33 to 45. The IJC expressed concern that there are too many indicators, insufficient supporting backup data, and a lack of commitment and funding from EPA to implement and make operational the agreed upon SOLEC baseline data collection and monitoring techniques. The IJC recommended in its last biennial report that any new indicators should be developed only where resources are sufficient to access scientifically valid and reliable information. The ultimate successful development and assessment of indicators for the Great Lakes through the SOLEC process are uncertain because insufficient resources have been committed to the process, no plan provides completion dates for indicator development and implementation, and no entity is coordinating the data collection. Even though the SOLEC process has successfully engaged a wide range of binational parties in developing indicators, the resources devoted to this process are largely provided on a voluntary basis without firm commitments to continue in the future. GLNPO officials described the SOLEC process as a professional, collaborative process dependent on the voluntary participation of officials from federal and state agencies, academic institutions, and other organizations attending SOLEC and developing information on specific indicators. Because SOLEC is a voluntary process, the indicator data resides in a diverse number of sources with limited control by SOLEC organizers. GLNPO officials stated that EPA has neither the authority nor the responsibility to direct the data collection activities of federal, state, and local agencies as they relate to the surveillance and monitoring of technical data elements that are needed to develop, implement, and assess Great Lakes environmental indicators. Efforts are underway for the various federal and state agencies to take ownership for collecting and reporting data outputs from their respective areas of responsibility and for SOLEC to be sustained and implemented; each indicator must have a sponsor. However, any breakdown in submitting this information would leave a gap in the SOLEC indicator process. EPA supports the development of environmental indicators as evidenced by the fact that, since 1994, GLNPO has provided about $100,000 annually to sponsor the SOLEC conferences. Additionally, GLNPO spends over $4 million per year to collect surveillance data for its open-lake water quality monitoring program, which also provides supporting data for some of the indicators addressed by SOLEC. A significant portion of these funds, however, supports the operation of GLNPO’s research vessel, the Lake Guardian, an offshore supply vessel converted for use as a research vessel. GLNPO also supports activities that are linked or otherwise feed information into the SOLEC process, including the following: collecting information on plankton and benthic communities in the Great Lakes for open water indicator development; sampling various chemicals in the open-lake waters, such as phosphorus for the total phosphorus indicator; monitoring fish contaminants in the open waters, directly supporting the indicator for contaminants in whole fish and a separate monitoring effort for contaminants in popular sport fish species that supports the indicator for chemical contaminants in edible fish tissue; and operating 15 air-monitoring stations with Environment Canada comprising the IADN that provides information for establishing trends in concentrations of certain chemicals and loadings of chemicals into the lakes. EPA uses information from the network to take actions to control the chemicals and track progress toward environmental goals. In November 2001, EPA committed to an agencywide initiative to develop environmental indicators for addressing the agency’s nationwide environmental conditions, stating that “indicators help measure the state of our air, water and land resources and the pressures placed on them, and the resulting effects on ecological and human health.” However, this initiative does not specifically relate to the Great Lakes. The short-term goal for this initiative is to develop information that will indicate current nationwide environmental conditions and to help EPA make sound decisions on what needs to be done. The long-term goal is to bring together national, regional, state, and tribal indicator efforts to describe the condition of critical environmental areas and human health concerns. Program officials frequently cite output data as measures of success rather than actual program accomplishments in improving environmental conditions in the basin. As a rule, program output data describe activities, such as projects funded, and are of limited value in determining environmental progress. For example, in reporting the accomplishments for Michigan’s Great Lakes Protection Fund, officials noted that the program had funded 125 research projects over an 11-year period and publicized its project results at an annual forum and on a Web site. Similarly, the Lake Ontario Atlantic Salmon Reintroduction Program administered by the U.S. Department of the Interior’s Fish and Wildlife Service listed under its accomplishments the completion of a pilot study and technical assistance provided to a Native American tribe. Of the 50 federal and state programs created specifically to address conditions in the basin, 27 reported accomplishments in terms of outputs, such as reports or studies prepared or presentations made to groups. Because research and capacity building programs largely support other activities, it is particularly difficult to relate reported program accomplishments to outcomes. The federal and state environmental program officials who responded to our evaluation generally provided output data or, as reported for 15 programs, reported that the accomplishments had not been measured for the programs. Only eight of the federal or state Great Lakes specific programs reported outcome information, much of which generally described how effective the programs’ activities or actions had been in improving environmental conditions. For example, EPA’s Region II program for reducing toxic chemical inputs into the Niagara River, which connects Lake Erie to Lake Ontario, reported reductions in priority toxics from 1986 through 2002 from ambient water quality monitoring. Other significant outcomes reported as accomplishments for the Great Lakes included (1) reducing phosphorus loadings by waste treatment plants and limiting phosphorus use in household detergents; (2) prohibiting the release of some toxicants into the Great Lakes, and reducing to an acceptable level the amount of some other toxicants that could be input; (3) effectively reducing the sea lamprey population in several invasive species-infested watersheds; and (4) restocking the fish-depleted populations in some watersheds. To fulfill the need for a monitoring system called for in the GLWQA and to ensure that the limited funds available are optimally spent, we recommended that the Administrator, EPA, in coordination with Canadian officials and as part of an overarching Great Lakes strategy, (1) develop environmental indicators and a monitoring system for the Great Lakes Basin that can be used to measure overall restoration progress and (2) require that these indicators be used to evaluate, prioritize, and make funding decisions on the merits of alternative restoration projects. Mr. Chairman, this completes my prepared statement. I would be happy to answer any questions that you or other members of the Subcommittee may have at this time. For further information, please contact John B. Stephenson at (202) 512-3841. Individuals making key contributions to this testimony were Willie Bailey, Karen Keegan, Rosemary Torres-Lerma, Jonathan McMurray, Margaret Reese, and John Wanska. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The five Great Lakes, which comprise the largest system of freshwater in the world, are threatened on many environmental fronts. To address the extent of progress made in restoring the Great Lakes Basin, which includes the lakes and surrounding area, GAO (1) identified the federal and state environmental programs operating in the basin and the funding devoted to them, (2) evaluated the restoration strategies used and how they are coordinated, and (3) assessed overall environmental progress made in the basin restoration effort. There are 148 federal and 51 state programs funding environmental restoration activities in the Great Lakes Basin. Most of these programs are nationwide or statewide programs that do not specifically focus on the Great Lakes. However, several programs specifically address environmental conditions in the Great Lakes. GAO identified 33 federal Great Lakes specific programs, and states funded 17 additional unique Great Lakes specific programs. Although Great Lakes funding is not routinely tracked for many of these programs, we identified a total of about $3.7 billion in basin-specific projects for fiscal years 1992 through 2001. GAO identified several Great Lakes environmental strategies being used at the binational, federal, and state levels. These strategies are not coordinated or unified in a fashion comparable to other large restoration projects, such as the South Florida ecosystem. Without an overarching plan for these strategies, it is difficult to determine overall progress. The Water Quality Act of 1987 charged EPA's Great Lakes National Program Office with the responsibility for coordinating federal actions for improving the Great Lakes' water quality, however, it has not fully exercised this authority to this point. With available information, it is not possible to comprehensively assess restoration progress in the Great Lakes. Current indicators rely on limited quantitative data and subjective judgments to determine whether conditions are improving, such as whether fish are safe to eat. The ultimate success of an ongoing binational effort to develop a set of overall indicators for the Great Lakes is uncertain because it relies on the resources voluntarily provided by several organizations. Further, no date for completing a final list of indicators has been established. |
The domestic production of oil and gas is essential to the nation’s energy portfolio. According to DOE’s Energy Information Administration (EIA), the United States consumed approximately 6.7 billion barrels of oil and 24 trillion cubic feet of gas in 2010. Together, oil and gas production supply over 60 percent of the nation’s total energy demand, and demand is expected to grow in the future. While domestic drilling for oil and gas can present risks to the environment, it also results in the creation of jobs and economic growth, as well as payments to the government in the form of royalties. Proven, or proved, reserves are defined as oil and gas that geological and engineering data demonstrate with reasonable certainty to be recoverable in future years from known reservoirs under existing economic and operating conditions. While the United States also has extensive offshore oil and gas reserves, the focus of this report is limited to onshore oil and gas production. and gas except in instances when supporting documentation or information we use from other sources makes this distinction necessary; in such cases, we provide the relevant definition in a footnote. Oil and gas is found within underground layers of rock referred to as formations. The geologic characteristics of the formations in which the various hydrocarbons are found vary widely, along with the characteristics of the hydrocarbons themselves. For example, shale oil and gas formations are generally tighter and much less permeable than other formations, causing the oil and gas to be much less free flowing. Coalbed methane formations, located at shallow depths of 1,000 to 2,000 feet, are more permeable formations through which gas can flow more freely than through shale formations. In addition, heavy oil, due to its higher viscosity, has much less ability to flow freely through a formation compared to lighter oil. various gas hydrocarbons. Viscosity is a measure of the resistance of a fluid to flow. A significant amount of water is produced daily as a byproduct from onshore drilling of oil and gas, but the volume produced by a given well will vary depending on the type of hydrocarbon being produced, the geographic location of the well, and the method of production used. Overall, most produced water is of poor quality and cannot be used for other purposes without prior treatment; however, produced water quality can also vary greatly depending on the hydrocarbon, geography, and production method. An estimated 56 million barrels of produced water are generated every day as a byproduct of onshore oil and gas production in the United States. This estimate is based on an Argonne National Laboratory study of produced water volumes generated during 2007—the most recent year for which such data were collected—and was derived from data collected from state agencies in 31 oil- and gas-producing states. The study is considered by agency officials, researchers, and other experts with whom we spoke to be the most comprehensive and accurate assessment of produced water volumes to date. However, because the Argonne study is based on limited and, in some cases, incomplete data, it likely underestimates the current total volume of produced water being generated by oil and gas operations today. Specifically, we noted the following limitations in the Argonne estimate: The reporting time frame for the study largely occurred prior to the recent, dramatic increase in shale gas production in the United States, which had an average annual growth rate of 48 percent from 2006 to 2010, according to the EIA. In 2010, EIA had estimated that shale gas accounted for approximately 23 percent of all gas production in the United States. The study is based on state data that were collected and maintained using a variety of methods, making them difficult to compare and aggregate at a national level. For example, in some states, producers are required to report produced water volumes as measured by flow meters, whereas in other states, producers are required to estimate produced water volumes using a method of their choosing, which may limit the precision of the data. Furthermore, in other states, producers are only required to report produced water volumes they dispose of in injection wells, which may not reflect the total produced water generated. Lastly, some states do not collect data on produced water volumes at all. In such cases, Argonne generated estimates of produced water volumes based on available information on oil and gas production in the state and made certain assumptions about water volumes based on produced water data from neighboring states. Although the Argonne study clearly demonstrates that a large amount of produced water is generated daily, the volume generated by a specific oil and gas well can vary significantly according to three key factors: the type of hydrocarbon being produced, the geographic location of the well, and the method of production used. First, according to literature we reviewed and stakeholders we spoke with, the type of hydrocarbon influences not only how much water a well generates, but also when the water is produced over the life of the well. This is because the geological formations for different hydrocarbons have different attributes, thus influencing the amount of water that is produced from a particular well. For example, coalbed methane wells produce large volumes of water in the early stages of production, because coal beds are essentially aquifers that contain coal rock and gas bound together from the pressure of the water present in the aquifer. By pumping out water, the resulting drop in pressure allows the gas to detach from the coal and flow to the surface. In contrast, one producer noted that their conventional gas wells produce much less water than their coalbed methane wells because the formations from which conventional gas is drawn contain much less water. Oil wells, on the other hand, typically generate less water during the first years of production, when formation pressure is high enough to allow the oil to flow freely to the surface. As these oil wells age, however, water volumes increase, as oil taken out is displaced by water flowing in from the surrounding formation. One producer that we spoke with noted that their older oil fields produce more than five times the volume of water produced by their younger oil fields. Second, according to the literature we reviewed and stakeholders we spoke with, the geographic location of a well also influences the volume of produced water it generates, due to differences in geology. For example, stakeholders noted that the Barnett Shale formation in Texas is generally known to be a “wetter” formation than the Marcellus Shale formation in the Northeast, with shale gas wells in the Barnett typically producing three to four times more water than shale gas wells in the Marcellus. Similarly, USGS reported that coalbed methane wells in the Powder River Basin in Wyoming and Montana produce, on average, 16 times more water than coalbed methane wells in the San Juan Basin in Colorado and New Mexico. In addition, produced water volumes can vary among wells in close proximity with one another. For example, at one site we visited in Wyoming, some gas wells were producing two to three times more water than other gas wells in the same field, for reasons that were, in general, not clear to the operator of those wells. Lastly, the method of production used to extract oil and gas also influences the volume of water generated, according to the literature we reviewed and stakeholders we spoke with. Specifically, stakeholders reported that methods of production that rely on the injection of water and other fluids into the formation in order to stimulate oil and gas production can generate more produced water than in cases in which oil and gas comes to the surface under existing pressure. For example, one stakeholder reported that the use of enhanced oil recovery methods such as steam injection can generate eight to nine barrels of water for every barrel of oil produced. At one enhanced oil recovery field we visited, produced water comprised more than 95 percent of the total liquids produced, with oil comprising the remainder. Similarly, the use of hydraulic fracturing to produce oil or gas can result in larger volumes of produced water than production in more porous formations, although the larger volumes associated with hydraulic fracturing are limited to the initial flowback of water and fracturing fluids. For example, with shale gas production, stakeholders reported that flowback volumes can range from approximately 10,000 to 60,000 barrels per well for each hydraulic fracture. However, once the initial flowback ceases, the volume of water produced by shale gas production may be relatively small, sometimes decreasing to just a few barrels per day. The quality of produced water from oil and gas production is generally poor, and in most situations, it cannot be readily used for other purposes without prior treatment. According to the literature we reviewed and stakeholders we spoke with, produced water may contain a wide range of contaminants in varying amounts. Most of the contaminants occur naturally in the produced water, but some are added through the process of drilling, hydraulic fracturing, and pumping oil and gas. The range of contaminants found in produced water can include, but is not limited to salts, which include chlorides, bromides, and sulfides of calcium, magnesium, and sodium; metals, which include barium, manganese, iron, and strontium, among oil, grease, and dissolved organics, which include benzene and toluene, among others; naturally occurring radioactive materials; and production chemicals, which may include friction reducers to help with water flow, biocides to prevent growth of microorganisms, and additives to prevent corrosion, among others. Exposure to these contaminants at high levels may pose risks to human health and the environment. For example, according to EPA, a potential human health risk from exposure to high levels of barium is increased blood pressure, and potential human health risks from exposure to high levels of benzene are anemia and increased risk of cancer. From an environmental standpoint, research indicated that elevated levels of salts can inhibit crop growth by hindering a plant’s ability to absorb water from the soil. Additionally, exposure to elevated levels of metals and production chemicals, such as biocides, can contribute to increased mortality among livestock and wildlife. The specific quality of water generated by a given well, however, can vary widely according to the same three factors that impact the volume of water produced from the well: the hydrocarbon being produced, the geographic location of the well, and method of production used. First, according to stakeholders we spoke with, the type of hydrocarbon is a key driver of produced water quality, due to differences in geology across the formations in which the hydrocarbons are found. Specifically, the depth at which the hydrocarbons are found influences the salt and mineral content of produced water, and, in general, the deeper the formation is, the higher the salt and mineral content will be. For example, produced water from shale gas wells drilled at depths generally ranging from 5,000 to 8,000 feet have salt and mineral levels 20 times higher than produced water from coalbed methane wells drilled at depths of 1,000 to 2,000 feet. Additionally, the amount of oil or gas that is mixed in with the produced water brought to the surface can also vary. For example, produced water typically blends more easily with oil than with gas. As a result, produced water from oil wells generally contains levels of oil, grease, and other organic compounds that are four to five times higher than water from gas wells. Second, the quality of produced water also varies depending on the well’s geographic location, also because of differences in geology. For example, producers we spoke with said that produced water from wells in the Marcellus Shale formation in the Northeast has higher levels of radionuclides than water from shale gas wells in the Barnett Shale formation in Texas. Similarly, according to research, produced water from coalbed methane wells in the Raton Basin in Colorado and New Mexico has a salt content, on average, roughly two and a half times higher than produced water from the Powder River Basin in Wyoming and Montana. In addition, produced water quality can vary within a given region, according to producers we spoke with. For example, some coalbed methane wells in the Powder River Basin in Wyoming contain barium levels five to six times higher than the barium levels found in wells less than 50 miles away. Additionally, produced water from wells in one oil field in California contains levels of boron four to five times higher than produced water from oil wells in neighboring fields. Lastly, the method of production can affect the quality of the water produced. These differences are largely attributable to the chemicals and other substances added during drilling or production processes, according to stakeholders we spoke with. Specifically, methods of production that rely on hydraulic fracturing or enhanced recovery methods can result in poorer quality produced water than other methods. For example, according to stakeholders, the range of chemicals, sand, and water that are added to facilitate the hydraulic fracturing process can lower the overall quality of the produced water from these kinds of operations. Similarly, the use of chemicals during enhanced recovery can also affect the quality of water produced. Stakeholders noted that enhanced recovery involves the addition of production chemicals such as biocides, corrosion inhibitors, and friction reducers, along with steam or carbon dioxide. For example, one stakeholder estimated that wells produced using these enhanced recovery methods can yield produced water with levels of some production chemicals three to four times higher than produced water from wells that do not use enhanced recovery techniques. Oil and gas producers have a number of options on how to manage produced water, but underground injection is the predominant practice. In addition to underground injection, a limited amount of produced water is managed by discharging it to surface water, storing it in surface impoundments, and reusing it for irrigation or hydraulic fracturing. With regard to treatment options, most produced water is minimally treated, although more advanced treatment methods are available if the end use of the water requires a higher level of treatment. Ultimately, cost is the primary driver in producers’ decisions about how to manage and treat produced water generated by oil and gas producers. Over 90 percent of the water produced during oil and gas operations is managed through underground injection practices; the remaining water is generally discharged to surface water, stored in surface impoundments, reused for irrigation, or reused for hydraulic fracturing. In its 2009 report, Argonne National Laboratory estimated, and EPA officials that we spoke with concurred, that most produced water is managed by injecting it underground into wells that are designated to receive this water. These wells, known as injection wells, must be constructed to protect underground sources of drinking water, and they are tested and monitored periodically to ensure no drinking water is being contaminated by well operations (see fig. 2). Injection wells can be used for enhanced recovery or permanent disposal of the water. When producers reuse produced water for enhanced recovery, they inject it into wells in the same producing formation to recover additional oil and, in limited applications, gas, thus prolonging the life of the production well. Some of this water will come back up as produced water in subsequent well operations. When producers inject produced water for permanent disposal into an underground formation, they inject it into wells in the same formation or a formation that is similar to the one the produced water was extracted from. According to EPA records, in 2010 there were 150,855 injection wells authorized for the injection of fluids brought to the surface during oil and gas production, including produced water, although EPA officials told us that not all are currently operating. About four-fifths of the wells— 124,837—are located in the nine states we reviewed (see table 1). According to state regulators we interviewed in the nine states, underground injection is common in most, but not all, of their states. Specifically, regulators in five states told us that all or almost all of the produced water is managed through underground injection, and in three other states, most of the produced water is managed this way. In the ninth state—Pennsylvania— many producers use underground injection for enhanced recovery, but the practice is not widely used for disposal, according to EPA officials. Produced water that is not injected into underground injection wells is disposed of or reused by producers in other ways, including the following. Discharge to surface water. According to Argonne National Laboratory’s report, less than 1 percent of produced water generated from onshore oil and gas operations in 2007 was managed by discharging it to surface water. Surface discharges of produced water directly from oil and gas production sites are prohibited in much of the United States, but produced water may be discharged from an off-site treatment facility. According to Argonne’s report, discharges of produced water to surface water are primarily limited to the western United States and generally occur only when the salt content of the water is low. While a current national estimate of this practice is not available, EPA officials and regulators in seven of the nine states we reviewed told us that surface discharges of produced water are limited or are nonexistent in their states. For example, officials we spoke with from Colorado said that a very small portion of the produced water generated in that state is discharged to surface water and only 24 of the approximately 9,900 discharge permits they have issued are for oil and gas producers. Among the states we reviewed, Wyoming and Pennsylvania were the only two where producers commonly use surface discharges to manage produced water. For example, an oil producer we spoke with in Wyoming told us his company discharges a small portion of the produced water from some of its fields directly to a nearby creek because the water quality is high enough to meet the state’s discharge limits without prior treatment; however, the majority of the water generated from these fields has a much higher salt concentration and cannot be discharged into surface waters. Surface impoundment. Surface impoundments are lined or unlined ponds used primarily to facilitate evaporation of produced water, and in the case of unlined ponds, allow it to infiltrate into the ground. According to DOE’s NETL, drier climates are favorable for evaporation and spray nozzles may be used to increase the rate of evaporation. The National Research Council reported that, in 2008, surface impoundments were used to manage about 64 percent of produced water generated by coalbed methane producers in the Wyoming portion of the Powder River Basin. According to the report, the water was generally untreated, but in some cases water was treated to meet requirements for discharging to the impoundment. For example, a coalbed methane producer we visited in Wyoming treats the produced water at a treatment facility to first remove barium to meet state water quality standards, then disposes of most of it in a surface impoundment, where it evaporates or infiltrates to the subsurface. Officials we spoke with from California, Colorado, and New Mexico also said that surface impoundments are used to manage produced water in those states, but it is not a significant practice in any of them. Irrigation. Some produced water from coalbed methane is reused for irrigation in certain parts of Wyoming and Colorado because the water is generally of high enough quality that it does not require extensive treatment in order to avoid damage to the crops or soil. According to the National Research Council, about 13 percent of the produced water generated from coalbed methane producers in Wyoming’s Powder River Basin was reused for managed irrigation or subsurface drip irrigation. For example, a coalbed methane production operation we visited in Wyoming disposes of almost all of its produced water from the Powder River Basin using a managed irrigation system, following minimal treatment of the water. Because its wells produce more water than can be disposed of under its surface discharge permit, it is a fairly economical option, and is allowable under state regulations. While there are examples of irrigation with produced water occurring elsewhere, it is not a widely used management practice. According to NETL, a significant challenge to using produced water for irrigation is the salt content of the water, which can decrease crop yields and damage the soil. In addition, the National Research Council reported that while reusing coalbed methane produced water for beneficial purposes such as irrigation would seem to be a desirable and relatively easy objective, in reality it is potentially economically and environmentally burdensome, complex, and challenging. The suitability of water for irrigation depends on a number of factors including the type of crops grown, the soil type, irrigation methods, and the types and quantity of salts dissolved in the water. In addition, the reliability of the produced water supply over time, proximity to the irrigation site, and costs also present challenges. Hydraulic fracturing. In recent years, some shale gas producers have begun reusing produced water for hydraulic fracturing of additional wells at their operations. The water is typically treated first, either on-site or off- site, and then mixed with freshwater if salt concentrations remain high. Although no national estimate of producers’ use of this practice is available, a 2009 report on shale gas development reported that interest in this type of reuse for produced water was high. However, the report also noted that certain water treatment challenges needed to be overcome to make this type of reuse more widespread. According to NETL, in order for reuse of produced water to become widespread, low- cost treatment technologies must be developed. In the last couple of years, reusing produced water for hydraulic fracturing has become more common among shale gas producers in Pennsylvania, according to state regulators and producers we spoke with in the state. The shift was motivated, in part, by a change in the state’s surface discharge standards that ultimately made treatment and discharge a comparatively more expensive practice. For example, one shale gas production site we visited in Pennsylvania currently reuses all of its produced water for hydraulic fracturing, although it had used other practices in the past. Other shale gas producers in the state are also adopting this approach, according to agency officials and an academic expert we spoke with. In addition, regulators in five of the other states we reviewed told us that producers in these states are reusing produced water for hydraulic fracturing, although generally to a lesser extent than in Pennsylvania. Because most produced water is managed through underground injection wells, it is minimally treated; however, if produced water is going to be reused or disposed of in some other manner, then more advanced treatment methods are available, depending on the level of treatment required. Treatment methods for produced water managed through underground injection. Produced water managed through underground injection generally does not need to be treated because injection wells are designed to confine the produced water to the receiving formation and prevent it from migrating to underground sources of drinking water. In some cases, however, to meet an injection well’s operating requirements or prevent premature “plugging” of the formation, the water may be treated to control excessive solids, dissolved oil, corrosion, chemical reactions, or the growth of bacteria and other microbes, according to NETL. Such treatment is generally minimal and can include storing the water in a tank to allow solids to settle out and passing the water through a screen or filter to remove additional solids. Chemicals may also be added to prevent corrosion of the injection well equipment and filtration or biocides may be used to prevent bacteria, algae, or fungi present in the water from clogging equipment or encouraging corrosion. Treatment methods for produced water reused for hydraulic fracturing. Producers who reuse produced water for hydraulic fracturing told us they treat the water to meet their own operating requirements. While producers we spoke with said that they had previously treated the water to a very high quality before reusing it for hydraulic fracturing, they are currently experimenting with lower levels of treatment. For example, one producer told us that 2 years ago his company treated the water so that it was nearly as clean as freshwater, but based on internal research, the company no longer removes salt from the produced water that it reuses for hydraulic fracturing. This lower level of treatment has reduced operating costs, and the producer is considering eliminating other treatment steps as long as doing so will not cause operational problems, such as equipment corrosion. Treatment methods for produced water discharged to surface water bodies or reused for irrigation. If produced water is going to be discharged to surface water or reused for irrigation, then treatment is often necessary to reduce hardness, salts, and other contaminants, in addition to settling and filtration methods to remove solids. Solids and hardness removal are sometimes referred to as “pretreatment” steps. Hard water contains dissolved constituents, mainly calcium and magnesium ions, which can cause scaling of pipes and equipment. Hardness is typically removed prior to removing salts by adjusting the pH of the water and adding chemicals that cause dissolved calcium and magnesium to form small solids, or precipitates, which then settle out or are filtered out of the water with the aid of additional processes. Alternatively, when produced water is going to be reused for irrigation, calcium or magnesium may be added to the water to address sodium levels. Treatment technologies, including distillation, reverse osmosis, and ion exchange, are then used to remove salt and other contaminants from produced water. Distillation is a treatment process that essentially boils produced water to evaporate and then condense the clean water, leaving behind concentrated brine. Reverse osmosis is a filtration process that forces water through a semi- permeable membrane, allowing water to pass through but trapping salt on the other side. Reverse osmosis generally requires a high level of pretreatment to prevent fouling of the membranes, and it is only feasible when salt concentrations in the produced water are less than approximately 25,000 parts per million, according to a study by the Colorado School of Mines. For example, produced water from a gas operation we visited in Wyoming had to first undergo pretreatment to remove solids, hardness, and other contaminants before being put through three stages of reverse osmosis. A third treatment technology, ion exchange, selectively captures sodium ions from produced water and replaces them with others. The water is passed through a large bed of resin beads and sodium ions are adsorbed to (i.e., concentrate on the surface of) the resin. Similar to reverse osmosis, ion exchange faces upper limits on salt concentrations of approximately 7,000 parts per million, according to the Colorado School of Mines.technologies to remove salt typically generates concentrated brine, which must then be properly disposed of as well. While a variety of factors influence how produced water is managed and the level to which it is treated, cost is the primary factor that oil and gas producers consider when making these decisions. According to producers and agency officials we spoke with, how produced water is managed and treated is primarily an economic decision, made within the bounds of federal and state regulations. In most cases, underground injection is the lowest-cost option and producers we spoke with said that their costs for underground injection range from $0.07 to $1.60 per barrel of produced water. However, if a producer is not operating in close proximity to injection wells, transporting the water via truck or pipeline can significantly increase these costs. Furthermore, producers told us that trucking is one of the most significant cost factors they face, and they seek to minimize this cost by managing the water closer to the production site when possible. For example, according to one producer, trucking costs in Texas range from $0.50 to $1.00 per barrel because injection wells in the area are plentiful, whereas costs in Pennsylvania range from $4.00 to $8.00 per barrel because injection wells are scarce, and the produced water often must be transported out of state. As a result, once trucking is factored in, underground injection may in fact become more costly than other management practices. Another significant component of cost is whether treatment will be required as part of the management practice being employed. Treatment costs depend heavily on the technologies used, which in turn depend on the quality of the produced water being treated and the level of treatment needed for the disposal or reuse option being considered. For example, there are a number of treatment methods to remove salt from produced water, but each option has a different cost and differing level of effectiveness. Distillation, while effective at removing salts, has significantly higher costs, and representatives from water treatment facilities and producers we spoke to said it can cost from $6.35 to $8.50 per barrel, on average. Reverse osmosis and ion exchange are less costly treatment options for removing salt, but their use is limited by the level of salt content they can remove from produced water, and reverse osmosis can require extensive pretreatment, which can significantly drive up costs. Producers we spoke with who use reverse osmosis and ion exchange to treat produced water told us that their costs range from $0.20 to $0.60 per barrel. In some cases, producers may be able to change the management practice they use to minimize their treatment costs. For example, state regulators told us that when more stringent discharge limits were put into place in Pennsylvania, many shale gas producers in the state stopped discharging produced water to the surface and started to reuse it for hydraulic fracturing because the latter requires a simpler, and less expensive, level of treatment. In addition to cost, according to our review of the literature and stakeholders we spoke with, produced water management decisions are also influenced by a number of other factors including the following: Poor water quality is a key reason most produced water is managed through underground injection, rather than reused or discharged to the surface. However, when water quality is relatively good, as some of it is in the Powder River Basin of Wyoming, management practices such as irrigation and infiltration from surface impoundments may become viable options. Nonetheless, adequate quantities of produced water are needed for irrigation to be a sustainable practice, and the water must be in close proximity to the land it will be used on or producers can face high transportation costs. Proximity and region-specific factors, such as geology, also influence which management practices are feasible in a given area. For example, some oil and gas producers are not located in close proximity to injection wells, or the number of available wells is limited by the underlying geology of the area, and therefore producers must manage their produced water some other way. An oil producer we spoke to told us his company would prefer to manage all of its produced water through underground injection for enhanced recovery and disposal. However, opportunities for enhanced recovery at one of the producer’s sites are limited by its level of oil production and disposal is constrained by the geology in the area. As a result, the producer told us he manages about 20 percent of the produced water from this site through treatment and discharge, which is significantly more costly and technically challenging than underground injection. Climate is also a factor in the decision-making process. Arid climates are favorable for managing produced water using surface impoundments for evaporation, and limited water supplies in certain regions can motivate producers to make the water available for other purposes, such as irrigation. Regulatory requirements at the federal or state level can also influence producers’ management decisions. As discussed earlier, the changes in discharge limits in Pennsylvania led to a change in management practices by shale gas producers in the state. Producers’ risk management policies can also influence how they manage the water. For example, regulators we spoke with from California told us that liabilities associated with surface discharges and impoundments are a driving factor in moving away from those practices and toward underground injection. According to these regulators, surface impoundments were commonly used in California to manage produced water in the past, but in the last few years hundreds of them have closed down and they are no longer widely used. The management of produced water is regulated by EPA and the states we reviewed through a variety of means, depending on how the water is disposed of or reused. EPA regulates the management of produced water that is injected underground under the Safe Drinking Water Act, while it regulates the management of produced water that is discharged into surface waters under the Clean Water Act. Other management practices, such as disposal of the water into surface impoundments, irrigation, and reuse of the water for hydraulic fracturing, are primarily regulated by the state authorities. The management of produced water through underground injection is subject to the Safe Drinking Water Act’s Underground Injection Control program. This program is designed to prevent contamination of aquifers that supply, or could supply in the future, public water systems by ensuring the safe operation of injection wells. Under this program, EPA or authorized states generally require producers to obtain permits for their injection wells by, among other things, meeting technical standards for constructing, operating, and testing and monitoring wells. Of the nine states we reviewed, all but Pennsylvania have received approval authority from EPA to implement this program for class II wells, including issuing permits and conducting oversight. In most of these states, the agency that oversees oil and gas activities is responsible for implementing this program. Regardless of whether EPA or the state has authority for implementing the program, EPA regional offices periodically review each state’s program and require states to submit an annual report on program activity, according to EPA officials from the regions we spoke with. As part of the Underground Injection Control program, producers generally must apply for a permit to drill an injection well and supply information, including the location and depth of the proposed well. Furthermore, once EPA or the state has issued an Underground Injection Control permit, producers must observe and record the injection pressure, flow rate, and cumulative volume each month and report this information to the permitting agency annually. In addition, the injection well permit also requires producers to conduct mechanical integrity tests on the wells at least once every 5 years, although EPA and some states require testing to be performed more often, according to officials we spoke with. Officials at many of the state agencies we spoke with said that they observe these tests in person to ensure that the well is mechanically sound. According to officials in each of the eight states we contacted, the state can levy penalties for noncompliance for violations ranging from a failure to submit a report to exceeding the pressure permitted in the well. Enforcement response to noncompliance can range from a warning letter to a fine. EPA can commence a separate action for penalties if it believes that a state’s imposition of penalties is insufficient, although EPA officials we spoke with stated that this is rare. The management of produced water through discharge into surface waters is regulated under the Clean Water Act’s National Pollutant Discharge Elimination System. Under this program, all facilities that discharge pollutants to surface waters must obtain a permit from EPA or the designated state agency, which is generally the agency responsible for environmental protection or quality. Permits can be tailored to individual facilities or cover multiple facilities within a specific geographic region. To obtain a permit, producers must complete an application that, among other things, describes the waste that will be discharged, where the discharge will take place, and the method of treatment or containment, if applicable. Once the state or EPA has issued a permit, producers must report any discharges, including the volume of effluent and the amount of each pollutant specified in the permit, to the permitting authority at least once per year. EPA has issued regulations establishing Effluent Limitations Guidelines for some onshore oil and gas extraction including shale gas, but these regulations do not apply to coalbed methane extraction. Of the nine states we reviewed, all but New Mexico have received approval authority from EPA to implement this program for industrial and municipal facilities. EPA requires states with approval authority to submit annual reports on program activity in their state and conduct program reviews every 2 to 5 years. Of these eight states, four—California, Colorado, Pennsylvania, and Wyoming—have issued permits for the discharge of produced water. This is in part because discharging produced water directly from a production site is generally prohibited by regulations implementing the Clean Water Act for locations east of the 98th meridian, which, in the United States, runs from near the eastern border of North Dakota through the eastern portion of Texas, passing near the Dallas-Ft. Worth area. Discharge of produced water from an off- site treatment plant, however, is allowed under the Clean Water Act provided the treated water meets applicable water quality standards, and some states have permitted this activity. For example, a commercial water treatment facility we visited in Pennsylvania treats produced water from shale gas to meet the state’s new, more stringent discharge limits and then releases it to a municipal sewer system. In addition, one producer we spoke with in Wyoming told us his company takes some of its produced water to a treatment facility, where it is treated with reverse osmosis and then discharged to a ravine that flows into the Powder River. The eight states that have approval authority to administer the discharge program may levy penalties if they find producers are not complying with their permit, or if they are discharging without a permit, according to officials we spoke with. As with the Underground Injection Control program, EPA may commence a separate action for penalties if it believes a state’s penalty determination to be inadequate, but EPA officials we spoke with stated that this is rare. The management of produced water through disposal into a surface impoundment or reuse for irrigation is regulated at the state level in the four states we reviewed where producers employ these practices. For example, the Oil and Gas Conservation Commissions in Colorado and Wyoming are among the state regulatory agencies that allow for disposal in surface impoundments; however, these states have set different standards for the quality of the water that may be placed in the ponds. For example, Colorado generally does not require these ponds to be lined, while Wyoming requires any pond with a total dissolved solids level of 10,000 parts per million or more to be lined. In addition, Wyoming also allows produced water to be used for irrigation or as water for livestock with approval from the Wyoming Department of Environmental Quality. Some of the states we reviewed also regulate other practices to reuse produced water. For example, regulations in Colorado, Pennsylvania, and Wyoming allow for the application of produced water to roads in certain circumstances. Specifically, Colorado regulations allow produced water to be spread on roads as long as it meets certain requirements and is authorized by the owner of the road. In addition, reuse of produced water for hydraulic fracturing is regulated at the state level for some states. Specifically, some states have regulations that apply to the temporary storage of hydraulic fracturing fluids, including flowback water, on drilling sites. For example, Oklahoma has recently adopted standards for the construction, operation, location, and maintenance of noncommercial ponds used for temporary storage of flowback water. In addition, some states have begun to require producers to disclose the chemical composition of their hydraulic fracturing fluids. Of the nine states we reviewed, four states—Louisiana, Pennsylvania, Texas, and Wyoming— currently require this disclosure. Over 100 federal research studies conducted during the last 10 years have addressed various aspects of using, managing, and treating produced water. Many federal research projects have focused on describing the characteristics of produced water, such as the volume of water produced from oil and gas activities and the quality of that water. Other research efforts have focused on describing strategies producers could use to manage produced water and the regulatory context for doing so. Federal research also has focused on developing and describing new and existing technologies for treating produced water. Appendix II of this report includes a compilation of the studies we identified. Several federal agencies, including USGS, the Bureau of Reclamation, a number of DOE national laboratories, and EPA, have issued or sponsored studies describing the characteristics of produced water from oil and gas operations––particularly the volume and quality of produced water. For example, in 2009, USGS published a fact sheet that, among other things, described water disposal issues associated with gas production in the Marcellus Shale. More recently, in 2010, USGS published an article describing the quality of produced water from coalbed methane production, and how untreated or partially treated produced water from these operations may threaten fish and aquatic resources.USGS also maintains a database that provides the location, geologic setting, and chemical composition of produced water samples from locations throughout the United States. The Department of the Interior’s Bureau of Reclamation participated in a study published in 2008 that describes the quantity of produced water generated and specific contaminants it contains from oil and gas production in the western United States. The study was designed to assist producers and others in determining viability of this water for beneficial reuse and for selecting appropriate treatment processes. John A. Veil, Markus G. Puder, Deborah Elcock, and Robert J. Redweik, Jr., Argonne National Laboratory, A White Paper Describing Produced Water from Production of Crude Oil, Natural Gas, and Coal Bed Methane (January 2004). two other projects. The first is designed to differentiate produced water from surface water or shallow groundwater in the Marcellus shale area. The other is a multiagency effort with industry, EPA, and USGS, among others, to establish baseline water quality data at a Marcellus drilling site that will be monitored 1 year prior to development and compared to data acquired for 1 year after production begins. Finally, EPA’s Office of Research and Development initiated a study in January 2010 to examine the potential impacts of hydraulic fracturing on drinking water resources and the quality of flowback and produced water. The study plan is currently being reviewed by EPA’s Science Advisory Board, and the agency anticipates issuing an interim report on the potential impacts of hydraulic fracturing on drinking water resources in 2012 and a final report in 2014. Federal research has also focused on providing information about options producers can use to manage their produced water and the regulations they must follow in doing so. For example, the Energy Policy Act of 2005 mandated that the Department of the Interior, in consultation with EPA, engage the National Academy of Sciences to conduct a study on the effect of coalbed methane production on surface and groundwater resources in selected northern and western states. The study was issued in 2010. Several DOE laboratories have also conducted studies or partnered with industry, universities, and other labs to provide information about managing produced water and associated regulations. For example, in May 2009, NETL, in cooperation with the Ground Water Protection Council, published a report summarizing produced-water-related regulations enacted by states for the purpose of protecting water resources.partnered with Clemson University and Chevron to study the efficacy of constructing wetlands to provide a low-cost, effective technology for the treatment and potential reuse of produced water. In addition, NETL has partnered with other DOE laboratories to conduct research on management of produced water. For example, from 2001 through 2004, NETL is also currently NETL collaborated with Idaho National Laboratory and others to analyze coalbed methane production on an Indian reservation and to evaluate options for managing the associated produced water in an effort to minimize the environmental impacts of the water. Other DOE national laboratories also have undertaken studies related to the management and regulation of produced water. For example, Argonne National Laboratory published a study in 2002 that described regulatory issues affecting the management of produced water from coalbed methane production. This was followed by a 2004 study that provided information on how produced water is managed and regulated, and the cost of various management practices. More recently, Argonne National Laboratory published a series of studies describing produced water management practices in different energy-producing regions, including the Marcellus formation in the Appalachians and the Fayetteville Shale in Arkansas. In addition, Sandia National Laboratories partnered with a producer to study options for managing produced water from coalbed methane, and published its analysis in 2008. Also, Oak Ridge National Laboratory has developed new approaches for produced water sampling, analysis, and remediation, and Los Alamos National Laboratory is currently conducting research to provide information about how produced water can be used to cultivate algae for biofuel production. Other federal agencies have also contributed to research on the management and regulation of produced water. Specifically, the Bureau of Reclamation partially funded a publication containing the proceedings from an April 2006 workshop on produced water, and the agency presented information about the beneficial use of produced water at the 2007 International Petroleum Environmental Conference. In addition, in 2006, USGS issued a bibliography of studies from across oil- and gas- producing areas that it had compiled from the last 80 years. These studies describe the effects of produced water on soils, water quality, and ecosystems. Federal research efforts, primarily conducted by DOE’s national laboratories, have also focused on new technologies and treatment methods for produced water. Sandia National Laboratories, for example, has partnered with a producer to conduct pilot testing of a new treatment system to lower the salt content of produced water from coalbed methane sources. NETL has partnered with Los Alamos National Laboratory, the New Mexico Institute of Mining and Technology, and the University of Texas on a long-term project to develop and test a prototype for a new treatment system that uses an innovative filtration method to remove problem contaminants and that would facilitate on-site treatment of produced water. From March 2003 through the end of 2005, NETL also partnered with Oak Ridge National Laboratory and industry to develop and test novel liquid solvents to remove organic substances from produced water. In addition, NETL is currently sponsoring a project to develop high-temperature nanofiltration technology to remove contaminants from produced water. According to the agency, the goal of this project is to minimize environmental impacts from coalbed methane and shale gas operations and allow cost-effective reuse of produced water that will reduce freshwater consumption and disposal costs. More recently, NETL sponsored research that led to the development of a new treatment system that, according to the agency, successfully treated flowback water from a hydraulic fracturing site in Pennsylvania. According to NETL, the treatment system significantly reduced the producer’s disposal costs. Other federal research efforts have been designed to improve existing techniques to treat produced water. For example, NETL partnered with Texas A&M, Argonne National Laboratory, and industry to develop improved reverse osmosis membrane filtration technology for the removal of salt from produced water. The desalination technology developed through this project led to the construction of a large-scale mobile unit and the development of a commercial oilfield treatment system at a site in Texas, according to NETL officials. Similarly, NETL has partnered with industry to develop a process that, when combined with existing reverse osmosis treatment, will facilitate the reuse of produced water by lowering energy requirements needed to treat produced water and by reducing membrane fouling. We provided a draft of this report to the Department of Energy, the Department of the Interior, and the Environmental Protection Agency for review and comment. None of these agencies provided written comments to include in our report; however, the Environmental Protection Agency and the Department of the Interior provided technical comments, which we incorporated as appropriate. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies of this report to the appropriate congressional committees, the Secretary of Energy, the Secretary of the Interior, the EPA Administrator, and other interested parties. In addition, the report will be available at no charge on the GAO website at http://www.gao.gov. If you or your staff members have any questions about this report, please contact us at (202) 512-3841 or mittala@gao.gov or ruscof@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix III. Our objectives for this review were to describe (1) what is known about the volume and quality of produced water from oil and gas production; (2) what practices are generally used to manage and treat produced water, and what factors are considered in the selection of each; (3) how the management of produced water is regulated at the federal level and in selected states; and (4) what federal research and development efforts have been undertaken during the last 10 years related to produced water. To address each of these objectives, we conducted a literature review of studies and other documents on produced water quality and volume, management, and regulation issued by federal agencies and laboratories, state agencies, the oil and gas industry, and academic institutions. These documents included peer-reviewed scientific and industry periodicals, government-sponsored research, and reports from nongovernmental research organizations. We identified this literature through a systematic search of databases such as ProQuest, EconLit, and BioDigest, and used an iterative process to identify the most relevant studies for our review. We believe we have included the key studies and have qualified our findings, where appropriate. However, we may not have identified all of the studies with findings relevant to our objectives. In addition, we reviewed studies that fit the following criteria for selection: (1) the research was of sufficient breadth and depth to provide observations or conclusions directly related to our objectives; (2) the research was targeted specifically toward the volume and quality of produced water, available management practices and treatment methods, regulation of produced water broadly and in selected states, and undertaken by the federal government; and (3) the research was typically published in the last 10 years. We examined key assumptions, methods, and relevant findings of major scientific articles primarily related to water volumes and quality, and treatment methods. Where applicable, we assessed the reliability of the data we obtained and found them to be sufficiently reliable for our purposes. In addition, we interviewed federal and state regulatory officials; federal scientists from the Environmental Protection Agency’s (EPA) Office of Research and Development and the Department of Energy’s (DOE) Argonne National Laboratory, Los Alamos National Laboratory, National Energy Technology Laboratory, Oak Ridge National Laboratory, and Sandia National Laboratories; officials from oil and gas exploration and production companies; officials from water treatment facilities; and other experts with experience related to produced water. The federal and state regulatory officials included those with responsibility over oil and gas regulation, as well as clean water and drinking water regulation. We focused our review of management techniques and produced water regulation on nine states—California, Colorado, Kansas, Louisiana, New Mexico, Oklahoma, Pennsylvania, Texas, and Wyoming. We selected eight of these states because the volume of produced water generated within their borders accounts for nearly 90 percent of the produced water generated in the United States as of 2007, the most recent year for which there were available data. In addition, we selected Pennsylvania because of the recent growth in shale gas development in the Marcellus shale formation and the expected potential for large-scale produced water management approaches in this area. While oil shale production has expanded and continues to expand in Texas, North Dakota, and other states, we did not look specifically at produced water from oil shale as part of this review. Furthermore, GAO will be conducting future work on the development of shale gas resources and the use of hydraulic fracturing for oil and gas development and will address these topics more fully in subsequent reports. We supplemented our literature review and stakeholder discussions with site visits to selected locations in Pennsylvania, Texas, and Wyoming, where we met with oil and gas producers and officials from produced water treatment facilities and discussed issues related to produced water management and treatment and the factors that influence these decisions. We selected these states because of the current and potential volumes of produced water generated, the range of hydrocarbons produced, and the different management and treatment practices employed. We also visited hydraulic fracturing drilling operations, underground injection control sites, and a number of different treatment facilities employing a variety of technologies. To determine what federal research and development efforts have been undertaken during the last 10 years related to produced water, we analyzed information supplied by and conducted interviews with federal officials from DOE and select national laboratories, EPA, and the Department of the Interior’s U.S. Geological Survey, Bureau of Land Management, and Bureau of Reclamation. We conducted this performance audit from October 2010 to January 2012, in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. The following is a list of federally sponsored research efforts undertaken in the last 10 years that we identified in consultation with officials from the Department of Energy and select national laboratories, the Department of the Interior, the Environmental Protection Agency, and the National Research Council. These research efforts include those related to the quantity, quality, management, treatment, and use of produced water. The following studies were published by or prepared for Argonne National Laboratory. Analysis of Data from a Downhole Oil/Water Separator Field Trial in East Texas. February 2001. Clark, C.E and J.A. Veil. Produced Water Volumes and Management Practices in the United States. September 2009. Harto, Christopher. Shale Gas – The Energy Water Nexus. April 2011. Argonne National Laboratory. An Introduction to Slurry Injection Technology for Disposal of Drilling Wastes. September 2003. Puder, Markus G., Bill Bryson, and John A. Veil. Compendium of Regulatory Requirements Governing Underground Injection of Drilling Wastes. February 2003. Puder, M.G. and J. A. Veil. Offsite Commercial Disposal of Oil and Gas Exploration and Production Waste: Availability, Options, and Costs. August 2006. Veil, John A. and John J. Quinn. Downhole Separation Technology Performance: Relationship to Geologic Conditions. November 2004. Veil, John A. and Maurice B. Dusseault. Evaluation of Slurry Injection Technology for Management of Drilling Wastes. May 2003. Veil, J.A. and M.G. Puder. Potential Ground Water and Surface Water Impacts from Oil Shale and Tar Sands Energy-Production Operations. October 2006. Veil, John A. Regulatory Issues Affecting Management of Produced Water from Coal Bed Methane Wells. February 2002. Veil, J., J. Gasper, M. Puder, and P. Leath. Summary of DOE/PERF Water Program Review. January 2006. Veil, John A. Thermal Distillation Technology for Management of Produced Water and Frac Flowback Water. Water Technology Brief #2008-1. May 13, 2008. Veil, J.A. and J.J. Quinn. Water Issues Associated with Heavy Oil Production. November 2008. Veil, John A. “Water Management Practices Used by Fayetteville Shale Gas Producers.” Oil & Natural Gas Technology (June 2011). Veil, John A. “Water Management Technologies Used by Marcellus Shale Gas Producers.” Oil & Natural Gas Technology (July 2010). Veil, John A., Markus G. Puder, Deborah Elcock, and Robert J. Redweik, Jr. A White Paper Describing Produced Water from Production of Crude Oil, Natural Gas, and Coal Bed Methane. January 2004. The following studies were published by or prepared for Los Alamos National Laboratory. Altare, Craig R., Robert S. Bowman, Lynn E. Katz, Kerry A. Kinney, and Enid J. Sullivan. “Regeneration and Long-term Stability of Surfactant Modified Zeolite for Removal of Volatile Organic Compounds from Produced Water.” Microporous and Mesosporous Materials, 105 (2007): 305-316. Kwon, Soondong, Enid J. Sullivan, Lynn E. Katz, Robert S. Bowman, and Kerry A. Kinney. “Laboratory and Field Evaluation of a Pretreatment System for Removing Organics from Produced Water.” Water Environment Research, vol. 83 (2011). Ranck, J. Michael, Robert S. Bowman, Jeffrey L. Weeber, Lynn E. Katz, and Enid J. Sullivan. “BTEX Removal from Produced Water Using Surfactant-Modified Zeolite.” Journal of Environmental Engineering (March 2005). Sullivan, E.J., C.A. Dean, T.M. Yoshida, B. Cordova, M. Rearick, P. Laur, A. Viszolay, L. Brown, and J. Brown. Chemical Quality Impacts of Oil and Gas Produced Water as a Growth Medium for Nannochloropsis Grown at Pilot Scale for Biofuel Production. LA-UR-11-11017. The following studies were published by or prepared for the National Energy Technology Laboratory. Billingsley, R.L. Identifying and Remediating High Water Production Problems in Basin-Centered Formations. December 2005. Brown, Terry, Carol D. Frost, Thomas D. Hayes, Leo A. Heath, Drew W. Johnson, David A. Lopez, Demian Saffer, Michael A. Urynowicz, John Wheaton, and Mark D. Zoback. Final Report: Produced Water Management and Beneficial Use. January 2009. Burnett, David B. and Mustafa Siddiqui. Recovery of Fresh Water Resources from Desalination of Brine Produced During Oil and Gas Production Operations. September 2003–December 2006. COAL BED METHANE PRIMER—New Source of Natural Gas – Environmental Implications: Background and Development in the Rocky Mountain West. February 2004. DOE Oil and Natural Gas Water Resources Program. December 2009. Feasibility Study of Expanded Coal Bed Natural Gas Produced Water Management Alternatives in the Wyoming Portion of the Powder River Basin Phase One. January 2006. A Guide to Practical Management of Produced Water from Onshore Oil and Gas Operations in the United States. October 2006. Handbook on Best Management Practices and Mitigation Strategies for Coal Bed Methane in the Montana Portion of the Powder River Basin. April 2002. Handbook on Coal Bed Methane Produced Water: Management and Beneficial Use Alternatives. July 2003. Modern Shale Gas Development in the United States: A Primer. April 2009. Policy Analysis of Produced Water Issues Associated With In-Situ Thermal Technologies. January 2011. Remson, Don J. Produced Water in the Rocky Mountain Region— Quantity and Quality. November 2005. Review of the U.S. Department of Energy’s Environmental Program. July 1, 2010. Siting, Design, Construction and Reclamation Guidebook for Coalbed Natural Gas Impoundments. May 2006. Ground Water Protection Council, State Oil and Natural Gas Regulations Designed to Protect Water Resources. May 2009. Use of Produced Water in Recirculated Cooling Systems at Power Generating Facilities. September 2006. Wang, Xixi, Bethany A. Kurz, and Marc D. Kurz. Subtask 1.18 – A Decision Tool for Watershed-Based Effluent Trading. February 2007. Welch, Robert A. and Dwight F. Rychel. Produced Water from Oil and Gas Operations in the Onshore Lower 48 States. December 2004. The following are past and current projects funded by the National Energy Technology Laboratory, but for which no studies have been published. Advanced Membrane Filtration Technology for Cost-Effective Recovery of Fresh Water from Oil and Gas Produced Brine. Project start: 2003. Project end: 2006. Anti-Fouling Reverse Osmosis Desalination System. Project start: 2009. Project end: 2010. Barnett and Appalachian Shale Water Management and Reuse Technologies. Project start: 2009. Estimated project end: 2011. Cleaning Agents for Produced Water Membrane Filters. Project start: 2004. Project end: 2006. Coal Bed Methane Best Management Practices Workshop. Project start: 2003. Project end: 2004. Coalbed Methane Research. Project start: 2006. Project end: 2008. Coalbed Natural Gas Produced-Water Treatment Using Gas Hydrate Formation at the Wellhead. Project start: 2005. Project end: 2009. Coalbed Natural Gas Produced-Water Treatment Using Gas Hydrates. Project start: 2006. Project end: 2008. Coalbed Natural Gas Research. Project start: 2003. Project end: 2006. Comprehensive Lifecycle Planning and Management System for Addressing Water Issues Associated with Shale Gas Development in New York, Pennsylvania and West Virginia. Project start: 2009. Estimated project end: 2012. Cost Effective Recovery of Low-TDS Frac Flowback Water for Re-use. Project start: 2009. Project end: 2011. Cost-Effective Treatment of Produced Water Using Co-Produced Energy Sources for Small Producers. Project start: 2008. Original project end: 2010 (extended). Effects of Irrigating with Treated Oil and Gas Product Water on Crop Biomass and Soil Permeability. Project start: 2008. Project end: 2010. Energy in the Environment-Initiatives 2004-09. Project start: 2004. Project end: 2009. Evaluations of Radionuclides of Uranium, Thorium, and Radium Associated with Produced Fluids, Precipitates, and Sludges from Oil, Gas, and Oilfield Brine Injection Wells in Mississippi. Project start: 2002. Project end: 2003. Field Validation of Toxicity Tests to Evaluate the Potential for Beneficial Use of Produced Water. Project start: 2004. Project end: 2008. GIS and Web-Based Water Resource Geospatial Infrastructure for Oil Shale Development. Project start: 2008. Estimated project end: 2012. Handbooks for Preparing, Evaluation Development, Environmental Plans and Background Development Pertinent to Coal Bed Methane Production. Project start: 2002. Project end: 2005. Hypoxia, Program Review, and Total Petroleum Hydrocarbon Workshop. Project start: 2006. Project end: 2008. Identification, Verification, & Compilation of Produced-Water Best Management Practices for Conventional Oil & Gas Production Operations. Project start: 2004. Project end: 2007. Improving Science-Based Methods for Assessing Risks Attributable to Petroleum Residues in Soil Transferred to Vegetation. Project start: 2002. Project end: 2005. Innovative Water Management Technology to Reduce Environmental Impacts of Produced Water. Project start: 2008. Estimated project end: 2012. An Integrative Framework for the Treatment and Management of Produced Water. Project start: 2008. Estimated project end: 2011. An Integrated Water Treatment Technology Solution for Sustainable Water Resource Management in the Marcellus Shale. Project start: 2009. Estimated project end: 2011. Integration of Water Resource Models with Fayetteville Shale Decision and Support Systems. Project start: 2009. Estimated project end: 2012. Life Cycle Assessment, Produced Water, and Waste Management Analyses. Project start: 2004. Project end: 2007. Long-term field Deployment of a Surfactant Modified Zeolite Vapor Phase Bioreactor System. Project start: 2004. Project end: 2007. Long Term Field Development of a Surfactant-Modified Zeolite/Vapor- Phase Bioreactor System for Treatment of Produced Waters for Power Generation. Project start: 2004. Project end: 2007. Management of Produced Water. Project start: 2003. Project end: 2006. Managing Coalbed Natural Gas Produced Water for Beneficial Uses, Initially Using the San Juan and Raton Basins as a Model. Project start: 2003. Project end: 2008. Membrane Technology for Produced Water at Lea County, NM. Project start: 2008. Estimated project end: 2011. Microbial Ecology of Shale Gas Production Waters. Project start: 2011. Estimated project end: not established. Modified Reverse Osmosis System for Treatment of Produced Water. Project start: 2000. Project end: 2004. Modeling of Water-Soluble Organic Content in Produced Water. Project start: 2002. Project end: 2005. NMWAIDS: A Produced-Water Quality and Infrastruture GIS Database for New Mexico Oil Production. Project start: 2002. Project end: 2005. Northeast National Petroleum Reserve-Alaska Reconnaissance-Level Airborne Contaminants Study. Project start: 2001. Project end: 2006. Northern Cheyenne Indian Reservation (NCIR) Coalbed Natural Gas Resource Assessment and Analysis of Produced-Water Disposal Options. Project start: 2001. Project end: 2004. Novel Cleanup Agents for Membrane Filters Used to Treat Oilfield Produced Water for Beneficial Purposes. Project start: 2004. Project end: 2007. Novel Fouling—Reducing Coatings for Ultrafiltration, Nanofiltration and Reverse Osmosis Membranes. Project start: 2004. Project end: 2008. Pilot Testing: Pretreatment Options to Allow Re-Use of Frac Flowback and Produced Brine for Gas Shale Resource Development. Project start: 2009. Estimated project end: 2011. Pretreatment and Water Management for Frac Water Reuse and Salt Production. Project start: 2009. Estimated project end: 2011. Produced Water Management and Beneficial Use. Project start: 2005. Project end: 2007. Produced Water Management and Beneficial Use/15549 Colorado School of Mines. Project start: 2005. Project end: 2007. Produced Water Management and Beneficial Use/15549 Colorado School of Mines. (Different portion of the preceding project with distinct project identification.) Project start: 2005. Project end: 2007. Produced Water Treatment and Decision Tool. Project start: 2008. Estimated project end: 2012. Provide Support to Produced Water: Osage-Skiatook Petroleum Environmental Research Project. Project start: 2001. Project end: 2006. Range Resources Baseline Monitoring Site for Marcellus Shale Gas. Project start: 2011. Estimated project end: not established. Recovery of More Oil-in-Place at lower Production Costs While Creating a Beneficial Water Resource. Project start: 2002 Project end: 2006. Research and Development Concerning Coalbed Natural Gas— Congressional Mandate. Project start: 2006. Project end: 2008. Research to Enhance Oil and Gas Development and Environmental Protection on Federal Lands: Joint Montana Regional Coalbed Natural Gas Ground-Water Monitoring Program. Project start: 2005. Project end: 2008. Risk Based Data Management System (RBDMS) and Cost Effective Regulatory Approaches (CERA) Related to Hydraulic Fracturing and Geologic Sequestration of CO-2. Project start: 2009. Estimated project end: 2012. Subsurface Drip Irrigation. Project start: 2007. Estimated project end: 2014. Sustainable Management of Flowback Water during Hydraulic Fracturing of Marcellus Shale for Natural Gas Production. Project start: 2009. Estimated project end: 2012. Treating Coalbed Natural Gas Produced Water for Beneficial Use by MFI Zeolite Membranes. Project start: 2004. Project end: 2008. Treatment and Beneficial Reuse of Produced Waters Using a Novel Pervaporation-Based Irrigation Technology. (NETL in-house project not yet awarded.) Treatment of Produced Water by FARADAVIC Electrodialysis and Reverse Osmosis. Project start: 2009. Project end: 2010. Treatment of Produced Waters using a Surfactant Modified Zeolite/Vapor- Phase Bioreactor. Project start: 2002. Project end: 2006. Treatment of Produced Waters Using a Surfactant-Modified Zeolite/Vapor-Phase Bioreactor System. Project start: 2003. Project end: 2006. Treatment of Produced Waters Using a Surfactant-Modified Zeolite/Vapor-Phase Bioreactor System. (Next phase.) Project start: 2004. Project end: 2006. Unconventional High Temperature Nanofiltration for Produced Water Treatment. (Phase I.) Project start: 2009. Project end: 2010. Unconventional High Temperature Nanofiltration for Produced Water Treatment. (Next phase.) Project start: 2010. Project end: 2012. Use of Ionic Liquids in Produced-Water Clean-up. Project start: 2003. Project end: 2005. Use of Stable Isotopes to Discern Marcellus Produced Water When Commingled with Surface Water or Shallow Groundwater. Project start: 2011. Estimated project end: not established. Use of Wetland Plant Species and Communities for Phytoremediation of Coalbed Natural Gas Produced Water and Waters of Quality Similar to that Associated with Coalbed Natutral Gas Deposits of the Powder River Basin. Project start: 2001. Project end: 2008. Using Helicopter Electromagnetic Surveys to Determine the Hydrologic Fate of Coalbed Methane Produced Water. Project start: 2002. Project end: 2004. Water Management Strategies for Improved Coalbed Methane Production in the Black Warrior Basin. Project start: 2009. Estimated project end: 2012. Water-Related Issues Affecting Conventional Oil and Gas Recovery and Potential Oil Shale Development in the Uinta Basin, Utah. Project start: 2008. Estimated project end: 2011. Water & Waste Regulatory Analysis. Project start: 2006. Project end: 2008. Zero Discharge Water Management for Horizontal Shale Gas Well Development. Project start: 2009. Estimated project end: 2011. The following studies were published by or prepared for Oak Ridge National Laboratory. Bostick, Debra T., H. Luo and B. Hindmarsh. Characterization of Soluble Organics in Produced Water. January 2002. Klasson, K. Thomas, Costas Tsouris, Sandie A. Jones, Michele D. Dinsmore, David W. Depaoli, Angela B. Walker, Sotira Yiacoumi, Viriya Vithayaveroj, Robert M. Counce, and Sharon M. Robinson. Ozone Treatment of Soluble Organics in Produced Water. Petroleum Environmental Research Forum Project 98-04. January 2002. McFarlane, J. “Application of Chemometrics to Modeling Produced Water Contamination.” Separation Science and Technology, 40 (2005): 593- 609. McFarlane, Joanna. Modeling of Water-Soluble Organic Content in Produced Water. May 2006. McFarlane, Joanna. New Approaches to Produced Water Sampling, Analysis and Remediation at ORNL. 2004. McFarlane, Joanna. Measurement, Characterization and Prediction of Organic Solubility in Produced Water. Presentation at Gas Technology Institute Natural Gas Technologies II Conference and Exhibition, February 8-11, 2004. McFarlane, Joanna. Offshore Versus Onshore Produced Water Characterization and Models. Presentation at Gas Technology Institute Natural Gas Technologies II Conference and Exhibition, February 8-11, 2004. McFarlane, Joanna, Debra T. Bostick, and Huimin Luo. Characterization and Modeling of Produced Water. 2002. Ren, R.X. Room Temperature Ionic Liquids for Separating Organics from Produced Water. Separation Science and Technology, 40 (2005): 1245- 1265. The following study was published by or prepared for Sandia National Laboratories. Cappelle, Malynda, Randy Everett, William Holub, Richard Kottenstette, and Allan Sattler. Coal Bed Natural Gas Produced Water Preliminary Pilot Plant Operation and Results. August 2008. The following studies were published by or prepared for the Bureau of Reclamation. Benko, Katie L. “Ceramic Membranes for Produced Water Treatment.” World Oil (April 2009): 1-3 Benko, Katie L and Jörg E. Drewes. “Produced Water in the Western United States: Geographical Distribution, Occurrence and Composition.” Environmental Engineering Science, vol. 25, no. 2 (2008): 239-246. Benko, Katie and Jörg Drewes, Pei Xu, and Tzahi Cath. “Use of Ceramic Membranes for Produced Water Treatment.” World Oil, Gulf Publishing Company, vol. 230, no. 4 (April 2009). Drewes, Jörg E, Pei Xu, Dean Heil, and Gary Wang. Multibeneficial Use of Produced Water Through High-Pressure Membrane Treatment and Capacitive Deionization Technology. Desalination and Water Purification Research and Development Program Report No. 133. February 2009. Dundorf, Steve and Katie Benko. “Geographical Assessment of Potential for Beneficial Use of Produced Water.” Presentation at International Petroleum Environmental Conference. November 2007. Colorado Waters Resources Research Institute, Colorado State University, Produced Water Workshop, (April 4-5, 2006). The publication of workshop results was partially funded by the Bureau of Reclamation. The following studies were published by or prepared for the U.S. Geological Survey. Engle, Mark A., Carleton R. Bern, Richard W. Healy, James I. Sams, John W. Zupancic, and Karl T. Schroeder. “Tracking solutes and water from subsurface drip irrigation application of coalbed-methane produced waters, Powder River Basin, Wyoming.” Environmental Geosciences, v. 18, no. 3 (September 2011): 1-19. Farag, Aida M., David D. Harper, Anna Senecal, and Wayne A. Hubert. “Potential Effects of Coalbed Natural Gas Development on Fish and Aquatic Resources.” Chapter 11 in Coalbed Natural Gas: Energy and Environment. Nova Science Publishers, Inc., 2010. Healy, Richard W., Cynthia A. Rice, Timothy T. Bartos, and Michael P. McKinley. “Infiltration from an impoundment for coal-bed natural gas, Powder River Basin, Wyoming: Evolution of water and sediment chemistry.” Water Resources Research, Vol. 44, W06424, June 2008. Healy, Richard W., Timothy T. Bartos, Cynthia A. Rice, Michael P. McKinley, and Bruce D. Smith. “Groundwater chemistry near an impoundment for produced water, Powder River Basin, Wyoming USA.” Journal of Hydrology, 403 (2011): 37-48. Kharaka, Y.K., and J.K. Otton, 2003, Environmental Impacts of Petroleum Production: Initial Results from the Osage-Skiatook Petroleum Environmental Research Sites, Osage County, Oklahoma: USGS Water- Resources Investigations Report 03-4260. Orem, William H., Calin A. Tatu, Harry E. Lerch, Cynthia A. Rice, Timothy T. Bartos, Anne L. Bates, Susan Tewalt, and Margo D. Corum. “Organic compounds in produced waters from coalbed natural gas wells in the Powder River Basin, Wyoming, USA.” Applied Geochemistry 22 (May 2007): 2240-2256. Otton, James K. Environmental Aspects of Produced-water Salt Releases in Onshore and Coastal Petroleum-producing Areas of the Conterminous U.S. – A Bibliography. Open-File Report 2006-1154. Peterman, Zell E., Joanna N. Thamke, Kiyoto Futa, and Thomas A. Oliver. Strontium Isotope Detection of Brine Contamination in the East Poplar Oil Field, Montana. Open-File Report 2010-1326. Rice, Cynthia A., Timothy T. Bartos, and Margaret S. Ellis. Chemical and Isotopic Composition of Water in the Fort Union and Wasatch Formations of the Powder River Basin, Wyoming and Montana: Implications for Coalbed Methane Development. Coalbed Methane of North America – II, 2002: 53-70. Rowan, E.L., M.A. Engle, C.S. Kirby, and T.F. Kraemer. Radium Content of Oil- and Gas-Field Produced Waters in the Northern Appalachian Basin: Summary and Discussion of Data. Scientific Investigations Report 2011-5135, 2011. Smith, Bruce D., Joanna N. Thamke, Michael J. Cain, Christa Tyrrell, and Patrica L. Hill. Helicopter Electromagnetic and Magnetic Survey Maps and Data, East Poplar Oil Field Area, Fort Peck Indian Reservation, Northeastern Montana, August 2004. Open-File Report 2006-1216, Version 1.0. Soeder, Daniel J. and William H. Kappel. Water Resources and Natural Gas Production from the Marcellus Shale. USGS Fact Sheet 2009-3032. May 2009. Environmental Protection Agency, Draft Hydraulic Fracturing Study Plan (April 28, 2011). Committee on Management and Effects of Coalbed Methane Development and Produced Water in the Western United States, Committee on Earth Resources, Board on Earth Sciences and Resources, Water Science and Technology Board, Division on Earth and Life Studies, National Research Council of the National Academies. Management and Effects of Coalbed Methane Produced Water in the Western United States. 2010. In addition to the contacts named above, Elizabeth Erdmann, Assistant Director; Colleen Candrl; Nancy Crothers; Randy Jones; Annamarie Lopata; Alison O’Neill; Stuart Ryba; Rebecca Sandulli; Rebecca Shea; Lindsay Taylor; and Barbara Timmerman made significant contributions to this report. | Water is a significant byproduct associated with oil and gas exploration and production. This water, known as produced water, may contain a variety of contaminants. If produced water is not appropriately managed or treated, these contaminants may present a human health and environmental risk. GAO was asked to describe (1) what is known about the volume and quality of produced water from oil and gas production; (2) what practices are generally used to manage and treat produced water, and what factors are considered in the selection of each; (3) how produced water management is regulated at the federal level and in selected states; and (4) what federal research and development efforts have been undertaken during the last 10 years related to produced water. To address these objectives, GAO reviewed studies and other documents on produced water and interviewed federal and state regulatory officials, federal scientists, officials from oil and gas companies and water treatment companies, and other experts. GAO focused its review on the nine states that generate nearly 90 percent of the produced water, and conducted site visits in three states. A significant amount of water is produced daily as a byproduct from drilling of oil and gas. A 2009 Argonne National Laboratory study estimated that 56 million barrels of water are produced onshore every day, but this study may underestimate the current total volume because it is based on limited, and in some cases, incomplete data generated by the states. In general, the volume of produced water generated by a given well varies widely according to three key factors: the hydrocarbon being produced, the geographic location of the well, and the method of production used. For example, some gas wells typically generate large volumes of water early in production, whereas oil wells typically generate less. Generally, the quality of produced water from oil and gas production is poor, and it cannot be readily used for another purpose without prior treatment. The specific quality of water produced by a given well, however, can vary widely according to the same three factors that impact volumehydrocarbon, geography, and production method. Oil and gas producers can choose from a number of practices to manage and treat produced water, but underground injection is the predominant practice because it requires little or no treatment and is often the least costly option. According to federal estimates, more than 90 percent of produced water is managed by injecting it into wells that are designated to receive produced water. A limited amount of produced water is disposed of or reused by producers in other ways, including discharging it to surface water, storing it in surface impoundments or ponds so that it can evaporate, irrigating crops, and reusing it for hydraulic fracturing. Managing produced water in these ways can require more advanced treatment methods, such as distillation. How produced water is ultimately managed and treated is primarily an economic decision, made within the bounds of federal and state regulations. The management of produced water through underground injection is subject to the Safe Drinking Water Acts Underground Injection Control program, which is designed to prevent contamination of aquifers that supply public water systems by ensuring the safe operation of injection wells. Under this program, the Environmental Protection Agency (EPA) or the states require producers to obtain permits for their injection wells by, among other things, meeting technical standards for constructing, operating, and testing and monitoring the wells. EPA also regulates the management of produced water through surface discharges under the Clean Water Act. Other management practices, such as disposal of the water into surface impoundments, irrigation, and the reuse of the water for hydraulic fracturing, are regulated by state authorities. Several federal agencies, including EPA; the Department of Interiors Bureau of Reclamation and U.S. Geological Survey; and a number of Department of Energy national laboratories, have undertaken research and development efforts related to produced water. These efforts have included sponsoring and issuing studies that describe the volume and quality of produced water, options for managing produced water and associated regulatory issues, as well as options for improving existing technologies for treating produced water and developing new technologies, such as more cost-effective filters. GAO is not making any recommendations. A draft was provided to the Departments of Energy and the Interior, and EPA for review. None of these agencies provided written comments. EPA and Interior provided technical comments, which we incorporated as appropriate. |
Our May 2009 report on the then-current round of ASP testing found that DHS increased the rigor of ASP testing over that of previous tests, and that a particular area of improvement was in the performance testing at the Nevada Test Site, where DNDO compared the capability of ASP and current-generation equipment to detect and identify nuclear and radiological materials. For example, unlike in prior tests, the plan for the 2008 performance test stipulated that the contractors who developed the equipment would not be involved in test execution. This improvement addressed concerns we previously raised about the potential for bias and provided increased credibility to the results. Nevertheless, based on the following factors, in our report we questioned whether the benefits of the new portal monitors justify the high cost: The DHS criteria for a significant increase in operational effectiveness. Our chief concern with the criteria is that they require only a marginal improvement over current-generation portal monitors in the detection of certain weapons-usable nuclear materials during primary screening. DNDO considers detection of such materials to be a key limitation of current-generation portal monitors. The marginal improvement required of ASPs to meet the DHS criteria is problematic because the detection threshold for the current-generation portal monitors does not specify a level of radiation shielding that smugglers could realistically use. Officials from the Department of Energy (DOE), which designed the threat guidance DHS used to set the detection threshold, and national laboratory officials told us that the current threshold is based not on an analysis of the capabilities of potential smugglers to take effective shielding measures but rather on the limited sensitivity of PVTs to detect anything more than certain lightly shielded nuclear materials. DNDO officials acknowledge that both the new and current-generation portal monitors are capable of detecting certain nuclear materials only when unshielded or lightly shielded. The marginal improvement in detection of such materials required of ASPs is particularly notable given that DNDO has not completed efforts to fine-tune PVTs’ software using a technique called “energy windowing” that could improve the PVTs’ sensitivity to nuclear materials. DNDO officials expect they can achieve small improvements in sensitivity through energy windowing, but DNDO has not yet completed efforts to fine-tune the PVTs’ software. In contrast to the marginal improvement required in detection of certain nuclear materials, the primary screening requirement to reduce the rate of innocent alarms by 80 percent could result in hundreds of fewer secondary screenings per day, thereby reducing CBP’s workload. In addition, the secondary screening criteria, which require ASPs to reduce the probability of misidentifying special nuclear material by one-half, address the limitations of relatively small handheld devices in consistently locating and identifying potential threats in large cargo containers. Results of performance testing and field validation. The results of performance tests that DNDO presented to us were mixed, particularly in the ASPs’ capability to detect certain shielded nuclear materials during primary screening. The results of performance testing at the Nevada Test Site showed that the new portal monitors detected certain nuclear materials better than PVTs when shielding approximated DOE threat guidance, which is based on light shielding. In contrast, differences in system performance were less notable when shielding was slightly increased or decreased: both the PVTs and ASPs were frequently able to detect certain nuclear materials when shielding was below threat guidance, and both systems had difficulty detecting such materials when shielding was somewhat greater than threat guidance. With regard to secondary screening, ASPs performed better than handheld devices in identification of threats when masked by naturally occurring radioactive material. However, the differences in the ability to identify certain shielded nuclear materials depended on the level of shielding, with increasing levels appearing to reduce any ASP advantages over the handheld identification devices. Other phases of testing uncovered multiple problems in meeting requirements for successfully integrating the new technology into operations at ports of entry. Of the two ASP contractors participating in the current round of testing, one has fallen behind due to severe problems encountered during testing of ASPs’ readiness to be integrated into operations at ports of entry (“integration testing”); the problems may require that the vendor redo previous test phases to be considered for certification. The other vendor’s system completed integration testing, but CBP suspended field validation testing in January 2009 after 2 weeks because of serious performance problems resulting in an overall increase in the number of referrals for secondary screening compared with existing equipment. DNDO’s plans for computer simulations. As of May 2009, DNDO did not plan to complete injection studies—computer simulations for testing the response of ASPs and PVTs to simulated threat objects concealed in cargo containers—prior to the Secretary of Homeland Security’s decision on certification even though delays to the ASP test schedule have allowed more time to conduct the studies. According to DNDO officials, injection studies address the inability of performance testing to replicate the wide variety of cargo coming into the United States and the inability to place special nuclear material and other threat objects in cargo during field validation. DNDO had earlier indicated that injection studies could provide information comparing the performance of the two systems as part of the certification process for both primary and secondary screening. However, DNDO subsequently decided that performance testing would provide sufficient information to support a decision on ASP certification. DNDO officials said they would instead use injection studies to support effective deployment of the new portal monitors. Lack of an updated cost-benefit analysis. DNDO had not updated its cost- benefit analysis to take into account the results of ASP testing. An updated analysis that takes into account the testing results, including injection studies, might show that DNDO’s plan to replace existing equipment with ASPs is not justified, particularly given the marginal improvement in detection of certain nuclear materials required of ASPs and the potential to improve the current-generation portal monitors’ sensitivity to nuclear materials, most likely at a lower cost. DNDO officials said they were updating the ASP cost-benefit analysis and planned to complete it prior to a decision on certification by the Secretary of Homeland Security. Our May report recommended that the Secretary of Homeland Security direct DNDO to (1) assess whether ASPs meet the criteria for a significant increase in operational effectiveness based on a valid comparison with PVTs’ full performance potential and (2) revise the schedule for ASP testing and certification to allow sufficient time for review and analysis of results from the final phases of testing and completion of all tests, including injection studies. We further recommended that, if ASPs are certified, the Secretary direct DNDO to develop an initial deployment plan that allows CBP to uncover and resolve any additional problems not identified through testing before proceeding to full-scale deployment. DHS agreed to a phased deployment that should allow time to uncover ASP problems but disagreed with GAO’s other recommendations, which we continue to believe remain valid. The results of DNDO’s most recent round of field validation testing, which it undertook in July 2009, after our May report was released, raise new issues. In July 2009, DNDO resumed the field testing of ASPs at four CBP ports of entry that it initiated in January 2009 but suspended because of serious performance problems. However, the July tests also revealed ASP performance problems, including two critical performance deficiencies. First, the ASP monitors had an unacceptably high number of false positive alarms for the detection of certain high-risk nuclear materials. According to CBP officials, these false alarms are very disruptive in a port environment in that any alarm for this type of nuclear material would cause CBP to take enhanced security precautions because such materials (1) could be used in producing an improvised nuclear device and (2) are rarely part of legitimate or routine cargo. Furthermore, once receiving an alarm for this type of nuclear material, CBP officers are required to conduct a thorough secondary inspection to assure themselves that no nuclear materials are present before permitting the cargo to enter the country. Repeated false alarms for nuclear materials are also causes for concern because such alarms could eventually have the effect of causing CBP officers to doubt the reliability of the ASP and be skeptical about the credibility of future alarms. Secondly, during the July testing the ASP experienced a “critical failure,” which stemmed from a problem with a key component of the ASP and caused the ASP to shut down. Importantly, during this critical failure, the ASP did not alert the CBP officer that it had shut down and was no longer scanning cargo. As a result, were this not in a controlled testing environment, the CBP officer would have permitted the cargo to enter the country thinking the cargo had been scanned, when it had not. According to CBP officials, resolving this issue is important in order to assure the stability and security of the ASP. In addition to these key performance problems, the ASP was not able to reduce referrals to secondary inspection by 80 percent as required by the DHS criteria for a significant increase in operational effectiveness. According to the report from the ASP Field Validation Advisory Panel, a panel made up of officials from CBP, DNDO, and a national laboratory, the ASP was able to reduce referrals to secondary inspection by about 69 percent rather than the 80 percent as required by the DHS criteria. While the performance of the ASP during the July field validation testing raises issues about its potential readiness for deployment, DNDO’s proposed solutions to address these performance problems raise additional questions about whether this equipment will provide any meaningful increase in the ability to detect certain nuclear materials. Specifically, to address the problem of false positive alarms indicating the presence of certain nuclear materials, according to DNDO officials, DNDO has modified the ASP to make this equipment less sensitive to these nuclear materials. While this may address the issue of false positive alarms, it will also diminish the ASP capability of detecting a key high-risk nuclear material. Since the ASP modification, DNDO conducted computer simulations using a vendor-provided system called a “replay tool” to examine the effect of the modification on the ASP’s performance. According to DNDO officials, the replay tool demonstrated that the modified ASP will still be able to detect certain nuclear materials better than the PVT. However, at this point, DNDO does not plan to retest the ASP at the Nevada Test Site where it can examine the effects of these modifications using actual nuclear materials. As we reported earlier this year, the results of the testing at the Nevada Test Site demonstrated that the ASPs represented a marginal improvement in detecting certain nuclear materials. By reducing the sensitivity to these materials and not retesting the modified ASPs against actual nuclear materials, it is uncertain exactly what improvement in detecting certain nuclear materials these costly portal monitors are providing. While DNDO is reducing the sensitivity of ASPs to certain nuclear materials, it has yet to complete efforts to improve the PVT’s ability to detect these same materials through energy windowing. For several years, CBP officials have repeatedly urged DNDO officials to complete this research. However, it was not apparent from our discussions with DNDO officials if this effort is making meaningful progress with the development of energy windowing or when it will be completed. Furthermore, CBP officials stated that, depending on the outcome of this research, energy windowing could be the more cost effective way to improve detection of certain nuclear materials. In our view, ASPs being modified to diminish their capabilities to detect certain nuclear materials raises questions about whether energy windowing might be able to achieve a similar level of performance against these same materials from the PVTs that are already in place. Beyond reducing the sensitivity of ASPs to certain nuclear materials, DNDO also plans to address the issue of critical failures by, among other things, installing an indicator light on the ASP that will alert CBP officers that the ASP has experienced a mission-critical failure and is no longer scanning cargo. While this should address the issue of CBP officers not knowing that the ASP has suffered a critical failure, CBP officials stressed to us the need for the ASP to be stable and secure enough to avoid these shutdowns. In closing, the issues raised by the results of the July 2009 field validation tests provide even greater reason for DNDO to address recommendations from our May 2009 report. In particular, we reiterate the importance of our prior recommendation for DNDO to assess whether ASPs meet the criteria for a significant increase in operational effectiveness based on a valid comparison with PVTs’ full performance potential, given that the ASPs will no longer be as effective in detecting certain nuclear materials. Mr. Chairman, this completes my prepared statement. I would be happy to respond to any questions that you or other Members of the Subcommittee may have at this time. For further information about this testimony, please contact me at (202) 512-3841 or aloisee@gao.gov. Dr. Timothy Persons (Chief Scientist), Ned Woodward (Assistant Director), Joseph Cook, and Kevin Tarmann made key contributions to this testimony. Kendall Childers, Karen Keegan, Carol Kolarik, Jonathan Kucskar, Omari Norman, Alison O’Neill, and Rebecca Shea also made important contributions. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The Department of Homeland Security's (DHS) Domestic Nuclear Detection Office (DNDO) is responsible for addressing the threat of nuclear smuggling. Radiation detection portal monitors are key elements in the nation's defenses against such threats. DHS has sponsored testing to develop new monitors, known as advanced spectroscopic portal (ASP) monitors, to replace radiation detection equipment being used at ports of entry. DNDO expects that ASPs may offer improvements over current-generation portal monitors, particularly the potential to identify as well as detect radioactive material and thereby to reduce both the risk of missed threats and the rate of innocent alarms, which DNDO considers to be key limitations of radiation detection equipment currently used by Customs and Border Protection (CBP) at U.S. ports of entry. However, ASPs cost significantly more than current generation portal monitors. Due to concerns about ASPs' cost and performance, Congress has required that the Secretary of Homeland Security certify that ASPs provide a significant increase in operational effectiveness before obligating funds for full-scale ASP procurement. In May 2009, GAO issued a report (GAO-09-655) on the status of the ongoing ASP testing round. This testimony (1) discusses the principal findings and recommendations from GAO's May report on ASP testing and (2) updates those findings based on information from DNDO and CBP officials on the results of testing conducted since the report's issuance. DHS, DNDO, and CBP's oral comments on GAO's new findings were included as appropriate. GAO's May 2009 report on ASP testing found that DHS increased the rigor in comparison with previous tests and thereby added credibility to the test results. However, GAO's report also questioned whether the benefits of the ASPs justify its high cost. In particular, the DHS criteria for a significant increase in operational effectiveness require only a marginal improvement in the detection of certain weapons-usable nuclear materials, which DNDO considers a key limitation of current-generation portal monitors. The marginal improvement required of ASPs is particularly notable given that DNDO has not completed efforts to fine-tune current-generation equipment to provide greater sensitivity. Moreover, the test results showed that ASPs performed better than current-generation portal monitors in detection of such materials concealed by light shielding approximating the threat guidance for setting detection thresholds, but that differences in sensitivity were less notable when shielding was slightly below or above that level. Finally, DNDO had not yet updated its cost-benefit analysis to take into account the results of ASP testing and did not plan to complete computer simulations that could provide additional insight into ASP capabilities and limitations prior to certification even though test delays have allowed more time to conduct the simulations. DNDO officials believed the other tests were sufficient for ASPs to demonstrate a significant increase in operational effectiveness. GAO recommended that DHS assess ASPs against the full potential of current-generation equipment and revise the program schedule to allow time to conduct computer simulations and to uncover and resolve problems with ASPs before full-scale deployment. DHS agreed to a phased deployment that should allow time to uncover ASP problems but disagreed with the other recommendations, which GAO believes remain valid. The results of DNDO's most recent round of field testing raise continuing issues. In July 2009, DNDO resumed the field testing of ASPs that it initiated in January 2009 but suspended because of serious performance problems. However, the July tests also revealed critical performance deficiencies. For example, the ASP had a high number of false positive alarms for the detection of certain nuclear materials. According to CBP, these false alarms are very disruptive in a port environment because any alarm for this type of nuclear material causes CBP to take enhanced security precautions. To address these false alarms, DNDO plans to modify the ASP to make these monitors less sensitive to these nuclear materials and thereby diminishing the ASPs' capability. As GAO reported earlier this year, previous testing results demonstrated that the ASPs represented a marginal improvement in detecting these materials. By reducing the sensitivity to nuclear materials even further, it is uncertain exactly what improvement in detecting these materials the ASPs are providing or whether DNDO might be able to achieve a similar level of performance as the modified ASPs by improving the current-generation portal monitors that are already in place. In addition, the July 2009 testing also identified a critical equipment failure, including an alert malfunction, which DNDO is taking steps to resolve for future testing. |
The process of issuing and enforcing regulations is one of the basic tools of government, and the process has generated considerable controversy. Some individuals and organizations have called for increasing regulation of businesses and other nonfederal entities to achieve certain goals, such as fairer competition, cleaner water, or safer consumer products and services. Others have recommended drastic reductions in federal regulatory activity and/or the imposition of constraints on how agencies develop or implement regulations, often because of the burden associated with regulatory compliance and questions about whether the regulations are actually achieving their stated purposes. A number of studies have attempted to analyze the effect of federal regulations on businesses or the economy as a whole, with some analysts claiming that federal regulations cost the economy hundreds of billions of dollars each year. Although these estimates are frequently cited, measurement of the effects of regulation on the economy is imprecise and controversial. Also, relatively little is known about the impact of all regulations on individual businesses or even how many regulations apply to a business. This type of information about the impact of regulations on individual businesses would provide a better understanding of the impact of regulations on the economy as a whole. Regulations generally start with an act of Congress and serve as the means by which statutes are implemented and specific requirements are established. These requirements tell people and businesses what must be done to comply with the law. The statutory basis for a regulation can vary dramatically, from (1) very broad grants of authority that state only the general intent of the legislation and leave agencies with a great deal of discretion as to how that intent should be implemented to (2) very specific requirements delineating what regulatory agencies should do and how they should do it. The Agricultural Adjustment Act is an example of a broad grant of authority, delegating to the Secretary of Agriculture wide discretion to make agricultural marketing “orderly.” The statute provides little guidance on which crops should have marketing orders or how to apportion the market among growers. The toxic air provisions of the Clean Air Act Amendments of 1990 are examples of very specific statutory requirements. The provisions specified that the Environmental Protection Agency (EPA) establish standards, on the basis of the best existing pollution control technologies, for major sources of 189 of the most prevalent and hazardous air pollutants. The provisions also established three interim milestones and a final milestone for setting the standards and specified certain consequences if EPA missed a milestone for any source category. Likewise, in the Safe Drinking Water Act Amendments of 1986, Congress specified 83 contaminants for which EPA was to promulgate standards within 3 years. The act also required EPA to regulate drinking water contaminants to be as close as technically feasible to a level at which no known or anticipated health effects occur. The federal government has long regulated economic activity, often through independent regulatory agencies established separate from traditional federal departments and agencies. Social regulation in such areas as environmental quality, workplace safety, and consumer protection is a relatively recent phenomenon. Beginning in the 1960s, a number of major new statutes were enacted in those areas, including amendments to the Clean Air Act (CAA) and the Clean Water Act (CWA), the Toxic Substances Control Act (TSCA), the Resource Conservation and Recovery Act (RCRA), the Occupational Safety and Health Act, the Truth in Lending Act, and the Consumer Product Safety Act. Those and other acts, as well as executive orders, also created new regulatory agencies, such as EPA, the Occupational Safety and Health Administration (OSHA), the National Highway Traffic Safety Administration, and the Consumer Product Safety Commission. By the 1980s, an array of federal regulations were in place that affected many decisions made by American businesses. Concerns then began to be raised about whether the benefits these regulations and regulatory agencies were attempting to achieve were worth the costs associated with compliance. Concerns were also being raised about the cumulative effect of all federal regulations on individual businesses. The basic rulemaking process is spelled out in section 553 of the Administrative Procedure Act, which, among other things, generally requires agencies to (1) publish notice of a proposed rulemaking in the Federal Register; (2) allow interested persons an opportunity to participate in the rulemaking by providing “written data, views, or arguments”; and (3) publish the rule 30 days before it becomes effective. Other procedural rulemaking requirements have been added through general statutes, executive orders, and judicial decisions. For example, certain executive orders since 1981 have required agencies (other than those considered to be independent regulatory agencies) to submit at least their significant regulations to the Office of Management and Budget (OMB) for its review before publication in the Federal Register. Because of the numerous processes involved, federal rulemaking can take years to complete. Once completed, federal regulations are compiled in the Code of Federal Regulations (CFR). Numerous attempts have been made legislatively and by the executive branch to reform federal regulatory processes. For example, Congress enacted the following two regulatory reform initiatives in 1980: (1) the Paperwork Reduction Act and (2) the Regulatory Flexibility Act. As its name implies, the Paperwork Reduction Act attempted to minimize the paperwork and reporting burdens agencies impose on nonfederal entities. The act also established the Office of Information and Regulatory Affairs (OIRA) within OMB to review and approve all agency information collection activities. The Regulatory Flexibility Act required agencies to assess the impact of their regulations on small entities (e.g., businesses and governments) and to publish their plans for new regulations. We have reported on the effects these laws have had on agencies’ regulatory programs and recommended improvements to their design and implementation. During the 104th Congress, numerous legislative initiatives have been introduced that attempted to reform the regulatory process and/or reduce businesses’ regulatory burden. As of July 1996, at least three major governmentwide reform initiatives had been enacted. The Unfunded Mandates Reform Act of 1995 established a mechanism for advising Congress of the nature and size of federal mandates in proposed legislation or regulations to allow congressional consideration of the appropriateness of such mandates on state, local, or tribal governments or the private sector. As part of that process, agencies are required to assess the anticipated costs and benefits of federal mandates. The Paperwork Reduction Act of 1995 reaffirmed the principles of the 1980 Act, required OIRA to establish governmentwide and agency-specific paperwork reduction goals, redefined key terms such as “collection of information,” and required agencies to establish their own paperwork review and clearance function. The Small Business Regulatory Enforcement Fairness Act of 1996 made several changes in regulatory procedures, including (1) amending the Regulatory Flexibility Act to allow for judicial review of agency decisions, (2) requiring the publication of “small entity compliance guides” to explain the actions a small business or other small entity must take to comply with a rule or a group of rules, and (3) establishing a congressional review process through which Congress can disapprove of final agency regulations. Some Members of Congress viewed this review process as necessary because they believed some agencies had issued regulations that went beyond the intent of Congress when it passed the underlying statutes. Every president in recent years also has taken steps intended to reduce the burden of federal regulations. In 1981, President Reagan issued Executive Order 12291, which gave OMB the authority to review all new regulations for consistency with administration policies. The order also required agencies to prepare a “regulatory impact analysis” for each major rule, describing the costs, benefits, and alternatives to the rule. In 1985, President Reagan issued Executive Order 12498, which required agencies subject to Executive Order 12291 to submit a list of significant regulatory actions they expected to propose during the upcoming year to OMB for clearance. The President also established a Task Force on Regulatory Relief, headed by then Vice President Bush. In turn, President Bush named his Vice President to head the Competitiveness Council, which was charged with advocating regulatory relief for business. In 1992, President Bush sent a memorandum to all federal departments and agencies calling for a 90-day moratorium on new proposed or final rules. During the moratorium, agencies were “. . . to identify and accelerate action on initiatives that will eliminate any unnecessary regulatory burden or otherwise promote economic growth.” The Clinton administration has also made a number of attempts to reform the federal regulatory process. Issued in September 1993, Executive Order 12866 revoked Executive Orders 12291 and 12498 and, among other things, established a number of “principles of regulation” (e.g., use the best scientific, technical, economic, or other information; specify performance objectives, not behaviors; make regulations simple and easy to understand; and use cost-benefit analysis and risk assessment) and reaffirmed the role of OMB in the regulatory review process (although only for “significant” rules). Vice President Gore’s National Performance Review also made a number of recommendations to improve the regulatory process, including (1) encouraging innovative regulatory approaches and negotiated rulemaking; (2) streamlining agency rulemaking procedures; and (3) ranking the seriousness of environmental, health, or safety risks. In March 1995, the President reiterated his interest in regulatory reform, calling on all agencies to (1) conduct a page-by-page review of all their regulations and eliminate or revise those outdated or in need of reform; (2) change the performance measures of agencies and regulators to focus on results, not process and punishment; (3) convene groups of regulators and the people affected by their regulations around the country and create “grassroots partnerships”; and (4) expand their efforts to promote consensual rulemaking. During 1995 and early 1996, the President also announced regulatory reform initiatives aimed at certain agencies or issues (e.g., the environment, pensions, and cancer drugs) and announced other reforms applicable to all federal agencies. For example, agencies were asked to halve many of their reporting requirements, reduce penalties for self-disclosed violations of certain regulations, and allow companies to change processes for certain low-risk manufacturing operations without agency preapproval. The level of federal regulatory activity, and the burden placed on businesses and others as a result of that activity, has been measured in a number of ways. For example, a number of commentors have used relatively simple, easy-to-understand indicators, such as the number of pages in the CFR, the length of the CFR on the bookshelf, the total weight of the rules, and even the length of all of the rules if each sheet of paper were placed end to end. Others have characterized federal regulatory burden in terms of federal spending on regulatory programs or the number of federal employees assigned to regulatory activities. Although these types of measures are relatively easy to develop and are appealing in some respects, they are at best only relative and indirect measures of regulatory burden and may not accurately reflect the difficulties experienced by the public or individual businesses in complying with federal regulations. These measures also require careful interpretation. Another indicator of regulatory burden that some analysts have used is the number of hours federal agencies estimate are needed to fill out their required paperwork. Although a more direct measure of regulatory burden than the measures previously described, a paperwork hour estimate has several limitations as a measure of overall regulatory burden. First, paperwork is but one element of the overall burden of federal regulations, and paperwork burden does not include other potentially relevant factors, such as labor costs unrelated to paperwork or capital expenditures. Second, paperwork burden is generally considered to be inaccurately measured. Not all paperwork burden is always counted in the data that agencies submit pursuant to the Paperwork Reduction Act, and many believe that the paperwork burden that is measured is underestimated. Finally, users of these paperwork estimates must be careful in their interpretation; changes in the burden hour totals may not reflect changes in the regulatory burden felt by businesses and individuals. Another common measure of regulatory burden is the cost borne by entities responsible for complying with the regulations involved. Studies of the costs associated with federal regulations vary in such terms as their scope (i.e., whether focused on individual businesses, sectors, or the economy as a whole) and the factors they consider (i.e., economic costs, social costs, paperwork costs, etc.). Types of regulatory cost studies are discussed in detail in chapter 3 of this report. Regardless of the nature of the study, though, the results must be carefully interpreted. At the request of five Members of Congress, we agreed to obtain information about the cumulative impact of federal regulations on selected businesses. Therefore, we focused our efforts on a limited number of businesses that could help us understand the issues and variables involved. Our specific objectives were to describe (1) what selected businesses and federal agencies believed were the federal regulations that applied to those businesses, (2) what those businesses believed was the impact (cost and other) of those regulations, and (3) the regulations those businesses said were most problematic to them and relevant federal agencies’ responses to those concerns. To accomplish these objectives, we first needed to identify the businesses that would be the focus of our study. Because some of the information we wanted to collect was proprietary in nature (e.g., information on the companies’ operating expenses) or involved regulatory enforcement actions that could be very sensitive, we recognized that some companies might not want to participate in our study unless their identities could be concealed. Therefore, we told the businesses we contacted that we would not disclose their identity unless we had their permission to do so or unless we were legally compelled or required to do so by Congress. We asked for and received pledges from our requesters that they would concur with and honor our pledge of confidentiality, and that they would oppose disclosure requests from other committees or Members of Congress. The requesters agreed that, although they would have access to summary data that would not identify individuals or firms, neither they nor their staff would have access to the individual business’ responses. We initially attempted to obtain nominations of businesses to participate in our study from two types of organizations: (1) business interest groups, including several that had testified before Congress and/or made public comments criticizing federal regulations, and (2) public interest groups, some of which had defended the need for federal regulatory action. The five business interest groups we contacted were the U.S. Chamber of Commerce, the National Federation of Independent Businesses (NFIB), the National Association of Manufacturers (NAM), the Chemical Manufacturers Association (CMA), and the Greater Washington Board of Trade. The four public interest groups we contacted were Public Citizen Litigation Group, OMB Watch, Business for Social Responsibility (BSR), and Global Environmental Management Initiative (GEMI). We initially contacted most of these organizations during June through August 1994 and asked them to nominate businesses that they believed would be good candidates for our study. Of these organizations, NAM and the Greater Washington Board of Trade provided nominees for the study (three and six nominees, respectively). Although the U.S. Chamber of Commerce initially indicated it would be able to provide nominees for our study, several months later a representative of the Chamber said that it would not provide any nominees because of concerns its member companies had about our ability to guarantee the confidentiality of their responses. During several months of telephone calls, the NFIB representatives said that they were not able to provide nominees because their efforts were then directed toward other legislative initiatives and priorities. GEMI’s board of directors declined to participate but did not provide a reason. BSR representatives initially appeared interested in providing nominees for the review, but they did not respond to any of our subsequent telephone calls. CMA, Public Citizen, and OMB Watch initially agreed to try to identify companies for us to contact, but we never received any nominees from either organization. Because we wanted to contact more companies than the interest groups identified, we turned to other sources for potential study participants. One such source was a list of companies that had participated in a March 1994 forum on regulatory reform sponsored by the Small Business Administration (SBA). SBA staff who worked on the forum identified seven companies that they believed would be good candidates for participation in our review, and we accepted the nomination of another company from one of the SBA nominees. Another source that we used to identify possible company participants (35 companies) was newspaper and magazine articles in which specific companies commented either positively or negatively about federal regulations or their federal regulatory experience. The periodicals we reviewed to identify these companies included INC., Nation’s Business, The Wall Street Journal, and the ABA Banking Journal. We also used a literature search to improve the diversity of our company selections, focusing on articles about companies in certain industrial categories and geographic areas that were not represented by the other nominees. The combination of all of these methods yielded a total of 51 companies as potential participants. Before contacting the 51 company nominees, we developed a standardized telephone interview guide as part of an initial screening process to (1) provide consistent descriptions of the purpose of our review, (2) collect preliminary information about the companies’ views regarding federal regulations, (3) explain the confidentiality guarantees we were able to offer, and (4) determine the companies’ interest in participating in our study and their ability to provide the information we needed. Of the 51 company nominees we contacted, 8 did not respond to repeated telephone calls made over the course of several months. Of the remaining 43 companies, 16 declined to participate in the study during the screening process. Officials from 12 of these 16 companies said they did not have the time or resources needed to participate in our study. Two companies’ officials said they did not have the kinds of documentation we were seeking. The other two companies did not specify the reason for their decision not to participate. Some companies decided not to participate in the study after we had been in discussions with them for several months. We then sent each of the 27 companies that agreed to participate in the study a standardized interview guide we developed for use in our site visits. The interview guide included an overall description of the study, a list of the questions we intended to ask, and definitions of what regulations and regulatory costs would and would not be considered applicable in the study. (See app. I for a reprint of this interview guide.) We asked that the companies review the guide’s instructions and prepare the requested information in advance of our visit. Of the 27 companies that initially agreed to participate in the study, 10 withdrew before we could visit them and collect any detailed information. These companies cited a variety of reasons for their withdrawal, such as a lack of resources needed to participate in the study, the review’s data requirements, and company personnel problems. Of the remaining 17 companies, we selected 15 for inclusion in the study. We did not select one company because of its remote geographic location, and another company was not chosen because we had already selected other companies in the same industry. Ten of the 15 companies requested that we not disclose their identity. Whenever we discussed those companies with federal regulators and whenever those companies are referred to in this report, we used 10 generic company descriptors. Those 10 descriptors are listed below: • a federally chartered community bank (“Bank A”), • a state-chartered community bank (“Bank B”), • a large commercial bank (“Bank C”), • a large teaching hospital (“hospital”), • a manufacturer of railway tank cars (“tank car company”), • a manufacturer of flexible plastic packaging (“packaging manufacturer”), • a manufacturer of consumer glassware and fiber optic systems (“glass • a manufacturer of paper and allied products (“paper company”), • a tropical fish farm (“fish farm”), and • a producer of crude oil and natural gas (“petrochemical company”). The following five companies allowed us to use their names. • Metro Machine Corporation, a ship repair and maintenance company located in Norfolk, VA, with 850 employees; • Minco Technologies Labs, Inc., a computer chip testing company located in Austin, TX, with 129 employees; • Multiplex Company, Inc., a beverage dispenser equipment manufacturer headquartered in St. Louis, MO, with 217 employees; • Roadway Services, Inc., a transportation and logistics company headquartered in Akron, OH, with about 50,000 employees; and • Zaclon, Inc., a chemical manufacturing company located in Cleveland, OH, with 52 employees. Table 1.1 shows the distribution of all 15 participating companies by size and industry category. We defined a company as small if it had 49 or fewer employees, medium if it had from 50 to 249 employees, and large if it had 250 or more employees. The industry category groupings are the nine major Standard Industrial Classifications defined by the Department of Labor (DOL). As table 1.1 indicates, larger manufacturing companies constitute the largest proportion of participating companies while companies with few employees and companies in the services, transportation, agriculture, and mining industries constitute the smallest proportion of participating companies. The 15 companies that participated in the review were geographically dispersed. They were located in California, the District of Columbia, Florida, Illinois, Iowa, Missouri, New York, Ohio, Tennessee, Texas, and Virginia. Using our standardized interview guide developed for the site visits, we visited 14 of the 15 companies, interviewed company officials, and obtained any available supporting documentation. Specifically, we asked each company for information on (1) the aggregate list of regulations with which the company must comply, (2) the aggregate impact (cost and other) of all of those regulations on the company, (3) the regulations the company viewed as most problematic, (4) what the company believed government and businesses could do to correct or mitigate those problematic regulations, and (5) what the company viewed as the benefits of federal regulations. After our discussions with the companies, we developed written summaries of the concerns they expressed about problematic regulations, sent them to the companies for their review and correction, and obtained their written agreement that the summaries accurately portrayed the concerns they expressed. Subsequently, the companies also reviewed and approved any other information in our report that we attributed to a named company. The companies provided more than 100 usable examples of regulations or regulatory actions that they considered problematic. To present a general summary of those concerns, we coded each concern according to 10 recurring themes that we developed by analyzing the companies’ comments. (See ch. 4 for a discussion of these 10 themes.) Most of the companies’ concerns contained expressions of more than one theme. To verify our coding, we had a staff reviewer, who was otherwise not involved in the job, select a random sample of about 26 percent of the concerns (29 of 111 concerns) and independently code each concern using our original theme definitions. The independent reviewer agreed with our original determinations as to whether a theme was present in more than 90 percent of the cases. We provided the verified summaries of the companies’ regulatory concerns to the appropriate federal regulatory agencies for their review and comment. The following agencies responded to the companies’ concerns: • Board of Governors of the Federal Reserve System; • Department of Health and Human Services’ (HHS) Food and Drug Administration (FDA) and Health Care Financing Administration (HCFA); • Department of Housing and Urban Development (HUD); • Department of the Interior’s Fish and Wildlife Service; • Department of Justice (DOJ); • Department of Labor’s Occupational Safety and Health Administration, Employment Standards Administration’s (ESA), and Pension and Welfare Benefits Administration (PWBA); • Department of Transportation (DOT); • Department of the Treasury’s Financial Crimes Enforcement Network (FinCEN) and Internal Revenue Service (IRS); • Environmental Protection Agency; • Equal Employment Opportunity Commission (EEOC); • Federal Deposit Insurance Corporation (FDIC); • Federal Emergency Management Agency; • Office of the Comptroller of the Currency (OCC); • Pension Benefit Guaranty Corporation (PBGC); and • United States Sentencing Commission (USSC). Each agency was allowed to decide how it would respond to the companies’ regulatory concerns. We used a content analysis, which was similar to the one we used for the companies’ concerns, to code each of the agencies’ responses to one of nine recurring themes to allow summarization of those responses. The agency response coding was also independently reviewed by a staff reviewer to ensure accuracy and consistency. We also asked the agencies to identify which of their regulations were applicable to each of the selected companies. However, several of the regulatory agencies said developing a regulatory inventory for each company would be very time consuming and would require detailed information about the companies. Because 10 of the 15 companies requested anonymity, and because detailed information (location, industry, and size) could lead to identification of those companies, we could not provide the agencies with the information they said they needed to identify the companies’ responsibilities. Therefore, we asked several of the regulatory agencies to identify (1) the general types of company information they would need to determine which of their regulations would apply to a specific company (i.e., regulatory determinants) and (2) the types of assistance they provide to businesses to help them identify their regulatory responsibilities and how to comply with those responsibilities (i.e., informational mechanisms). We also asked three of the agencies—EPA, DOL, and EEOC—to identify their regulatory responsibilities for two of the companies that did not request anonymity—Minco Technologies Lab, Inc., and Zaclon, Inc. The methodology we used in this study—focusing on a small group of nonrandomly selected businesses—prevents us from drawing statistical generalizations from the information we obtained. The 15 companies we selected were generally those that (1) were identified by interest groups, SBA officials, or in the literature and (2) were willing to participate in our study and to provide the information we requested. Therefore, we make no inferences about the representativeness of their responses to how other companies would respond. For example, even though 8 of our 15 companies were manufacturers, we cannot conclude that their responses are typical of how other manufacturing companies would have responded. However, the comments the companies we contacted made during this study were similar in many respects to comments made by companies in some of our previous reports and in the literature. Therefore, we believe that these 15 companies are not atypical and their comments and experiences provide insights regarding issues common to organizations beyond the limited sample of companies. Because the purpose of our review was to determine businesses’ and federal agencies’ views regarding regulatory issues, we did not collect information from individuals and organizations outside of those groups. For example, we did not discuss companies’ regulatory responsibilities or their regulatory concerns with labor unions or other employee organizations. Neither did we collect information from individuals and organizations that were the potential beneficiaries of the regulations cited by the companies as problematic. Collecting the views of all such organizations for all of the regulations cited in this report would have been difficult, if not impossible. Therefore, this report does not reflect the full range of opinions that may exist regarding the issues raised during this review. However, it does reflect the views of the two stakeholders in which we were most interested—certain elements of the regulated community and the regulators themselves. One of our objectives was to describe what the selected businesses believed was the impact (cost and other) of all existing federal regulations that applied to them. This portion of the study does not address the development of cost or cost-benefit analysis information for individual regulations, such as the analyses agencies are required to perform under Executive Order 12866. Although we attempted to obtain documentation wherever possible, we were unable to verify most of the data companies provided on the cost of regulatory compliance, their regulatory concerns, and other issues. Companies frequently provided little documentation to support their cost estimates, and we had no basis to judge whether the costs they identified were reasonable, comparable to costs incurred by similar companies, or even whether such costs were, in fact, the direct result of a specific federal regulatory requirement. Neither did we evaluate the accuracy of the information we obtained from federal regulatory agencies. Our approach was to present the views of both the businesses and the agencies without attempting to resolve the many differences in opinions or attempting to independently determine whether sufficient evidence was available to support either view. In this report, when we indicate that “some” of the companies met a certain condition, we mean that at least three and no more than five companies met that condition. When we use the term “many companies” we mean either 6 or 7 companies, and the term “most companies” refers to between 8 and 14 companies. We conducted our review from June 1994 to July 1996 in accordance with generally accepted government auditing standards. We invited comments on a draft of this report from the OIRA Administrator because of OIRA’s governmentwide regulatory responsibilities, but an OIRA official said OIRA had no comments. We also invited comments from the top officials or their designees in the previously listed 19 federal departments and agencies responsible for the companies’ federal regulatory concerns. Between August 19, 1996, and September 19, 1996, we received comments from top officials or their designees in 13 of these 19 departments and agencies, but officials from the Board of Governors of the Federal Reserve System, FDA, HUD, the Department of the Interior’s Fish and Wildlife Service, the Federal Emergency Management Agency, and USSC said they had no comments. In general, the agencies’ comments indicated that the report was an accurate characterization of their regulatory operations and their positions regarding the companies’ concerns. Most of the agencies suggested technical corrections or additions of text, which were incorporated as appropriate. As noted in chapter 1, representatives from both government and industry have described federal regulatory burden in terms of the sheer volume of regulations with which businesses and other regulated entities must comply. Several of the companies participating in this review also made such comments to us in the course of our discussions with them. For example, an official from the fish farm compared the range of regulatory requirements to “getting pecked to death by ducks—each bite may not hurt, but all together they are very painful.” In recognition of the large number of federal regulations and the burden they impose, an element of both the Clinton administration’s and Congress’ recent regulatory reform initiatives has been the review of existing regulations and, where possible, the elimination of certain requirements. Although the total number of regulations is only a rough indication of regulatory burden, developing an inventory of those requirements is the first step in developing an accurate measure of an organization’s regulatory burden. Therefore, we asked the companies participating in this review to develop a list of all of the federal regulations with which they had to comply at the time of our review. We also asked a number of federal regulatory agencies to identify which of their regulations they believed were applicable to those businesses. We generally provided the companies with a copy of our data collection instrument several weeks in advance of our visit, and each company agreed to develop a list of regulations applicable to their firm. We recognized that, in preparing such a list, the companies might find it difficult to identify the specific names or legal citations of regulations. Therefore, we told the businesses that their list of applicable regulations should, at a minimum, cite the major federal statutes governing the regulations. For example, we said the list of statutes in the health and safety area of workplace regulations might include the Occupational Safety and Health Act or the Drug Free Workplace Act. We also said that other categories of workplace regulations could include labor standards (e.g., the Fair Labor Standards Act); employee benefits (e.g., Employee Retirement Income Security Act (ERISA)); civil rights (e.g., title VII of the Civil Rights Act of 1964); and labor relations (e.g., the National Labor Relations Act). Finally, we noted that other categories of regulations (e.g., environmental and tax regulations) could also be listed. We told the companies not to include certain types of regulations on their lists, such as federal regulations that had been proposed but had not been published as a final rule. We also said that they should not include state or local regulations, but we said that any state or local requirement that they believed was mandated by federal law or regulation should be included. Although all 15 of the companies participating in the review identified at least some regulations that they believed were applicable to their organizations, none of the companies provided us with a complete list of applicable federal regulations. The companies’ lists varied substantially in the degree to which they covered the general regulatory areas that would probably be applicable to the companies (e.g., tax, wage and hour, and workplace rules). Several companies listed regulations in only certain functional areas or for certain agencies. For example, officials from the paper company identified what they believed were applicable environmental, health and safety, and transportation regulations—areas that they said were their company’s greatest concern. However, they did not identify any regulations in the employee benefits, civil rights, labor relations, or tax areas. Some companies provided what they described as a partial list of regulations and indicated they would provide additional information, but never did so. Two companies’ lists reflected only the problematic regulations we asked them to identify for another portion of this review. (See ch. 4 of this report.) Although officials from several companies said they believed their lists contained 75 to 90 percent of the regulations applicable to them, most of the companies’ officials acknowledged that their lists were incomplete. The companies’ lists also varied in the level of detail they provided. Some companies identified only broad regulatory areas or general regulatory requirements (e.g., “workmen’s compensation” or “IRS”) but other companies’ lists were more specific. For example, one section of Roadway’s list focused on civil rights and employee benefits. Within that section, the company officials cited title VII of the Civil Rights Act of 1964 and listed its associated regulations—29 C.F.R. Parts 1601 and 1602 (Subparts A-E) and 29 C.F.R. Parts 1604-1606, 1608, and 1610-1612. Most of the companies did not maintain lists of applicable regulations, so they had to compile the information they provided in response to our request. Although officials from each company told us they would prepare a list of applicable regulations, several companies had not done so at the time of our site visit. Other companies made a more extensive effort to respond to our requests for information. For example, a number of officers and staff within Roadway conducted research and developed documentation. The hospital provided several lists of applicable regulations, one that the administrative staff had developed covering a variety of issues and another from hospital health protection staff. A few companies compiled the information we requested on the basis of lists of regulations that had been previously developed for certain areas of their operations. However, even these lists were not comprehensive. For example, the petrochemical company had developed a list of environmental, workplace safety, and other regulations for use by their internal auditing staff. However, company officials told us that the list was not necessarily complete. The paper company used an EPA publication entitled Federal Environmental Regulations Potentially Affecting the Commercial Printing Industry. However, company officials noted that this document contained a disclaimer that said it should “not be relied on by companies in the printing industry to determine applicable regulatory requirements.” Some of the companies said it was difficult for them to produce a complete list of applicable regulations because they had limited resources and higher priorities. For example, Multiplex officials said that to compile a complete list of regulations would “use so much time and so many resources that it would be a burden on the company and adversely affect its business operations.” Bank C’s official said the bank could not devote the time and staff resources needed to produce a complete list of regulations due to higher priority bank-related work. We recognized that producing an aggregate list of regulations would be an expensive and time-intensive endeavor because of the complex analysis required to identify every regulation affecting the business. Some of the companies also said that some federal regulatory requirements were hard to identify because they had become part of the companies’ standard procedures. For example, Roadway’s officials said that developing a comprehensive list of regulations was difficult because many regulations have been around for so long they are now part of everyday operations of the company. Officials from the fish farm said that some regulatory requirements (e.g., payroll recordkeeping standards) are now considered part of their everyday business operations. The officials also said that some outside organizations (e.g., insurance companies) require the company to follow certain safety procedures, and they were not sure whether those procedures were also required by regulations. Officials from the petrochemical company said regulations often cause a fundamental shift in business processes that later becomes less distinctive. The officials noted that industry incurred start-up costs associated with the requirement to produce unleaded gasoline, but because the entire industry was required to be in compliance the identification and capture of these costs became less relevant over time. Some of the companies also said it was difficult to distinguish between federal requirements and those of other governmental jurisdictions. Officials from the paper company said making this distinction was “difficult, if not impossible.” The petrochemical company indicated that it was particularly difficult to separate the requirements when state or local governments enforce federal standards and can add additional requirements. One California company noted that all OSHA regulations and many EPA regulations are enforced by the state, and that California often adds stricter state requirements. The companies also cited other reasons why their lists of applicable regulations were incomplete or difficult to compile. For example, a fish farm official said that the regulators themselves are sometimes unable to inform the company of all applicable regulations. Petrochemical company officials were reluctant to characterize their list as complete because of a concern that if a regulator saw certain requirements missing from their list they might assume the company was not complying with the missing regulations and pursue some type of enforcement action against the company. Other companies simply noted that their lists were incomplete, but did not provide a reason for this characterization. We also planned to ask each of the agencies whose regulations were cited by the companies to provide a list of regulations applicable to each of the companies. However, officials from several of the agencies we initially contacted said they could not provide such lists without first obtaining a great deal of specific information about the company. For example, a DOT official said that the agency would need such information as whether the company uses rail transportation, the types of material the company transports, and the nature of the business enterprise (e.g., whether it was a partnership or a corporation). IRS officials said they would need to know whether the business was privately or publicly held, whether it imported or exported materials or products, and whether it had foreign as well as domestic operations. The IRS officials said operational and administrative decisions made throughout the existence of the companies affect their tax status and, therefore, the applicable tax regulations. EPA officials also said that the collection of the information they needed to identify applicable regulations would require a large expenditure of resources at a time when their budget was uncertain. The officials said they would have to conduct a site visit at each company to identify their applicable regulations. Another reason we did not ask the agencies to provide lists of regulations for each of the participating companies was that 10 of the 15 participating companies wanted to remain anonymous. Therefore, the agencies could not contact them directly to collect information, and we could not provide the agencies with detailed information about the companies (e.g., industry, size, location, etc.) that could disclose their identities. Because we were unable to receive lists of applicable regulations from the agencies, we changed the nature of our inquiry to focus on three related issues. First, we asked 17 of the regulatory agencies that the companies had cited in their lists of regulations to describe the kinds of information they needed to be able to determine the applicability of their regulations to a particular company. Second, we asked each of these agencies to describe the kinds of assistance they provide to companies to help them determine which regulations were applicable to them and how to comply with the regulations. Finally, we asked six agencies and offices to identify which of their regulations were applicable to two of the participating companies that had not asked for anonymity. Fifteen agencies provided information on their regulatory determinants. Officials from many of the 15 agencies said that unique characteristics of a business or a business activity determine whether their regulations are applicable to that business. See the following examples of agencies’ regulatory determinants. • DOT officials said that the applicability of its regulations generally varied by industry, transportation mode, location, and other factors, including the type of material being shipped and the nature and ownership of the transportation firm. • OCC officials said that a bank’s specific activities determined the applicability of its regulations. Similarly, the Federal Reserve Board’s (FRB) officials said that coverage of its regulations “is determined by either the nature of a particular company or the nature of the activities in which a particular company engages or intends to engage.” • EPA officials said that “there is no single set of regulatory determinants that would cover all situations in which a facility may be covered by EPA’s regulations. Many of our regulations are event driven, some factors are related to facility location, and many regulations are triggered by physical and operational characteristics of a particular facility.” IRS officials said it was difficult to come up with criteria for the development of a list of regulations applicable to a particular company because of decisions that companies make in the course of their business. For example, IRS officials said that if a company chooses to provide a qualified retirement plan, it must comply with the statutory provisions and regulations applicable to such plans. • OSHA’s officials said that it “regulates occupational safety and health hazards, not specific industries.” Therefore, the “applicability of individual standards depends on whether or not the hazard addressed by the standard is present in the workplace.” Officials from DOL indicated that companies’ specific reporting requirements also varied according to specific criteria. For example, the officials said that the requirements for the Form 5500 used by employee benefit plan administrators to satisfy their reporting obligations under title I of ERISA, title IV of ERISA, and the Internal Revenue Code depend on (1) the type of plan (i.e., whether the plan is a pension or welfare plan); (2) the size of the plan (i.e., whether the plan has fewer than 100 participants or 100 or more participants); and (3) how the benefits are funded (i.e., through a trust or insurance or from the general assets of the employer). On the other hand, some agencies’ officials said that determining the applicability of their regulations is relatively straightforward. See the following examples: • EEOC officials said that, with the exception of reporting requirements, EEOC regulations apply to all entities covered by the statutes it enforces, and the officials indicated that the applicability of those statutes is primarily a function of company size. For example, EEOC officials said that title VII of the Civil Rights Act of 1964 and the Americans with Disabilities Act (ADA) apply to any company with 15 or more employees. The officials also said that the Age Discrimination in Employment Act (ADEA) applies to all employers with 20 or more employees. According to the officials, any private employer with 100 or more employees must complete their EEO-1 reporting form indicating the race, ethnicity, and sex of employees by job category.• Officials from the Office of Federal Contract Compliance Programs (OFCCP) in DOL said Executive Order 11246, which OFCCP administers, applies only to contractors and subcontractors who perform government contracts that total at least $10,000 in a 12-month period. However, the officials said that nonconstruction contractors with 50 or more employees and contracts greater than $50,000 have additional obligations. • The PBGC officials said PBGC’s insurance program and regulations apply only with respect to defined benefit pension plans as described in section 4021 of ERISA. The agency officials said that section 4021 generally covers all defined benefit pension plans voluntarily established by private sector employers, and specifically states what plans are excluded from coverage. • The FDIC officials said that coverage by FDIC regulations “is determined basically by whether an institution is FDIC-insured, and as a subset of that status, whether the institution is an insured nonmember bank for which the FDIC has primary supervisory responsibility at the federal level.” Sixteen agencies described how they provide information to companies and the public on their regulatory requirements. Most of the agencies identified telephone numbers and listed various publications, handouts, brochures, informational pamphlets, and notifications containing regulatory information. Half of these agencies had or were developing (1) special programs to communicate regulatory requirements to affected businesses and (2) outreach efforts to gather feedback on their regulatory requirements. Six agencies indicated that they used some type of electronic bulletin board as an informational mechanism. OSHA described a variety of informational resources and programs, including the following examples: • more than 80 different publications available from OSHA’s Publications Office, some of which were also available in OSHA’s 100 field offices; • safety and health standards in CD-ROM format (available for purchase from the Government Printing Office) containing all OSHA standards, compliance directives, and standards interpretations; • a DOL-operated Labor News Electronic Bulletin Board and an OSHA Computer Information System accessible through the Internet, which contains some of the information on the CD-ROM; the OSHA Consultation Program, which offers free, on-site, expert assistance to small employers in all 50 states to help them comply with OSHA requirements and establish effective safety and health programs (according to OSHA, more than 100,000 employers have used this service and priority is given to small firms in high-hazard businesses); and • courses on specific safety and health issues for employers who want intensive information about specific safety and health issues at OSHA’s Training Institute in Des Plaines, IL, and at OSHA Education Centers in 12 states. OSHA’s officials said OSHA is also piloting several other ways to use computer technology to provide assistance to employers, including the development of interactive compliance tools. EPA also cited dozens of sources of information about its regulations, including brochures, pamphlets, fact sheets, booklets, letters, hotlines, regional contact numbers, guidance manuals, posters, question and answer sheets, and catalogues of informational materials. EPA listed informational sources for each of its program offices and, within those offices, the sources were also often differentiated by issue. For example, the Office of Prevention, Pesticides and Toxic Substances said it had many of the informational modes listed above as well as information lines for pesticide questions, “PR Notices” providing detailed information to regulated industries, registration kits for those interested in how to register new pesticide products, and outreach efforts to help growers that rely on minor use pesticides. Among the initiatives EPA particularly noted were the following: • Compliance assistance centers were being established by EPA’s Office of Compliance for four industry sectors—automotive, metal finishing, printing, and agriculture. EPA officials said the centers offer “one-stop shopping” for understandable guidance materials, waste minimization and pollution prevention assistance, and advice in reducing regulatory compliance costs. • Sector Notebooks, which are profiles of 18 industries (e.g., metal fabrication, petroleum refining, and printing), were designed to assist firms in understanding what multimedia regulations apply to them. Each profile includes, among other things, applicable federal statutes and regulations as well as compliance assistance information. Notebooks are available through the Government Printing Office and on an electronic bulletin board. • The Office of the Small Business Ombudsman was established to provide a variety of information mechanisms to help communication between the small business community and EPA. EPA officials said the Office has a hotline service that receives nearly 20,000 calls per year, serves as the “one-stop shop” for EPA technical assistance and information, maintains an informal dialogue with over 45 trade associations, and advocates small business positions inside EPA. EPA also described the mechanisms one of its regional offices (Kansas City) used to communicate regulatory information to the public, including an Agricultural Compliance Assistance Center, public meetings and workshops, “availability sessions” in which regional staff privately meet with citizens one-on-one to discuss issues, state- and trade association-sponsored meetings, mass mailings, public speaking, a toll-free Action Line, and an Iowa RCRA Hazardous Waste Helpline. Other agencies cited many of the same kinds of mechanisms. According to its officials, EEOC (1) conducted training and outreach seminars during fiscal year 1993 that reached an estimated 4,000 private sector employers and more than 94,000 individuals; (2) published and distributed millions of copies of training materials; (3) mailed out 477,933 publications during the first three quarters of fiscal year 1995; and (4) responded to thousands of public inquiries per year. DOT’s officials said DOT public information efforts include electronic bulletin boards, toll-free hotlines, free guidance materials, news releases, mailing lists, and briefings to industry associations. The DOJ officials cited their toll-free ADA Information Line; technical assistance materials; a computer bulletin board and Internet connections; a speakers’ bureau, which provides technical assistance at about 120 events each year; more than 40 technical assistance grants to trade associations; an ADA Information File containing more than 30 technical assistance publications placed in 15,000 public libraries throughout the country; and informational notices about the ADA, which are distributed to 6 million businesses through IRS’ quarterly mailing to employers. Some of the agencies’ regulatory informational mechanisms are proactive, providing information to businesses and others at the agencies’ initiative. However, many agencies, if not most, require businesses to take the initiative in obtaining compliance information. For example, OSHA officials said that although the agency offered a variety of informational mechanisms, it was up to each business to understand its own regulatory compliance responsibilities. However, the businesses that we talked to sometimes indicated a reluctance to approach regulatory agencies for information. For example, an official from Minco said that it was difficult to stay aware of the changes in regulatory requirements because they could not ask “enforcers” to provide information without potentially calling Minco’s actions into question. Other indications of the businesses’ reluctance to address regulators directly were the decisions by 10 of the 15 companies to remain anonymous during this review. Also, it is not always readily apparent to businesses which agency of the federal bureaucracy is responsible for a particular program. For example, DOJ officials explained that the ADA defines separate responsibilities for EEOC and DOJ. EEOC provides information about ADA employment regulations, but information about titles II and III of the act (architectural barriers) is available only from DOJ. Sometimes multiple information sources exist within a particular agency, with businesses frequently required to make more than one contact to gather information about that agency’s regulatory requirements. As previously noted, EPA listed informational mechanisms by program office—Air and Radiation; Prevention, Pesticides, and Toxic Substances; Solid Waste and Emergency Response; and Water—as well as within a regional office. Therefore, a business would need to be aware of EPA’s structure and programmatic configuration to obtain information about all EPA programs. However, EPA has taken some steps to consolidate this information by providing single points of contact for small businesses in its Small Business Ombudsman office and compliance assistance centers for certain industries. Some agencies are attempting to develop methods by which businesses can obtain information about regulatory requirements and other topics in a more efficient and understandable way. For example, in June 1995, the President asked SBA to cochair an effort to make government information more accessible to government customers. SBA developed the idea of a World Wide Web site on the Internet that would, among other things, allow companies to know what federal regulations apply to their operations. Working with the Lawrence Livermore National Laboratory, the University of Massachusetts, and more than two dozen federal departments and agencies, SBA developed the “U.S. Business Advisor” home page, a version of which was formally unveiled in February 1996. Using the Advisor, businesses can access a regulatory assistance center to search an electronic database of current and proposed regulations within particular subject areas. For example, a business can type in the term “chlorine production” and get a listing of chlorine-related regulations and proposed regulations. Businesses can also use the Advisor to obtain the full text of the current or proposed rule. SBA officials said the Advisor is still being developed, and they hope that future iterations will be even more user-friendly. Some agencies are developing their own World Wide Web sites on the Internet with regard to particular issues. For example, in October 1995, OSHA worked with the business community to create an on-line “asbestos advisor” program that helps businesses determine whether their company is complying with regulations on asbestos exposure. The program solicits information about users’ workplaces and tasks, and automatically provides guidance to ensure compliance. As of July 1996, more than 6,100 people had downloaded copies of the program. OSHA said that because of further distribution of the asbestos advisor by major corporations and trade associations, actual circulation could be 10 times greater than the number of downloaded copies. OSHA has developed on-line advisors for other standards (e.g., permit-required confined spaces and cadmium) and plans to create other interactive expert advisors on other issues (e.g., lead in construction and control of hazardous energy sources). DOL officials said the Department has also developed an Internet web site that includes copies of its statutes and regulations and a small business handbook that provides information about all DOL workplace requirements in nontechnical language. We asked EPA, EEOC, and four agencies and offices within DOL—OSHA, PWBA, and the Wage and Hour Division and OFCCP within ESA—to identify which of their regulations were applicable to two of the companies participating in this review—Minco and Zaclon. Neither of these two companies requested anonymity, and they agreed to provide any information needed by the agencies in determining their applicable regulations. Minco is located in Austin, TX; has 129 employees; and is a federal subcontractor that tests computer chips for federal contractors involved in military, space, and medical industries. Zaclon is located in Cleveland, OH; has 52 employees; and manufactures organic and inorganic chemical compounds. Officials from some of the agencies we contacted initially expressed concerns about providing a list of regulations applicable to the companies. For example, EPA officials said that if the list was incomplete in any way (e.g., because new regulations were issued after they provided a list or because EPA did not know about an element of a company’s operations that was covered by its regulations), the absence of a regulation from its list could be construed to mean that the company did not legally have to comply with that requirement. An official at OSHA questioned whether developing a list of applicable regulations was a useful method to measure the impact of OSHA’s regulatory requirements. The official said that a list of all OSHA regulations a company must comply with could appear extensive, but would not provide any information on the beneficial results produced by these requirements. The official also said that this exercise might create an unfair impression of OSHA outreach efforts or a company’s knowledge of its regulatory responsibilities. Despite these concerns, EPA, EEOC, and each of the four DOL agencies provided a list of regulations they said were applicable to the companies. Several of the agencies indicated that the two companies’ regulatory responsibilities varied substantially. One indicated that the companies’ regulatory responsibilities were the same in some respects and different in others. One agency said the two companies’ responsibilities were the same in all respects. EPA limited its description of applicable regulations to four major environmental statutes—RCRA, CAA, CWA, and TSCA. The agency also emphasized that its observations were not meant to indicate there had been a formal compliance review or audit of the companies and, therefore, its observations were not an explicit or implied assessment of the companies’ compliance status. With regard to Minco, EPA officials said that the company is not regulated at the federal level with regard to any of the above-mentioned statutes because it has no air permits, does not formulate new or existing chemicals, does not manufacture or handle herbicides or pesticides, and does not have any underground storage tanks. However, the officials said that EPA’s limited review of Zaclon’s operations indicated that it faced a number of federal regulatory requirements. • Because the company produces chemicals, EPA officials said Zaclon is responsible for reporting under TSCA’s sections 5 and 8 and is subject to annual Emergency Planning and Community Right-to-Know Act (EPCRA) Section 313 emissions reporting. • EPA officials said Zaclon’s discharges to the local waterways make the company responsible under CWA and its permits for (1) monthly discharge monitoring reports; (2) the Spill Prevention, Control, and Countermeasures’ (SPCC) revision deadline; and (3) the National Emissions Standards for Hazardous Air Pollutants’ (NESHAP) benzene waste water report. • EPA officials said Zaclon’s stack emissions make the company responsible under CAA, including NESHAP fugitive and point source emission reporting for benzene, and air compliance reports. • As a hazardous waste generator EPA officials said Zaclon is subject to RCRA, including quantifying amounts of annual hazardous wastes, annual financial assurance reporting, and waste minimization reports. EPA officials also said that Zaclon is required under many of the statutes to report any spills or releases when they occur. The officials said that certain chemicals could be considered a pesticide under the Federal Insecticide, Fungicide, and Rodenticide Act; a drug or cosmetic under the Federal Food and Drug Act; a chemical under TSCA; or a waste under RCRA. However, because EPA assumed that the company knew whether any chemicals it produces or uses fall under one of these statutes, it did not describe applicable regulations in those situations. EEOC divided its response into reporting and recordkeeping issues. The agency’s officials said that the only reporting requirement EEOC imposed on Minco was that it annually submit a complete Form 100 (also known as an EEO-1 report) because the company had at least 100 employees. However, the agency officials said that Zaclon does not have to submit the form because it has less than 100 employees. EEOC officials said both Minco and Zaclon are covered by the agency’s recordkeeping requirements, but the length of time for which the records must be maintained varied by statute. The officials said 29 C.F.R. 1602.14 requires both companies to preserve all personnel and employment records for 1 year after the preparation of the record or the date of a related personnel action, whichever is later. Also, records must be retained for any employee involuntarily terminated. EEOC officials said 29 C.F.R. 1620.32 requires both companies to preserve records relevant to the payment of wages under the Equal Pay Act for 2 years. The officials said 29 C.F.R. 1627.3 requires each company covered by the ADEA to make and keep records for each employee (name, address, date of birth, occupation, rate of pay, and compensation earned each week) for 3 years. Employers also must keep a copy of their employee benefit plans on file. If an enforcement action is initiated under title VII, the ADA, or the ADEA, EEOC requires the employer to retain any related records until a final disposition. Finally, EEOC noted that Minco and Zaclon are covered by the recordkeeping requirements in the Uniform Guidelines on Employee Selection Procedures (29 C.F.R. 1607), but that each company has different obligations under these requirements. Since Minco has more than 100 employees, it is required to maintain records that would disclose whether its selection procedures have an adverse impact on the basis of race, gender, or ethnic group. The Guidelines also require Minco to annually evaluate whether its selection process is having an adverse impact. If so, it must maintain and have available evidence supporting the validity of its selection process. However, because Zaclon has less than 100 employees, it can use simplified recordkeeping procedures that use the existing statistical information present in the company’s personnel files. OSHA officials indicated that the regulatory responsibilities of Minco and Zaclon are identical. The officials cited 29 C.F.R. 1903 (Inspections, Citations and Proposed Penalties); 1904 (Recording and Reporting Occupational Injuries and Illnesses); 1910.20 (Access to Employee Exposure Records); and various other subparts of section 1910 as applicable to both companies. They also said OSHA’s General Industry Standards would generally apply to each company, depending on the hazards in the workplace. DOL’s Wage and Hour Division stated that most of the regulations it cited applied to both Minco and Zaclon. For example, it said the following: • The Fair Labor Standards Act (29 U.S.C 201 et seq.) and its applicable regulations (29 C.F.R. Parts 510-794) apply to both companies because they are covered “enterprises” with sales in excess of $500,000 per year. • The Employee Polygraph Protection Act (29 U.S.C. 2001-2009) and its applicable regulations (29 C.F.R. Part 801) apply to both companies because they are “. . . engaged in or affecting commerce or in the production of goods for commerce . . . .” • FMLA (29 U.S.C. 2601 et seq.) and its applicable regulations (29 C.F.R. Part 825) apply to both companies because they are both “. . . engaged in commerce or in any industry or activity affecting commerce . . .” and employ at least 50 employees during 20 or more work weeks during the year. However, the Wage and Hour Division’s officials said that the McNamara-O’Hara Service Contract Act (41 U.S.C. 351 et seq.) and its regulations (29 C.F.R. Part 4) applied only to Minco because Zaclon does not contract to provide services to the federal government or the District of Columbia. OFCCP’s officials said that OFCCP administers and enforces three equal employment opportunity programs that pertain to government contractors and subcontractors and to federally assisted construction contractors and subcontractors. The programs OFCCP officials noted are listed below: • Executive Order 11246, as amended (41 C.F.R. Parts 60-1 through 60-60), which prohibits discrimination in employment on the basis of race, color, religion, sex, or national origin and requires affirmative action; • section 503 of the Rehabilitation Act of 1973, as amended (29 U.S.C. 793), and its implementing regulations (41 C.F.R. Part 60-741), which require affirmative action and prohibit discrimination in employment against qualified individuals with disabilities; and • provisions of the Vietnam Era Veterans’ Readjustment Assistance Act of 1974, as amended (38 U.S.C. 4212), and its implementing regulations (41 C.F.R. Part 60-250), which require affirmative action and nondiscrimination with respect to special disabled and Vietnam-era veterans. The programs apply to contractors and subcontractors who perform government contracts or federally assisted construction contracts that total at least $10,000 in a 12-month period. Because Zaclon was not a federal contractor or subcontractor at the time of our review, OFCCP officials said the programs’ requirements did not apply to the company. However, Minco was a federal subcontractor with contracts in excess of $10,000, so OFCCP officials said that all three programs applied to the company. Also, because Minco had more than 50 employees and performed government contracts worth at least $50,000, OFCCP’s officials said it was obligated to develop a written affirmative action program under each of these laws. OFCCP officials said these plans could be separate documents containing the different analysis that each law requires or a single document consolidating each of the laws’ required analyses into one affirmative action plan. PWBA summarized the companies’ basic reporting and disclosure requirements under Part 1 of ERISA. PWBA’s officials stated that PWBA did not attempt to summarize all conceivable reporting and disclosure requirements because the requirements vary according to the size and nature of the benefit plan. For example, the officials said PWBA did not discuss the Consolidated Omnibus Budget Reconciliation Act of 1985, Part 6 of title I, or disclosure requirements related to fiduciary regulations. The agency officials also noted that ERISA contains general recordkeeping requirements. For example, adequate records must be maintained to verify the accuracy of benefit calculations and information reported in the annual report. PWBA said that the two companies’ 401(k) regulatory reporting and disclosure responsibilities differed somewhat because of differences in the companies’ plan size and operations. For example, it said because Minco has a single-employer defined contribution 401(k) profit-sharing pension plan subject to Part 1 of title I of ERISA covering 105 of Minco’s 129 eligible individuals with assets held in trust, the plan administrator must file the Form 5500 (with all applicable schedules) for each plan year. An annual audit is required and an opinion of an independent, qualified public accountant must also be filed. However, because Zaclon has fewer than 100 eligible employees, PWBA’s regulations allow Zaclon’s plan administrator to file the Form 5500-C/R (with all applicable schedules), which is an abbreviated Form 5500. Also, under PWBA’s regulations for such small plans, an audit and opinion of an independent, qualified public accountant is not required. Although in both companies, participants covered by the pension plans and beneficiaries receiving benefits under the plans must be furnished a summary description of the plan, a summary of any material plan modifications, and a summary of each year’s annual report. Both companies must also file their pension plan’s description and modifications with DOL. PWBA also said that because Minco’s welfare plans cover 100 or more employees, the administrator of Minco’s plans must file the Form 5500 (and all applicable schedules) for each plan year, but Minco is exempt from the audit requirements because the plan was fully insured or unfunded. Zaclon, with less than 100 participants in its fully insured or unfunded welfare plans, is exempt from the Form 5500 report and the audit requirements. PWBA officials said that both companies must furnish their welfare plan participants a summary plan description; a summary of any material modifications; and, for Minco’s welfare plan, a summary of each year’s Form 5500 report. Minco must also file its welfare plan’s description and modifications with DOL. Our work suggests that the number of federal regulations applicable to a particular company may be substantial. However, producing a complete inventory of those regulations is a very difficult undertaking for both businesses and federal regulatory agencies. A business must have a sophisticated level of knowledge of its regulatory environment and be able to devote the time and resources needed to develop a comprehensive inventory. Our efforts to acquire this information from federal agencies demonstrated that regulatory requirements are contingent on a variety of factors and can vary substantially from one company to the next, thereby making this task more difficult than it appears to be. Although most of the companies initially told us that they could develop a complete list of regulations applicable to their companies, none of them ultimately did so. The partial lists the companies did develop often focused on only certain functional areas or certain problematic regulations. The companies said their lists were incomplete because of time and resource constraints and because of difficulties they experienced disentangling federal regulatory requirements from their regular operating procedures, state or local requirements, and other nonregulatory requirements. Other participating companies did not provide a reason why their lists were incomplete. Because of their day-to-day involvement in regulatory matters, it may seem logical that regulatory agencies would be able to determine the applicability of their regulations to particular companies quite easily. However, several of the agencies we contacted said they could not make that determination without expending a substantial amount of their limited resources. Several of the agencies said that unique characteristics of a business or a business activity determine whether their regulations apply to that business. Therefore, they said they would have to collect detailed information about each company to determine regulatory coverage—information such as whether the firm (1) was a federal contractor or subcontractor, (2) had a qualified retirement plan, (3) had an underground storage tank, (4) used certain types of compressed gases, and (5) discharged water into a local waterway. The agencies also indicated that they made extensive amounts of information available to the public, so the businesses themselves could determine their regulatory responsibilities. However, these sources of information often appear to be fragmented both within and across agencies. As a result, a business attempting to determine its regulatory responsibilities may find it necessary to contact multiple agencies, and sometimes multiple offices within particular agencies, to collect the information it needs. In some cases, responsibility for an issue may be spread between two or more agencies, making it difficult for companies to determine which agency or agencies should be called regarding that issue. The increasing complexity of the federal regulatory environment makes effective communication between regulatory agencies and the regulated community even more important, and some agencies are taking steps in that direction. The difficulties businesses and agencies experienced in developing a list of applicable regulations also suggest two other conclusions—one is an issue of compliance and the other is a research concern. First, a business that finds it difficult to list its regulatory compliance responsibilities may not be fully aware of those responsibilities. As a result, the business runs the risk of being out of compliance with regulations that it did not know were applicable. Second, the development of a list of a company’s compliance responsibilities is the first step in determining the impact of all regulations on that company. If the list of regulations applicable to a company is incomplete, any assessment of the impact of regulations on that company will be equally incomplete. The difficulties businesses and agencies described in developing company-specific lists of regulatory compliance responsibilities suggest that the development of information on the costs and benefits of regulations to those companies will be at least as difficult. To fairly assess the impact of regulations on businesses, one must consider both the burden and the benefits of compliance. Of the two issues, regulatory burden is generally considered to be easier to measure than benefits. As mentioned in chapter 1, regulatory burden has been described and measured in various ways, including the number of pages in the CFR, the number of federal employees involved in regulatory activities, and the number of paperwork burden hours imposed by federal information collection requirements. One commonly cited measure of regulatory burden is the cost associated with compliance with regulations. In this chapter, we briefly discuss the ways regulatory costs could be assessed, what we attempted to measure in this review, and what company officials told us when we asked about the costs and other types of burden associated with compliance with federal regulations to their businesses. This chapter also discusses what businesses told us about the benefits associated with federal regulations. Studies of regulatory costs vary in a number of ways, one of which is the types of costs the studies attempt to assess. Direct costs are those that regulated entities incur in the course of complying with regulatory requirements. These direct costs include the wages and salaries of workers carrying out regulatory responsibilities; capital expenditures (e.g., wastewater treatment facilities or safety equipment); employee training expenses; and other costs incurred as a direct result of regulatory requirements. Indirect or secondary costs include costs such as lost productivity, decreased competitiveness, construction delays, or resource misallocation. Still other costs include those associated with the development and enforcement of regulations, not just costs borne by regulated entities; therefore, these studies could include the budgets of regulatory agencies. Regulatory cost studies also vary in terms of the range of regulations included within the scope of the study. Studies may focus on costs associated with compliance with any regulatory requirements issued by any agency, or on only selected regulations, such as those within certain subject areas (e.g., environmental rules) or those issued by certain agencies. Cost studies also vary in terms of the types of costs considered attributable to regulatory requirements. Studies could include all expenditures by the regulated entity that are in any way related to the regulatory requirements at issue. In such studies, for example, if a company spent a total of $1 million during the course of a year on worker safety training and equipment, the full $1 million would be counted toward the company’s regulatory costs. However, because the cost study includes all of the company’s expenditures in this area, such an approach implicitly assumes that the company would have spent nothing on worker safety training and equipment during that year in the absence of regulatory requirements. Because many companies probably spend some money to protect their workers in the normal course of business, attributing those expenditures to regulatory requirements is erroneous and overstates the burden of regulations. Another approach does not include all expenditures in the measurement of regulatory costs, focusing only on the incremental costs directly attributable to the regulations in question. If, in the above example, the company had spent $600,000 on worker safety training and equipment, regardless of any regulatory requirements, the incremental cost attributable to regulations in that year would be $400,000 ($1 million minus $600,000). “. . . must be measured against a baseline. The baseline should be the best assessment of the way the world would look absent the proposed regulation. . . . All costs calculated should be incremental, that is, they should represent changes in costs that would occur if the regulatory option is chosen compared to costs in the base case (ordinarily no regulation or the existing regulation) or under a less stringent alternative.” Therefore, OMB recommends calculation of regulatory costs in incremental terms, not the total expenditures in a regulatory area. The scope of the cost studies we reviewed also varied widely. Some studies, such as the work of Thomas Hopkins, attempted to estimate the cost of regulations to the economy as a whole. Hopkins included the following five expenditure categories in his estimates of cumulative regulatory costs: • direct costs of environmental regulations; • direct costs of other social regulations, including consumer safety, nuclear safety, worker health, and worker security and pensions; • direct costs of economic regulations, which include agricultural, communications, transportation, energy, financial, construction, and international trade regulations; transfers stemming from economic regulations, including transfers stimulating exports and agricultural price supports; and • process costs of paperwork and reporting obligations, based upon the OMB estimate of the number of hours spent on federal paperwork requirements. Totaling data from all five cost categories, Hopkins estimated in 1993 that the cumulative cost of federal regulations to the economy would be $607 billion in 1995. However, some economists believe that one of the elements of Hopkins’ study—transfer costs—should not be considered part of the cost of regulations to the economy because transfers represent a loss to one group and a corresponding benefit to another. There are also concerns about the accuracy of some of the data included in Hopkins’ analysis. For example, we have noted that not all burden is counted in the preparation of OMB’s paperwork burden-hour estimate, and that which is counted is often underestimated. Other studies have attempted to measure the costs of regulatory compliance on individual businesses, not the economy as a whole. One such study by Arthur Andersen and Company focused on 48 large companies’ compliance with federal regulations during 1977. The study did not attempt to assess the companies’ cost of complying with all federal regulations, focusing instead on six federal agencies and programs. Arthur Andersen auditors went to each company, instructed company officials on what they considered to be allowable and unallowable regulatory costs, and required company officials to estimate the amount the company spent complying with the identified regulations. The study used an incremental measure of regulatory cost, subtracting from the companies’ total expenditures the amount the company would have spent to achieve the objectives of the regulations had those regulations not been in force. Using this methodology, Arthur Andersen estimated that the 48 companies’ total compliance costs in 1977 were $2.6 billion—an average of more than $54 million per company. This cost can be compared with the companies’ 1977 net income after taxes of $16.6 billion—an average of about $346 million per company. To assess the burden of federal regulations on the 15 companies participating in our review, we used an approach that was similar in some respects to the approach used in the Arthur Andersen study. Like the Arthur Andersen study, we asked each of the companies to provide information on their direct incremental costs. However, unlike the Arthur Andersen study, we focused on costs associated with complying with all federal regulations during 1994, not just selected agencies and programs. Also, we did not require the companies to estimate the costs associated with regulatory compliance. Instead, we left it to the companies to provide what information they believed was appropriate. Because we wanted to be sure that the companies described their regulatory costs consistently, we provided extensive instructions to the companies regarding what should and should not be included in their tabulations. (See app. I for the instructions provided to the companies and the data collection instrument.) For example, because our focus was on incremental costs, we told the companies that any costs that would have been incurred in the normal course of business during that period should not be included in their cost measures. Because indirect costs are more difficult to measure, we told the companies to provide cost data on direct costs and asked for examples of indirect costs. We also delineated other types of costs that should be excluded from their tabulations, including lobbying costs; costs associated with nonfederal rules; and payments to the federal government, such as taxes and fines for noncompliance. We asked the companies to include costs for all regulations that they had identified in developing their aggregate list of federal regulations (presented in ch. 2). We attempted to collect the cost data in total and in each of three cost classifications: (1) capital costs, (2) labor costs, and (3) other costs. We told the companies that their accounting and financial records should be their primary source of the cost data, and that we would like to collect, or at least review, any documentation of these costs. Although all of the 15 companies participating in our review provided at least some data on their compliance costs, none of the companies provided cost data that were both comprehensive and incremental. Some of the officials with whom we met recognized that the data provided were not what we had asked them to provide. Our interviews with these company officials and our review of the information they provided revealed various reasons why the companies’ cost data were neither comprehensive nor incremental. Although we asked each of the companies to provide cost data for all the regulations they faced, none of the companies provided comprehensive cost data. The uncomprehensive nature of the cost data provided was sometimes a function of the difficulty company officials had in citing all applicable regulations. (See ch. 2.) Because we asked each company to provide cost information for all applicable federal regulations, and because company officials generally said they could not identify all applicable regulations, they, therefore, could not provide a measurement of their regulatory costs that they believed reflected all of their responsibilities. Another reason company officials said they had difficulty providing data on their companies’ federal regulatory compliance costs was because they found it difficult to distinguish between federal requirements and those of other governmental jurisdictions. For example, officials from Roadway told us that the intertwining of federal, state, and local requirements made it difficult to separate the effects of each type of requirement. Also, officials from the paper company said that determining their federal regulatory costs was “difficult if not impossible” because they could not distinguish between costs to comply with federal regulations versus state and local requirements. Making this distinction was particularly difficult for the companies in regulatory areas where state governments enforced federal standards and could also attach additional requirements. Officials from a company operating in California said that California enforces all of OSHA’s regulations and some of EPA’s regulations—often adding stricter state requirements. In other cases, company officials recognized federal regulatory requirements in certain functional areas but still did not provide any cost data for those functional areas. For example, officials from the petrochemical company provided data on environmental, health, and safety costs, but did not provide data on costs associated with other types of regulations with which they said they had to comply (e.g., transportation and shipping). The cost data the companies provided were also incomplete in other ways. For example, one company provided data on its incremental regulatory costs, but only for a portion of its labor expenses. The company did not provide comprehensive data on capital or other types of costs, and the labor cost data did not include the company’s hourly workers. Also, although one company said its incremental cost estimate covered the entire company, the manner in which the estimate was calculated revealed that all units were not actually represented. Company officials estimated the company’s total incremental regulatory costs at $52.4 million—$44.9 million for labor costs, $6.8 million in capital costs, and $700,000 in other costs. The officials said they developed these cost figures by first generating these costs for their largest operating unit, then doubling this figure to arrive at the estimated total cost. They said they used this method because the operating unit generated one-half of the company’s revenue and accounted for one-half of the company’s costs. Therefore, they said that doubling the unit’s regulatory costs was a reasonable estimate of the whole company’s incremental regulatory costs. However, we believe that if their largest operating unit’s regulatory costs were atypical in any way, simply doubling this estimate could result in an underestimate or an overestimate of the company’s total incremental costs. Although we requested that companies provide incremental compliance costs because we believe they are more accurate measures of regulatory burden than total expenditures in areas covered by regulations, most companies did not provide incremental cost data. As previously noted, calculation of incremental costs requires company officials to decide what actions their company would have taken in the absence of the identified regulations—a determination that can be difficult, if not impossible, to make in retrospect. For example, officials from Bank C indicated that it would be very difficult, in most cases, to estimate what expenses they would have incurred if the federal regulations did not exist. An official from the glass company said that it was difficult to determine what percentage of its costs were due to regulations and what part would have been incurred as a normal part of business. Officials from the paper company said a substantial amount of their costs were costs they would have incurred even if federal regulations did not exist. They said that they would still have formidable environmental and health and safety programs simply as a good business practice, and they could not separate what was required from what they were doing voluntarily. Reflecting the difficulty in separating regulatory and business-related costs, the officials also said their companies’ accounting and financial records did not capture the information necessary to determine incremental compliance costs. For example, officials from the petrochemical company told us their company’s accounting systems were not designed to uniquely categorize the costs of new and ongoing regulatory requirements. These officials said that there is little incentive to isolate and monitor these costs because such information has little business value. Officials from Multiplex also said that their financial records did not itemize many of the administrative costs associated with specific regulatory compliance activities. Multiplex officials estimated that for environmental compliance, about 60 to 70 percent of their costs were captured using the company’s financial records. Company officials also said they could not provide incremental regulatory cost data because the companies’ regulatory responsibilities were sometimes difficult to distinguish from their regular processes and functions. For example, officials from the glass company said regulatory responsibilities were woven into individuals’ jobs, and it was difficult to separate what was being done strictly for regulatory reasons. Officials from the tank car company said it would take a significant amount of time and resources to separate these compliance costs from their day-to-day operations costs. Officials from the petrochemical company said there is little incentive to isolate and uniquely monitor the explicit costs associated with new and ongoing regulatory requirements because they generally view regulations as nonrevenue-producing mandates. Although we did not ask the companies that participated in our review to quantify their indirect costs of complying with federal regulations, company officials provided a number of examples of those costs that they said their companies had experienced. The examples indicate that indirect costs can be substantial and are probably the most difficult types of costs to measure. The types of indirect costs that the companies provided included lost productivity, decreased competitiveness, lost business opportunities, delays in the expansion of new products or businesses, misallocation of resources, and delays in construction of new plants and/or equipment. Examples of indirect costs the companies cited included the following. • Officials from the paper company said that regulations tie up company resources and staff that could be better used in other ways, such as developing new products or processes. They said the company’s international competitiveness is also affected by regulations. The officials said that, contrary to popular belief, European countries are less regulated than U.S. companies, and as a result, U.S. companies are at a competitive disadvantage. • Representatives from the petrochemical company said that the company had curtailed or forgone facility expansions because of regulations that discourage voluntary emissions reductions at the plant site or emissions trading programs that are too administratively burdensome. Company officials also expressed concerns about regulations that restrict access to potential natural gas markets even though the increased use of natural gas is a cost-effective means of achieving emissions reductions. The officials also referred to the unavailability of cash flow to capture business opportunities because of the allocation of funds to compliance requirements. • An official from Minco, a federal subcontractor, said that regulatory activity tends to drive out other human resource-related actions the company would like to take. For example, she said that the company would like to offer more training and employee development but cannot do so because January and February are generally dedicated to development of affirmative action plans. Another Minco official said that the company could make more money as a prime contractor but intentionally remains a subcontractor because of the complexity of federal procurement regulations. • A Multiplex official commented on how responding to regulatory requirements causes a delay of management information and responsiveness to business-related needs. For example, the official said that a company management team’s required involvement with an 18-month IRS audit slowed company management’s response to internal requests and to other company business. He said a large company can easily absorb a certain level of recordkeeping requirements, but the same requirements can cause many problems for a smaller company like Multiplex. The assessment of benefits is of equal importance to the measurement of costs and other types of burden in assessing the impact of a given regulation or regulations in general. However, as previously noted, accurate measures of the benefits of regulations appear even more difficult to develop than cost measures. For example, in 1991 we reported that, although environmental controls have resulted in substantial and valuable benefits, assigning a monetary value to these benefits was much more difficult than estimating costs. Much of the literature deals with the assessment of costs of regulations, and relatively few reports and studies have addressed the benefits of regulations. In June 1995, Public Citizen, a national consumer advocacy organization, released a report that sought to document the benefits of federal health and safety regulations. The report recognized that the benefits of health and safety standards cannot always be measured in dollar terms, and it highlighted the difficulty involved in attempting to calculate the exact number of lives saved, injuries prevented, and costs averted by a regulation. The report also criticized cost-benefit analysis by maintaining that many of the variables used in the analysis are unquantifiable, and in many cases the primary source of cost data is from industry itself. Despite the concerns the businesses expressed about the costs of regulatory compliance, most of the company officials we interviewed generally recognized that regulations provide benefits to society as a whole, to certain groups and individuals, and even to their own businesses. Company officials said federal regulations had helped protect air and water quality, created safer workplaces, promoted fair competition, and improved both the manufacturing process and product quality. Specific examples of regulatory benefits the companies cited included the following. • Officials from the paper company said that compliance with federal regulations had helped to improve their manufacturing process. They said some of the dioxin regulations would make their paper manufacturing process more effective and less costly, even though short-term costs could be high. Company representatives added that solid waste regulations were leading the company to use chemicals that are not as hazardous. • Representatives of the hospital indicated that OSHA’s Blood-borne Pathogens Standard had helped reduce the number of needlestick injuries experienced in the hospital. They also said that the Clinical Laboratory Improvement Amendment regulations encouraged laboratories to look more closely at the quality of their work. • Officials from the glass company said federal regulations had created business opportunities for their company. They said the company created its environmental products and pharmaceutical services businesses to assist others in meeting their regulatory requirements of air pollution control and product safety testing. Company officials also said that federal regulations protected environmental quality, created safer workplaces for employees, and protected businesses from unfair business practices by their competitors. • A Minco official said she believes the company’s requirement as a federal contractor to create an affirmative action plan has aided in the employment of a diverse workforce and fair employment practices. Another official at the company said OSHA’s regulations have brought about a greater awareness of job safety for both management and employees. • Officials from Roadway, Zaclon, Bank B, and the glass company indicated that federal regulations provide a level playing field of uniform requirements for businesses. With this level playing field, federal regulations preempt multiple and often different state and/or local standards, making compliance easier and less costly. As pointed out by the glass company, it can be far more costly to track and comply with 50 different regulations than to comply with a single federal regulation in a given area. Our work suggests that measuring the incremental impact—direct costs, indirect costs, and benefits—of federal regulations on individual companies is an extremely problematical endeavor. Although the companies clearly experienced indirect costs associated with federal regulations and recognized that those regulations provided benefits to society and to themselves, our discussions with the companies also indicated that measuring indirect costs would be extremely difficult. We encountered a number of serious obstacles in our effort to assess even the most straightforward of such costs—direct incremental costs. Although all of the 15 companies participating in our review provided some data on their regulatory costs, none could provide comprehensive, verifiable measures of their direct incremental costs of complying with federal regulations. As shown below, the companies described several obstacles in the development of such information. • As discussed in chapter 2, the companies did not produce a comprehensive inventory of federal regulations applicable to their operations. For example, the companies found it difficult to distinguish federal regulatory requirements from those of other governmental jurisdictions and from their normal business practices. Therefore, there was no comprehensive basis for cost assessment. • Companies could not isolate incremental regulatory costs because they were unable to identify what costs would have been incurred in the normal course of business operations without federal regulations. • Companies’ financial information systems were not geared to identifying costs associated with regulation. These difficulties do not necessarily mean that regulatory costs are not substantial or that the measurement of those aggregate costs is impossible. However, they do suggest that serious conceptual and methodological questions need to be raised and answered before studies that attempt to measure total current regulatory costs are used to guide public policy. The following are examples of such questions. • Which regulations are included in the universe for which cost measures are developed? If federal regulations are the focus, how were they distinguished from those of other jurisdictions? • Are incremental regulatory costs being measured? If so, how did the researcher determine what businesses would have spent in the absence of regulations? • What financial records were used to substantiate the cost figures, and what assumptions guided the collection of the data? Users of studies of regulatory costs need to be aware of the inherent difficulties and assumptions involved in producing such measures. To this point, this report has attempted to assess the impact of federal regulations on selected businesses by focusing on aggregates—the total number of regulations applicable to a company and the aggregate burden (cost and other) of those regulations. Another way to understand the impact of regulations is to examine the concerns those businesses have about the particular regulations that comprise that aggregate. In a June 1994 study, we used this type of approach to obtain comments on a defined set of federal regulations. Employer and union representatives in selected businesses were asked about their experiences in dealing with 26 statutes and 1 executive order on workplace regulation, including the ADA, the Equal Pay Act, and the Service Contract Act. In summary, both groups generally supported the need for workplace regulations but voiced concerns about the operation of the overall regulatory process. For example, many of the employers said that certain paperwork requirements had questionable value. These employers also said that the regulatory approach used by many agencies was largely adversarial, characterized by poor communication, unfair and inconsistent enforcement, and vague laws and regulations. Both employers and union representatives called for agencies’ providing a more service-oriented approach to workplace regulation; improving information access and educational assistance to employers, workers, and unions; and permitting more input into agency standard-setting and enforcement efforts. In this study, we asked officials representing the 15 participating companies to identify the specific federal regulations that they considered most problematic for their organizations. We also asked those officials what government (Congress or federal agencies) and businesses could do to address either the problems they identified or the federal regulatory process in general. In our instructions to the companies, we defined “problematic regulations” as any federal program, regulation, or law that the officials viewed as causing their companies the greatest difficulty. We said a regulation could be considered problematic for a variety of reasons, such as being too costly, too vague, unnecessary, or duplicative. In contrast to the difficulties the companies experienced in compiling an aggregate list of regulations and determining incremental regulatory costs, all 15 companies provided examples of what they considered to be their most problematic regulations. In total, we received more than 100 such concerns from the participating companies. We developed written summaries of each of the companies’ concerns and verified the accuracy of those summaries with company officials. We then sent the verified summaries to the appropriate regulatory agencies for their review and comments. In some cases, the agencies said that they needed more information to allow them to respond to the companies’ concerns. For example, one agency said it needed to know the state in which a company was located so that it could be certain that the company’s concern involved a federal regulation and not a state or local regulation. In those cases, we attempted to obtain additional information from the companies or permission to disclose information we already had that could address the agencies’ questions. We were usually able to provide the agencies with the additional information they said that they needed. However, in some cases we could not provide the information because doing so could have led to the identification of companies that wanted to remain anonymous. In those cases, the agencies responded as best they could with the information that was available to them. By obtaining and presenting agencies’ responses to the companies’ concerns, we attempted to present a balanced picture of the regulatory issues involved. However, it is important to note the limitations of this methodology and presentation sequence. The companies were able to set the agenda by specifying the topics to which the agencies had to respond. Also, although agencies could question or dispute the companies’ concerns about regulatory issues, we did not give companies a comparable opportunity to respond to the agencies’ assertions. Lastly, agencies had the final word regarding the companies’ concerns, but this presentation should not be interpreted to imply our agreement with the agencies’ positions regarding these issues. The 15 companies’ regulatory concerns varied substantially. Many of these concerns were about specific federal regulations, but others focused on federal regulatory agencies’ actions and the interrelationship between regulations at other levels of government. After analyzing them, we grouped the companies’ concerns into the following 10 broad themes: (1) Compliance with a regulation was costly and/or those costs outweighed the benefits provided by the regulation. (2) The compliance costs associated with a regulation affected the companies’ competitiveness. (3) The regulation at issue was unreasonable (e.g., it was not scientifically based). (4) The requirements associated with a regulation were difficult to understand, either because of the technical language involved or because the requirements kept changing. (5) Certain regulatory requirements were unnecessarily rigid or inflexible. (6) The paperwork or process requirements associated with a regulation were excessive and costly. (7) The penalties imposed on companies for noncompliance were too severe. (8) Regulators were overly deficiency-oriented or had a “gotcha” enforcement approach. (9) Regulators lacked knowledge of industries and provided little assistance to businesses trying to comply with the regulations. (10) Regulations from different agencies or levels of government were poorly coordinated or duplicative. About half of the concerns that the businesses expressed included elements of more than 1 of these 10 themes. For example, 14 of the companies’ concerns indicated that certain regulations were both too costly and unreasonable. Similarly, a number of concerns involved both paperwork issues and regulatory costs. Some individual concerns included as many as five themes. Each of the 10 company concern themes is discussed below along with the associated comments from the regulatory agencies. Like the companies’ concerns, the agencies’ comments varied substantially. In some cases, the agencies agreed with the companies’ concerns and said that actions had been taken or needed to be taken to address those concerns. In other cases, the agencies disagreed with the companies’ portrayal of an enforcement action or of a regulation’s requirements. In still other cases, the agencies said the companies’ concerns were a function of the regulations’ underlying statutory requirements. Appendix II contains a sample of the companies’ concerns and the applicable agencies’ responses for each of the 10 themes. Company officials most frequently expressed concerns about the cost of complying with particular federal regulations. Representatives of 14 of the 15 companies mentioned this concern about at least one regulation. For example, one official said DOT-required hazardous materials training cost the company $475,000 annually. Another company official said that as a result of changes to CAA’s regulatory requirements, the cost of air quality testing, which was needed to get approval of a construction permit, increased from $10,000 to $30,000. In response to these concerns, the agencies most frequently said that the companies had overstated regulatory compliance costs. For example, EEOC and DOJ raised questions about the $750,000 Roadway said it spent to comply with ADA requirements. Both agencies said that practical experience to date indicates that the cost of ADA compliance is limited. EEOC cited a study commissioned by Sears, Roebuck and Co. showing that less than 3 percent of the accommodations that Sears made to comply with the ADA cost the company more than $1,000. In several cases, agencies said the costs incurred by the companies might be a function of the way that they complied with the regulations. For example, one company said that OSHA required them to replace certain electrical receptacle boxes with more expensive ones. In response, OSHA said the company only had to abate the electrical hazard, and it was up to the company to decide how to accomplish that goal. According to OSHA officials, the company could have found other ways to power its equipment or used other, less expensive receptacle boxes available for indoor use. OSHA officials added that their standard is consistent with the national electrical code, which is recognized by most authorities as a safe way to provide electrical energy to buildings. A number of companies also said that they believed the costs associated with compliance with certain regulations outweighed any regulatory benefits. Many companies said they found it difficult to see any benefits associated with the regulations they mentioned, either to themselves or to society in general. Other companies said they did not mind spending money to comply with federal regulations, but not when costs exceeded the benefits. For example, some companies said that although they had substantially met the goals set in particular regulations, total compliance would be extremely costly and would far outweigh any marginal benefits provided by those increased costs. In other cases, two companies mentioned having to retrofit machinery and/or train more employees than necessary to satisfy certain regulatory requirements—expenditures that they said were unnecessary and yielded no apparent benefits. Agencies responding to these concerns frequently said that the companies did not recognize the benefits that the regulations provided and/or that they overstated the costs. Another common agency response was that the company had misstated or misinterpreted the regulatory requirement. In several instances, the agencies said that the particular regulatory procedure mentioned by the company was required by law. For example, one company said that the nondiscrimination test IRS requires in the administration of the company’s 401(k) thrift savings plan was costly, and that the IRS requirement for a separate audit of the plan was an unnecessary expense. However, IRS said that both of these requirements were imposed by statute. In other cases, though, the agencies agreed with the companies’ cost concerns and said they had taken or were taking action to make regulatory compliance less costly. Examples 1 through 4 in appendix II illustrate companies’ concerns and agencies’ responses relating to cost and cost-benefit issues. Nine companies indicated that the costs associated with compliance with certain regulations were a disincentive to their investment or expansion decisions, or otherwise affected the companies’ competitiveness in the marketplace. Some of the companies said that particular regulations prevented them from expanding their operations because their compliance costs would increase disproportionately to the profits they expected to generate from the expansion. Other companies said that compliance costs imposed a potential liability that discouraged investment. Still other company concerns were that some regulations applied to only certain types of businesses, thereby giving a competitive advantage to similar businesses that were exempt from those regulations. In response, the agencies often acknowledged that regulatory compliance costs could affect companies’ competitiveness and/or flexibility of decisionmaking. However, they frequently said that the requirements in question were statutorily mandated. Several agencies also indicated that the regulations in question were necessary and yielded benefits that the companies did not mention. In several cases, the agencies said they were working to improve the operation of the regulations the companies cited as inhibiting competition, making them more consistent, flexible, or less burdensome. For example, one company said it would not purchase property that had previously been the site of industrial operations because of the potential cleanup liability under the Comprehensive Environmental Response, Compensation, and Liability Act. Thus, the company felt that its options for choosing new sites were restricted at a time when it needed property for business expansion. EPA responded to this concern by indicating that it has initiatives under way to encourage economic redevelopment through environmental cleanup. Also, EPA said it has encouraged redevelopment of these sites by reassuring prospective new owners that they may not have to face Superfund liability. Also, some agencies’ responses indicated that the way in which companies chose to comply with the regulations could have affected their costs. For example, Bank C said that it had to create 15 new forms as a result of a regulatory requirement, but FRB said that there was no requirement that bank personnel had to complete forms to accomplish the goal of the regulation. Examples 5 and 6 in appendix II illustrate two of the companies’ concerns and agencies’ responses relating to the effect of regulatory costs on competitiveness. Twelve of the 15 companies said that certain regulations they dealt with were unreasonable because they were either (1) not based on sound scientific research, (2) outdated, (3) unlikely to achieve their intended goals, or (4) unreasonable for some other reason. For example, officials of the paper company objected to a DOT requirement that each of the company’s several hundred locations submit drivers’ logs and other documents to the company’s headquarters to help DOT’s review. The company thought it was impractical and unnecessary to maintain all of the drivers’ logs in one place. In another example, the paper company alleged that title V of CAA requires the regulation of extremely low levels of emissions, and that the company was required to obtain a permit for methanol emissions at the company’s fence line that are no more concentrated than in a person’s breath. The regulatory agencies frequently disagreed with the companies’ characterization of their regulations as unreasonable. In about half of these areas of disagreement, the agencies said that the regulations did, in fact, have a rational basis and/or provided benefits the companies did not acknowledge. For example, EPA officials said that the paper company’s concern creates the misleading impression that paper mills are subject to title V only for low levels of methanol emissions when, in fact, large emissions of other pollutants would easily justify the need for a title V permit even if the mill had no methanol emissions. In the other half of these cases, the agencies said that the companies had mischaracterized the regulation or the incident involved. For example, hospital officials questioned the reasonableness of what they described as a revised Federal Aviation Administration (FAA) rule that shortened the shift lengths of helicopter pilots. However, FAA said that the shift length rules had not recently changed. In several other instances, the regulatory agencies said that the regulatory provisions the companies characterized as unreasonable were required by law. For example, HUD officials said that disclosure requirements that bank officials said were unreasonable were established by Congress in specific provisions of the Real Estate Settlement Procedures Act (12 U.S.C. 2601 et seq). However, in some cases the agencies agreed with the companies’ concerns about the reasonableness of regulations, and said that they or Congress had taken or were taking steps to minimize the problems the companies had experienced. Examples 7 through 10 in appendix II illustrate companies’ concerns and agencies’ responses relating to the reasonableness of regulatory requirements. Most companies said they did not understand certain regulatory requirements because they were vague or complex. As a result, the companies said they had difficulty determining whether regulations applied to them and, if so, what they needed to do to be in compliance. Some companies cited confusing, ambiguous, or conflicting terminology used in the regulations themselves or on the required forms. Some companies said they were not able to obtain clarification of the regulations’ requirements from agency staff. Other companies said they had to hire outside consultants to explain the requirements or complete the forms, but even those experts are sometimes not able to help them. For example, paper company officials said OSHA regulations require their paper machines to have mechanisms that will stop the machines “quickly,” but the rule does not define what “quickly” means, and experts do not agree on a single definition. In many instances, the regulatory agencies agreed with the companies that the regulations cited in their concerns were vague and/or complex. The agencies often said that some kind of action had been taken or was being taken to clarify the regulations and make them less complex. For example, regulators frequently cited direct compliance assistance, written guidance, simplified processes, toll-free telephone numbers, computerized bulletin boards, and presentation of examples within the regulations themselves as ways that they have tried to make the requirements more understandable. In several instances, the regulators said the complexity of the regulations in question—particularly IRS regulations—was a function of the law or actions taken by Congress. In other cases, the regulators indicated that the regulations were complex because of the inherent complexity of the subject matter being regulated. Companies also said they did not understand regulatory requirements because of frequent changes to the regulations, thereby making it difficult to stay up to date and know what was required to be in compliance. For example, officials from the hospital said that it was difficult to keep pace with the frequently changing Medicare and Medicaid billing rules, which caused the hospital’s computer programmers to spend numerous hours attempting to update the automated patient billing system. In virtually every case where a company complained about frequent regulatory changes, the agencies said that the changes were caused by congressional, not agency, action. For example, in response to bank officials’ concerns about frequent changes in the tax code, Department of the Treasury officials indicated that the changes were initiated by Congress. Treasury officials said they had urged Congress to stabilize the tax code so that taxpayers and their advisers could understand its requirements. FDIC officials said that the number of changes and the level of detail in their call reports were driven in part by statutory requirements. Examples 11 through 13 in appendix II illustrate companies’ concerns about regulations that were vague, complex, and/or frequently changing and agencies’ responses to those concerns. Many companies said some federal regulations that they have to comply with were unnecessarily inflexible or rigid. The companies expressed frustration with the same standards or regulations being applied to all companies, locations, or situations without any consideration to other factors that they believed should be taken into consideration. Some companies felt that in certain regulatory situations, a variety of factors should be considered in determining whether a company should have to comply with that regulation. Examples of the factors the companies believed should be considered included the number of employees in a company, the extent to which a company uses a particular chemical, and the amount of pollutant discharged as the result of a company’s process. For the most part, the agencies disagreed that the regulations or the situations the companies cited were examples of unnecessary inflexibility. In some of the situations the companies described, the agencies said that the requirements were justifiable regardless of other factors that the companies believed should be taken into consideration. In other cases, the agencies said that the regulations the companies cited already had some flexibility built in or that the companies had misunderstood the process or the regulation cited. In some cases, the regulators said that there was no need for the regulations to be more flexible because the standards the company cited were not even applicable to the situation the company described. Examples 14 and 15 in appendix II illustrate companies’ concerns and agencies’ responses about regulations that are considered to be inflexible. Officials from 14 of the 15 companies said certain regulations’ paperwork or other procedural requirements were excessive. The paperwork that company officials cited as problematic included (1) forms or reports that had to be periodically submitted to federal agencies and/or kept for their own records and (2) permit applications that had to be submitted to an agency to obtain approval for certain company activities. Most of the paperwork or process costs the companies mentioned related to labor costs associated with having their employees complete the required forms or reports. For example, the hospital said it had to hire a consultant for $50,000 to help complete the annual Medicare cost report. Two of the companies indicated that they would not mind bearing the expense of preparing regulatory reports if they felt the reports were actually used by federal agencies. For example, Bank B considered the reporting requirements under the Bank Secrecy Act to be of negligible value to law enforcement agencies. Bank officials said they had seen little evidence of law enforcement agencies’ using the information and few prosecutions resulting from information in these reports. Other companies’ concerns focused on the costs associated with having to prepare similar reports for regulators at the federal, state, and/or local levels. Agencies responding to these concerns most frequently said they agreed that the paperwork or procedural requirements could be expensive. They also said their agencies had taken or were taking action to address the companies’ concerns. In several other instances, the agencies said that the costly paperwork or procedures were required by law. For example, in response to the hospital’s concern about the annual Medicare cost report being a burden to prepare, HHS officials said that the Social Security Act requires the agency to maintain a system of cost-reporting for prospective payment system hospitals, including annual information to settle costs associated with health care services rendered to Medicare beneficiaries. In about half of the other responses, the agencies said that the companies had misinterpreted or misstated the paperwork or procedural requirements; therefore, they were incurring unnecessary expenses. For example, although one company complained that they had to keep certain employee safety training records “forever,” OSHA said that no such employee training records were required and, therefore, no retention requirement existed. Examples 16 through 19 in appendix II illustrate companies’ concerns and agencies’ responses relating to paperwork and process issues. Many companies expressed concerns that the penalties imposed on them for noncompliance with regulations or their requirements were too severe. For example, the glass company said sizable penalties had been imposed on them for procedural “mistakes,” such as not filing pension-related paperwork on time. Another company said it was fined several hundred thousand dollars for not obtaining a federal wastewater discharge permit. The company thought the penalty for this offense was too severe given that the company had a state permit that it believed was sufficient. In addition, a Metro Machine Corporation official said OSHA currently holds companies, rather than individual employees, accountable for violations caused by employee negligence or willful removal of company-installed safety devices. He said OSHA should differentiate between corporate negligence and employee responsibility. In response to these concerns, the agencies most frequently said that the penalties the companies cited would be imposed only in the most egregious circumstances. For example, in response to the Metro Machine Corporation concern, OSHA said that when it conducts an inspection and determines that a company’s management is attempting all reasonable steps to comply and get employees to comply but the employees are systematically refusing to comply with safety and health standards or rules, OSHA will excuse the employer from a violation. OSHA officials added that the Occupational Safety and Health Act does not permit citing employees for violations. In another response, EPA said two companies’ concerns about possible imprisonment of company officials for failing to disclose certain information would be imposed only in instances where those officials had knowingly falsified information or willfully failed to provide the required public notice of the release of a hazardous substance. Agencies also said that the penalties the companies complained about were established in the underlying statutes and, in other cases, that the agency had taken or would take action to address these concerns about the severity of certain regulatory penalties. Examples 20 and 21 in appendix II illustrate companies’ concerns and agencies’ responses to penalty issues. Officials from more than half of the companies cited incidents in which regulators evidenced a “gotcha” manner or were more interested in finding companies in noncompliance with regulatory requirements than helping companies comply with the regulations. For example, one company said an IRS official unexpectedly visited its facility and, in a “nasty” manner, threatened to close the company down if the company did not immediately remit taxes that were reportedly unpaid. In another instance, a company said that EPA could initiate enforcement actions even when companies self-report deficiencies. Company officials also cited examples in which regulators were more focused on procedural or administrative issues (e.g., filing timely reports) than on whether the objective of the regulation was being achieved (e.g., less air or water pollution from their manufacturing processes). The agencies responded to the companies’ comments in a variety of ways. The agencies often said that (1) their enforcement approaches were reasonable and consistent with their policies, (2) the companies had mischaracterized the incidents or the rules involved, or (3) they have or will take action to minimize these problems. For example, OSHA said that it had used citations and penalties as workforce performance measures in the past, but said it has now “put a stop to that practice.” EPA said that in 1995, it revised its policy to generally reduce or eliminate penalties when violations are self-disclosed and corrected. In response to the company’s concern about the IRS employee’s demand for immediate payment, IRS said it did not approve of employees who do not follow IRS procedures that require employees to do their jobs in a professional, ethical, and fair manner. Examples 22 and 23 in appendix II illustrate the types of concerns companies had about how agencies enforce their regulations and the agencies’ responses to those concerns. Some companies said regulators were uninformed about the regulations they enforced and did not understand the business practices of the companies they regulate. For example, one company said IRS auditors who conducted an audit of their company in 1994 were not knowledgeable about business accounting practices or IRS rules. Also, some companies said regulators were not very helpful when the companies sought assistance. For example, although the tank car company repeatedly tried to obtain clarification from EPA about the meaning of the term “approaching atmospheric,” the company said it was unable to get any assistance from EPA. In another case, officials from the fish farm said they had difficulties getting assistance from DOL on how to interpret the Family and Medical Leave Act (FMLA) regulations when an employee is on leave under FMLA and does not intend to work after the leave period. Relatedly, officials from some companies said agencies do not always provide companies with sufficient opportunities for input into the rulemaking process or adequately consider the comments they receive during that process. For example, Bank B said that although the bank provided comments on Regulation C and Regulation DD, it did not believe the Federal Reserve addressed its concerns before finalizing the two regulations. Agencies responded in various ways to the companies’ concerns about regulators’ lack of knowledge and assistance. In a few cases, the agencies indicated efforts were under way to improve their staffs’ knowledge of the industries they regulate. For example, IRS said it was working to develop a highly skilled frontline workforce that is more knowledgeable about different industries. In a response to a concern about lack of assistance, OSHA said it has implemented a number of information-dissemination projects and plans to undertake new initiatives to improve the availability of safety and health data to the public. In response to the rulemaking concerns, agencies said companies usually get the opportunity to provide comments as a rule is being developed. Also, one agency said that while all comments received are considered, these comments cannot always be incorporated in the final regulation. For example, OSHA said it had been working with stakeholders to identify the most pressing new priorities for agency action and had stepped up its efforts to involve business and labor in the entire regulatory process. Examples 24 and 25 in appendix II illustrate companies’ concerns and agencies’ responses about regulators’ level of knowledge, the rulemaking process, and the availability of assistance from regulators. Some of the companies described what they believed were conflicting regulatory requirements from different federal agencies or, in some cases, from the same agency. For example, officials from the paper company claimed that sections of OSHA’s pulp and paper standards (29 C.F.R. 1910.261) conflicted with other OSHA regulations, leaving company officials confused about what to do. Officials from this company also said DOT and OSHA had different and conflicting standards for defining corrosive materials. We also heard concerns about overlapping and duplicative regulations between federal and state or local agencies. Multiplex said that EPA required the local sewer district to test the company’s sewer effluent—the cost of which was charged to the company—and then the local sewer district required Multiplex to perform the same tests. In response to the companies’ concerns about coordination and duplication issues, several of the agencies’ responses indicated that action has been or will be taken to remedy any further or potential confusion about regulatory requirements. For example, OSHA said it had been working closely with EPA to develop uniform process safety management standards to protect workers from accidental chemical releases. Also, some agencies indicated that the companies had mischaracterized the regulations or other factors in the companies’ concerns. In response to a company’s concerns about the overlap between FMLA and the ADA, DOL said the laws contain differing employee protections that serve distinctly different purposes. However, OSHA officials also said that on July 22, 1996, OSHA published proposed regulations to eliminate duplicate or redundant standards from its rules. Examples 26 through 29 in appendix II illustrate the companies’ concerns and agencies’ responses about coordination and duplication issues. We also asked companies for suggestions on what government and businesses can do to address their regulatory concerns. Although most of the companies’ proposals focused on actions that they believed regulatory agencies and/or Congress should take, they also suggested some steps businesses could take to make federal regulations less problematic. Some of the companies said they simply want fewer regulations, or at least a halt in the growth of regulatory requirements. They suggested that the federal government could accomplish these goals by taking the following actions: • eliminate one old regulation for every new regulation issued, • review existing regulations for their relevance, and • eliminate paperwork and other regulatory requirements that are not related to the intent of the underlying statute. Officials from some of the companies said that if the federal government cannot reduce the total number of regulations, at a minimum, they wanted an assurance that the benefits of compliance justify the costs. To do this, officials suggested that regulators do cost-benefit analyses before issuing regulations. Another official said cost-benefit analyses could address the issue of “bad science” underlying some regulations, and could result in agencies’ implementing only regulations that have proven benefits. However, another company official said he had difficulty envisioning how a sound cost-benefit analysis could work. He suggested that each regulation should have a sunset date and, before reauthorization, the responsible agency should determine whether the regulation was achieving its original intent. “. . . government should move away from the current ’specification based’ regulatory process and toward a new approach in which government and business jointly establish performance-based environmental, health, and safety standards. Government and business should both be accountable for achieving measurable, quantifiable objectives. Goals would be accomplished in a stepwise fashion, improving cost effectiveness by allowing parties to learn from what works. Government and business should work cooperatively and share the burden for obtaining information and demonstrating results. A peer review procedure could be used to maintain the quality and integrity of the process. Over time, the process would force industry and regulators toward low-cost, high impact solutions with proven effectiveness. Market-based incentives could be widely used, as there are currently few incentives for business to remedy the environmental impact of its operations. The new paradigm would allow those closest to a problem to solve it in the most cost-effective manner . . . .” Officials from another company agreed with this assessment, adding that the federal regulatory process could be improved if laws were developed that address general goals and objectives and were of long duration. The officials suggested developing long-term strategies that make laws less susceptible to short-term political whims. Another company official said Congress’ tendency to be prescriptive and specific in writing legislation is driven even further by lobbyists on both sides of the issues. The official said that agencies would do a better job of writing sensible regulations if the legislation were less constraining. Many of the company officials said that regulators should offer companies more assistance as they try to comply with federal regulations. Specifically, the officials said that regulators could • adopt a partnership approach with companies to help them to comply with • serve as consultants to companies, • provide companies with compliance training, • support companies that make reasonable attempts to comply with • give companies a chance to correct regulatory violations before being • hire credible and technically competent staff, and • use review commissions to assist businesses in compliance with regulations. Although officials of several companies clearly believed that the federal government could make changes to improve the regulatory environment, some officials also believed that companies had a role to play in that regard. They suggested several actions that businesses could take to make regulations less problematic, including the following: • devote more time to commenting on proposed regulations during the • devote sufficient resources to becoming more knowledgeable about • ensure that company management and employees are trained to properly • participate in trade and professional organizations that interact with Congress and federal agencies; • ensure that top management supports a regulatory compliance strategy; • modernize and make better use of information processing systems; and increase their employees’ awareness of the seriousness of complying with federal regulations, the potential for problems related to noncompliance (e.g., an increase in job-related injuries), and the fines that could be imposed on them or the company for failure to comply. Many of the suggestions from the companies were consistent with the agencies’ regulatory improvement goals. For example, several of the agencies said they have, or were planning to implement, active outreach programs that disseminate information to the companies through written communications, seminars, toll-free telephone numbers, and computer bulletin boards. In addition, several agency officials said they are in the process of systematically eliminating outdated and impractical regulations. Two agencies said they were shifting to a results-oriented focus because in the past their agencies focused too heavily on processes and activity. DOL officials said they intend to make greater use of negotiated rulemaking—a process in which representatives of the government and all interested parties, including employers, actually draft the proposed rule for public comment. The 15 participating companies provided us with a lengthy and varied list of regulatory concerns, the most common of which involved the cost of regulatory compliance. The companies also frequently said that federal regulations were unreasonable or inflexible, paperwork was excessive, regulatory requirements were difficult to understand, and regulators had a “gotcha” enforcement approach. Many of the regulatory agencies indicated that they were aware of the companies’ concerns and, in a number of cases, the agencies said that they were taking or had already taken action to alleviate the problems. In several other cases, the agencies said that the companies’ concerns were a function of the statutes underlying the regulatory requirement. However, in many instances, the agencies disagreed with the companies’ comments, frequently saying that the companies did not recognize the benefits of the regulations or the companies mischaracterized, misstated, or misinterpreted the regulations involved. The companies’ concerns and the agencies’ responses indicate that communications between the companies and the agencies are not always effective. Companies do not seem to have enough information about their regulatory responsibilities, and they may be reluctant to seek that information from regulatory agencies. Agencies, on the other hand, have an array of information about their regulatory requirements; however, they do not appear to be getting the information to companies in such a way that the companies understand what regulations are applicable to them and how to comply with those regulations. The companies and the agencies had several suggestions to alleviate this communication gap and improve relations between them. Some of these suggestions were consistent with agencies’ regulatory improvement goals. For example, the companies indicated they wanted more information made available to them about regulatory compliance. Meanwhile, the agencies said they were using or were planning to implement a number of outreach programs, including seminars, computer bulletin boards, and toll-free telephone numbers. However, the degree to which these various informational mechanisms will improve the flow of communication will depend to at least some extent on whether they are integrated, easy to use, and provided in a manner that businesses are willing to use them. Government regulation, particularly at the federal level, has long been the subject of public debate. At times, that debate has been extremely contentious, with opponents and defenders of federal regulatory policy staking out very different positions. Opponents contend that some federal regulations are not needed and those that are needed often have become too burdensome. Therefore, opponents believe agencies’ regulatory authority should be limited and closely scrutinized. Those defending federal regulations do not want to unnecessarily impose burdens but contend that federal regulatory standards are needed to provide certain societal benefits, such as safer transportation, cleaner air and water, greater workplace safety, and protection for some individuals and groups. Nevertheless, a consensus has emerged within government that the federal regulatory process needs reexamination. Both Congress and the executive branch have initiated efforts to improve that process. A number of legislative proposals have been introduced in the 104th Congress to change the federal regulatory process. Several of those changes have been enacted, including new paperwork reduction goals and new judicial review processes. Congress has even made itself part of the regulatory review process by instituting an expedited process to reject agency rules that it finds objectionable. The administration’s regulatory reform efforts have addressed a number of areas as well, including eliminating and revising existing regulations; changing the performance measures of agencies and regulators to focus on results, not process and punishment; and working with the regulated community in the development of new regulations. A great deal of the debate about federal regulation has centered on whether the burdens associated with federal regulations outweigh the benefits that those regulations are intended to provide. A number of attempts have been made to measure the burden of federal regulations. Some of these measures are only indirect indicators of regulatory burden (e.g., the number of regulators and the number of pages in the CFR). Other, more direct measures are of questionable validity (e.g., paperwork burden-hour estimates). For example, one study of the direct cost of federal regulations estimated that their cost to the economy would be $607 billion in 1995. However, the validity of this estimate has been questioned by economists and others. Other studies of regulatory costs within particular sectors of the economy have been similarly criticized.The benefits of federal regulations are generally regarded as even more difficult to measure than regulatory burdens, so measures of those benefits may also be of questionable validity. In this review, we did not try to measure the overall burdens or benefits of federal regulations. We focused our analysis on a limited number of businesses so that we could better understand the issues and the variables involved in the federal regulatory process. Specifically, we attempted to develop information on the impact of federal regulations on selected businesses by determining (1) which federal regulations the businesses participating in our review and relevant federal agencies believed were applicable to those businesses, (2) the impact (particularly cost but also other effects) the businesses believed those regulations had on them, and (3) the regulations those businesses considered were most problematic and relevant federal agencies’ responses to those concerns. Although our objectives were modest in comparison to those of many previous studies of regulations, we experienced a number of difficulties in conducting this review. First, most of the interest groups we contacted—both critics and defenders of federal regulations—did not provide the names of businesses with whom we could discuss these issues. Some of these groups cited concerns about confidentiality or the priority of other business as the reasons why they would not provide nominees. Second, most of the 51 businesses that we were able to identify and contact through other means refused to participate in our study, frequently citing time and resource constraints. Some of these businesses had publicly criticized federal regulations or regulatory processes. Third, many of the 15 businesses we contacted, which had agreed to participate in the review, did not provide information that they initially indicated that they would be able to provide. For example, although nearly all of the businesses said they could develop a list of applicable federal regulations, none of the businesses provided a complete listing. None could provide comprehensive data on the incremental costs of regulations. Our experience and the information we were able to collect led us to several conclusions. First, we believe that comprehensive, empirically based data about the cost of federal regulations to individual businesses do not readily exist and could not be developed without a great deal of time and effort on the part of both the regulators and the regulated community. Without such data, the cost of regulations to a business or the economy as a whole can only be roughly estimated. Second, the agencies also said that they are reexamining at least some of the federal regulations and processes that businesses found most problematic. If so, the businesses may feel some reduction in the burden associated with those regulations in the future. Third, we believe that communication between businesses and federal regulators about which regulations are applicable to particular businesses and how to comply with specific regulatory requirements is not always effective. Finally, although the agencies frequently said that problematic regulations were statutorily driven, some Members of Congress believe agencies sometimes establish regulations that go beyond the intent of Congress. This suggests that an opportunity exists for improved communication between Congress and federal regulators—communication that may occur through recently enacted congressional regulatory review procedures. As previously noted, the burden associated with federal regulations is generally considered to be easier to measure than the benefits of regulations. Within the burden category, direct regulatory costs are generally regarded as easier to measure than other types of burden (e.g., negative effects on competitiveness or productivity). Conceptually and logistically, it would seem to be easier to calculate regulatory costs with regard to a single business than for an entire industry or the economy as a whole. However, our work suggests that it is extremely difficult to develop a comprehensive, data-based measure of direct incremental regulatory costs, even for an individual business. The first step in determining the cost of federal regulations to a business is to identify all of the regulations applicable to that business. Any comprehensive tally of a business’ regulatory costs will only be as complete as its list of applicable regulations. Although nearly all of the businesses we contacted said they could develop a complete list of applicable federal regulations, none did so. As we discussed in chapter 2, development of such a list is very difficult, requiring a sophisticated understanding of the circumstances in which federal regulations apply as well as a detailed understanding of a company’s business processes and products. The businesses we contacted generally did not have prepared lists of applicable regulations, and most found it difficult to distinguish between federal, state, and local regulations or said they did not have the time or resources to develop such a list. Although it may seem logical to assume that regulatory agencies would be able to easily develop a list of applicable regulations, in many cases the agencies would have to gather detailed information about the companies’ business processes and products to determine regulatory applicability. Another step in determining the cost of regulations to a business is to obtain data on its expenditures that are directly traceable to federal regulatory requirements and estimate its incremental regulatory costs. Because most businesses would have incurred some expenses related to regulatory goals, such as a safe workplace or the prevention of environmental damage, even if no regulations existed, the incremental measure is the most appropriate gauge of a company’s regulatory costs. To identify their incremental regulatory costs, businesses must subtract expenses they would have incurred in the absence of regulatory requirements from their total expenditures in such areas as worker safety or environmental protection. However, the businesses we contacted did not collect or retain data on their incremental regulatory compliance costs, probably because there is no business reason for them to do so. The businesses also indicated that they could not develop incremental cost data because they could not say what actions they would have taken or what expenses they would have incurred in the absence of regulatory requirements. In summary, determining the actual cost of federal regulations to a single company or the economy as a whole requires data—data that the companies we visited did not have and that our work showed would be extremely difficult for them to obtain or develop. The universe of regulations for which cost data should be gathered was difficult for the companies to identify and incremental cost data could at best be roughly estimated because the businesses could not determine what expenses they would have incurred if current federal regulations did not exist. We believe that these problems are unlikely to be unique to the companies we visited. Therefore, unless the breadth of companies’ regulatory responsibilities is made clear and the incremental costs of regulations can be accurately gauged, measures of the cost of regulations to a company or the economy as a whole should be viewed as estimates, not precise measures of regulatory burden. The more these estimates rely on empirical data and sound assumptions, the greater value they hold for decisionmakers. Public policymakers should use regulatory cost estimates only with a clear understanding of their underlying conceptual and methodological bases. The agencies’ responses to the companies’ regulatory concerns also indicated that many of the regulations or regulatory processes underlying those concerns were being reviewed or had recently been changed. Listed below are some agencies’ responses to companies’ concerns. • EPA said it had proposed allowing oil recovered from collocated and/or commonly owned organic chemical plants to be exempt from its RCRA hazardous waste regulations. • Treasury officials said their agency was issuing a proposed regulation that would dramatically reduce the reporting obligations of banks under the Bank Secrecy Act.• DOL, IRS, and PBGC said that they had made a number of regulatory changes to make it easier for businesses to establish pension plans. They also said that new pension simplification proposals announced by the President in June 1995 would, if enacted, simplify the rules even further.• EPA said it was reexamining its data needs and ways to improve its data gathering systems for the RCRA hazardous waste management program. The agency also said that, for certain companies, it had completed a number of actions that significantly reduced the reporting burden associated with the EPCRA Form R report, thereby reducing companies’ reporting requirements from nine pages to two pages. • EPA said it has published guidelines for reducing or eliminating penalties when violations are self-disclosed and corrected. EPA also said that its policy is to generally provide penalty reductions for such matters as good faith efforts to comply, ability to pay, and other factors. The number of times that the agencies indicated that they were taking action or had already taken action regarding areas of concern to the companies suggests that a variety of regulatory reform initiatives are under way within the federal government. As noted in chapter 1, every president in recent years has attempted to reform the regulatory process, and both the Clinton administration and Congress have recently taken a number of reform actions. The administration’s National Performance Review and other initiatives in this area, such as those aimed at eliminating certain reporting requirements and reducing penalties for self-disclosed violations of certain regulations, are part of this larger effort. Also, Congress’ recent assumption of a role in the review of proposed regulations also presages possible modification of regulations that Congress concludes are uneconomic or otherwise objectionable. We have not examined the initiatives the agencies described to determine whether they have been implemented or whether they will actually afford the businesses we talked with in this study the kind of regulatory relief they sought. If the changes the agencies described are made, the businesses may see a reduction in at least some of the regulatory burden that they viewed as most problematic. The information that we collected regarding applicable federal regulations and what the businesses viewed as their most problematic regulations strongly indicated that communication between businesses and federal regulatory agencies has not always been adequate to meet their respective needs. Both businesses and agencies need information to determine which regulations are applicable to particular companies—information that both parties said would require substantial time and resources to obtain. However, agencies have the information that businesses need, and vice versa; what seems to be lacking is an effective exchange of that information. Poor communication can also lead to businesses’ misunderstanding regulatory requirements, which can in turn lead to compliance problems or unnecessary expenditures. Finally, opportunities appear to exist for improved communication between Congress and federal regulatory agencies regarding the consistency of regulations with their statutory underpinnings. Recently enacted regulatory review procedures may provide the vehicle for that communication. The information that we obtained from federal regulatory agencies on the factors that determine the coverage of their regulations and on the regulatory compliance responsibilities of two of the companies in this review—Minco and Zaclon—clearly indicated that different businesses can have substantially different compliance responsibilities. Most of the agencies said that companies’ compliance responsibilities are highly situational, dependent on such factors as the companies’ size, location, and decisions they make in the course of conducting their business (e.g., whether to have an underground storage tank or to have a qualified retirement plan). As a result, most of the regulatory agencies we contacted said they needed a great deal of information about a business to identify the regulations applicable to that business. Businesses also need a great deal of information about the factors that trigger regulatory coverage to identify their own regulatory compliance responsibilities. However, both the agencies and the businesses told us they could not devote the time and resources needed to develop the information they need to make those determinations. Interestingly, each party in the regulatory process already has the information the other party needs. Agencies know about their regulations and what characteristics of companies can determine the applicability of those regulations, but are unfamiliar with individual companies’ operations and, therefore, would need to expend substantial amounts of time and energy learning about them to provide regulatory counsel. Businesses know how their organizations are configured regarding relevant regulatory determinants, but they do not always understand those determinants and, therefore, may not know about all of their regulatory compliance responsibilities. More effective communications could help bridge the informational gap between businesses and federal regulatory agencies and, as a result, could help achieve the agencies’ regulatory goals. Federal agencies’ comments regarding the companies’ most problematic regulations reinforced our conclusion regarding the adequacy of communication between those agencies and companies about regulations. In a number of cases, the agencies said that the companies had misstated or misinterpreted the statute or regulatory requirement involved. In some of those cases, the agencies said the companies were incurring unnecessary expenses because they had misconstrued the requirements or had taken steps that the regulations did not necessarily require. Listed below are examples of inadequate communication between the agencies and companies. • Officials from Bank B said that some regulations required banks (but not investment firms) to disclose the risks that consumers face regarding certain investment products, and that about a quarter of the advertising time Bank B purchased was used to publicize these risks. However, FDIC officials said there is no regulation requiring the disclosure of investment risks when advertising nondeposit investment products. • Hospital officials complained about a “costly rule change” that limits helicopter pilots to 12-hour schedules during a 24-hour period, thereby forcing the hospital to hire two additional pilots at a cost of $100,000 per year. However, FAA officials said that the duty time rules had not recently changed and that no 12-hour shift limit existed. • Officials from the paper company said OSHA lead exposure standards require even routine maintenance workers to put on personal protective equipment and be “fit tested,” a process the company said was extremely expensive. However, OSHA said the regulations the company cited did not apply to the type of routine maintenance activities the company described. • Bank B officials said that BSA requires banks to complete a report on cash transactions of $10,000 or more while the depositor is still in the bank. However, Treasury officials said that once the bank has verified certain basic information about a customer, it can rely on this information in the future and need not require the customer to remain on-site each time a reportable transaction is conducted. • Bank A officials complained about an FDIC requirement that all banks—even small ones—should have a detailed contingency plan. However, FDIC officials said the bank was referring to a policy statement by the Federal Financial Institutions Examination Council, not a statutory or regulatory requirement. The FDIC officials said the policy statement sets forth areas for management to consider when developing a contingency plan but sets no requirements. Therefore, communication problems in the regulatory arena can result in misunderstandings of responsibilities and, ultimately, compliance problems and unnecessary costs. One of the reasons that government regulators and businesses have not always communicated effectively may be the nature of agencies’ regulatory informational mechanisms. A business’ regulatory compliance responsibilities can originate from any of the several dozen federal regulatory agencies, each of which separately provides information about its own regulatory programs. Therefore, although the agencies have a dizzying array of brochures, toll-free numbers, and other methods to inform businesses of their regulatory requirements, a business attempting to determine its governmentwide compliance responsibilities may have to contact each agency to obtain this information. In some cases, multiple agencies have responsibility for implementing a single statute, with each agency specifying its own functions and requirements. Regulatory compliance responsibilities may also differ within a particular agency. For example, EPA regulatory requirements originate from several different program units within the agency (Air and Radiation; Water; Solid Waste and Emergency Response; and Prevention, Pesticides, and Toxic Substances). Each of these units has its own informational mechanisms. EPA has established an Office of the Small Business Ombudsman and compliance assistance centers for certain industry sectors for the purpose of consolidating information from these different program offices. However, we did not evaluate the effectiveness of the Office or centers to determine whether they eliminate the need for a small business or a company within a covered industry to contact the program units directly. Also, these offices by definition do not cover large businesses or companies in industries not covered by the centers. We have not thoroughly analyzed agencies’ regulatory informational mechanisms. However, the information that we obtained in this review suggests that the federal government’s overall approach to the dissemination of regulatory information is fragmented and may be contributing to ineffective communication between regulatory agencies and the business community. We recognize that there is a natural tension that exists between regulators and those in the regulated community and that no amount of information or communication will completely eliminate disagreements and compliance problems. However, a better understanding of which regulations are applicable to a business and the requirements those regulations impose on the business is fundamental to an improved relationship between these parties. Agencies have recently taken steps to make information about their regulations more centralized and accessible to businesses. For example, the U.S. Business Advisor and the Asbestos Advisor programs are designed to make it easier for businesses to determine their federal regulatory compliance responsibilities. Other steps agencies have taken to be more “user friendly” include (1) the previously mentioned EPA Office of the Small Business Ombudsman and EPA’s compliance assistance centers and (2) OSHA’s consultation program, which offers free, on-site expert assistance to small employers in all 50 states. Congress has also attempted to make information more available to businesses. The Small Business Regulatory Enforcement Fairness Act requires agencies to, among other things, publish “small entity compliance guides,” explaining the actions a small business is required to take to comply with a rule or group of rules. These kinds of efforts may help improve communication between federal regulatory agencies and regulated businesses. Although virtually all regulations have some kind of statutory basis, the extent to which particular regulatory requirements are driven by the underlying statutes varies. Some statutes grant agencies the authority to issue rules within broad parameters whereas other statutes provide agencies with little discretion regarding what should be regulated and how the regulations should be developed and implemented. The federal regulatory agencies responding to the businesses’ concerns about problematic regulations frequently said that the specific requirements the businesses were concerned about were statutorily driven. Listed below are examples of statutorily driven business concerns. IRS officials said that the requirements (1) to conduct a nondiscrimination test in the administration of a 401(k) thrift savings plan and (2) to audit the plan were required by statute rather than by IRS regulations.• HCFA officials said that the frequent changes in Medicare and Medicaid billing rules were, in a number of situations, “due to enhancements or changes made by Congress.” • FDIC officials said that the level of detail and the number of changes in its call reports were driven by, among other things, statutory requirements. • EEOC officials said that its record retention requirements vary because they are tied to the different discrimination complaint filing periods established in each civil rights statute. We did not review the regulations and statutes that the agencies cited to determine whether the regulatory provisions of concern to the companies are required by the underlying statutes. If those provisions are required by the statutes, agencies will not be able to revise them significantly without changes in the underlying legislation. As previously noted, several agencies said they were recommending statutory changes to address some of the companies’ concerns. In doing so, agencies can communicate to Congress the degree to which their regulations are required by the statutory language that Congress enacted. However, some Members of Congress clearly perceive that federal regulatory agencies have sometimes established regulatory requirements that go beyond the intent of Congress when it passed the underlying statutes. This perception in part led to the establishment of expedited congressional regulatory review procedures through the Small Business Regulatory Enforcement Fairness Act of 1996. If the regulatory provisions that are of concern to companies are, in fact, not required by the statutes, the agencies have a responsibility to address those concerns on their own and not shift the responsibility to Congress. If congressional committees of jurisdiction or individual Members of those committees believe that an agency’s regulations do not reflect the intent of the underlying statute or its legislative history, those committees or Members can communicate their concerns to the agencies informally. If Congress as a whole believes an agency’s regulation is inconsistent with the intent of the underlying statute, Congress can amend the statute to reflect current congressional intent and, in effect, require the agency to amend its regulation. The expedited congressional regulatory review procedures in the Small Business Regulatory Enforcement Fairness Act may also serve as a vehicle for that communication—at least for new or revised regulations. If Congress believes a new or revised regulation is inconsistent with the intent of the underlying statute, it can pass a resolution disapproving the rule. On the other hand, Congress can allow regulations to take effect if it believes that a rule is in keeping with statutory intent. Although no substitute for straightforward discussions between agencies and Congress, the congressional review procedures in the act have the potential to lead to a better understanding between major players in the federal rulemaking process. | Pursuant to congressional requests, GAO reviewed the cumulative impact of federal regulations on a limited number of businesses, focusing on: (1) what selected businesses and federal agencies believed were the federal regulations that applied to those businesses; (2) what those businesses believed were the cost and other impacts of those regulations; and (3) the regulations those businesses said were most problematic to them and relevant federal agencies' responses to those concerns. GAO found that: (1) most of the businesses contacted declined to participate in the study; (2) none of the 15 participating companies developed a complete list of regulations that were applicable to them or provided comprehensive data on the cost of regulatory compliance; (3) time and resource constraints and the difficulty of disentangling federal regulatory requirements from those of other jurisdictions and other nonregulatory procedures proved to be major obstacles for the companies; (4) most federal regulatory agencies said that they could not detail which regulations applied to a particular company without a great deal of company-specific information and the expenditure of a substantial amount of resources; (5) measuring the incremental impact of all federal regulations on individual companies is extremely difficult and, therefore, decisionmakers need to be aware of the conceptual and methodological underpinnings of studies that attempt to measure total current regulatory costs; (6) many of the 15 participating companies recognized that regulations provide benefits to society and their own businesses, but all of them provided GAO with a varied list of concerns about regulatory costs and the regulatory process; (7) these concerns included perceptions of high compliance costs, unreasonable, unclear, and inflexible demands, excessive paperwork, and a tendency of regulators to focus on deficiencies; (8) the agencies responsible for the regulations the companies viewed as problematic often said that the companies misinterpreted regulatory requirements; (9) the agencies and some congressional members do not always agree on the extent to which problematic regulations are statutorily driven; and (10) the agencies said that they were aware of and were responding to a number of the companies' concerns. |
This section describes (1) WIPP’s layout and base operations, (2) the February 2014 truck fire and radiological release accident investigation reports, (3) DOE’s requirements for operations activities and for project management applicable to the WIPP Recovery effort, (4) best practices for cost and schedule estimating, (5) best practices for conducting an AOA, (6) Revisions in 2014 and 2015 to DOE’s requirements for AOAs, and (7) DOE’s AOA process for WIPP’s new permanent ventilation system. As shown in figure 1, WIPP’s layout consists of surface facilities, underground facilities, and four shafts that connect the surface with the underground. WIPP’s underground facilities currently include seven waste disposal units or panels where waste containers are placed for final disposal, an area designated for constructing future disposal units, and an area for experimental research. The four shafts connecting the surface and underground facilities are (1) the air intake shaft, which is the primary air supply to the underground; (2) the exhaust shaft, through which all air exits the underground; (3) the salt handling shaft, which is used to remove mined salt from the underground and also supplies air to the underground; and (4) the waste handling shaft which is used to transport waste containers to the underground. Figure 1 also shows the locations of the February 2014 accidents that led to the suspension of WIPP’s operations. Under normal operating conditions, base operations at WIPP include waste handling to receive waste from generator sites and prepare and place the waste, ongoing maintenance of the WIPP surface facilities and underground, as well as support functions such as program management, engineering, quality assurance, safety and security, and environmental management. In addition, NWP reviews and certifies waste containers on DOE’s behalf at DOE’s transuranic waste generator sites. During implementation of the WIPP recovery plan, DOE continued aspects of these base operations such as maintenance of the surface facilities and program management. In addition, NWP continued to review and certify waste packages at select sites. The salt truck fire, which occurred on February 5, 2014, created substantial smoke and soot that damaged key equipment and facilities in the underground. On February 14, 2014, the radiological release occurred when a transuranic waste container was breached. The breach was caused by a chemical reaction inside the container between materials that DOE later determined should not have been packaged together. The reaction generated enough heat to increase pressure in the container. The pressure forced open the container’s lid and propelled its radioactive waste contents, combustible gases, and other materials into the air and on to adjacent waste containers. The radioactive contents, gases, and other materials ignited and triggered a fire in the disposal room by igniting other materials in the room. WIPP’s ventilation system failed to contain all of the airborne radiological material underground and allowed a small amount to enter the environment. As a result of the release, portions of the WIPP underground and the existing ventilation system were radiologically contaminated. DOE issued an accident investigation report on the salt truck fire in March 2014 and an accident investigation report on the radiological release in April 2014 called “phase one report.” The phase one report focused on the release to the environment and the response at WIPP to the release. One year later, in April 2015, DOE issued a phase two report that focused on the radiological release from inside the waste container. The reports determined that the causes of the accidents included the degraded condition of critical equipment, inadequacies in the design and operations of WIPP’s ventilation system, and deficiencies in the management of WIPP’s safety programs that are intended to control and discipline operations to protect workers, the public, and the environment from radiological and other hazards. In total, the three reports made 100 recommendations to DOE and NWP to complete corrective actions. For example, the phase one report identified a number of weaknesses with the WIPP nuclear safety documentation and recommended that this documentation be revised prior to resuming limited waste disposal operations to ensure the repository can be operated safely with respect to workers, the public, and the environment. DOE is managing the recovery activities to restart limited waste disposal operations at WIPP and to design and build the new permanent ventilation system capital asset project following two separate project management requirements. DOE requires that the recovery activities to restart limited waste disposal operations and overall base operations of WIPP be managed following EM’s protocol for operations activities. This protocol defines operations activities to include activities that are project- like with defined start and end dates and reoccurring facility or environmental operations. The new permanent ventilation system project must follow DOE’s Order 413.3B, which governs project management for the acquisition of capital assets. The following is a description of the two different requirements: EM operations activities protocol. EM’s protocol establishes a framework for managing and reporting the progress of cleanup operations by requiring, among other things, that project performance be measured objectively and that management actions be taken to mitigate risks and manage costs. EM’s protocol directs EM sites to develop performance baselines—or estimates—for cost, schedule, and scope to use in assessing the project’s performance over the fiscal year, multiyear contract period of performance, and the life cycle of the project. The protocol directs that the project develop estimates but does not include specific requirements that must be followed, such as the steps that must be followed to develop cost and schedule estimates. DOE Order 413.3B. DOE’s order provides program and project management direction for the acquisition of capital assets, with the stated goal of delivering fully capable projects within the planned cost, schedule, and performance baseline. The order establishes five critical decision processes of project development that each end with a major approval milestone—or “critical decision” point—that cover the life of a project. The order specifies the requirements that must be met, including for developing and managing project cost and schedule estimates to move a project past each critical decision milestone. DOE also provides suggested approaches for meeting the requirements contained in the order through a series of guides, such as a guide for cost estimating. In June 2015, the Secretary of Energy directed that this order be revised to require that cost and schedule estimates be developed, maintained, and documented in a manner consistent with methods and the best practices identified in GAO’s cost and schedule assessment guides and other published standards and best practices, such as the Federal Acquisition Regulation and Office of Management and Budget guidance. The GAO Cost Estimating and Assessment Guide and the GAO Schedule Assessment Guide compiled best practices corresponding to the characteristics of high-quality and reliable cost and schedule estimates. A high-quality, reliable cost estimate has the following four characteristics: comprehensive (e.g., has enough detail to ensure that cost elements are neither omitted nor double counted), well-documented (e.g., allows for data it contains to be traced to accurate (e.g., is based on an assessment of most likely costs and has been adjusted properly for inflation), and credible (e.g., discusses any limitations because of uncertainty or bias surrounding data or assumptions). Similarly, a high-quality, reliable schedule has four characteristics: comprehensive (e.g., captures all government and contractor activities necessary to accomplish a project’s objectives), well-constructed (e.g., sequences all activities using the most straightforward logic possible), controlled (e.g., is updated periodically to realistically forecast dates for activities), and credible (e.g., uses data about risks to predict the level of confidence in meeting a completion date and necessary schedule contingency and high-priority risks are identified based on conducting a robust schedule risk analysis). The characteristics of a high-quality and reliable cost estimate are supported by best practices listed in appendix II, and the characteristics of a high-quality and reliable schedule are supported by best practices listed in appendix III. In a December 2014 report we compiled 24 best practices for conducting an AOA. As noted above, the AOA is a key first step in acquisition of a capital asset such as WIPP’s new permanent ventilation system. The process entails identifying, analyzing, and selecting a preferred alternative. Conforming to these best practices helps ensure that the preferred alternative selected is the one that best meets the agency’s mission needs. We grouped these 24 best practices into the following four categories: General principles. This category includes best practices to be applied in planning, conducting, and documenting the AOA, such as defining functional requirements based on the mission need and conducting the analysis without a predetermined solution. Identifying alternatives. The identifying alternatives category includes best practices that help ensure the alternatives to be analyzed are sufficient, diverse, and viable. Analyzing alternatives. The analyzing alternatives category contains best practices related to estimating the costs and benefits of each alternative over its life cycle. Selecting a preferred alternative. The selecting a preferred alternative category includes best practices to help ensure a preferred alternative is selected that best meets the mission need by comparing alternatives based on their costs and benefits and independently reviewing the AOA process. Appendix IV lists the 24 best practices organized by the four categories. DOE’s procedures for conducting an AOA have changed twice since DOE began the AOA process for the WIPP permanent ventilation system. DOE started the AOA process for this system in October 2014. At that time, DOE was operating under AOA requirements defined in Order 413.3B. The AOA process, as defined in the order, occurs during the span of the first two critical decisions—Critical Decision-0 and Critical Decision-1 (Approve Alternative Selection and Cost Range). In December 2014, the Secretary of Energy established a new requirement that for projects estimated to cost $50 million or more, the responsible program office is to conduct an AOA independent of the contractor responsible for the project. Later, in June 2015, the Secretary of Energy directed the department to develop guidance for conducting AOAs consistent with the AOA best practices that GAO has published. The Secretary further required that AOAs be conducted and documented in a manner consistent with the guidance when it is complete. A DOE official told us in June 2016 that the department expected the guidance to be completed by December 2016. The Secretary’s action followed a recommendation that we made in December 2014 that DOE revise its Order 413.3B to adopt AOA best practices. DOE’s changes in its requirements for the AOA process under Order 413.3B took effect in the midst of DOE’s AOA process for WIPP’s new permanent ventilation system. DOE’s process was conducted as follows: DOE EM approved the mission need for the new ventilation system project in October 2014. Also in October 2014, NWP started an initial AOA and completed it in January 2015 shortly after the Secretary’s December 2014 directive on the independence of AOAs and prior to the Secretary’s June 2015 directive about the use of the AOA best practices; therefore the new requirements did not apply to NWP’s AOA. To respond to the Secretary’s December 2014 directive, DOE contracted in spring 2015 with a contractor unaffiliated with WIPP— Trinity Engineering Associates (Trinity)—to do a second AOA. DOE specifically directed Trinity in its contract to implement all of the AOA best practices that we compiled in our December 2014 report. The WIPP program manager in the Carlsbad Field Office told us that he included the best practices in the contract to provide more formality to their process. Trinity completed its AOA in October 2015. DOE evaluated the preferred alternatives proposed by NWP and Trinity, and it selected and approved a final alternative to complete Critical Decision-1 in December 2015. DOE did not meet its initial cost and schedule estimates for the efforts needed to restart WIPP disposal operations, resulting in about $64 million in added costs and a delay of nearly 9 months. Two primary factors contributed to the cost increase and schedule delay. First, DOE only partially followed best practices in developing its initial cost and schedule estimates, which made them unreliable and increased the likelihood that they would be exceeded. Second, DOE did not successfully manage key project risks it had identified that had potential to cause delay. DOE incurred a cost increase of about $64 million and a nearly 9-month schedule delay in its efforts to restart WIPP waste disposal operations. Specifically, in January 2016, approximately a year after approving the initial project baseline, DOE approved a new project management baseline that increased the estimated costs for the recovery project by $2 million (from $242 million to $244 million) and added 8.5 months to the project schedule, extending the date when limited waste disposal operations might begin from March 2016 to December 2016. According to DOE officials, the project was unable to meet its estimated completion date due to delays, including delays associated with procuring components for the interim ventilation system upgrade that were found to be faulty, and delays associated with DOE’s decision to require the project to adhere to new nuclear safety requirements. According to the operations activity manager for WIPP in the Carlsbad Field Office, the recovery project only exceeded the $242 million baseline by $2 million in part because DOE overestimated the cost of some project activities and spent contingency funds to make up for cost increases in certain parts of the project. For example, DOE overestimated the cost for decontaminating the WIPP underground by more than $6 million, which offset a $5 million cost increase caused by delays associated with the faulty ventilation system components. However, we identified an additional $61.6 million cost increase in base operations that was attributable to the delay in completing the recovery project. Specifically, DOE’s justification supporting the President’s fiscal year 2016 budget request to Congress from February 2015 estimated about $130.6 million for WIPP’s base operations in fiscal year 2016. According to DOE officials, as DOE revised its cost and schedule estimates for the WIPP recovery project in the fall of 2015, DOE also revised its cost estimate for WIPP’s base operations and provided the updated information to the congressional appropriations committees. In December 2015, DOE received $192.1 million for fiscal year 2016 base operations, an increase of $61.6 million (47 percent) over the initial estimate for base operations in fiscal year 2016. According to the operations activity manager for WIPP in the Carlsbad Field Office and representatives from NWP, this increase to base operations occurred because DOE estimated the cost of base operations on the assumption of restarting WIPP in March 2016 following its initial schedule estimate. The delay in the recovery efforts required DOE to keep base operations running alongside the recovery efforts for a longer period of time than initially planned. DOE did not meet its initial schedule and cost estimates for restarting waste disposal operations at WIPP, in part because DOE did not develop the estimates following all best practices, rendering the estimates unreliable and increasing the likelihood that they would be exceeded. More specifically, of the four characteristics of a high-quality and reliable cost estimate, DOE’s initial cost estimate substantially met two characteristics—comprehensive and well-documented—but partially met or minimally met the other two characteristics—accurate and credible. We made the following observations: DOE substantially met best practices for a comprehensive cost estimate by, for example, including a majority of life-cycle costs and identifying the cost estimating ground rules and assumptions. DOE substantially met best practices for a well-documented estimate by including the actual sources for the cost data and documenting that the cost estimate was reviewed and accepted by DOE management. DOE partially met best practices for an accurate estimate by, for example, only documenting a portion of the process they used to adjust costs to account for inflation. DOE minimally met best practices for a credible estimate by, for instance, not developing an independent cost estimate. Had DOE completed an independent cost estimate, the department would have had an unbiased and objective benchmark to assess whether the cost estimate prepared by NWP could be achieved, and thus would have been positioned to reduce the risk that the project would proceed underfunded. Regarding DOE’s initial schedule for restarting waste disposal operations, DOE substantially met two of the four characteristics of a high-quality schedule—comprehensive and controlled—but partially met the characteristics—well-constructed and credible—resulting in a schedule that was also unreliable and unrealistic. We made the following observations: DOE substantially met best practices for a comprehensive schedule because, for example, the schedule reflected the activities in the recovery work breakdown structure, which defined the work necessary to accomplish the project’s objectives. In addition, work scope was assigned in the schedule as the responsibility of NWP or DOE. DOE substantially met best practices for a controlled schedule by, for example, regularly updating its master schedule and using the schedule as the basis for measuring performance. DOE partially met best practices for a well-constructed schedule because, for instance, a significant number of activities in the schedule had incorrect or missing logic relationships that are important for determining how delays or accelerations in one activity would affect the start or finish of other activities later in the schedule. DOE partially met best practices for a credible schedule, in particular, because its schedule did not include extra time, or contingency, to account for known project risks. As noted above, DOE acknowledged in its WIPP recovery plan that the schedule did not include contingency that may be needed due to unanticipated difficulties or delays with the project. However, DOE did not acknowledge in the plan that the schedule also did not include contingency for the occurrence of known and quantified risks that had been anticipated in its project risk analysis—such as the risks that were the primary causes of the project delays. Notably, DOE’s risk analysis predicted that the March 2016 date to restart operations had a less than 1 percent confidence level, meaning in effect that DOE had a less than 1 percent chance of meeting the March 2016 deadline. EM officials said the department used a less than 1 percent confidence level because it wanted to have an aggressive goal for restarting operations. According to EM officials, when EM managers presented this schedule to senior DOE decision-makers for approval, they described the schedule as “aggressive” but did not clarify that there was less than a 1 percent chance of meeting the schedule. According to these officials, they did not think that senior managers would understand the project management terminology regarding confidence levels. In January 2016, DOE approved revised cost and schedule estimates for restarting WIPP’s limited waste disposal operations. The operations activity manager for WIPP in the Carlsbad Field Office said the revised schedule included contingency and the revised restart date of December 2016 had an 80 percent confidence level. However, according to DOE officials, they did not follow other best practices. For example, DOE did not provide evidence of having an independent cost estimate which would have provided DOE an unbiased and objective benchmark to validate the estimates prepared by NWP. As noted above, we did not assess DOE’s revised estimates against all of the best practices. The full results of our analysis of the initial cost and schedule estimates can be found in appendixes II and III. DOE did not follow all best practices in developing the initial cost and schedule estimates for the WIPP recovery project or in developing new estimates because, unlike DOE’s requirements for capital asset projects under Order 413.3B, DOE’s EM operations activities protocol that governs cleanup operations such as WIPP recovery does not require the use of best practices in developing such estimates. As mentioned above, DOE EM’s protocol requires that EM sites develop or approve baselines for each project’s cost and schedule to judge project performance. However, the protocol does not specify any best practices in terms of the steps to follow in developing the cost and schedule estimates in these baselines. The absence of a requirement to follow best practices for cost and schedule estimating is in contrast to a decision in June 2015 by the Secretary of Energy to require that all capital asset projects with an estimated cost of $10 million or more follow such practices under Order 413.3B. Operations activities such as the WIPP recovery activities are not considered capital asset projects and therefore are not required to follow these requirements even though EM’s operations activities share many of the same characteristics as capital asset projects. Without requiring EM’s operations activities such as WIPP to follow all best practices when developing cost and schedule estimates DOE cannot have confidence that it is producing reliable baselines needed to monitor its performance in managing these activities. Without reliable baselines, DOE will also continue to be at risk of cost overruns and delays in achieving its cleanup missions such as the permanent disposal of transuranic waste from sites across the country. DOE has reported progress in completing the activities needed to restart limited waste operations using its revised cost and schedule estimates, although it still faces challenges and it remains unclear whether DOE identified and analyzed the risks associated with these challenges in revising the estimates. In terms of progress, for example, in December 2015, NWP submitted revisions to WIPP’s nuclear safety documentation to EM for review. In addition, DOE reported in March 2016 that NWP completed most of the construction of the interim ventilation system upgrade in the underground. However, as noted above, in March 2016, the Defense Nuclear Facilities Safety Board reported that additional work was needed to revise WIPP’s nuclear safety documentation to prevent recurrence of a radiological release accident. In addition, in April 2016, DOE’s Office of Enterprise Assessments, which provides independent internal oversight of DOE’s management of safety, issued two reports that found that although NWP had made improvements in its operational safety and emergency management programs and procedures, significant challenges remained to fully meet DOE requirements and effectively plan and implement these programs and procedures. As we noted above, DOE officials acknowledged that the department did not follow all best practices in developing the revised estimates to restart WIPP operations. Without having followed all best practices, including having an independent cost estimate conducted to validate the estimates, DOE cannot be confident that NWP sufficiently accounted for these challenges in revising its risk analysis and that DOE set an appropriate allowance for contingency to reduce the risk of cost overruns and delays in restarting WIPP’s operations. DOE did not successfully manage key risks that it had identified and that contributed to the project’s cost increase and schedule delay but has taken steps to revise the risk management process for the WIPP recovery project. DOE’s EM operations activities protocol requires the use of a risk management process to identify and mitigate risks to completing a project within its baseline cost and schedule estimates. For the WIPP recovery project, DOE and NWP developed a project risk register to support the initial project management baseline for WIPP recovery. The risk register listed the risks that could increase costs or delay completion. DOE and NWP assessed the likelihood and consequence of each risk and identified a mitigation plan for each risk that described the actions to reduce the impact on the project if the risk were realized. In a public statement explaining the need to revise its baseline for the recovery project, DOE cited problems with three activities that the department had previously identified in its project risk register: (1) completing revisions to WIPP’s nuclear safety documentation; (2) installing the interim ventilation system upgrade; and (3) completing the corrective actions that address the recommendations in the phase two accident investigation report. DOE had identified risks related to each of these activities in the project risk register but was not able to effectively mitigate them, as discussed below: WIPP nuclear safety documentation. NWP identified the risk of a potential delay of up to 6 months in restarting operations if DOE decided that the WIPP nuclear safety documentation should be revised according to standards that were in the process of being updated rather than according to the existing standards. According to DOE, when NWP started to revise the safety documentation in the summer of 2014, the contractor assumed it should follow the existing standard. NWP added this risk to its project risk register to account for the possibility that EM could require NWP to revise the nuclear safety documentation according to the new standards. To mitigate this risk, NWP’s plan was to ensure they had concurrence with DOE to start the revisions using the existing standards. DOE issued the new standards in November 2014 and notified NWP in December 2014 that the WIPP safety documentation would need to comply with them. According to an NWP recovery project manager, this change in policy resulted in a 7-month delay in NWP’s schedule for revising the WIPP safety documentation. Interim ventilation system. NWP identified the risk of a potential delay of 3 months to complete the installation of the interim ventilation system upgrade if NWP faced difficulties acquiring the components for the system. To mitigate this risk, NWP’s plan was to identify difficult- to-procure equipment early in the project, ensure it had a valid and up-to-date list of qualified suppliers, and monitor and review equipment purchases. According to NWP officials, this plan was partially effective. Specifically, NWP officials said that they discovered that components for the interim ventilation system were damaged when they inspected the components before formally accepting them. In addition, they said that the costs to return the components to the manufacturer and repair them were paid for by the manufacturer. However, the NWP officials said that as a result of receiving the faulty equipment, the installation of the interim ventilation system was delayed about 4 months because NWP needed to send the components back to the manufacturer and make corrections to its shipping process to prevent a reoccurrence of these issues. Corrective actions from the phase two accident investigation report. NWP identified the risk of potential delay of up to 6 weeks regarding the length of time DOE needed to complete the phase two accident investigation. In the summer and fall of 2014, when NWP was developing the initial project management baseline, DOE’s phase two accident investigation was still underway. According to NWP officials, they needed to wait for DOE to complete the accident investigation and issue the final report with recommendations before they could begin developing and implementing corrective actions. DOE issued the final report in April of 2015. According to the operations activity manager for WIPP in the Carlsbad Field Office, completing the corrective actions in response to the report’s recommendations resulted in a 3-month delay. NWP officials said that in recognition of the need for a more effective risk management process, they revised their process as part of revising the recovery cost and schedules estimates. These officials explained, for example, the revised process now involves more frequent discussions and updating the status of potential risks with managers at WIPP who oversee the departments where the project risks are generated. We did not conduct an assessment of the effectiveness of this new risk management process. DOE did not follow all best practices in analyzing and selecting an alternative for the new ventilation system at WIPP. As a result, DOE’s analysis was not reliable and DOE cannot be confident that the selected alternative will best provide the needed capabilities. Of the four categories of the best practices, DOE fully met the category for identifying alternatives. However, DOE partially or minimally met the other three categories of best practices—general principles, analyzing alternatives, and selecting the preferred alternative—because of key limitations. (See app. IV for the detailed results of our analysis of DOE’s AOA process.) DOE’s AOA process fully met the best practices in the identifying alternatives category. To identify alternatives, DOE relied on analyses conducted by NWP, the contractor responsible for the project, and the second contractor—Trinity. These analyses were completed in January and October 2015, respectively. NWP and Trinity identified a broad range of alternatives. Specifically, they identified nine ventilation system alternatives that the contractors determined should be analyzed because the alternatives could meet the mission need and other screening criteria that they defined: NWP identified six and Trinity identified three. The alternatives included continued use of the existing exhaust shaft as well as constructing a new shaft or more than one shaft, such as a new exhaust shaft and a new air intake shaft. In addition, the studies identified different ventilation filtration capacities and modes of operation that included continuous filtering of all exhaust air or filtering air only when a radiological release is detected underground. DOE’s AOA process partially met the best practices in the general principles category, which covers how the AOA process is planned, conducted, and documented. In particular, DOE partially or minimally followed the best practices of defining the mission need for the project, analyzing alternatives without a predetermined solution, and defining the functional requirements based on the mission need. As the AOA related to the best practices of defining the mission need for the project and analyzing alternatives without a predetermined solution, DOE did not define the mission need for the new ventilation system by focusing only on the capabilities needed for the project, but instead defined the need for the system to include a particular solution, which included constructing a new ventilation exhaust shaft. Because DOE defined the mission need by mentioning a new exhaust shaft, both contractors appeared to have analyzed alternatives with a preference for the alternatives that included constructing a new shaft. Specifically, NWP completed its analysis of the six alternatives by proposing two for further consideration by DOE, one that did not include a new shaft and one that did. NWP’s alternative that did not include a new shaft was its highest-scoring alternative and the alternative that included a new shaft was its third-highest scoring alternative. NWP officials told us that they proposed the second alternative along with the highest-scoring one because they believed the mission need statement discussed a new shaft. Regarding Trinity, in its final AOA report, Trinity stated that the mission need statement endorsed a new exhaust shaft. Trinity then analyzed three alternatives, one of which included using the existing shaft, and proposed an alternative to DOE that included constructing a new shaft. In its final assessment, DOE assessed the two alternatives proposed by NWP and the one proposed by Trinity and selected NWP’s alternative that included constructing a new shaft, shown as the second of the three alternatives in table 1. According to best practices, the AOA process should be an unbiased inquiry into the costs, benefits, and capabilities of all alternatives. By conducting the AOA with a predetermined solution, DOE undermined the credibility of its final decision. Another limitation under the general principles category was that DOE partially followed the best practice of defining the functional requirements that the ventilation system would need to satisfy. DOE did not consistently define two key functional requirements of the project—(1) the rate of airflow needed to support full operations at WIPP, and (2) the expected operational life of the new system—which limited the reliability of the overall AOA process as follows: Regarding the estimated airflow, DOE did not specify a functional requirement in its mission need statement for the rate of airflow that the new ventilation system must be capable of providing for full operations at WIPP. Instead, the mission need statement described the minimum airflow required by WIPP’s hazardous waste facility permit, which was 260,000 cfm, and the actual airflow at WIPP before the accidents, which was 425,000 cfm. In conducting its AOA, NWP calculated its own airflow that it thought would be necessary for full operations at WIPP, which they estimated was about 540,000 cfm. NWP estimated the higher airflow based on its analysis of underground operations and used this amount to analyze a range of alternatives. In contrast, Trinity used the airflow associated with the WIPP hazardous waste facility permit, 260,000 cfm, as the estimated airflow in conducting its AOA. The federal project director in the Carlsbad Field Office told us that he did not direct Trinity to redo its analysis using NWP’s higher airflow because he believed that airflow was needed for examining the scale of the system and would be defined lower during design. Moreover, he said that he wanted to maintain Trinity’s independence from NWP to be consistent with the December 2014 DOE requirement for conducting independent AOAs. However, as a result of not defining the same functional airflow requirement for both contractors, DOE officials explained that they needed to conduct additional analysis to compare the alternatives proposed by NWP and Trinity. Regarding the estimated operational life of the new ventilation system, DOE did not specify a functional requirement in its mission need statement for the estimated operational life of WIPP. Without such a requirement, NWP used 2030 as the estimated end date for WIPP operations—this date was based on DOE’s approved life-cycle plan for ending WIPP’s operations current at the time NWP conducted its analysis. NWP completed its AOA in January 2015 when it presented two alternatives to DOE for further evaluation—as described above, NWP had proposed these two alternatives from an initial group of six alternatives. In February 2015, EM revised the estimated operational life of WIPP, extending it from 2030 to 2050 to more accurately reflect DOE’s schedules for transuranic waste cleanup at DOE sites. Trinity did not start its AOA process until after DOE had revised the estimated operational life of WIPP to 2050 and therefore used the 2050 date in conducting its AOA. In analyzing NWP’s two proposed alternatives, DOE used the new estimated operational life of 2050 and did not reassess the four previously eliminated alternatives using the revised date. By not following the best practice of specifying functional requirements at the start of the AOA process, including specifying the airflow that the system needed to deliver and the estimated end date of operations for WIPP, it is unclear whether the AOAs conducted by NWP and Trinity and DOE’s subsequent analyses of these AOAs allowed DOE to select the alternative that best meets mission needs. DOE’s AOA process minimally met the best practices in the analyzing alternatives category. Significant limitations in this category included DOE and its contractors not consistently examining life-cycle costs for each alternative and not quantifying their benefits. Regarding examining life- cycle costs: NWP did not examine full life-cycle cost estimates for all of the alternatives it examined. Specifically, NWP’s AOA completed in January 2015 examined the estimated costs for the design and construction of each of the six alternatives, but did not examine the full life-cycles of these alternatives. As a result, NWP eliminated four of its six principal alternatives from further evaluation before examining each of them in terms of full life-cycle costs. According to the NWP project manager who led NWP’s AOA team, the team did not have sufficient time to examine the full life-cycle costs of each alternative. In addition, according to DOE officials, NWP was not required to examine the full life-cycle costs of each alternative under DOE Order 413.3B because they completed the analysis in January 2015, which was several months before the Secretary’s June 2015 directive to incorporate AOA best practices in the order. According to AOA best practices, the team conducting the AOA should be given enough time to complete the AOA process to ensure a robust and complete analysis. NWP and Trinity used different assumptions in developing life-cycle cost estimates, which prevented DOE from directly comparing the life- cycle estimates for the three proposed alternatives. We found that DOE relied on NWP and Trinity to define the assumptions they used rather than providing the contractors consistent guidance or direction. As a result, NWP assumed certain costs should be excluded in estimating the life-cycle costs of its two proposed alternatives. These costs included the costs for major equipment replacements or upgrades which NWP assumed would not be needed during the 30- year operational life of the system; costs for providing facility security, quality assurance, and nuclear safety that are provided to all facilities at WIPP; and costs for the final closure of the new exhaust shaft after 2050 when the system will be decommissioned. In contrast, Trinity assumed these costs should be included in estimating the life-cycle costs of its alternatives. As a result of including these additional costs, Trinity’s estimated life-cycle cost was substantially more than NWP’s estimate—Trinity’s estimate for its proposed alternative was $3.45 billion and NWP’s estimates for its two proposed alternatives were $368.8 million and $467.6 million. Regarding quantifying benefits, DOE and its contractors did not provide any measures of the benefits or effectiveness for each of the alternatives in their analyses. By not ensuring that the life-cycle costs of each alternative were developed in a consistent manner and not ensuring that the benefits or effectiveness of each alternative were quantified, DOE did not have an accurate and complete picture of all the alternatives to make reliable comparisons between them. DOE’s AOA process partially met the best practices in the selecting a preferred alternative category which covers selecting an alternative that best meets the mission need and also independently reviewing the overall AOA process. Most notably, DOE and its contractors did not follow the best practice to select the alternatives based on a cost-benefit analysis and only partially followed the best practice to independently review the AOA process, as follows: Regarding cost-benefit analysis, DOE and its contractors did not select the alternatives based on such an analysis which compares the life-cycle costs and benefits or effectiveness of each alternative. As noted above, DOE and its contractors did not consistently examine life-cycle cost estimates for each alternative and did not quantify their benefits. Therefore, DOE did not produce the information needed for a cost-benefit analysis. By not selecting the alternatives based on a cost-benefit analysis, DOE did not adequately justify that the selected final alternative would best provide the capabilities needed at WIPP. Regarding independently reviewing the AOA process, DOE conducted an independent review of the AOA process (which assessed whether best practices were followed) but did not implement the recommendation identified by the review. Specifically, an independent review conducted by DOE’s Office of Project Management Oversight and Assessments found that the project team did not adequately document a cost-benefit analysis and that, as a result, the selection of the preferred alternative was not supported by compelling information. The review recommended that DOE perform a cost-benefit analysis to be consistent with best practices and support the selection of the final alternative. However, DOE and NWP did not conduct the recommended cost-benefit analysis and document it before DOE selected the final alternative. According to the DOE officials who led the independent review of the AOA process, the project team was not required by DOE’s Order 413.3B to implement the recommendation or justify and document the reasons for not doing so. Therefore, DOE EM approved the selection of the preferred alternative for the permanent ventilation system project in December 2015 without implementing the independent review’s recommendation. Conducting an independent review of the AOA is a best practice because it is one of the most reliable means to ensure that bias does not influence the AOA process and that the AOA is sufficiently thorough to ensure that a preferred solution is chosen and not a favored solution. DOE’s independent review was an internal control to help DOE meet its goal for the capital asset project, in this case selecting the alternative that will best provide the capabilities to meet the mission need. Under federal standards for internal control, management is to evaluate and document internal control issues and determine appropriate corrective actions. By not implementing the recommendation from the independent review to do a cost-benefit analysis or not justifying and documenting the reason for not doing so, DOE cannot provide assurance that the final alternative selected would best provide the capabilities needed at WIPP. Restarting WIPP’s waste disposal operations is a top priority for DOE, and it has made progress in its efforts to restart limited waste disposal operations. However, DOE did not meet its initial cost and schedule estimates to restart operations and incurred a cost increase of about $64 million—$2 million to the recovery project and $61.6 million in base operations—and a delay of nearly 9 months. DOE incurred the cost increase and schedule delay, in part, because DOE’s initial estimates did not follow all best practices and were therefore unreliable. Notably, DOE did not include any contingency in its schedule, giving itself less than a 1 percent chance of success—said another way DOE gave itself a 99 percent chance of failure in meeting its schedule to restart operations. Moreover, when DOE revised its cost and schedule estimates for WIPP recovery it still did not follow all best practices. DOE did not develop its estimates for WIPP recovery following best practices, in part because DOE does not require its cleanup operations activities, such as WIPP, to follow them. This lack of a requirement is in contrast to a new policy put into effect in June 2015 by the Secretary of Energy, which established the requirements that DOE develop cost and schedule estimates for its capital asset projects following best practices—such projects include the new permanent ventilation system at WIPP. Without similar requirements for EM operations activities to follow best practices, DOE cannot have confidence that it is producing reliable baselines needed to monitor its performance in managing these activities and reduce the risk of cost overruns and delays in achieving its cleanup missions, such as the permanent disposal of transuranic waste from DOE sites across the country. By also requiring cleanup operations to follow best practices, DOE would have more confidence in the estimates for its cleanup operations activities and its capital asset projects. In selecting the preferred alternative for the new permanent ventilation system at WIPP, DOE relied on an AOA process that did not follow all best practices. In particular, DOE only partially followed or did not follow best practices to define functional requirements based on the mission need, compare alternatives using a cost-benefit analysis, or independently review the AOA process. By not following these and other practices, DOE’s AOA process was not reliable. Notably, DOE conducted an independent review of the process which found that the project team did not adequately document a cost-benefit analysis and that, as a result, the selection of the preferred alternative was not supported by compelling information. Nonetheless, DOE did not implement the recommendation of the review to conduct a cost-benefit analysis before the department selected the preferred alternative. DOE officials explained that the project team was not required by DOE’s Order 413.3B to implement the recommendation. By not implementing the recommendation of the independent review or not justifying and documenting the reason for not doing so, DOE cannot provide assurance that the final selected alternative would best provide the capabilities needed at WIPP. More broadly, without requiring in its Order 413.3B that recommendations from independent reviews of AOAs be implemented or that the reason for not doing so be justified and documented, DOE cannot have assurance that it selects, for all capital asset projects, preferred alternatives from its AOAs that best meet the mission need. To help ensure that DOE develops and uses reliable cost and schedule estimates and AOAs, we recommend that the Secretary of Energy take the following three actions: Direct EM to revise its protocol governing cleanup operations activities to require use of best practices in developing cost and schedule estimates. Direct EM to implement the recommendation made by DOE’s Office of Project Management Oversight and Assessments in its independent review of the AOA for WIPP’s new permanent ventilation system to perform a cost-benefit analysis consistent with best practices for conducting an AOA, or justify and document why the office does not intend to do so. Direct DOE to revise its Order 413.3B to require that DOE offices implement any recommendations from an independent review of the extent to which an AOA followed best practices, or justify and document the rationale for not doing so. We provided a draft of this report to DOE for review and comment. In written comments, reproduced in appendix V, DOE concurred with the report’s recommendations. DOE stated that our recommendations were consistent with its commitment to continuous improvement in project management. In addition, DOE stated that it had two issues that the department believed needed to be addressed. DOE also provided technical comments that were incorporated, as appropriate. DOE identified the following two issues: DOE stated that WIPP recovery activities were under way concurrent with DOE-wide efforts to improve project management, including revisions to its project management order (DOE Order 413.B). DOE stated that, consequently, EM project management practices were separate from evolving department project management guidance and that it is now in the process of bringing EM practices into conformance with the department’s project management guidance. DOE stated that the report needs to be clear, in each instance, regarding the $64 million cost increase for WIPP attributed to schedule delays in recovery activities, to indicate that $2 million of the cost increase was WIPP recovery project-related and that $61.6 million was due to reexamination and assessment of the cost of base operations. We do not believe changes are needed to the report for either of these issues. The report explains that WIPP recovery activities were based on EM’s operations activities protocol and that actions were under way within the department to improve project management, including revisions to DOE’s project management order. In addition, the report includes a breakout of the cost increase. We did, however, revise the conclusions to reflect this breakout. As noted above, DOE concurred with the report’s three recommendations. Regarding the first recommendation—that EM revise its protocol governing cleanup operations activities to require use of best practices in developing cost and schedule estimates—DOE stated in its written comments that it concurred with clarification. DOE stated that EM is transitioning from the operations activities protocol to a new directive that is expected to include a key decision approving a cost and schedule baseline. As EM develops the guidance for this key decision, it will include the use of cost and schedule best practices. DOE stated that EM plans to finalize this new directive by September 2016 and seek departmental approval by December 2016. Regarding the second recommendation— that EM implement the recommendation made by DOE in its independent review of the AOA for WIPP’s new ventilation system to perform a cost-benefit analysis consistent with best practices, or justify and document why it does not intend to do so—DOE stated in its written comments that it concurred with clarification. DOE stated that in accordance with GAO best practices, further cost-benefit analysis will be conducted on the project prior to approval of Critical Decision-2 (Approve Performance Baseline). DOE stated that several alternatives remain to be evaluated including the size of the ventilation system and the location of the exhaust shaft. Regarding the third recommendation—that DOE revise its project management (Order 413.3B) to require that DOE offices implement recommendations from independent reviews of the extent to which an AOA followed best practices, or justify and document the rationale for not doing so—DOE concurred with the recommendation. DOE stated that it will prepare a project management policy on how DOE offices should respond to recommendations from independent reviews. DOE stated that it will prepare this policy by December 2016 and update DOE Order 413.3B at the next available opportunity. We are sending copies of this report to the appropriate congressional committees, the Secretary of Energy, and other interested parties. In addition, the report is available at no charge on the GAO website at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-3841 or trimbled@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix VI. Our report examined the extent to which the Department of Energy (DOE) (1) met its initial cost and schedule estimates for restarting waste disposal operations at the Waste Isolation Pilot Plant (WIPP) and (2) followed best practices in analyzing and selecting an alternative for the new ventilation system. To address both objectives, we conducted a site visit to WIPP in January 2015. During the site visit we obtained documentation and interviewed officials from DOE’s Office of Environmental Management (EM), which is responsible for oversight of WIPP and exercises this responsibility primarily through its Carlsbad Field Office. We also interviewed representatives at WIPP from Nuclear Waste Partnership LLC (NWP), which is the private contractor that manages and operates WIPP for DOE. To examine the extent to which DOE met its initial cost and schedule estimates for restarting waste disposal operations, we compared DOE’s initial cost and schedule estimates to restart limited waste disposal operations contained in its February 2015 WIPP recovery project management baseline with DOE’s revised estimates in its January 2016 integrated project management baseline for WIPP. We reviewed DOE’s budget justification supporting the President’s fiscal year 2016 budget request to Congress from February 2015 and the amounts DOE received for WIPP in the 2016 Consolidated Appropriations Act. We reviewed DOE’s reports on the reasons it exceeded the initial estimates for restarting waste disposal operations and DOE’s risk management plans for WIPP recovery prepared by NWP and used in developing the February 2015 estimates. We also compared DOE’s initial cost and schedule estimates to restart limited waste disposal operations from the February 2015 WIPP recovery project management baseline with the best practices described in our cost and schedule guides that identified the characteristics of high-quality, reliable cost and schedule estimates because these were the approved estimates when we conducted our analysis. Specifically, we compared DOE’s initial WIPP recovery cost estimate presented in DOE’s project management baseline document and supporting documents and data with the best practices in our Cost Estimating and Assessment Guide. In addition, we compared DOE’s initial WIPP recovery schedule presented in DOE’s project management baseline document and supporting documents and data with the best practices in our Schedule Assessment Guide. We interviewed the Carlsbad Field Office officials who oversee the recovery project and NWP’s cost estimator and scheduler. We provided a draft of our cost and schedule assessments to the Carlsbad Field Office and NWP and revised the draft, as appropriate, after discussing our assessment with the federal officials and the contractor. We reviewed documentation on the revised (January 2016) baseline and interviewed DOE and NWP officials on the approach followed to develop the revised cost and schedule estimates to restart WIPP’s waste disposal operations in the baseline, but we did not assess the revised estimates against the best practices because of the time frame of our review. To examine the extent to which DOE followed best practices in analyzing and selecting an alternative for WIPP’s new ventilation system, we used as criteria the best practices for conducting an AOA identified in GAO-15-37 issued in December 2014. GAO developed the best practices identified in this report by reviewing AOA policies and guidance used by seven public and private-sector entities with experience in the AOA process, and verified these practices with subject-matter experts. DOE’s AOA process for the WIPP ventilation system consisted of three elements: NWP’s January 2015 AOA that resulted in the selection of two preferred alternatives proposed for further analysis, a second AOA completed by Trinity Engineering Associates (Trinity) in October 2015 which resulted in a single preferred alternative proposed to DOE, and DOE’s final alternative evaluation process led by the Carlsbad Field Office that considered the three alternatives proposed by the initial studies. Our analysis assessed the overall AOA process considering each element. We compared the process with the best practices and determined a score for the overall process. We reviewed project documentation from the Carlsbad Field Office, NWP, and Trinity and interviewed Carlsbad Field Office officials and NWP representatives in charge of the AOA. In addition, in October 2015, as we were conducting our engagement, EM and DOE’s Office of Project Management Oversight and Assessments completed separate assessments of the WIPP AOA for the new ventilation system. We reviewed documentation of these reviews and interviewed the DOE officials who worked on them. We examined the extent that the independent assessments followed the best practice to have an independent entity assess the extent that a project’s AOA followed all best practices. To score DOE’s AOA process, a GAO analyst examined the AOA documentation received from the agency and then assigned a score for each of the 24 best practices. Following this, a GAO AOA specialist independent of the engagement team reviewed the AOA documentation and the scores assigned by the analyst for accuracy and cross-checked the scores in all the analyses for consistency. We used a five-point scoring system to determine the extent to which DOE’s AOA process conformed to the best practices. After determining a score for each individual best practice, we calculated the score for each category—(1) general principles, (2) identifying alternatives, (3) analyzing alternatives, and (4) selecting a preferred alternative—by calculating the average of the scores for the best practices that fall under each category. If the score for each best practice and the average score for each category was “fully met” or “substantially met,” we concluded that the AOA process conformed to best practices and therefore could be considered reliable. In contrast, if the score was “partially met,” “minimally met,” or “not met,” we concluded that the AOA process did not conform to best practices and therefore could not be considered reliable. We shared our analysis with DOE officials and representatives from NWP for review and incorporated their technical comments and any additional evidence they provided in our analysis, as appropriate. We also interviewed officials from DOE’s Office of Enterprise Assessments, which provides internal oversight of DOE facilities; the Defense Nuclear Facilities Safety Board, which provides external oversight of DOE defense nuclear facilities; the U.S. Mine Safety and Health Administration, which provides external oversight of mining activities at WIPP; as well as the U.S. Environmental Protection Agency and the New Mexico Environment Department, both of which provide external regulation of WIPP. We conducted this performance audit from November 2014 to August 2016 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Table 2 below assesses the Department of Energy’s (DOE) initial cost estimate in its February 2015 project management baseline for the Waste Isolation Pilot Plant (WIPP) recovery project compared to best practices for cost estimating published in the GAO Cost Estimating and Assessment Guide. Overall, of four characteristics of a high-quality and reliable cost estimate, DOE’s initial cost estimate substantially met two characteristics—comprehensive and well-documented—but partially met or minimally met the other two characteristics—accurate and credible. Table 3 below assesses the Department of Energy’s (DOE) initial schedule in its February 2015 project management baseline for the Waste Isolation Pilot Plant (WIPP) recovery project compared to best practices for developing a schedule estimate published in the GAO Schedule Assessment Guide. Overall, DOE’s schedule substantially met two of the four characteristics of a high-quality schedule—comprehensive and controlled—but partially met the characteristics of well-constructed and credible. Table 4 below compares the Department of Energy’s (DOE) analysis of alternatives (AOA) process completed in December 2015 for the new permanent ventilation system at the Waste Isolation Pilot Plant (WIPP) to best practices for conducting an AOA published in a December 2014 GAO report. DOE’s overall AOA process consisted of an initial AOA by DOE’s WIPP contractor, Nuclear Waste Partnership LLC, completed in January 2015; a second AOA by a contractor unaffiliated with WIPP, Trinity Engineering Associates, completed in October 2015; and DOE’s final evaluation of the three preferred alternatives proposed by the contractors’ analyses. Overall, DOE’s AOA process fully met the category for identifying alternatives and partially or minimally met the other three categories of best practices—general principles, analyzing alternatives, and selecting the preferred alternative. David C. Trimble, (202) 512-3841 or trimbled@gao.gov. In addition to the contact named above, Daniel Feehan, Assistant Director; Cheryl Arvidson; Mark Braza; Richard P. Burkard; Jennifer Echard; Brian M. Friedman; Carly Gerbig; Jason Lee; Eli Lewine; Cynthia Norris; and Katrina Pekar-Carpenter made key contributions to this report. | DOE's WIPP is the only deep geologic repository for the disposal of U.S. defense-related nuclear waste. In February 2014, waste operations were suspended following a truck fire and an unrelated radiological release. DOE estimated in February 2015 that it would complete recovery activities and restart limited waste operations by March 2016. To resume full operations, DOE planned to build a new ventilation system at WIPP. DOE completed an AOA to identify the best solution for this system in December 2015. The Senate Report accompanying a bill for the National Defense Authorization Act for Fiscal Year 2015 included a provision for GAO to review WIPP operations. This report examines the extent to which DOE (1) met its initial cost and schedule estimates for restarting waste disposal operations, and (2) followed best practices in analyzing and selecting an alternative for the new ventilation system. GAO examined documentation on the WIPP recovery estimates. GAO compared DOE's February 2015 cost and schedule estimates and AOA with best practices GAO published. The Department of Energy (DOE) did not meet its initial cost and schedule estimates for restarting nuclear waste disposal operations at the Waste Isolation Pilot Plant (WIPP), resulting in a cost increase of about $64 million and a delay of nearly 9 months. DOE incurred this cost increase and delay partly because it did not follow all best practices in developing the cost and schedule estimates. In particular, DOE's schedule did not include extra time, or contingency, to account for known project risks. Instead, DOE estimated it would restart waste operations in March 2016 based on a schedule with no contingency that gave DOE less than a 1 percent chance of meeting its restart date. In January 2016, DOE approved new estimates that added 8.5 months to the schedule, extending the restart to December 2016; increased the estimated cost of recovery by $2 million; and resulted in an additional $61.6 million in costs for operating WIPP in fiscal year 2016. DOE's WIPP operations activity manager said the revised schedule included contingency. However, according to DOE officials, they did not follow other best practices. For example, DOE did not provide evidence of having an independent cost estimate to validate the revised estimate. DOE did not follow all best practices for cost and schedule estimates in part because DOE does not require that its cleanup operations, such as WIPP, follow these practices. Therefore, DOE cannot have confidence that its estimates are reliable. In contrast, DOE established new requirements in June 2015 that its capital asset projects, such as the new ventilation system at WIPP, follow these best practices. By also requiring cleanup operations to follow them, DOE would have more confidence in the estimates for cleanup operations and capital asset projects. DOE did not follow all best practices in analyzing and selecting an alternative for the new ventilation system at WIPP. As a result, DOE's analysis was not reliable and DOE cannot be confident that the alternative it selected in December 2015 will best provide the needed capabilities at WIPP. The analysis of alternatives (AOA) process entails identifying, analyzing, and selecting a preferred alternative to best meet the mission need. Of the four categories of best practices for AOAs, DOE's process fully met the category for identifying alternatives. For example, DOE identified a broad range of ventilation alternatives. However, DOE only partially or minimally met the other three categories: general principles, analyzing alternatives, and selecting the preferred alternative. DOE did not follow the best practice to select the preferred alternative based on a cost-benefit analysis that assesses the difference between the life-cycle costs and benefits of each alternative. In addition, an independent review that DOE commissioned consistent with best practices found that DOE's AOA did not adequately document a cost-benefit analysis and that, as a result, the selection of the preferred alternative was not supported by compelling information. The independent review recommended that DOE conduct a cost-benefit analysis consistent with best practices. However, DOE did not conduct the recommended analysis and document it before selecting the final alternative because there was no requirement to do so. In June 2015, the Secretary of Energy directed DOE to develop guidance for conducting AOAs consistent with AOA best practices. A DOE official said the department expected to issue the new guidance by December 2016. GAO recommends that DOE require cleanup operations to follow best practices for cost and schedule estimates and require projects, including the WIPP ventilation system, to implement recommendations from independent AOA reviews or document the reasons for not doing so. DOE concurred with the recommendations. |
In our June 2015 report, we found that Interior uses a multistep review process to help ensure that tribal-state compacts, and any compact amendments, comply with IGRA, other federal laws not related to jurisdiction over gaming on Indian lands, and the trust obligation of the United States to Indians. Interior’s Office of Indian Gaming is the lead agency responsible for managing the multistep process for reviewing compacts submitted by tribes and states. The Office of Indian Gaming coordinates its compact reviews with Interior’s Office of the Solicitor. The Office of Indian Gaming submits a final analysis and recommendation regarding compact approval to the Assistant Secretary of Indian Affairs, who makes a final decision on whether to approve the compact. Interior has 45 days to approve or disapprove a compact once it receives a compact package from a state and tribe. Under IGRA, any compacts Interior does not approve or disapprove within 45 days of submission are considered to have been approved (referred to as deemed approved), but only to the extent they are consistent with IGRA. From 1998 through fiscal year 2014, Interior reviewed and approved most of the 516 compacts and compact amendments that states and tribes submitted. Specifically, 78 percent (405) were approved; 12 percent (60) were deemed approved; 6 percent (32) were withdrawn or returned; and about 4 percent (19) were disapproved. In the decision letters we reviewed for the few disapproved compacts (19 out of 516), the most common reason for disapproval was that compacts contained revenue sharing provisions Interior found to be inconsistent with IGRA. For example, Interior found the revenue sharing payment to the state in some compacts to be a tax, fee, charge, or assessment on the tribe, which is prohibited by IGRA. For one compact, Interior found the state’s offer of support for the tribe’s application to take land into trust did not provide a quantifiable economic benefit that justified the proposed revenue sharing payments. Consequently, Interior viewed the payment to the state as a tax or other assessment in violation of IGRA. Interior also disapproved compacts for other reasons, including that compacts were signed by unauthorized state or tribal officials, included lands to be used for gaming that were not Indian lands as defined by IGRA, or included provisions that were not directly related to gaming. Interior did not approve or disapprove 60 of the 516 compacts submitted by tribes and states within the 45-day review period. As a result, these compacts were deemed approved to the extent that they are consistent with IGRA. According to Interior officials, as a general practice, the agency only sends a decision letter to the tribes and state for deemed approved compacts to provide guidance on any provisions that raised concerns or may have potentially violated IGRA. We reviewed the decision letters for 26 of the 60 deemed approved compacts. In 19 of the 26 letters we reviewed, Interior described concerns about the compact’s revenue sharing provisions, and most of these letters also noted concerns about the inclusion of provisions not related to gaming. The remaining 7 letters we reviewed cited other concerns, such as ongoing litigation, that could affect the compact. As we found in our June 2015 report, the roles of states and tribes in regulating Indian gaming are established in two key documents: (1) compacts for class III gaming and (2) tribal gaming ordinances for both class II and class III gaming. Compacts lay out the responsibilities of both tribes and states for regulating class III gaming. For example, compacts may include provisions allowing states to conduct inspections of gaming operations, certify employee licenses, review surveillance records, and impose assessments on tribes to defray the state’s costs of regulating Indian gaming. Under IGRA, tribal gaming ordinances—which outline the general framework for tribes’ regulation of class II and class III gaming—must be adopted by a tribe’s governing body and approved by the Commission’s Chair before a tribe can conduct class II or class III gaming, as required under IGRA. Tribal ordinances must contain certain required provisions that provide, among other things, that the tribe will have sole proprietary interest and responsibility for the conduct of gaming activity; that net gaming revenues will only be used for authorized purposes; and that annual independent audits of gaming operations will be provided to the Commission. IGRA allows states and tribes to agree on how each party will regulate class III gaming, and we found that regulatory roles vary among the 24 states that have class III Indian gaming operations. We identified states as having either an active, moderate, or limited role to describe their approaches in regulating class III Indian gaming, primarily based on information states provided on the extent and frequency of their monitoring activities. Monitoring activities conducted by states ranged from basic, informal observation of gaming operations to testing of gaming machine computer functions and reviews of surveillance systems and financial records. We also considered state funding and staff resources allocated for regulation of Indian gaming, among other factors, in our identification of a state’s role. Based on our analysis of states’ written responses to questions and interviews with states we found the following: Seven states have an active regulatory role: Arizona, Connecticut, Kansas, Louisiana, New York, Oregon, and Wisconsin. These states monitor gaming operations at least weekly, with most having a daily on-site presence. Over 17 percent (71 of 406) of class III Indian gaming operations are located in these seven states, accounting for about 25 percent of gross gaming revenue in fiscal year 2013. These states perform the majority of monitoring activities, including formal and informal inspection or observation of gaming operations; review of financial report(s); review of compliance with internal control systems; audit of gaming operation records; verification of gaming machines computer functions; review of gaming operator’s surveillance; and observation of money counts. Eleven states have a moderate regulatory role: California, Florida, Iowa, Michigan, Minnesota, Nevada, New Mexico, North Dakota, Oklahoma, South Dakota, and Washington. Most of these states monitor operations at least annually, and all collect funds from tribes to support state regulatory activities. About 75 percent (303 of 406) of class III Indian gaming operations are located in these states and generated 69 percent of all gross Indian gaming revenue in fiscal year 2013. States with a moderate regulatory role have the broadest range of regulatory approaches. For example, according to Nevada officials, Nevada conducts comprehensive inspections of gaming operations once every 2 to 3 years and performs covert inspections, as needed, based on risk. In contrast, North Dakota officials told us they conduct monthly inspections of gaming operations and an annual review of financial reports. Six states have a limited regulatory role: Colorado, Idaho, Mississippi, Montana, North Carolina, and Wyoming. The role of these states is largely limited to negotiating compacts with tribes, and they do not incur substantial regulatory costs or regularly perform monitoring activities of class III Indian gaming operations. Eight percent (32 of 406) of class III Indian gaming operations are located in these states, and the operations accounted for about 4 percent of gross Indian gaming revenue in fiscal year 2013. Tribes take on the primary day-to-day role of regulating Indian gaming. For example, each of the 12 tribes that we visited had established tribal gaming regulatory agencies that perform various regulatory functions to ensure that their gaming facilities are operated in accordance with tribal laws and regulations and, for class III operations, compacts. The tribes’ regulatory agencies were similar in their approaches to regulating their gaming operations. For example, all of the tribes’ regulatory agencies had established procedures for developing licensing procedures for employees, obtaining annual independent outside audits, and establishing and monitoring gaming activities to ensure compliance with tribal laws and regulations. Among other things, representatives from tribal associations we contacted emphasized that tribal governments have worked diligently to develop regulatory systems to protect the integrity of Indian gaming and have dedicated significant resources to meet their regulatory responsibilities. For example, according to representatives of the National Indian Gaming Association, in 2013, tribal governments dedicated $422 million to regulate Indian gaming, including funding for tribal government gaming regulatory agencies, state gaming regulation, and Commission regulation and oversight of Indian gaming collected through fees required by IGRA. In our June 2015 report, a key difference we found between class II and class III gaming is that IGRA authorizes the Commission to issue and enforce minimum internal controls standards for class II gaming but not for class III gaming. Commission regulations require tribes to establish and implement internal control standards for class II gaming activities— such as requirements for surveillance and handling money—that provide a level of control that equals or exceeds the Commission’s minimum internal control standards. But, in 2006, a federal court ruled that IGRA did not authorize the Commission to issue and enforce regulations establishing minimum internal control standards for class III gaming. However, Commission regulations establishing minimum internal control standards, including standards for class III gaming, that were issued before the ruling were not struck down by the court or withdrawn by the Commission. The Commission issued these regulations in 1999 and last updated the standards in 2006, which we refer to as the 2006 regulations.gaming, the Commission continues to (1) conduct audits using the 2006 regulations at the request of tribes and (2) provide monitoring and enforcement of these regulations for 15 tribes in California with approved Since the court decision, for operations with class III tribal gaming ordinances that call for the Commission to have such a role. The Commission plans to issue guidance with updated minimum internal control standards for class III gaming and withdraw its 2006 regulations. Commission officials told us they have authority to issue such guidance, and tribes could voluntarily adopt them as best practices. According to Commission officials, issuing such guidance would be helpful because updated standards could be changed to reflect technology introduced since the standards were last updated. Commission officials told us that before the agency can make a decision on how to proceed with issuing guidance for class III minimum internal control standards, it first needs to consult with tribes. In February 2015, the Commission notified tribes of plans to seek comments on its proposal to draft guidance for updated class III minimum internal control standards during meetings in April and May 2015. States involved in the regulation of Indian gaming are also impacted by the Commission’s proposal to draft updated guidance and withdraw its 2006 regulations; however, the Commission’s plans for obtaining state input on this proposal are unclear. We found that many tribal-state compacts incorporate by reference the Commission’s 2006 regulations establishing minimum internal control standards. For example, three states have tribal-state compacts that require tribes to comply with the Commission’s 2006 regulations. If the Commission withdraws its 2006 regulations, it is not clear what minimum internal control standards the compacts would require tribes to meet. In addition, nine states have tribal- state compacts that require tribal internal control standards to be at least as stringent as the Commission’s 2006 regulations. If the Commission withdraws its 2006 regulations, these states and tribes would no longer have a benchmark against which to measure the stringency of tribal internal control standards. Standards for Internal Control in the Federal Government call for management to ensure that there are adequate means of communicating with, and obtaining information from, external stakeholders that may have a significant impact on the agency achieving its goals. According to a Commission official, the Commission is considering conducting outreach to the states on its proposal but did not have any specific plan for doing so. Consistent with federal internal control standards, seeking state input is important, as it could aid the Commission in making an informed decision on how to proceed with issuing such guidance and whether withdrawal of its 2006 regulations would cause complications or uncertainty under existing tribal-state compacts. As a result of this finding, we recommended that the Commission seek input from states regarding its proposal to draft updated guidance on class III minimum internal control standards and withdraw its 2006 regulations. In its comments on our draft report, the Commission concurred with this recommendation. To help ensure compliance with IGRA and Commission regulations, the Commission conducts a broad array of monitoring activities—such as reviewing independent audit reports submitted annually by tribes, conducting site visits to tribal gaming operations to examine compliance with applicable Commission regulations, and assessing tribes’ compliance with minimum internal control standards as part of Commission-led audits. In addition, as required by IGRA, the Commission’s Chair reviews and approves various documents related to both class II and class III gaming operations, including tribal gaming ordinances or resolutions adopted by a tribe’s governing body. Under its ACE initiative, the Commission has emphasized providing tribes with training and technical assistance as a means to build and sustain their ability to prevent, respond to, and recover from weaknesses in internal controls and violations of IGRA and Commission regulations. For instance, the Commission hosts two regular training events in each region. Commission staff also provide one-on-one training on specific topics, as needed, during site visits and offer technical assistance in the form of guidance and advice to tribes on compliance with IGRA; Commission regulations; and day-to-day regulation of Indian gaming operations through written advisory opinions and bulletins. Commission staff also respond to questions by phone and e-mail, among other activities. However, the effectiveness of the Commission’s training and technical assistance efforts remains unclear. The Commission’s strategic plan for fiscal years 2014 through 2018 includes two goals corresponding to its focus on training and technical assistance to achieve compliance with IGRA and Commission regulations: one for continuing its ACE initiative; and another for improving its technical assistance and training to tribes. Yet, the Commission’s performance measures for tracking progress toward achieving these two goals are largely output-oriented rather than outcome-oriented, and overall do not demonstrate the effectiveness of the Commission’s training and technical assistance efforts. Specifically, 12 of the 18 performance measures for these two goals include output-oriented measures describing the types of products or services delivered by the Commission. For example, they include the number of audits and site visits conducted and the number of training events and participants attending these training events. In prior work, we found that these types of measures do not fully provide agencies with the kind of information they need to determine how training and development efforts contribute to improved performance, reduced costs, or a greater capacity to meet new and emerging transformation challenges. In that work, we concluded that it is important for agencies to develop and use outcome-oriented performance measures to ensure accountability and assess progress toward achieving results aligned with the agency’s mission and goals. This is consistent with Office of Management and Budget guidance, which encourages agencies to use outcome performance measures—those that indicate progress toward achieving the intended result of a program— where feasible. The Commission’s remaining 6 measures include outcome-oriented measures that track tribes’ compliance with specific requirements, including the percentage of gaming operations that submit audit reports on time and have a Chair-approved tribal gaming ordinance. They do not, however, indicate the extent to which minimum internal control standards are implemented or reflect improvements in the overall management of Indian gaming operations. In addition, they do not correlate such compliance with the Commission’s training and technical assistance efforts. Additional outcome-oriented performance measures would enable the Commission to better assess the effectiveness of its training and technical assistance efforts and its ACE initiative. Commission officials told us that they recognize they have more work to do on performance measures and are interested in taking steps to ensure that their ACE initiative is meeting its intended goals. In our June 2015 report,recommended that the Commission review and revise, as needed, its performance measures to include additional outcome-oriented measures. In its comments on our draft report, the Commission concurred with our recommendation. The Commission amended its regulations in August 2012 to formalize an existing practice of sending letters of concern to prompt tribes to voluntarily resolve potential compliance issues. A letter of concern outlines Commission concerns about a potential compliance issue and, according to Commission regulations, is not a prerequisite to an enforcement action.concern specify a time period by which a recipient must respond but do Commission regulations require that letters of not address which compliance issues merit a letter of concern or indicate when a letter should be sent once a potential compliance issue is discovered. The Commission also has not issued guidance or documented procedures on how to implement its regulation regarding letters of concern. In our review of letters of concern sent by the Commission in fiscal years 2013 and 2014, we found that the Commission sent 16 letters of concern to 14 tribes. Six of the 16 letters of concern did not include a time period by which the recipient was to respond, as required by Commission regulations. In addition, 12 letters did not specify in the subject line, or elsewhere in the letter, that they were letters of concern. By not including a time period for a response as required by Commission regulations and not consistently identifying its correspondence as a letter of concern, the Commission may not be able to ensure timely responses, and tribes may find it difficult to discern the significance of these letters. In addition, the Commission provided us with documentation to demonstrate whether a tribe took action to address the issues described in 8 letters of concern, but it did not provide documentation for the remaining 8 letters. Under federal internal control standards, federal agencies are to clearly document transactions and other significant events, and that documentation should be readily available for examination. Without guidance or documented procedures to inform its staff about how to complete letters of concern or maintain documentation tracking tribal actions, the Commission cannot ensure consistency in the letters that it sends to tribes, and it may be difficult to measure the effectiveness of the letters in encouraging tribal actions to address potential issues. As a result of these findings, we recommended and in its comment letter the Commission in our June 2015 report,generally agreed, that the Commission should develop documented procedures and guidance for letters of concern to (1) clearly identify letters of concern as such and to specify the type of information to be contained in them, such as time periods for a response; and (2) maintain and track tribes’ responses to the Commission on potential compliance issues. IGRA authorizes the Commission Chair to take enforcement actions for violations of IGRA and applicable Commission regulations for both class II and class III gaming. Specifically, the Commission Chair may issue a notice of violation or a civil fine assessment for violations of IGRA, Commission regulations, or tribal ordinances and, for a substantial violation, a temporary closure order. The most common enforcement action taken by the Commission Chair in fiscal years 2005 through 2014 was a notice of violation. The Chair issued 107 notices of violations that cited 119 violations during this period. We found that the Chair issued 100 out of 107 notices of violation prior to fiscal year 2010. Since fiscal year 2010, fewer enforcement actions may have been taken because recent Commission chairs have emphasized seeking voluntary compliance with IGRA. Chairman Barrasso, Vice Chairman Tester, and Members of the Committee, this completes my prepared statement. I would be pleased to answer any questions that you may have at this time. If you or your staff members have any questions about this testimony, please contact me at (202) 512 -3841 or fennella@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Other individuals who made key contributions to this testimony include Jeff Malcolm (Assistant Director), Cheryl Arvidson, Amy Bush, Jillian Cohen, John Delicath, Justin Fisher, Paul Kazemersky, Dan Royer, Jeanette Soares, Kiki Theodoropoulos, Swati Sheladia Thomas, and Lisa Turner. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Over the past 25 years, Indian gaming has become a significant source of revenue for many tribes, reaching $28 billion in fiscal year 2013. IGRA, the primary federal statute governing Indian gaming, provides a statutory basis for the regulation of Indian gaming. Tribes, states, Interior, and the Commission have varying roles in Indian gaming. This testimony highlights the key findings of GAO's June 2015 report ( GAO-15-355 ). Accordingly, it addresses (1) Interior's review process to help ensure that tribal-state compacts comply with IGRA; (2) how states and selected tribes regulate Indian gaming; (3) the Commission's authority to regulate Indian gaming; and (4) the Commission's efforts to ensure tribes' compliance with IGRA and Commission regulations. For the June 2015 report, GAO analyzed compacts and Commission data on training, compliance, and enforcement; and interviewed officials from Interior, the Commission, states with Indian gaming, and 12 tribes in six states GAO visited based on geography and gaming revenues generated. In its June 2015 report, GAO found that the Department of the Interior (Interior) has a multistep review process to help ensure that compacts—agreements between a tribe and state that govern the conduct of the tribe's class III (or casino) gaming—comply with the Indian Gaming Regulatory Act (IGRA). From 1998 through fiscal year 2014, Interior approved 78 percent of compacts; Interior did not act to approve or disapprove 12 percent; and the other 10 percent were disapproved, withdrawn, or returned. GAO also found that states and selected tribes regulate Indian gaming in accordance with their roles and responsibilities established in tribal-state compacts for class III gaming, and tribal gaming ordinances, which provide the general framework for day-to-day regulation of class II (or bingo) and class III gaming. In addition, the 24 states with class III Indian gaming operations vary in their approaches for regulating Indian gaming, from active (e.g., daily or weekly on-site monitoring) to limited (e.g., no regular monitoring). Further, all 12 selected tribes GAO visited had regulatory agencies responsible for the day-to-day operation of their gaming operations. In GAO's June 2015 report, GAO found that the National Indian Gaming Commission (Commission)—an independent agency within Interior created by IGRA—has authority to regulate class II gaming, but not class III gaming, by issuing and enforcing gaming standards. The Commission is considering issuing guidance with class III standards that may be used voluntarily by tribes and has held consultation meetings to obtain tribal input. However, in June 2015, GAO found the Commission does not have a clear plan for conducting outreach to affected states on its proposal. Federal internal control standards call for managers to obtain information from external stakeholders that may have a significant impact on the agency achieving its goals. Along with tribes, state input could aid the Commission in making an informed decision. Even with differences in its authority for class II and class III gaming, GAO found that the Commission helps ensure that tribes comply with IGRA and applicable federal and tribal regulations through various activities, including monitoring gaming operations during site visits to Indian gaming operations and Commission-led audits. In addition, since 2011, the Commission has emphasized efforts that encourage voluntary compliance with regulations, including providing training and technical assistance and alerting tribes of potential compliance issues using letters of concern. However, the effectiveness of these two approaches is unclear. GAO found in June 2015 that the Commission had a limited number of performance measures that assess outcomes achieved. With such additional measures, the Commission would be better positioned to assess the effectiveness of its training and technical assistance. Further, GAO found the Commission does not have documented procedures, consistent with federal internal control standards, about how to complete or track letters of concern to help ensure their effectiveness in encouraging tribal actions to address identified potential compliance issues. Without documented procedures, the Commission cannot ensure consistency or effectiveness of the letters it sends. In its June 2015 report, GAO recommended that the Commission: (1) obtain input from states on its plans to issue guidance on class III minimum internal control standards; (2) review and revise, as needed, its performance measures to better assess its training and technical assistance efforts; and (3) develop documented procedures and guidance to improve the use of letters of concern. The Commission generally agreed with GAO's recommendations. |
Approximately 8,500 projects that have been financed with Federal Housing Administration (FHA)-insured mortgages are also supported by project-based Section 8 housing assistance payments contracts. In 1996, a HUD contractor estimated that for approximately 63 percent of these multifamily projects, the rents are higher than those of comparable unassisted rental units in the same housing rental market, which increases the costs of the Section 8 program to the federal government. However, if the Section 8 assistance is reduced, owners of many of the FHA-insured housing projects are likely to default on their mortgage payments, resulting in substantial claims to the FHA General Insurance Fund. In addition to the cost of insurance claims, defaults on mortgage payments could leave tenants without adequate affordable housing. The original project-based Section 8 contracts that were entered into in the 1970s and 1980s were typically for 15, 20, or 40 years and began expiring in the early 1990s. The Congress and HUD have worked together to renew expiring contracts for 1 year. As its long-term Section 8 contracts expire and its 1-year contracts are renewed annually, HUD estimates that its annual renewal costs will increase steadily. For example, HUD estimated that if no actions were taken, the annual cost of renewing project-based Section 8 contracts would rise to over $7 billion—or approximately one-third of HUD’s current total budget—by 2007. To address these escalating costs, the Congress appropriated $30 million to conduct a mark-to-market demonstration program in fiscal year 1996. This demonstration was intended to test various methods of restructuring the financing of insured Section 8 properties, generally by adjusting rents to market levels and reducing the mortgage debt to reflect any decline in net operating income resulting from the rent adjustment. Near the end of fiscal year 1996, the Congress repealed the demonstration program but authorized a new demonstration for fiscal 1997. To deal with Section 8 contract expirations occurring in fiscal year 1999 and thereafter, the Congress legislated a permanent mark-to-market program in October 1997. This legislation also extended the fiscal year 1997 demonstration program through fiscal 1998 with modifications as a transitional program while HUD developed regulations to implement the permanent program. The intended goals of the permanent mark-to-market program are to reduce the costs of expiring Section 8 contracts, address financially and physically troubled projects, correct management and ownership deficiencies, and preserve the affordability and availability of low-income rental housing. (See app. I for an overview of the mark-to-market process.) The act directed HUD to select capable entities to carry out restructuring under the mark-to-market program on behalf of the federal government. These entities are referred to as participating administrative entities (PAE). Entities eligible to apply to become PAEs include public agencies, such as state housing finance agencies, local housing agencies, nonprofit organizations, and these entities in partnership with each other or with private entities. The act specifies selection criteria such as experience in working directly with tenant organizations and other community-based organizations; experience with multifamily housing financing and restructuring; a history of stable, financially sound, and responsible administrative performance; and financial strength. The act established OMHAR as an entity within HUD and required the Office to be under the management of a presidentially appointed and Senate-confirmed Director. Section 572(a) of the act specifies that not later than 60 days after its enactment (Oct. 27, 1997), the President should submit a nomination for the Director for OMHAR to the Senate. However, the Director was not nominated until September 29, 1998. The Senate confirmed the nomination on October 21, 1998. The act authorizes the Director to hire personnel to carry out OMHAR’s functions. However, because the Director was not appointed until October 21, 1998, HUD staff began formulating organization and staffing plans for OMHAR in the Director’s absence. HUD anticipates very targeted hiring of staff for OMHAR with expertise to carry out critical policy development, oversight, and decision-making functions. Section 574(a),(b) of the act allows the Director to pay OMHAR’s employees (1) without regard to certain provisions of pay schedules used to hire most federal employees and (2) comparably to the officers and employees of the Federal Deposit Insurance Corporation. In effect, this authority allows the Director to compensate OMHAR employees at a higher level than other HUD employees to obtain the skills and expertise needed to accomplish the program’s purposes. HUD plans to hire these staff in phases. OMHAR will have approximately 75 to 85 staff at its peak; however, this number may increase if additional hiring authority is obtained. Some HUD personnel are expected to be detailed to OMHAR, but according to mark-to-market officials, most will be hired from outside HUD. For the first quarter of fiscal year 1999, the 10 existing HUD staff working on the mark-to-market program will be detailed to OMHAR and additional support will be provided by HUD staff with collateral assignments. HUD envisions that these senior core staff will be assisted by 10 staff hired for 1-year terms. HUD intends on hiring additional staff with 4-year terms in three phases—at the end of October and December 1998 and the end of February 1999. Of the 75 additional staff expected to be hired during fiscal year 1999, approximately 15 will be distributed among five HUD field offices, and the remainder will be located in headquarters. Because mark-to-market is a relatively short-term effort—authority for both the program and OMHAR terminates on September 30, 2001—HUD plans to focus staff resources on inherently governmental functions and will generally not hire internal staff to carry out commercial functions that are readily obtainable through private-sector vendors. Instead, HUD expects to obtain capacity for such functions through contracts to support internal staff. While HUD’s plan to oversee the contractors is not yet final, HUD anticipates having several staff dedicated to managing the contractors. These staff would include the chief of a newly created office of contracts, compliance, and controls; three contract analysts/government technical representatives; a records management analyst; two internal review specialists; and a clerical support staff member. Figure 1 illustrates OMHAR’s proposed organizational structure. Housing experts involved with mark-to-market issues, including representatives of advocacy groups, whom we contacted were concerned that a Director of OMHAR had not yet been appointed at the time of our discussions. For example, one representative was concerned that the delay in the Director’s appointment limited the Director’s opportunity to participate directly in developing the program’s regulations. In addition, according to some representatives, the delay in appointing the Director and implementing the program sent a negative message to stakeholders about the administration’s support of the program. In response to this concern, HUD mark-to-market officials said that the administration has been supportive of the program and has provided access to any resources necessary to prepare for its implementation. As mandated in the act, HUD is required to implement the mark-to-market program for Section 8 contracts expiring in fiscal year 1999 (beginning Oct. 1, 1998) and thereafter. Despite delays in hiring a management studies contractor and a voluminous number of tasks to complete, HUD has made considerable progress toward this end. For example, HUD has published interim regulations for the mark-to-market program and is expecting to develop an operating procedures manual and to select its partners (the PAEs) to perform the restructurings in October 1998. However, in spite of HUD’s efforts, some of these tasks either have been or will be completed behind their original schedule. For example, in May 1998, HUD had planned to issue the program’s interim regulations by August and to have final regulations in place by October 1998. Now, HUD expects to have only the interim regulations in place by October 1998. Final regulations are to be developed 3 months after the OMHAR Director’s appointment. Also, while HUD originally planned to begin briefing PAEs on their responsibilities in September 1998, because of delays in selecting the PAEs, briefings are not scheduled to begin until October. Of eight key tasks, such as preparing briefing manuals and conducting technical assistance briefings for field office staff, HUD had completed one as of September 8, 1998, the most recent date for which this information was readily available. The remaining seven were in progress, although three of these were behind HUD’s original schedule. Appendix II provides a more detailed description of HUD’s progress in carrying out eight key tasks needed to implement the mark-to-market program. In addition to the operational tasks that HUD must perform, the act places some reporting and other requirements related to the mark-to-market program on HUD. For example, the act requires the OMHAR Director to report certain information to the Secretary of HUD and for the Secretary to submit information on OMHAR’s operations to the Congress and the Office of Management and Budget. (See app. III for information on the mark-to-market reporting requirements.) The act also requires that prior to issuing final regulations, HUD seek recommendations on the implementation of the mark-to-market program from individuals and organizations affected by the program and convene at least three public forums so that those individuals and organizations can express their views concerning the proposed disposition of the recommendations. This requirement is discussed in greater detail later in this report. While most representatives of the groups involved with mark-to-market issues whom we contacted were generally complimentary of mark-to-market officials’ efforts to implement the program, many expressed concerns about the delays HUD has experienced in issuing the program’s regulations and soliciting and selecting its administrative partners. For example, there were concerns that the delay in publishing the interim regulations could limit the time that affected parties have to review and comment on them before they become effective and that entities that are eligible to serve as PAEs might not be able to develop a sufficient understanding of the regulations before having to decide if they want to participate in the program. According to HUD mark-to-market officials, the public will have 45 days to comment on the interim regulations, which they believe is a sufficient amount of time. Although the regulations become effective 30 days after publication, which will be before the end of the comment period, these officials said that the comments received will be considered in developing the program’s final regulations. Furthermore, these officials said that the published notice soliciting PAEs contains an adequate description of the scope of work for entities to decide whether they want to participate. Because HUD is responsible for establishing effective management controls over the program’s implementation, the HUD officials developing the mark-to-market program were establishing or planning several procedures and systems to oversee its implementation. However, as of October 14, 1998, most of these procedures and systems were not yet in place, and many of the related details remained to be developed. Among other things, management controls can include (1) controls over program operations, which are policies and procedures that management implements to reasonably ensure that a program meets its objectives, and (2) controls over compliance with laws and regulations, which are policies and procedures to reasonably ensure that resources are used consistent with laws and regulations. Because many of the functions necessary under the mark-to-market program will be carried out by the PAEs rather than by HUD, it will be particularly important for HUD to establish procedures ensuring that all the parties involved are carrying out their responsibilities in ways that meet the program’s objectives and are in compliance with the program’s requirements. To its credit, HUD has focused on developing oversight procedures prior to the program’s implementation. In general, to oversee the operations of the HUD field offices and PAEs in the mark-to-market program, HUD is developing a plan for centralized review and oversight. Within its general oversight system, HUD has either developed or is planning three key components to help ensure that field offices and PAEs carry out the program in a way that meets its objectives and is in compliance with requirements: (1) performance measures to judge the effectiveness of PAEs’ activities and a system of compensating PAEs, (2) an Internet-based tracking system to monitor and analyze actions taken by HUD field offices and PAEs in carrying out mark-to-market functions, and (3) an oversight and audit guide to test compliance with the program’s requirements and objectives. As of October 14, 1998, HUD was still in the process of developing all three of these components. The components and their status are presented in table 1. In addition to the processes described in table 1 for HUD’s general oversight of field offices’ and PAEs’ performance under the mark-to-market program, HUD is responsible for monitoring several specific components of the program as they are implemented. These components include (1) screening owners and projects to determine their eligibility for restructuring, (2) setting projects’ rents, (3) determining projects’ rehabilitation needs, (4) restructuring projects’ mortgages, (5) recapturing Section 8 funds, (6) reviewing restructured properties, and (7) using technical assistance funds. As of October 14, 1998, HUD had developed procedures to oversee two of these components but was still in the process of developing oversight procedures for the other five. The program’s components and the description and status of HUD’s procedures to oversee their implementation are presented in appendix IV. HUD’s ability to evaluate the results of the various restructuring approaches allowed under the mark-to-market demonstration programs is limited because a relatively small number of demonstration transactions have been fully completed. The Congress authorized these demonstrations for fiscal years 1996-98 to explore approaches for restructuring the financing of and reducing the Section 8 assistance provided for properties eligible for the mark-to-market program. Despite the fact that relatively few transactions have been fully completed, HUD mark-to-market officials believe that the experience and process of carrying out the demonstrations have yielded information useful for implementing the permanent program. In April 1996, the Congress passed legislation authorizing the fiscal year 1996 mark-to-market demonstration—a voluntary program. The repeal of this demonstration near the end of fiscal year 1996 did not nullify any agreements or proposals that had already been considered under the program, and HUD continued to process proposals that had been received prior to its termination. For project-based Section 8 contracts expiring in fiscal years 1997 and 1998, the Congress again authorized optional demonstration programs to explore approaches to restructuring the debt secured by these properties while minimizing adverse impacts on tenants, owners, and communities. As authorized by the legislation, restructuring transactions were performed by HUD field offices, by HUD in partnership with state and local housing finance agencies (HFA), and by HUD in joint ventures with nonprofit entities. As of October 2, 1998, the latest date for which information was readily available, HUD had executed contracts with 25 HFAs and entered into 4 joint venture agreements with nonprofit entities to restructure projects under the demonstrations. As of October 2, 1998, nearly 300 properties were participating in the demonstration programs. Restructuring had been completed for 42 of these properties, and all but 4 (which were processed under a joint venture arrangement) of these had been processed by HUD staff. None of the restructurings had been completed under HFA arrangements as of that date. Consequently, information that HUD could use to assess outcomes under the various restructuring approaches has been limited. Table 2 shows the number of projects participating in each component of the demonstrations and their status. Despite the relatively few demonstration transactions that have been fully completed, according to HUD mark-to-market officials, they have used information from the demonstrations to identify, evaluate, and improve, as necessary, the processes used by HUD field offices and third parties in carrying out restructuring activities. Specifically, information from the demonstrations was used to help develop process flow charts and an operating procedures manual for the permanent mark-to-market program. These charts and the manual were compiled by mark-to-market officials, with assistance from HUD’s management studies contractor, by reviewing and documenting HUD’s policy and program requirements under the demonstrations, identifying weak policies and procedures in the process, and recommending improvements in the program’s design and delivery. In addition, a front-end risk assessment of the fiscal year 1997 demonstration program identified some concerns, which HUD mark-to-market officials told us that they had considered in establishing the permanent program. For instance, the front-end risk assessment noted that concerns over potential tax liabilities kept many owners from participating in the program, owing either to an inability or an unwillingness to handle the tax consequences or a resistance to participating until anticipated changes to the tax legislation became known. (See app. II for more information on resolution of the tax issue.) The 1997 front-end risk assessment also identified communication and the coordination of HUD’s functions as an area of concern. For instance, the risk assessment stated that HUD’s information technology staff ideally should have been involved in developing information systems for the fiscal year 1997 demonstration program earlier in the process to ensure that all departmental documentation and other information technology requirements were met. The risk assessment concluded that, in general, the program’s staff should be representative of all applicable areas of HUD or should involve appropriate personnel from the earliest stages of the process to avoid the need to substantially modify processes late in the program. According to HUD mark-to-market officials, they have applied this “lesson learned” by involving other areas within HUD’s Office of Multifamily Housing, such as Section 8 and Multifamily Claims staff, and other HUD offices, such as the Office of Policy Development and Research and the Office of General Counsel, throughout the development of the permanent program. According to mark-to-market officials, the demonstration programs also provided other information useful for working out the details of the permanent program. For instance, mark-to-market officials believe that their efforts in developing agreements with state and local HFAs laid the foundation for HFAs’ participation in the permanent program as PAEs. Similarly, issues raised by the nonprofit partners under the joint venture arrangements have been instructive as mark-to-market officials consider details and policies under the permanent program, such as issues related to adjusting rents to market levels without mortgage restructuring and the compensation structure for nonprofit partners. With respect to the lessons learned from the component of the demonstration program carried out by HUD field offices, mark-to-market officials said that they have compiled an ongoing question-and-answer document to respond to the issues raised by the field offices involved. These mark-to-market officials said that they have communicated with the field offices to discuss the advantages and disadvantages of certain approaches and to develop responses to the questions. According to these officials, this effort has assisted in developing the permanent program by giving them a better understanding of the potential effects of policy decisions on the participating field offices. Under the mark-to-market legislation, HUD must establish procedures to provide an opportunity for parties affected by restructuring to participate in the process. The act also requires HUD to hold at least three public forums to obtain recommendations on the implementation of certain legislative provisions. In accordance with these requirements, HUD’s program guidance includes steps for involving affected parties in the restructuring process, and, on October 1, 1998, mark-to-market staff held the three required public forums. In addition, in preparing to implement the program, HUD has sought input from a variety of groups representing those affected by the program. Section 514(f) of the act requires HUD to establish procedures to provide an opportunity for tenants, neighborhood residents, local government, and other affected parties to participate in the restructuring process. These procedures must require consultation with affected parties in connection with, at a minimum, a project’s mortgage restructuring and rental assistance sufficiency plan, any proposed sale or transfer of the project to another entity, and the rental assistance assessment plan developed to determine whether to renew the project’s Section 8 assistance with project-based or tenant-based assistance. The act further directs that, to the extent practicable and consistent with the need to accomplish the restructuring of Section 8 projects in an efficient manner, the procedures should give all such parties an opportunity to provide the PAE with comments in writing, in meetings, or in another appropriate manner. HUD has addressed this legislative requirement by outlining procedures in the program’s draft operating procedures manual for affected parties’ participation. The manual identifies several points when affected parties must be given the opportunity to participate. For instance, the manual requires a PAE to conduct at least one consultation meeting before a project’s physical needs assessment is completed. The consultation meeting is to provide the PAE with an opportunity to receive oral presentations and comments and respond to tenants’ and communities’ concerns regarding the mortgage restructuring and rental assistance sufficiency plan, the rental assistance assessment plan, and any proposed sale or transfer of the project. In addition, within 10 days of the PAE’s completion of the proposed rental assistance assessment plan, the owner must send a notice and provide an opportunity to comment to the project’s residents, a local government representative, a representative of the public housing authority, and representatives of neighborhood residents or other affected parties, as determined by the PAE. Section 514 of the act also authorized up to $10 million annually in funding for tenant groups, nonprofit organizations, and public entities for building the capacity of tenant organizations, technical assistance, and tenant services. HUD has designated a portion of these funds to enhance opportunities for affected parties to participate in the mark-to-market process through the Outreach and Training Grants (OTAG) program. Under the program, $6 million is available to provide technical assistance for tenants of eligible mark-to-market projects so that the tenants can (1) participate meaningfully in the mark-to-market program and (2) affect decisions about the future of their housing. Grant funds can be used for, among several other activities, organizing residents of eligible low-income housing so that tenants can effectively participate in the mark-to-market process. The act also specifically requires that, in addition to obtaining public comments before publishing the program’s final regulations, HUD seek recommendations from various parties on how it should implement the legislative provisions governing the selection of PAEs and requiring that certain Section 8 contracts be renewed as project-based assistance. To seek views concerning HUD’s proposed disposition of these recommendations, the legislation requires HUD to convene at least three public forums involving organizations representing state and local HFAs, other potential PAEs, tenants, project owners and managers, state and local governments, and mortgagees. HUD held these forums on October 1, 1998, in New York, Chicago, and San Francisco; these locations were selected because, in part, of the levels of contract expirations in these areas. However, when the forums were held, the process of selecting PAEs—one of the forum topics on which HUD was soliciting recommendations—was already under way. According to HUD mark-to-market officials, interested parties had extensive opportunities to provide input prior to the start of the PAE selection process. Going beyond these legislative requirements, on December 18, 1997, HUD’s Office of Multifamily Housing began a consultation process with representatives from a wide variety of national advocacy groups interested in HUD’s implementation of the permanent mark-to-market program. Organizations representing owners and managers of FHA-insured assisted housing, tenant groups, nonprofit organizations, lenders, coalitions of local and state government agencies, and other advocacy groups were invited to submit ideas and comments on developing the new program. HUD received concept papers from housing groups representing several interested parties and from February 2 through 27, 1998, held 13 full-day meetings to discuss various issues related to mark-to-market’s implementation. According to HUD mark-to-market officials, the information presented during these consultations was used in developing the mark-to-market program and considered by the Department in drafting the program’s interim regulations. In addition to the concept papers and meetings, mark-to-market officials have also coordinated separately with the National Council of State Housing Agencies and individual housing finance agencies to obtain feedback from them regarding HFAs’ involvement in the mark-to-market program. For instance, in May 1998, the mark-to-market official responsible for the HFA component of the 1997 and 1998 demonstration programs met with the Council to obtain HFAs’ views on the compensation structure for their participation in the program. According to mark-to-market officials, they have been meeting with HFAs on a continuing basis to discuss the program and hear their concerns because their involvement as PAEs is key to the program’s success. The representatives of groups involved with mark-to-market issues whom we contacted generally believed that HUD’s efforts to obtain input on the program’s implementation were productive. However, three of these representatives were concerned that, since the meetings in February, HUD had not given any indication of its tentative decisions on the issues discussed. Some of these representatives said that, without knowing HUD’s intentions prior to the Department’s publication of the interim regulations for public comment, their groups may be less able to provide meaningful input on the program’s implementation, particularly if the interim regulations govern the program for a period of time. According to HUD mark-to-market officials, it would have been inappropriate to solicit information from the public during the process of writing the regulations. Accordingly, at the meetings in February, mark-to-market officials notified participants that there would be no further discussions until the regulations were completed and that any further comments would have to be submitted during the comment period. We provided HUD with a draft of this report for its review and comment. HUD agreed that our report accurately described the Department’s efforts in developing and implementing the mark-to-market program. HUD stated that our review was helpful to the Department in confirming that it was focusing its attention and efforts on issues that needed to be addressed. HUD also noted that it had developed an extremely aggressive work plan to accomplish the development and implementation of the mark-to-market program and that while some key tasks were accomplished later than provided for in its original schedule, a small but efficient core group of mark-to-market staff members have accomplished several significant work items, even in the absence of a Director. HUD suggested some minor technical changes, which we have incorporated. The complete text of HUD’s comments appears in appendix V. Section 577 of the act requires us to audit the operations of OMHAR annually for the first 2 fiscal years following the date of enactment (Oct. 27, 1997). For this assignment, our work focused on the status of OMHAR’s development and HUD’s plans for implementing the permanent mark-to-market program. To carry out the assignment’s objectives, we interviewed the HUD officials responsible for mark-to-market operations and reviewed documentation related to their plans for developing OMHAR and implementing the program, including their work plans, time frames, and draft program guidance. We also discussed HUD’s efforts to develop OMHAR and prepare for the program’s implementation with representatives of groups involved with issues related to mark-to-market, including two independent housing consultants, the National Council of State Housing Agencies, the National Leased Housing Association, the National Housing Law Project, the National Alliance of HUD Tenants, and the National Housing Conference. Because OMHAR was not yet established and HUD’s plans for implementing the permanent mark-to-market program were still being developed, we did not attempt to evaluate the effectiveness of OMHAR’s operations or HUD’s implementation plans. We performed our work at HUD in Washington, D.C., from March through September 1998 in accordance with generally accepted government auditing standards. We are sending copies of this report to the Secretary of Housing and Urban Development. We will make copies available to others on request. Please call me at (202) 512-7631 if you or your staff have any questions. Major contributors to this report are listed in appendix VI. As the Department of Housing and Urban Development (HUD) plans to implement it, the process to be used under the permanent mark-to-market program has nine phases: (1) screening owners and projects for eligibility, (2) assigning projects for restructuring, (3) preserving affordable housing, (4) collecting data, (5) underwriting, (6) approving the loan, (7) closing, (8) distributing postclosing documents, and (9) servicing and monitoring. This appendix describes the key activities that are to occur during each of the nine phases, as outlined in HUD’s draft operating procedures manual. 1. Screening Owners and Projects for Eligibility: The first phase of the restructuring process is determining a project’s eligibility. To be eligible to participate in the mark-to-market program, a project must be financed by a Federal Housing Administration (FHA)-insured or Secretary-held mortgage, receive project-based assistance, and have rents that exceed comparable market rents. HUD has established additional criteria for determining if a project is eligible to (1) renew its Section 8 contract at market rents and restructure the project’s debt, (2) renew its Section 8 contract at market rents without restructuring the project’s debt, or (3) renew its Section 8 contract at the lesser of the project’s current rents with an operating cost adjustment factor, budget-based rents with an operating cost adjustment factor, or base rents with an operating cost adjustment factor. HUD will extend the Section 8 contracts for projects with contracts that are due to expire while they are in the mark-to-market “pipeline,” such as projects undergoing eligibility determinations, or while they are participating in the development of a mortgage debt restructuring plan. Under these circumstances, a project’s Section 8 contract may be extended at the project’s current contract rent for up to 9 months, with the option of a further extension of 3 months. A project may not be restructured, and its contract may not be renewed if its owner has engaged in material adverse financial or managerial actions or omissions with regard to the project or other federally assisted, financed, or insured projects. Furthermore, a project may not be restructured if its condition is too poor to be rehabilitated at a reasonable cost or if the owner fails to comply with the requirements for restructuring. Such projects are to be directed into HUD’s enforcement program. 2. Assigning Projects for Restructuring: During the second phase of the mark-to-market process, HUD assigns multifamily projects to its administrative partners, who will actually perform the restructuring, which are referred to as participating administrative entities (PAE). Assets are first assigned to state and local housing finance agencies (HFA) that HUD has selected as PAEs. After this first round of assignments, non-HFA entities that HUD has selected to be PAEs can bid for the rights to restructure projects that remain. 3. Preserving Affordable Housing: One of the goals of the act is to preserve the affordability and availability of low-income rental housing through mortgage restructuring. As part of the restructuring process, a mortgage restructuring and rental assistance sufficiency plan is to be developed for each project. The development of this plan is expected to take, on average, 9 months. One element of the mortgage restructuring and rental assistance sufficiency plan is a rental assistance assessment plan, which the PAE develops to determine whether Section 8 assistance should be renewed as project-based or tenant-based assistance. The plan must also include a use restriction whereby the project’s owner agrees to maintain affordability for at least 30 years. Other elements required in the mortgage restructuring and rental assistance sufficiency plan, which will be developed in later phases, include a new rent subsidy level, a plan to restructure the debt, and a plan to finance the rehabilitation needs of the project. 4. Collecting Data: During the fourth phase of the process, the PAE collects and assesses project-specific data from project owners, lenders, servicers, or other third parties to be used during the underwriting and loan approval phases. The PAE is also to collect and analyze market studies, appraisals, and physical needs assessments to determine market rents, expenses, and the projects’ rehabilitation needs. 5. Underwriting: Once the data collection is complete, the fifth phase in the mark-to-market process is underwriting. During the underwriting phase, the size of the restructured first and the new second mortgage will be determined. The existing project mortgage will be restructured to provide a first mortgage that is sustainable at restructured market rents. The second mortgage may not exceed the difference between the restructured or new first mortgage and the current unpaid principal balance and must be an amount that can reasonably be expected to be repaid. The term of the second mortgage shall equal the term of the first mortgage. However, the only payments that need to be made on the second mortgage before repayment of the first are an amount equal to 75 percent of excess project income. 6. Approving the Loan: The sixth phase of the process is the loan’s approval. It is in this phase that the PAE reviews the mortgage restructuring and rental assistance sufficiency plan to ensure that all required HUD forms have been filled out completely and accurately and then delivers the plan to the OMHAR hub for approval.7. Closing: After the OMHAR hub grants approval of the plan, the seventh phase of the process is loan closing. Closing is the phase during which HUD, the PAE, the owner, and affected lenders or servicers fulfill the legal requirements for (1) executing the mortgage restructuring and rental assistance sufficiency plan, (2) restructuring the project’s debt, and (3) renewing the project’s Section 8 contract. 8. Distributing Postclosing Documents: During the eighth phase of the process, the postclosing documents are distributed to the parties (such as servicers, asset managers, and Section 8 contract administrators) who are responsible for loan servicing and asset management functions after restructuring. 9. Servicing and Monitoring: The ninth phase is loan servicing and monitoring. During this phase, servicers, asset managers, escrow agents, Section 8 contract administrators, and monitors have responsibilities to oversee the loans. The servicers of the first and second mortgages are responsible for the cash management function, which includes billing and collecting payments, and accurately accounting for and reporting payments. The asset managers of the first and second mortgages are responsible for ensuring that the underlying collateral is maintained in decent condition. The asset managers of the second mortgage are also responsible for ensuring that only eligible expenses have been deducted from project revenue to maximize excess project income, which can be used to pay off the second mortgage. The escrow agents are responsible for managing the accounts that will be used to finance the immediate rehabilitation needs of the projects, identified through the restructuring process. The Section 8 contract administrators are responsible for ensuring that the projects are maintained in decent, safe, and sanitary conditions by the projects’ owners. Monitors, referred to as restructuring use agreement monitors, are responsible for ensuring that the owners or future purchasers of the projects maintain the affordability and use restrictions agreed to during the restructuring process. mark-to-market program. The final regulations are required to be issued by the later of October 27, 1998, or 3 months after the Director of OMHAR has been appointed (§522(a)(1)(2)). Behind original schedule. The interim regulations were published September 11, 1998, and became effective 30 days later. HUD had originally expected to issue the interim regulations by August 1998 so that the final regulations could be issued in October 1998. According to mark-to-market officials, now the final regulations are not expected to be published until 3 months after the appointment of the OMHAR Director. To allow HUD to work with PAEs, which will actually restructure the mortgages and rental assistance payments of eligible multifamily projects (§513(a)(1)). Behind original schedule. HUD published the Request for Qualifications (RFQ) to solicit PAEs on August 17, 1998. HUD had originally planned to publish the RFQ in June 1998 in order to have the PAEs selected by August 1998. HUD will select the PAEs in two phases and, as of September 8, 1998, expected to complete the selection process by October 29, 1998. To familiarize the PAEs with the restructuring process and their responsibilities, which include determining owners’ eligibility, determining rent levels, restructuring loans, underwriting new or modified loans, managing the closing process, distributing documents after closing, and servicing the loans. Behind original schedule. The briefing sessions for PAEs will begin in October 1998. HUD had originally planned to begin briefing the PAEs in September 1998. To set forth a uniform process for restructuring FHA-insured Section 8 projects. As of September 8, 1998, the operating procedures manual was in draft form, but sections of the draft were expected to be ready for departmental clearance by September 11, 1998. (continued) To establish the obligations and requirements of the PAEs in developing mortgage restructuring and rental assistance sufficiency plans in accordance with the act (§513(a)(1)). Also, to identify the eligible multifamily housing projects for which a PAE is responsible and to clarify the duties that a PAE will typically perform. HUD will enter into portfolio restructuring agreements with the PAEs as soon as possible after they are selected. However, the time period for completing this process will depend on the length of time involved in negotiating the agreements. To instruct the field offices on their responsibilities, which include coordinating the flow of cases to PAEs and Section 8 contract administrators, determining projects with Section 516 violations that require enforcement action, and monitoring the performance of the PAEs. As of September 8, 1998, HUD expected the briefing manuals to be completed by the end of September, so that the briefing sessions could commence in October 1998, after the PAEs are selected. To obtain a favorable tax ruling regarding potential tax consequences for project owners whose mortgages are restructured. In late 1997, HUD initiated discussions with the Treasury Department to obtain a ruling on the proper income tax treatment of the components of the mortgage refinancing contemplated in section 517(a) of the act, which provides for a restructured or new first mortgage that is sustainable at market rents and a second mortgage that is an amount equal to the difference between the restructured first mortgage and the indebtedness under the existing insured mortgage. The act specifies that interest charged on the second mortgage cannot exceed the applicable federal rate (AFR). Since (1) section 7872 of the Internal Revenue Code, in general, defines a below-market loan as any loan on which the interest rate charged is less than the AFR and (continued) (2) section 7872(b) provides that the borrower of a below-market term loan is to be treated as having received cash from the lender in an amount equal to the excess of the amount loaned over the present value of all payments required under the loan, the ruling was required to determine if a project’s owner, whose second mortgage bears interest below the AFR, would be responsible for paying taxes on this amount. In August 1998, IRS ruled on this request, holding that the second mortgage loans made in accordance with the Multifamily Assisted Housing Reform and Affordability Act of 1997 are exempt from section 7872 of the Internal Revenue Code. In the ruling, IRS concluded that the legislative history of section 7872 indicates that most government-subsidized loans, such as government-insured residential mortgage loans, were intended to be exempted from section 7872 and that the factors justifying an exemption of second mortgages under the 1997 act were similar to the factors justifying the exemption of government-subsidized loans. Furthermore, IRS concluded that the interest arrangements of the second mortgage loans were not structured with a principal purpose of avoiding federal tax. To identify risks related to fraud, waste, and abuse of federal resources and to document both the existing program controls and management’s plans for implementing additional controls that mitigate the identified risks. HUD received proposals in response to the Request for Contract Services for the FERA on August 17, 1998. As of September 8, 1998, review of these proposals was complete, and the contract was expected to be awarded by September 11, 1998. The FERA was expected to be finished 60 days after the selection of a contractor. (Table notes on next page) Briefing sessions will be held for staff in the five OMHAR hubs. In addition, HUD scheduled a session for October 6, 1998, to discuss the Section 8 renewal policy with field office staff in its multifamily program centers and multifamily hubs. The Multifamily Assisted Housing Reform and Affordability Act of 1997 (MAHRAA) requires the OMHAR Director to report certain information to the Secretary of HUD and requires the Secretary of HUD to submit information on OMHAR’s operations to the Congress and the Office of Management and Budget. Table III.1 provides a summary of the reports required by the legislation and the status of actions taken by HUD. For these mandated reports, it is HUD’s interpretation of the act that the reporting requirements become effective once the permanent mark-to-market program is operational. The table also includes information on three other required reports: one on equity-sharing partnerships and the two others on demonstration program activity. As the table shows, HUD is in the process of issuing a report on the possible ways that equity-sharing partnerships may be used as options in implementing the mark-to-market program and has submitted reports on the 1996 and 1997 demonstration programs. Table III.1: Summary of HUD’s Reporting Requirements for the Mark-To-Market Program OMHAR’s Director must submit a report to the Secretary of HUD regarding the activities, determinations, and actions of the Director. Semiannually (§573(b)) According to HUD’s interpretation of the act, this report will be initiated once the OMHAR Director is in place. The Secretary of HUD must report figures to the Congress identifying (1) each project for which the participating administrative entity has developed a rental assistance plan that determines that the tenants generally supported tenant-based assistance but under which the assistance was renewed with project-based assistance and (2) each project for which the participating administrative entity has developed a plan under which the assistance is renewed as tenant-based assistance. Semiannually, starting April 27, 1998, for 2 years and annually thereafter (§520(b)) According to HUD’s interpretation of the act, this reporting requirement becomes effective once the permanent mark-to-market program is operational, by the later of October 1998 or appointment of the OMHAR Director. To ensure compliance with the legislation, the Secretary of HUD must conduct reviews and report to the Congress on actions taken under the legislation and on the status of eligible multifamily housing projects. Annually (§520(a)) According to HUD’s interpretation of the act, this reporting requirement becomes effective once the permanent mark-to-market program is operational, by the later of October 1998 or appointment of the OMHAR Director. The Secretary of HUD must submit a copy of the financial operating plans and forecasts for OMHAR to the Office of Management and Budget. These annual plans and forecasts are supposed to be included in the federal budget and in HUD’s congressional justifications for each fiscal year. Annually, before the beginning of each fiscal year (§575(a)) HUD is planning to prepare this report for the beginning of fiscal year 1999 to be included in the federal budget for fiscal year 2000 and HUD’s congressional justifications. (continued) The Secretary of HUD must submit a report on the results of OMHAR’s operations to the Office of Management and Budget. Quarterly and at the end of each fiscal year (§575(b)) HUD is planning to prepare this report for the beginning of fiscal year 1999 to be included in the federal budget for fiscal year 2000 and HUD’s congressional justifications. The Secretary of HUD must submit a report to the Congress on possible ways that equity-sharing partnerships may be used as options in implementing the mark-to-market program if the prohibition is lifted. February 15, 1998 (directed by the conference report of P.L. 105-65) As of September 8, 1998, this report was circulating through the Department for concurrence and signature. In their formal comments on our report draft on October 2, 1998, HUD officials said that the internal clearance of the equity-sharing report had been completed and letters transmitting the report to the Congress were being prepared. The Secretary of HUD must submit reports to the Congress describing and assessing the status of the projects in the demonstration programs. Semiannually for the 1996 demonstration (P.L. 104-134, §210(g)); quarterly for the 1997 and 1998 demonstrations (P.L. 104-204, §212(m)(1)(A)) HUD issued reports for the 1996 and 1997 demonstrations through the fourth quarter of 1997. HUD submitted the last report, covering demonstration program activities through the fourth quarter of fiscal year 1997, on March 28, 1998. The Secretary of HUD must submit a final report on the demonstration programs upon their completion. Not later than 6 months after the end of the demonstration program (P.L. 104-134, §210(g); P.L. 104-204, §212(m)(1)(B)) HUD is planning to prepare these reports for the 1996 and 1997 demonstration programs. As of September 8, 1998, HUD was in the process of updating its data to obtain the information needed for these reports. Explanation or related legislative provision(s) —It must have rents that, on an average per-unit or per-room basis, exceed the rents of comparable properties in the same market area. —It must have a HUD-insured or HUD-heldmortgage. HUD field offices initially screen projects to determine their eligibility on the basis of information provided by the owners, field office staff, third-party appraisers working for the field offices, and HUD’s Assessment and Enforcement Offices. Field offices are to make their screening determinations on the basis of legislative criteria, which are identified in HUD’s mark-to-market draft operating procedures manual. Field offices assign eligible projects to OMHAR, which will then assign the projects to PAEs for further processing. For projects with contract rents below the owners’ estimate of market rents, field offices are to review the owners’ estimate, based on their knowledge of the market area. Assistance from HUD appraisal staff or a third-party review should be obtained if there is any doubt as to the actual market rents. —Directors in HUD’s multifamily restructuring, a project must meet certain legislatively mandated criteria, including the following: hubs and program centers are required to authorize, by signature, the initial eligibility determination for each project. —It must consist of more than four dwelling units. —For projects identified as having below-market rents, field office supervisory review is required to confirm that contract rents are actually below market. —It must receive project-based Section 8 assistance. —For projects identified as potentially ineligible because of poor condition or adverse owner determination, the field offices and/or PAEs must recommend potential remedies to OMHAR; OMHAR makes determinations regarding whether to assign such projects to PAEs for restructuring or to deny the owners the right to restructure. In addition, under the legislation, a project is ineligible for restructuring if —the owner has engaged in material adverse financial or managerial actions or omissions or —Owners have 30 days to dispute decisions of the field offices regarding project eligibility with respect to poor condition or adverse owner determination. At the end of this period, OMHAR may affirm, modify, or reverse decisions. —the property is in poor condition that cannot be remedied in a cost-effective manner. HUD is to establish an administrative review process to appeal any final decision regarding rejection of an owner or project for restructuring. (§512(2), 516(a), 516(b)(2)(C)) (continued) Explanation or related legislative provision(s) The PAE, in consultation with the owner, determines whether to renew Section 8 assistance as project-based or tenant-based. On the basis of its assessment of comparable rents and input from an appraiser and the project owner, the PAE also determines restructured rent levels. If the PAE determines that a project would have negative net operating income at market-level rents, the PAE may issue a “Finding of Special Need” and calculate exception rents for the property. —After the PAE, in consultation renewed as either project-based or tenant-based. —Project-based renewals are mandatory for projects located in tight rental markets, that have a predominant number of units occupied by the elderly or disabled, or that are held by a nonprofit cooperative ownership housing corporation or trust. with the owner and lender, develops the draft mortgage restructuring and rental assistance sufficiency (MRRAS) plan for a project, including the determination of project-based or tenant-based Section 8 assistance and the restructured rent levels, the OMHAR hub must review the draft plan and will either approve or reject it. Restructured rents are to be based on equivalent market rents charged for at least two comparable properties in the same market area. If rents based on two comparable properties cannot be determined, rents can be set equal to 90 percent of the fair market rent (FMR). —According to mark-to-market officials, oversight of the exception rent authority is being worked on and will be addressed in the oversight and audit guide, which was still being developed as of October 14, 1998. —PAEs can approve exception rents up to 120 percent of FMR for no more than 20 percent of all units covered by the portfolio restructuring agreement between the PAE and HUD with contracts that expire in a fiscal year.(§515(c)(1), 515(c)(2)(A), 514(g)(1), 514(g)(2)) (continued) Explanation or related legislative provision(s) The owner or purchaser of a project must evaluate its rehabilitation needs and take actions as necessary to rehabilitate and maintain the project in decent, safe condition. The PAE, with support from a qualified inspector, must review the owner’s evaluation and conduct an independent assessment of the project’s rehabilitation needs. The PAE is ultimately responsible for determining the rehabilitation actions necessary to maintain the project in decent and safe condition, for determining their cost, and for identifying the source(s) of funding. HUD can delegate to state and local governments the responsibility for administering capital grants. —The PAE must submit a draft project must be evaluated. Rehabilitation may be paid from project accounts not required for project operations, increases in budget authority for Section 8 assistance contracts, capital grants, or through the debt-restructuring transaction. mortgage restructuring and rental assistance sufficiency plan, which includes the determination of a project’s rehabilitation needs, to the OMHAR hub. The OMHAR hub must review the draft plan and will either approve or reject it. —HUD may make grants for the capital costs of rehabilitation to owners of projects if the owners demonstrate that capital grant assistance is needed for rehabilitation of the projects and that project income is not sufficient to support such rehabilitation. —According to mark-to-market officials, details on capital grants will be generally discussed in the operating procedures manual. Instructions for the oversight of the grant funds will be addressed in the oversight and audit guide, which was still being developed as of October 14, 1998. Rehabilitation will be only for the purpose of restoring the project to a nonluxury standard adequate for the rental market intended at the original approval of the project-based assistance. Each owner or purchaser of a project to be rehabilitated under mark-to-market must contribute, from nonproject resources, at least 25 percent of the amount of rehabilitation assistance received. (§514(e)(3), 517(b)(7), 531) (continued) Explanation or related legislative provision(s) The PAE, in consultation with the owner and lender, determines the size of the restructured first mortgage and second mortgage. —a second mortgage no greater than the difference between the restructured or new first mortgage and the indebtedness under the existing mortgage, in an amount that can reasonably be expected to be repaid. The PAE submits the project’s mortgage restructuring and rental assistance sufficiency plan, which includes the PAE’s conclusions regarding the new first mortgage and second mortgage, to the OMHAR hub. Depending upon the details of the MRRAS plan, the OMHAR hub will perform either an administrative review or a technical review, after which it can accept or reject the plan. The size of a project’s restructured first mortgage and second mortgage must be determined. (§517(a)) HUD will recapture budget authority not required for contracts amended or terminated as part of restructuring and use it to provide housing assistance for the same number of families that were covered by that contract for its remaining term; any budget authority saved by shifting to project-based or tenant-based assistance will be rescinded. (§523(c)) Within HUD’s Office of Multifamily Housing, the Office of Program Management and Oversight will be responsible for administering recaptured Section 8 budget authority. officials, the procedures for monitoring this function will be defined in the oversight and audit guide, which was still being developed as of October 14, 1998. At least annually, PAEs qualified as Section 8 contract administrators must review the status of all restructured projects, including on-site inspections to determine compliance with housing codes and other requirements of the legislation and restructuring agreements. (§519(b)) PAEs that are qualified to be Section 8 contract administrators are responsible for annual project reviews. If a PAE is not qualified to be a Section 8 contract administrator, either HUD or a qualified state or local housing agency will be responsible for this required review. officials, these procedures will be addressed in the oversight and audit guide, which was still being developed as of October 14, 1998. In general, HUD currently intends to require the PAEs to submit monthly reports and to have HUD field staff conduct semiannual monitoring reviews. (continued) Explanation or related legislative provision(s) HUD can provide up to $10 million annually in funding to tenant groups, nonprofit organizations, and public entities for capacity-building and technical assistance in furthering the purposes of the mark-to-market program. (§514(f)(3)) HUD will select grantees on a competitive basis either to provide direct technical assistance or to act as intermediaries by administering a program of technical assistance grants to subrecipients. must submit quarterly performance reports to OMHAR. HUD will ensure that this reporting requirement is met by including provisions in the grant agreement and addressing this issue in the oversight and audit guide, which was still being developed as of October 14, 1998. John T. McGrail The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a legislative requirement, GAO reviewed the status of the Office of Multifamily Housing Assistance Restructuring's (OMHAR) development, focusing on: (1) its organization and staffing and how it relates to the Department of Housing and Urban Development's (HUD) overall structure; (2) whether it is on schedule to meet its key operational and reporting requirements; (3) the procedures and systems it will use to oversee the mark-to-market program's implementation; (4) the status of projects included in the three mark-to-market demonstration programs and how HUD is using information gathered from these programs; and (5) the actions it has taken to obtain information and feedback from parties that will be affected by the mark-to-market program. GAO noted that: (1) because a Director for OMHAR was not appointed until October 21, 1998, HUD had not made final decisions at the time of GAO's review on the Office's staffing and organization or on how the Office would relate to HUD's overall structure; (2) in addition to the 10 staff currently assigned to work on the mark-to-market program, HUD's preliminary plans call for hiring approximately 75 staff for OMHAR; (3) the Departments of Veterans Affairs and Housing and Urban Development, and Independent Agencies Appropriations Act of 1998, provides the Director special compensation authority to pay employees of OMHAR at a higher level than other employees in order to obtain the skills and expertise needed to accomplish the program's purposes; (4) HUD has made considerable progress toward meeting its key operational requirements for implementing the mark-to-market program; (5) because of delays in obtaining contract support and the sheer volume of tasks that HUD needs to complete, some of these tasks either have been or will be completed behind their original schedule; (6) the HUD officials responsible for the mark-to-market program are in the process of establishing or planning several procedures and systems designed to oversee the implementation of the program; (7) these include: (a) a system for measuring the performance of the entities responsible for carrying out restructuring transactions on HUD's behalf; (b) an Internet-based system to track the actions taken by these entities in carrying out mark-to-market functions; and (c) an audit guide to test these entities' compliance with the program's requirements and objectives; (8) as of October 14, 1998, most of these procedures and systems were still being developed; (9) in accordance with legislative requirements, HUD mark-to-market staff have taken actions to obtain information and feedback from affected parties; (10) specifically, they: (a) have developed program guidance that includes steps to involve affected parties including tenants and neighborhood associations, at various points in the restructuring process; and (b) have held three public forums to obtain recommendations from organizations representing affected parties on implementing certain legislative provisions; and (11) in addition to the legislatively required actions, HUD invited organizations representing project owners and managers, tenant groups, nonprofit organizations, lenders, coalitions of local and state government agencies, and other advocacy groups to submit ideas and comments on developing the program and to participate in meetings to discuss these issues. |
Labor and VA have had a long-standing relationship formalized by agreements acknowledging their mutual concern and responsibility for helping veterans with service-connected disabilities transition to the civilian workforce. Their past efforts to coordinate and collaborate have faced difficulties, however. In 2005, at the recommendation of a VA task force and encouragement of congressional staff, the two agencies forged a new memorandum of agreement. In this agreement, the agencies agreed to share information, including wage data, and establish and maintain management information systems to enable accurate yearly reporting. The agreement also called for three joint work groups to develop performance measures to assess partnership program results, design a training curriculum, and establish joint data collection, analysis, and reports. In addition, the agreement outlined means to promote cooperation and eliminate duplication of efforts between the agencies. These include developing an approach to serving veterans that involves both VA and state workforce agency staff from the early stages, advising all VA clients about the employment services offered by the state workforce agency, and establishing an effective process for referring VA clients who are seeking employment to the state workforce agency. While many of the specific elements of the agreement need to be implemented at a national level within Labor and VA, other activities, by their nature, would take place on the state or local level. The agreement provides for separate memorandums of agreement, containing common goals and measures, to be executed among Labor, VA, and the states. Table 1 summarizes the key elements of the national agreement, the specific actions to which Labor and VA agreed, and the level at which implementation generally would occur. Around the same time the national agreement was signed, VA rolled out the Five-Track employment program aimed at standardizing rehabilitation services for veterans with service-connected disabilities and providing a greater focus on employment options early in the rehabilitation process. The five employment tracks from which veterans can select are reemployment with their previous employer, rapid access to employment through job readiness preparation and training opportunities, self- employment, employment through long-term services that include education programs and formal training leading to employment, and independent living services for those who are currently unable to work due to their disabilities. These tracks were available prior to the rollout, but the program changed the way they were presented to veterans. The Five-Track program was piloted in four regional offices starting in October 2004, and a national rollout began in November 2005. Figure 1 describes the Five-Track model. As part of the Five-Track program, VA established the employment coordinator position and drew up plans to install job resource labs. The position description for the employment coordinator contains duties aimed at providing veterans with assistance preparing for and finding employment. These duties include helping veterans make informed choices about their employment track, assessing veterans’ readiness to seek employment, and assisting veterans with job networking, development, and placement. The position description also includes responsibilities for marketing the VA program to employers and developing partnerships with other agencies to assist veterans with employment services. Also as part of the Five-Track program, VA allocated funds for each location to install a job resource lab with computers, Internet access, and other materials for veterans to use in their career exploration and job search. VA made plans to have employment coordinators in most of its 57 regions, and a job resource lab in each regional office and in every satellite location. Although the missions of Labor’s VETS and VA’s VRE programs to provide employment assistance to disabled veterans are similar, the organizational structure of each is somewhat different. VA administers its VRE programs through regional offices—roughly one in each state, with multiple offices in larger states—and its staff are federal employees. The regional offices have some autonomy in deciding the operations of the office, including their working relationships with the Labor programs and the state workforce agency. Labor administers its programs through the Disabled Veterans’ Outreach Program (DVOP) and the Local Veterans’ Employment Representative (LVER) program, which are funded by a grant from Labor to the state workforce agencies. DVOP specialists and LVER staff are employees of the state and are typically housed in state employment service offices. Although there is evidence LVER staff and DVOP specialists often perform similar duties, the DVOP specialist’s role was designed to provide intensive services to veterans with employment barriers, including those with disabilities, while the LVER staff role was designed to market veterans to employers. Labor provides oversight and monitoring of the state grants through a state VETS program director (DVET) in each state. The Labor and VA programs serve similar clientele, but there are differences in eligibility requirements and the measure of successful completion. To participate in VA’s VRE program, the veteran must have disabilities that affect a minimum percentage of daily activities, as determined by VA, and must have an employment handicap related to the disability. In addition, there is a time limit for receiving services, generally 12 years following the date the veteran was discharged or separated from the military or received a disability percentage rating from VA. The primary outcome measured by VA’s VRE program is whether the veteran obtained and maintained suitable employment, that is, work that is within a veteran’s emotional and physical capabilities and consistent with the veteran’s abilities, aptitudes, and interests. By contrast, any veteran who was discharged because of a service-connected disability or who served more than 180 days and was not dishonorably discharged can receive services through Labor’s programs. The primary outcome measured by Labor is whether or not the veteran entered employment, without requiring determination of suitability. VA and the state workforce agencies—including DVOP specialists and LVER staff—work together on the local level. In many states, DVOP specialists work with veterans who have completed their VA training to help them find employment. The process by which these referrals are made is not determined nationally; it is left up to VA regional officials to develop local policies and procedures. In addition, some state workforce agencies provide VA staff with state unemployment insurance wage data to verify the employment status of their clients. Additionally, many VA regional offices have a DVOP specialist colocated in the office at least part of the time. However, cooperation between Labor and VA has historically been better in some states than in others, reportedly in part because of variation in the number and quality of services provided by DVOP specialists. As a result, VA’s VRE Task Force concluded in its 2004 report that VRE should consider using DVOP specialist services as one of many options to help its clients, and not view the DVOP specialist as the sole provider of employment services. Labor and VA have fulfilled some elements of their agreement to coordinate efforts, but the agencies face a variety of challenges to fully executing the agreement, including the lack of a comprehensive plan for implementing the agreement and measuring progress. The national offices of Labor and VA have implemented one of the elements of the agreement for which they are responsible, have begun to address four elements, and have not taken action on the others. All of the states we visited had taken action to implement some of the elements of the national agreement, but faced challenges implementing others due to state privacy laws, staffing limitations, and other obstacles. In addition, Labor and VA lack a long- term plan for implementing the agreement that includes timelines and benchmarks against which progress can be judged, as well as procedures for dealing with states that are not carrying out activities in which they have a role. Further, the agencies have provided states with limited guidance on implementation. As of July 2007, the extent of nationwide implementation was uncertain because the agencies had not thoroughly reviewed the implementation of the agreement at the state level. Labor and VA have fully implemented the element of their agreement that calls for the establishment of joint work groups composed of staff from both agencies. The agencies have formed three joint work groups to address issues related to shared performance measures, staff training, and joint data collection, with the goal of improving the quality of employment services and suitable job placements for veterans with disabilities. According to a Labor official, the implementation of the agencies’ agreement revolves around the efforts of these work groups. The work groups met for the first time in April 2006, 6 months after the agreement was signed. In January 2007, they conducted a survey of Labor and VA field staff in 24 states to gather preliminary information about the working relationship between the agencies at the local level and identify areas needing improvement. On the basis of the survey results, the work groups plan to launch a pilot project in eight regional offices in August or September of 2007 to explore new strategies to improve the partnership between Labor and VA. Labor and VA officials expect to complete the pilot project in January 2008, and plan to make recommendations thereafter regarding the implementation of these strategies nationwide, but the work groups have not yet made definitive recommendations for program changes. Four of the elements that Labor and VA are responsible for implementing—staff evaluations, yearly reporting, performance measures, and monitoring of state-level memorandums of agreement—are in process but not yet completed. The agreement between Labor and VA calls for agency managers to consider the effectiveness of partnership activities between VA and the state workforce agency when evaluating staff. Although Labor has implemented such standards for state VETS program directors, these provisions are not specifically outlined in the performance standards for other VETS program staff or for VA staff. The agreement also calls for the agencies to have yearly reports that include measures of the number of VA program participants referred to the VETS program who registered for state workforce agency services, the number of participants who entered suitable employment, and the number of participants who maintained suitable employment for 180 days or more. While the agencies are collecting data on most of these measures, they are not currently collecting data on the last measure. With regard to performance measures, one of the joint work groups has started to develop new shared measures to better assess partnership activities between the two agencies. For example, one proposed measure is the percentage of veterans referred from VA to VETS program DVOP specialists who receive an employment services assessment within 30 days. In addition, the agreement calls for the agencies to monitor common goals and measures within memorandums of agreement executed at the state level and stipulates that corrective action will be taken when such goals are not met. According to information provided by VA, the agency obtained copies of state-level agreements from all but one of its 57 regional offices in June 2007, but officials have not yet reviewed the agreements for common goals and measures. Labor and VA have not yet acted on the other elements of their agreement related to information sharing and management information systems. The agreement calls for the agencies to share information about veterans, including all information from interviews, counseling, testing, and assessment. In addition, the agreement stipulates that the agencies will develop and maintain management information systems that enable accurate yearly reporting. Although Labor and VA officials told us they are considering the possibility of establishing national data-sharing practices and a joint management information system through which information could be shared, no action has been taken on the national level. The 5 states we visited had all implemented a number of the elements of the agreement between Labor and VA, including advising veterans of the benefits of registering for state workforce agency services, establishing a referral process, establishing state-level memorandums of agreement, and appointing VETS points of contact for VA staff. All 5 states had processes in place whereby VA advised program participants of the benefits of registering for employment services with the state workforce agency. All of the states we visited had also established a process for referring veterans from VA to the state workforce agency for employment services, although the process was not the same in every state. In addition, all 5 states we visited had a memorandum of agreement between the regional VA office serving the state, the state VETS program director, and the state workforce agency to coordinate their services for disabled veterans. All of the states had recently revised their state agreement or were in the process of doing so. All 5 states we visited had also appointed a VETS program contact person for VA staff, and a national Labor official told us that all 50 states have appointed such points of contact. Figure 2 summarizes the implementation of the elements of the agreement in the 5 states we visited. All five states included key VA and state workforce agency staff in the rehabilitation process, but not all states fully coordinated the Labor and VA programs by involving DVOP specialists early in the process as outlined in the agreement between Labor and VA. Typically, VA clients seeking employment would meet with a VA counselor, VA employment services staff, and a DVOP specialist from the state workforce agency. However, the states varied in terms of the point at which the DVOP specialist became involved. In three of the states we visited, DVOP specialists provided assistance to veterans soon after they entered the VA program, typically by presenting labor market information to veterans to incorporate into their vocational rehabilitation plans. In the two other states we visited, DVOP specialists did not typically meet with veterans until they had neared the end of their VA training programs or were otherwise prepared to seek employment. In one of these states, we were told that this was due to negative past experiences in providing veterans services from both VA and DVOP specialist staff at the same time. The five states we visited were collecting data on all but one of the yearly reporting measures outlined in the agreement between Labor and VA, and staff identified challenges to implementing this remaining reporting measure. In all five states, agency staff told us that they were gathering data on the number of VA program participants referred to the VETS program, and of those referred, the number who registered for state workforce agency services and the number who entered suitable employment. In these states, VA staff said that they tracked program participants for 60 days after they obtained suitable employment, and we learned that state workforce agencies tracked veterans for varying lengths of time. However, none of the states were tracking the number of VA program participants who maintained suitable employment for 180 days or more. In four states, some agency staff told us that they would face challenges implementing this additional reporting measure, and staff in three states said that this was because following up with veterans for 180 days would likely require additional staff time and resources. In four states, workforce agencies were able to share UI wage data with VA, but in one state, they faced challenges sharing such data because of state laws. UI wage data are managed by the states. According to VA officials, if agency staff want to use this information—for example, to verify that a client is employed—they need to obtain the data from the state workforce agency. In Illinois, VA and the state workforce agency had signed a shared data agreement that allowed the workforce agency to provide VA with UI wage data. In Alabama, state workforce agency staff told us that they had given VA staff direct access to their database, which contained UI wage information. In California’s San Diego region, VA and the state workforce agency had developed a centralized process in which VA sent UI wage data requests to the state workforce agency headquarters. And in South Dakota, workforce agency staff told us that they were able to share UI wage data with VA, but said that they rarely received requests for such data. However, in Michigan, we were told that they were unable to share UI wage information because state law prevents the workforce agency from sharing an individual’s wage data with VA. Staffing limitations prevented some states from having a full-time colocated DVOP specialist at the VA office, although most had at least part-time coverage from an itinerant DVOP specialist. The agreement states that to the extent feasible and appropriate, a DVOP specialist or other designated individual will be colocated at the VA office or will otherwise provide itinerant coverage to VA participants. Only one of the five states we visited had a full-time colocated DVOP specialist at the VA office. Three of the other states we visited had part-time colocated DVOP specialists, but one state had no coverage at the VA office at the time of our visit. VA staff in the states with a DVOP specialist on site part-time told us that such an arrangement was useful, and staff in two states said that they would like to increase the amount of colocation time per week. However, officials in three states told us that funding constraints limited the number of workforce agency staff and thus the workforce agency’s ability to share a DVOP specialist with the VA office. National VA officials also told us that fluctuating state budgets made it challenging for states to commit to providing a colocated DVOP specialist from year to year. None of the five states had fully implemented the information-sharing element of the national agreement between Labor and VA, and staff in all five states told us that some of this information sharing was unnecessary. The agreement states that both agencies will share all information about veterans gathered from interviews, counseling, testing, and assessment. VA staff in the five states we visited told us that they did not regularly share all of the information outlined in the agreement with the state workforce agency, and staff in all states said that it is not necessary to share all information about veterans in order to help them find employment. In all five states, VA staff shared a standardized, but limited, set of information with the state workforce agency when referring a veteran for employment services. For example, in one state, VA staff told us that this standardized set of information included the veteran’s contact information, employment goal, and level of education, as well as general information related to the veteran’s barriers to employment. However, VA staff in all five states told us that they did not regularly share specific information about veterans’ disabilities with state workforce agency staff, and staff in two states said that this was a result of privacy concerns. None of the five states we visited were using partnership activities specifically to evaluate staff, as outlined in the agreement between Labor and VA, and some officials said it would be impractical. In the three states that had a centralized referral process and thus had minimal contact between state workforce agency and VA staff, some local agency staff stated that it did not make sense to evaluate staff on the effectiveness of their partnership activities. In all five states, we were told that staff evaluations contained a general category related to teamwork or cooperation, but did not include specific performance measures to evaluate DVOP specialists and VA staff on the effectiveness of their partnership activities. In one state, workforce agency managers told us that they were considering implementing specific performance measures related to the effectiveness of partnership activities. Labor and VA lack a comprehensive plan for implementing their agreement, and have not set benchmarks or long-term time frames for implementation. While Labor and VA have designed a pilot project, agency officials have not developed a plan to guide the full implementation of the agreement. Although the joint work groups have set short-term time frames for the pilot project, and plan to make recommendations regarding pilot project expansion in May 2008, the pilot project plan does not include long-term time frames for the implementation of the agreement, and Labor and VA officials told us that they considered implementation to be a work in progress. In addition, the pilot project plan does not outline steps for taking action when states do not implement the agreement. It is unclear whether national VA and Labor officials have provided sufficient direction to states on the implementation of the agreement. In one state, workforce agency staff told us that guidance from national officials on implementing the agreement was lacking. Both agencies sent letters to regional and state officials notifying them of the agreement, but have done very little to assist with implementation. A national VA official told us he has discussed implementation in conference calls with VA program managers and employment services staff, but a national Labor official told us his agency has not done any follow-up. In addition, the agreement calls for the agencies to take corrective action when common goals outlined in state-level memorandums of agreement are not met. However, both VA and Labor officials told us that they have not yet defined steps for corrective action. Labor also told us that it does not have much leverage to induce state workforce agencies to implement the agreement given its limited authority over such agencies. Further, as of July 2007, Labor and VA had not yet fully assessed state actions to implement the agreement, so the extent to which the national agreement has been implemented across all states was unclear. The agencies collected some data on states’ activities in the January 2007 survey of agency field staff conducted by the joint work groups. VA officials told us that the survey identified several areas for improved collaboration between the agencies, including clarifying definitions of terms, increasing staff training, and facilitating communication among staff. However, the results cannot be used to understand the extent of nationwide implementation because the survey only included about half of the states and did not cover all aspects of the national agreement. In June 2007, VA conducted a second survey asking all of its 57 regional offices to submit copies of their state-level memorandums of agreements to national VA officials, and received these agreements from all but one regional office. A VA official told us that the agency plans to review these agreements to assess their alignment with the national agreement and offer assistance to states that are having difficulties, but it had not done so as of July 2007. Although VA has almost fully implemented the Five-Track employment program, the agency may face challenges serving the employment needs of recently returning veterans because of the types and severity of their disabilities. VA officials told us that they have fully implemented four of the seven components of the Five-Track employment program and the other three components are mostly complete. VA officials have noted that many recently returning veterans have serious injuries and they anticipate higher caseloads as increasing numbers of veterans apply for program benefits. In response to these challenges, VA has taken some initial steps to address the specific employment and training needs of these veterans, and Labor has made similar efforts. According to VA, officials have completed most of the seven components needed to implement the Five-Track program. While four components have been completed, three other components remain in process. Figure 3 shows the seven components of the Five-Track program and their respective completion status. VA officials told us that three of the initial components of the new Five- Track employment program—the national distribution of orientation materials, the pilot study, and introductory staff training—have been fully implemented. According to VA, officials have also completed the distribution of Five-Track orientation materials nationally, including a new video and a booklet that describes program eligibility requirements. VA officials have also distributed information nationally on each of the five tracks to employment—reemployment, rapid access to employment, self- employment, employment through long-term services, and independent living services. In October 2004, VA officials began a pilot study in four regional offices to test several proposed components of the Five-Track program. At the pilot study’s completion in September 2005, VA issued a status report based on the study results and made some changes to the program before beginning the full launch a few months later. VA also completed introductory training for VRE staff members nationwide on the Five-Track program, between February and April of 2006. The VA staff who received this national training were expected to return to their local offices and train other staff on the program. According to VA, the training included simulated case management exercises and a review of the logistics and application of the new job resource labs. VA has also hired and trained employment coordinators. This position was developed to revamp the existing employment specialist position and provide more direct assistance to veterans with job readiness and job- seeking skills, as well as helping them with job placement. Training for employment coordinators began in November 2005, and a few months later, 52 employment specialists were reclassified as employment coordinators. Since then, VA national officials have hired additional employment coordinators, and as of April 2007 there were 74 employment coordinators nationally. Qualifications for the employment coordinator position include knowledge of marketing concepts, thorough knowledge of the VRE program and its objectives, and diverse presentation skills. Of the 6 employment coordinators we spoke with, 3 had prior experience providing employment assistance to veterans, either as a DVOP specialist or an employment specialist. Information provided by VA shows that the agency has mostly completed the implementation of 160 job resource labs. The job resource labs were intended to be a supportive tool to help VA staff provide veterans with employment resources and job readiness assistance. As of April 2007, 144 labs (90 percent) labs were operational, although some were utilizing borrowed computers. Sixteen of the labs (10 percent) were further postponed due to construction issues, lack of appropriate space, difficulty securing Internet access, or complications obtaining computers. Before providing funding, VA conducted a survey of its field offices to assess existing resources. Then, between June and September 2006, VA provided field offices funding to purchase noncomputer items such as furniture and books. VA transferred $1.5 million to the VA Office of Information and Technology for procurement of 359 computers, but there were delays in obtaining them. VA officials told us the computers should be available in summer 2007. According to VA, officials have begun to implement a new employment resource Web site and a program manual, but they are not yet completed. VA launched the employment resource Web site, known as vetsuccess.gov, in 2005 to be a new supportive tool for veterans and VA staff. Currently, veterans may access the Web site via the Internet and can utilize the site to view program videos, employment resources, employment search links, and information about VA partner organizations. VA is planning to install additional interactive features on the Web site, including a log-in option that would allow veterans to post résumés and employers to search and view them. However, a VA official told us the installation has been delayed indefinitely until security concerns can be resolved. Additionally, as part of the Five-Track program implementation, VA officials plan to distribute a program manual to staff. However, a VA official told us the agency is postponing the release of the manual until the employment Web site features have been completed and can be included as part of the manual. Officials in some states we visited raised concerns about the ability of employment programs—including the Five-Track program—to address the needs of severely disabled program participants returning from recent conflicts in Afghanistan and Iraq. According to VA officials, many recently returning veterans have multiple and severe disabilities, such as speech, hearing, and visual impairments as well as loss of limbs and brain injuries, and behavioral issues due to the stress of combat. Additionally, veterans from recent conflicts are surviving with more of these serious injuries that would have been fatal in past conflicts, a fact that can present major challenges to providing training and securing appropriate job placements. For example, VA officials in one state told us that from their experiences assisting veterans with traumatic brain injuries, they have found that these veterans may find it difficult to filter their thoughts and actions and may act and speak inappropriately, making employment placement more difficult. In addition, VA officials anticipate increased caseloads because of expanded outreach efforts to veterans and service members separating from the military, and increasing disability claims from veterans of recent conflicts. To address these challenges, VA told us that officials have developed programs to provide additional resources for recently returning veterans to receive employment services. VA officials have an early outreach program, Coming Home to Work, to provide civilian work experience to eligible service members pending medical separation from active duty at military treatment facilities. VA has also designated a particular individual in each VA regional office to coordinate vocational rehabilitation and employment case services for recently returning service members located in military treatment facilities. According to VA, officials have also made training available to staff on the specific disabilities that may be more prevalent among recently returning veterans. The agency has developed training materials on traumatic brain injuries, amputations, and transferable work skills, which an official told us have been made available to VA counselors and employment coordinators through VA internal satellite broadcasts and other means. Additionally, VA plans to develop an online training curriculum that would include training specific to the needs of recently returning veterans that will be available to VA staff. In addition to training, a VA official told us the agency used internal staff meetings and conference calls to address the immediate concerns of VA staff related to recently returning veterans. Labor has also taken some initial steps to address the challenges of serving severely disabled Five-Track program participants returning from recent conflicts, and we were told staff who work with these veterans have access to VA training. Labor has a program, called Recovery and Employment Assistance Lifelines (REALifelines), to provide individualized job preparation, counseling, and reemployment services to veterans seriously injured in recent conflicts. VETS program staff, including DVOP specialists and LVER staff, can also access the National Veterans’ Training Institute VETS NET online newsletter, which often includes links to information on serving recently returning veterans. In addition, VA told us that state VETS program directors have access to training materials related to the needs of recently returning veterans such as traumatic brain injuries and amputations. Similarly, VA told us that state workforce agency staff will also have access to its planned online training curriculum, including courses specific to the employment needs of recently returning veterans. A list of selected VA and Labor initiatives related to addressing the employment needs of recently returning veterans is in table 2. Employment coordinators and job resource labs in the five states we visited provided employment services to veterans, but some of these services were available elsewhere. Employment coordinators generally provided direct employment services to veterans and performed job development activities and outreach to employers and the community. Despite the fact that the employment coordinator position description outlines a variety of services to be provided to veterans across each VA region, these activities were largely limited to their local areas. In addition, employment coordinators performed some services that duplicated other available resources. Although job resource labs afforded some additional opportunities for VA staff to assist veterans with employment activities, they generally were not used by many veterans. Job resource labs also provided some resources that veterans were able to access elsewhere. The employment coordinators in the five states we visited provided some direct assistance to veterans. The services to veterans included helping with veteran employment plans, résumé preparation, interview preparation, and job search activities. In addition, employment coordinators with whom we spoke typically provided veterans and VA counselors with labor market information, which generally included average salary, to assist veterans in selecting potential career fields. Other duties varied by state and included managing the referral process to DVOP specialists, conducting follow-up with veterans after they have been placed in employment, arranging for the distribution of a stipend given to veterans during their employment search, and focusing on providing services to veterans who were having difficulty finding employment. One employment coordinator reported that she assisted employers with making workplace accommodations for disabled veteran employees who were participants in the VA program, while two others reported that they had not had the opportunity to provide such accommodations. In one case, the employment coordinator had carved out the specific role of assisting veterans with federal job applications. Most of the employment coordinators we met with also performed some job development and outreach activities. These activities included promotion of VA programs, developing partnerships with businesses and other agencies, and marketing and providing VA employer incentives. For example, an employment coordinator in one state we visited told us she partners with other agencies to obtain employment leads for veterans and another employment coordinator told us that he markets an incentive program to employers through which VA pays half of the veteran’s salary for the first 6 months of employment. Some employment coordinators also told us they conduct presentations about VA at community events. Employment coordinators in the five states we visited primarily provided services to veterans in their local areas, even though VA originally intended that they provide a variety of employment services across each VA region. In the five states we visited, the employment coordinators seldom traveled to offices outside their local area, and when some did, they met with a much smaller number of veterans. VA staff in several locations told us the lack of face-to-face contact with an employment coordinator disadvantaged some veterans. For example, VA staff in one office without an employment coordinator on-site found communication via phone and e-mail to be less effective in providing employment services to veterans than in-person interactions. Similarly, an employment coordinator in another state told us that veterans outside her local area do not receive the same level of services from her as those located in her immediate area. VA acknowledged it was not aware of the extent to which employment coordinators were able to work with veterans outside their local areas. In addition, VA national officials acknowledged that the current number of employment coordinators is not enough to provide a full array of employment services to all program participants, as they had originally intended when they wrote the position description. They told us they tried to compensate for this deficiency by assigning employment coordinators to offices in the most populous areas and expected that state workforce agency staff would serve veterans in other locales. In one state, the employment coordinator was stationed over 200 miles away from the office we visited and was only able to provide services for 15 of the approximately 300 veterans served in that office at the time of our interview. The employment coordinator, VA counselors, and some veterans in this state told us they considered local DVOP specialists to be the primary employment services providers for program participants. According to veterans and VA staff in the five states we visited, employment coordinators provided some unique services to veterans in their local areas, but other services were also available from alternative sources. Employment coordinators provided some services that were not available elsewhere, such as marketing VA employer incentives, promoting the VA program through networking and ongoing contacts with employers, distributing veterans’ employment search stipend, and assisting with employer accommodations. However, according to veterans and VA staff in the five states we visited, some of the direct services provided by employment coordinators were similar to those offered by others. For example, according to some VA staff and veterans, college career centers, DVOP specialists, and VA contractors all provide veterans assistance with résumé writing and interviewing techniques. Additionally, some veterans and VA staff told us that both employment coordinators and state workforce agency staff can assist veterans and VA counselors with labor market information. Similarly, in two states we visited employment coordinators were available to provide more services to veterans who were having difficulty finding employment, while DVOP specialists were also responsible for facilitating services for veterans with special employment needs. One veteran told us he met with the employment coordinator daily for assistance with résumé and interview preparation and obtaining employment leads while also working with a DVOP specialist weekly. The outreach activities employment coordinators conducted were also sometimes performed by others. In some states we visited, both employment coordinators and DVOP specialists marketed to employers on behalf of disabled veterans in the VA program by identifying employer hiring contacts and obtaining information on job vacancies and hiring prerequisites. Similarly, in some states employment coordinators and DVOP specialists performed job development activities, such as fostering partnerships through participation in job fairs and networking with local businesses. VA staff in two states told us that state workforce agency staff often performed job development and outreach activities. Although Labor and VA national officials acknowledged some similarities between the employment services performed by employment coordinators and those available to program participants elsewhere, VA officials have not collected information on the full extent of the duplication nationally. A summary comparing the services provided by employment coordinators and employment services available elsewhere in the states we visited is in table 3. Job resource labs provided equipment and materials that VA staff used to assist veterans with employment-related activities, but the labs were not typically staffed throughout the day. All of the labs in the five states we visited were equipped with computers that had Internet access, desks, printers, and employment resource libraries. The number of computers available for veterans’ use in the labs ranged from one to four. All of the labs also had at least three additional resources such as copiers, fax machines, informational pamphlets, job postings, and televisions and DVD players located directly in the lab or available for use elsewhere in the office. According to the veterans and VA staff in the five states we visited, the job resource labs were used primarily to conduct Five-Track program orientation and to allow for veterans to engage individually in career exploration activities. In three of the states we visited, the labs were also used as a meeting space, for example, to host workshops and veteran employment networking groups. The labs sometimes also served as an office for the part-time colocated DVOP specialist or as an extra space to conduct activities such as mock interviews. We were told VA staff were available to assist veterans as needed, but only one of the labs we visited was staffed throughout the day. Although some veterans were able to use the labs with minimal guidance, others told us they needed the assistance provided by VA staff when they were first introduced to the labs. Figure 4 shows photographs of a sample of the job resource labs we visited. Veterans in the five states we visited did not typically use the labs for any activities other than those that were part of the program orientation. VA staff in the five states we visited told us the labs were not used by many veterans. Four of the labs we visited tracked usage informally, and a month of data from these labs showed as many as 34 veterans using one lab and as few as 3 using another. Additionally, none of the veterans we spoke with used the labs regularly or more than a few times during the program. For example, 1 veteran said he only used the lab to occasionally look at his grades online and another veteran used printed resources in the lab once to prepare for an interview. Some of the low levels of usage may be explained by factors associated with the lab. In two of the states we visited, the labs had only been operational for 2 months or less. In three other sites, VA staff lamented that the labs were housed in buildings that were not conveniently located or required security searches to enter. While VA officials told us they plan to assess the usage and effectiveness of the job resource labs nationally, they have not done so yet because they are waiting for all labs to have permanent computers. Veterans and VA officials told us that program participants have access to equipment and materials similar to those in the job resource labs at other locations. In the five states we visited, job resource labs provided veterans with some additional VA services, for example, one-on-one computer assistance from VA staff and the opportunity to network with other VA participants. However, according to veterans and VA counselors with whom we spoke, veterans typically had computer and Internet access available at home, school, the public library, or the local one-stop career center. Similarly, one veteran told us he used printed materials found in the job resource labs, such as résumé and interviewing technique books, at his local one-stop career center. While some veterans and veteran service organization representatives we spoke with found that having access to resources at multiple locations allows veterans more flexibility, some also told us the labs were duplicative of most of the basic resources available at local one-stops, such as computers, printers, faxes, printed materials, job postings, and brochures. VA staff and their counterparts in Labor are charged with the responsibility of helping veterans achieve the best possible employment outcomes—each in their own capacity but also in cooperation with each other. Given their similar missions and clientele, and to ensure the seamless delivery of services and efficient use of resources, it is critical that these agencies work together effectively. The October 2005 agreement between Labor and VA stated that both parties commit themselves to active cooperation and coordination in meeting the goals of the agreement. However, without additional efforts—developing a comprehensive plan that outlines long-range time frames and benchmarks for its implementation, thoroughly reviewing implementation at the state level, providing guidance to states, and outlining plans for taking action if states do not implement the agreement—it is difficult to determine the extent to which the agreement has been implemented or what, if any, progress has been made toward its goals. In addition, VA officials were not fully aware of how the employment coordinator position was serving veterans outside of the employment coordinator’s local area. As a result, VA may not be providing all veterans in a regional office’s jurisdiction with equal access to the full array of services from an employment coordinator. Furthermore, Labor and VA have not collected systematic information on employment coordinators and job resource labs to help eliminate all duplication in the delivery of needed employment services and provide a seamless employment transition. Without this information, there may be unnecessary duplication of services, which is not in keeping with the agreement. Finally, VA has not examined the usage or effectiveness of the job resource labs nationwide, and as a result, VA may end up spending money to sustain a resource that few veterans are using. To ensure the complete and timely implementation of the agreement, we recommend that the Secretary of Labor and the Secretary of Veterans Affairs direct VETS and VRE to take the following actions: develop a written plan for the full implementation of the agreement that includes long-term time frames, benchmarks by which to track implementation at the state level, and plans for taking action in instances when states are not fully implementing the agreement; provide additional direction to the states on implementing the agreement, including examples of promising practices from states, such as strategies for sharing information; provide technical assistance to states that are facing difficulties implementing the agreement; and collect and assess complete information on the progress of the states in implementing the agreement using well-designed and appropriate methodology, such as a systematic review of state-level memorandums of agreements or a comprehensive survey of all locations. To ensure the employment coordinator role is being used in the most effective and efficient way possible without duplication of other available services, we recommend that the Secretary of Veterans Affairs, in consultation with the Secretary of Labor, direct VRE to take the following actions: determine how best to use the employment coordinator in serving veterans located outside the employment coordinator’s local area; undertake additional efforts to review how the employment coordinator role has been carried out at the regional level, especially vis-à-vis staff of other workforce agencies; determine how this position could best be used in light of other services available to VA program participants; and modify the national employment coordinator position description accordingly. To ensure that resources spent on job resource labs are used efficiently and effectively, we recommend that the Secretary of Veterans Affairs direct VRE to undertake additional efforts to: review the number of veterans using the job resource labs and ways in which veterans are using them; assess and offer regional offices direction on how the labs could be determine whether there are additional opportunities to coordinate with other agencies and organizations, such as local one-stop career centers. We provided a draft of this report to Labor and VA for their review and comments. In their comments, the agencies agreed with our recommendations. The agencies said they would work together to develop a plan to fully implement the agreement that focuses on time frames and benchmarks and would implement a systematic review of state-level agreements. Both agencies also said they will continue to develop joint training for improving the coordination and delivery of employment services for veterans with disabilities. In an effort to provide technical assistance to states that are facing difficulties implementing the agreement, Labor said it would review possible ways of providing VA confirmation that a client is employed without actually providing personal wage data. In addition, VA said it will continue to evaluate the effectiveness of the employment coordinator position and develop methodology to assess how the job resource labs are used. The Department of Labor’s comments are in appendix II and VA’s comments are in appendix III. We will send copies of this report to the Secretary of Labor, the Secretary of Veterans Affairs, other relevant congressional committees, and other interested parties and will make copies available to others upon request. In addition, the report will be available at no charge on GAO’s Web site at http://www.gao.gov. A list of related GAO products is included at the end of this report. If you or your staff have any questions about this report, please contact me at (202) 512-7215. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Staff acknowledgments are listed in appendix IV. Our overall research objectives were to describe the status of the implementation of the October 2005 memorandum of agreement between the Department of Labor (Labor) and the Department of Veterans Affairs (VA) aimed at coordinating efforts to advance the employment opportunities for veterans with service-connected disabilities, and to describe the status of the implementation of VA’s Five-Track employment program, with a particular emphasis on the role of employment coordinators and job resource labs. To address these objectives, we conducted interviews with national Labor and VA officials and national representatives of veterans’ service organizations. We also visited five states, where we met with Labor, VA, and the state workforce agency officials and staff, and toured job resource labs and one-stop career centers. During our site visits, we spoke with veterans who were past or current participants in VA’s Vocational Rehabilitation and Employment services (VRE) and local representatives of veterans’ service organizations. In addition, VA provided us with information on the number of job resource labs that were completed, the number still in process, and the reasons for the delays. Finally, we reviewed prior GAO reports and other relevant documentation. We conducted our work from November 2006 to September 2007 in accordance with generally accepted government auditing standards. We interviewed national Labor and VA officials to determine the status of implementation of the agreement, in particular, time frames for completion and any challenges to implementation. With VA officials, we also discussed the status of the implementation of the Five-Track program, including any implementation challenges that remain, and the roles of employment coordinators and job resource labs in the employment and training of disabled veterans. We conducted site visits in five states—Michigan, Alabama, Illinois, South Dakota, and California. California is divided into three VA regions; we visited only the San Diego region. We selected a mix of states based on the following criteria: (1) dispersion across the four VRE geographic areas (Eastern, Southern, Central, and Western), (2) both pilot and nonpilot sites for the Five-Track program, and (3) states with large and small veteran populations. Table 4 lists our selected site visit locations and summarizes these selection criteria for each state. In making our selections, we also factored in the states’ reputation for coordination between Labor and VA at the state and local levels, which we determined based on input from national Labor and VA officials and other sources. In each state, we met with VA regional office staff and officials from Labor and the state workforce agency. At VA regional offices, we spoke with VA counselors, employment coordinators, and regional office management. We also toured the job resource labs. In each state, we also spoke with either the state VETS program director or assistant director assigned to the region. In addition, we met with staff and officials from the state workforce agency, including Disabled Veterans’ Outreach Program specialists. We visited one-stop career centers in four states and met with local office managers in three of these. In some cases, we followed up on our interviews by phone, e-mail, or in person to collect additional information. In two of the states, we visited locations that have satellite offices of the VA regional office. In California, we visited Anaheim, which is a satellite location of the San Diego VA regional office. There, we met with the VA vocational rehabilitation counselor and state workforce agency local office management and staff. We also observed the job resource lab. We chose Anaheim based on its proximity to San Diego, which was one of our site visit locations. In Alabama, we interviewed VA and state workforce agency staff in Huntsville, a satellite location of the Montgomery VA regional office, and also visited the job resource lab. We selected Huntsville based on the recommendation of a Montgomery VA official and its proximity to a GAO field office. To gain additional perspectives on the implementation of the agreement and the Five-Track program, on each of our site visits, we met with veterans who were currently participating in the Five-Track program or who had recently completed training through VA’s Vocational Rehabilitation and Employment program. We also spoke with local representatives of veterans’ service organizations. In addition, we conducted phone interviews with representatives of a number of veterans’ service organizations who are based in the Washington, D.C. area. VA provided us with a list of the job resource labs and the status of their implementation. From this list, we calculated the total number of labs, the number of labs that were completed, and those that were still in process. We also categorized and sorted the reasons for delays in implementing the labs that were not yet complete. We assessed the reliability of the VA data and determined it was suitable for the purposes of this report. The information provided by VA was current as of May 2007. Sigurd R. Nilsen, Director Heather Hahn and Kathryn Larin, Assistant Directors Caitlin Croake and Amber Yancey-Carroll also made significant contributions to this report in all facets of the work. In addition, Walter Vance assisted in the review of external data and in developing site visit selection criteria; Elizabeth Curda and Gregory Whitney lent subject matter expertise; Doreen Feldman and Jessica Botsford provided legal support; Letisha Jenkins and John Ortiz assisted with data collection; and Charles Willson provided writing assistance. Veterans’ Employment and Training Service: Labor Could Improve Information on Reemployment Services, Outcomes, and Program Impact. GAO-07-594. Washington, D.C.: May 24, 2007. Trade Adjustment Assistance: Labor Should Take Action to Ensure Performance Data Are Complete, Accurate, and Accessible. GAO-06-496. Washington, D.C.: April 25, 2006. Veterans’ Employment and Training Service: Greater Accountability and Other Labor Actions Needed to Better Serve Veterans. GAO-06-357T. Washington, D.C.: February 2, 2006. Veterans’ Employment and Training Service: Labor Actions Needed to Improve Accountability and Help States Implement Reforms to Veterans’ Employment Services. GAO-06-176. Washington, D.C.: December 30, 2005. Workforce Investment Act: Labor and States Have Taken Actions to Improve Data Quality, but Additional Steps Are Needed. GAO-06-82. Washington, D.C.: November 14, 2005. Veterans’ Employment and Training Service: Preliminary Observations on Changes to Veterans’ Employment Programs. GAO-05-662T. Washington, D.C.: May 12, 2005. Workforce Investment Act: States and Local Areas Have Developed Strategies to Assess Performance, but Labor Could Do More to Help. GAO-04-657. Washington, D.C.: June 1, 2004. Veterans’ Employment and Training Service: Flexibility and Accountability Needed to Improve Service to Veterans. GAO-01-928. Washington, D.C.: September 12, 2001. Veterans’ Employment and Training Service: Proposed Performance Measurement System Improved, but Further Changes Needed. GAO-01-580. Washington, D.C.: May 15, 2001. Veterans’ Employment and Training Service: Strategic and Performance Plans Lack Vision and Clarity. GAO/T-HEHS-99-177. Washington, D.C.: July 29, 1999. Veterans’ Employment and Training Service: Assessment of the Fiscal Year 1999 Performance Plan. GAO/HEHS-98-240R. Washington, D.C.: September 30, 1998. Veterans’ Employment and Training: Services Provided by Labor Department Programs. GAO/HEHS-98-7. Washington, D.C.: October 17, 1997. Vocational Rehabilitation: Better VA Management Needed to Help Disabled Veterans Find Jobs. GAO/HRD-92-100. Washington, D.C.: September 4, 1992. | To better assist veterans with service-connected disabilities seeking employment, in 2005, the Departments of Labor (Labor) and Veterans Affairs (VA) signed an agreement to coordinate employment services for disabled veterans. Around the same time, VA rolled out a redesigned employment program for these veterans, known as the Five-Track program, which also established an employment coordinator position and job resource labs. To help Congress understand the status of these initiatives, GAO was asked to provide information on (1) the progress in implementing the 2005 agreement and challenges to implementation, (2) the status of implementation of VA's Five-Track program and challenges posed by recently returning veterans, and (3) the role of employment coordinators and job resource labs in serving veterans. To obtain this information, GAO interviewed Labor and VA officials and national veterans' service organizations, and conducted site visits in five states. Labor and VA have implemented some elements of their agreement to coordinate efforts, but face challenges executing the agreement on the state level and lack a complete plan for implementing and assessing the progress of the agreement. Labor and VA have implemented one element of the agreement--establishing three joint work groups--but have only partially implemented or taken no action on the others. In addition, all five states we visited had implemented at least some elements of the agreement that need to be carried out on the state level, but faced challenges implementing others. Labor and VA have not developed an implementation plan that includes long-range time frames and benchmarks to measure progress. Further, Labor and VA have not fully assessed state actions to implement the agreement and may not have provided states with sufficient guidance. While VA has mostly implemented its Five-Track employment program for disabled veterans, the employment needs of newly returning veterans may pose challenges. VA officials have completed a pilot study, trained staff, distributed orientation materials, and deployed employment coordinators, but other components remain in process. Some officials expressed concerns that employment programs for disabled veterans--including the Five-Track program--may not be prepared to meet the needs of participants returning from recent conflicts, who VA predicts will be more likely than previous returning veterans to have certain disabilities, such as those associated with traumatic brain injuries. VA has begun efforts to address these needs. VA employment coordinators and job resource labs in the five states we visited provided employment assistance to a limited number of veterans and some of their functions were available elsewhere. Employment coordinators provided direct employment services for veterans and also performed job development and outreach activities. However, employment coordinators we met with provided limited services to veterans outside their local areas, and similar services were available elsewhere. Job resource labs provided some additional resources for veterans, but according to some staff, not many veterans are using the labs. Job resource labs also appeared to duplicate other available services. |
In response to budget reductions in the 1990s, SBA streamlined its field structure by downsizing the 10 regional offices, shifting its workload to district or headquarters offices, eliminated most of the region’s role as intermediate management layer between headquarters and the field (district offices), and created the Office of Field Operations to assume the intermediary role. Prior to this restructuring, SBA had changed its service delivery model—moving from making loans directly to small businesses to guaranteeing loans made by commercial lenders. SBA previously had provided loans directly to small businesses largely through its district offices. SBA’s 7(a) loan guaranty program exemplifies how the service delivery model changed. The program has several delivery methods—such as the certified and preferred lender programs—in which SBA has delegated different functions to lenders. For instance, certified lenders must perform a thorough credit analysis on the loan application packages they submit to SBA so that SBA can heavily rely on that analysis in making SBA’s credit and eligibility decisions, thereby shortening the time for SBA loan processing. Preferred lenders have delegated authority to make SBA- guaranteed loans, subject only to a brief eligibility review and assignment of a loan number by SBA. SBA provides final approval for loans made under the regular 7(a) program for loans made by lenders who have not been approved as certified or preferred lenders. SBA has 68 district offices that operate as the frontline provider or conduit of SBA services and programs. Specifically, district offices provide counseling and training services to individuals or businesses that aid in the formation, management, financing, or operation of a small business enterprise. District offices provide information on and promote SBA products to lending partners, the small business community, and groups such as chambers of commerce and trade associations. Finally, district offices are charged with completing statutorily mandated reviews, such as section 8(a) program participant reviews that ensure participants continue to qualify for the program and are meeting program requirements. When we reported on SBA in 2003, the agency had started or was planning transformation efforts to better align resources and functions with the changed organizational structure. The transformation effort was to be done in three phases that included two key initiatives. We reported on the first phase of these efforts. The first initiative focused on changing the role of district offices (through a pilot in three district offices) to emphasize outreach to small businesses about SBA’s products and services and linking these businesses to the appropriate resources, including lenders. For instance, SBA provided training to employees on marketing and outreach, developed new job descriptions for its marketing and outreach specialist positions, and reorganized staff in the offices. However, we found that budget constraints impeded work on this initiative and could continue to do so. We also found that the agency’s budget requests for transformation were inconsistent and lacked a detailed plan that showed priorities and linked resources to desired results. The second initiative focused on centralizing loan functions to improve efficiency and consistency of the loan approval, guaranty purchase, and liquidation processes for SBA’s 7(a) loan program. In March 2003, the liquidation center in Santa Ana, California, that typically worked on disaster loans began a pilot program to process new 7(a) liquidations and guaranty purchase cases from the three pilot district offices. According to SBA, the centralization achieved processing efficiencies. In phases two and three of its transformation, SBA had planned to establish a new 7(a) liquidation and guaranty purchase center near Washington, D.C., and expand the centralization initiative until all of the loan functions performed by its 68 district offices were centralized. However, we found that SBA’s centralization efforts also could be impeded by the challenge of trying to ensure that job realignments and relocations from multiple field offices would result in having experienced staff at the centralized locations. We also noted that relocations could prove disruptive for employees by decreasing morale and productivity and, thus, could negatively affect SBA’s operations. Between 2004 and 2006, SBA created two centers to conduct work that had been previously performed in the district offices. As part of their respective missions, these centers provide assistance and oversight, as necessary, to lenders. Specifically, the NGPC in Herndon, Virginia, processes guaranty purchase requests and assists lenders during loan liquidation and the Standard 7(a) Loan Guaranty Processing Center (operating out of two locations in Citrus Heights, California, and Hazard, Kentucky) has responsibility for processing 7(a) loan guaranty applications for lenders who have not been approved as preferred lenders. SBA’s workforce has declined significantly since the 1990s. During the restructuring in the 1990s, SBA’s workforce decreased from over 3,800 employees to about 3,100 employees—a decrease of about 19 percent. Between 2000 and 2007, when additional restructuring occurred, SBA’s workforce decreased further by about 26 percent. As of September 2007, SBA had 2,166 employees. Our 2003 report noted that transforming an organization is challenging. We further concluded that while SBA had made progress with its endeavors, weaknesses we identified in SBA’s approach could be mitigated by using the key practices discussed in the report. Moreover, comprehensive and strategic planning would minimize challenges, including problems with employee morale and productivity. In particular, we said that more two- way communication, transparency, and employee involvement would improve the effort as the agency moved forward with transformation. Accordingly, we made recommendations related to how SBA could use the key practices to increase the potential for a successful transformation. We discuss our recommendations related to best practices, including actions taken by SBA, in this report. SBA has made progress in applying key practices to implement change and has begun addressing remaining employee concerns about involvement and continued leadership commitment. SBA’s current leadership has emphasized transparency and communication. Additionally, it refocused agency priorities with a new performance management system, which prioritizes agency goals and links them to operating plans for offices, performance scorecards, and employee performance plans. SBA also has made efforts to solicit employees’ ideas and feedback and has begun to examine employees’ continued concerns about their involvement in decisions that affect their work. Finally, some SBA employees expressed concerns about the commitment of future leadership to the current initiatives, but senior career officials said that they planned to continue positive efforts. SBA’s current leadership has emphasized transparency and communication, addressing our prior recommendation that the agency improve its communication strategy. In 2003, during the early stages of its transformation, we made this and five other recommendations to SBA because we found the agency’s transformation plans were not transparent enough to allow related progress to be tracked and frequent two-way communication was lacking within SBA (see table 1). In July 2006, 3 years after transformation began and following the unsuccessful outcome of some transformation initiatives and other agency challenges that we discuss elsewhere in this report, SBA underwent a change in leadership with the appointment of a new Administrator. Under its new Administrator, SBA’s leadership took steps to increase transparency in the agency’s actions and improve its communication efforts. For fiscal year 2008, SBA developed a comprehensive communications plan that addressed challenges from the previous year with regard to negative reports about its disaster program, primary loan program, and other aspects of the agency. The plan included strategies for improving internal and external communications and detailed steps for making these improvements. It also assigned responsibility for each communication area such as SBA’s Web site, marketing, and internal communications to specific individuals within SBA’s communications office. Senior leaders at SBA acknowledged that effective communication had become a key area of focus when the Administrator began his term, and identified communication deficiencies. When we met with managers and staff in SBA’s district offices, many told us that under new leadership, agencywide communication had improved significantly and that employees were more informed about ongoing agency initiatives. Both managers and staff attributed these improvements to the Administrator’s commitment to transparency and communication, noting that he had personally taken steps to directly communicate agency plans and updates to all employees. In previous work reviewing practices that empowered and involved federal employees, we noted that top leadership commitment is crucial to instilling a common vision across the organization and creating an environment that is receptive to innovation. SBA officials told us that, early in his term, the Administrator had visited the district offices and informed employees that they could contact him directly by e-mail, and in two of the locations that we visited, district directors told us that employees had communicated with the Administrator. The Administrator also continued town hall meetings, as the previous Administrator had done, and implemented video broadcast meetings and messages that were accessible to all employees, including those in the field. In addition to the Administrator’s individual efforts, SBA has used e-mail and conference calls to improve its internal communications. For example, employees said that SBA distributes agencywide e-mails on procedural notices. Regular conference calls also provide headquarters and field employees with opportunities to share information. For example, the Office of Field Operations—which acts as a liaison between headquarters and the field—participates in a weekly conference call with the district offices that various program offices in headquarters also sometimes participate in, and the program offices hold conference calls with field employees who work in those areas. On the local level, the district offices also participate in weekly regional conference calls with their respective regional Administrators. As part of its efforts to improve communications specifically between headquarters and the field, SBA also developed a 10-member field advisory council (FAC), comprising one district director from each region, which gives district office management an opportunity to communicate field issues and concerns directly to headquarters. The FAC’s purpose is to provide input and feedback directly to the Administrator and senior management regarding the implementation of agency policy and programs on operations in the field level and on SBA’s customers. District directors serve on the FAC on a rotational basis for an initial period of 18 months and can be reconfirmed for subsequent terms. The FAC participates in weekly conference calls with senior leaders and meets with the Administrator several times a year. During our visits to the district offices, several district directors told us that the FAC was an effective tool for two- way communication between headquarters and the field, particularly for sharing the field perspective with headquarters officials. For example, one district director said that the FAC was useful for discussions related to staff resources in the field, and several identified their regional representatives or said that they could share issues with their regional representatives to be raised in FAC meetings. SBA has also updated its internal (intranet) and external (Internet) Web sites to provide more useful information than provided in the past, such as program area resources to its employees and revised program guidance to its customers. Although some employees indicated that SBA’s Internet site was more user-friendly than its intranet site in searching for information, others provided examples of intranet improvements, such as resources to help with conducting marketing and outreach with lenders—a major aspect of the district offices’ role following centralization. As noted, many district office employees we interviewed indicated that SBA’s current leadership had improved agencywide communication significantly. Employees said they were aware that communication was now a priority for the agency and said that in the past, leadership’s actions had been unclear or not communicated to them. Partly because of more open communication from the top and because district directors were more informed than in the past, some district office employees said that communication had improved in their offices. Some also noted that communication with other areas in the agency, such as the loan centers, had improved. However, several district office employees said they experienced challenges communicating with headquarters officials. For example, they said that headquarters officials were not consistently responsive to their communications or were unable to answer their questions during scheduled conference calls. In addition to a new Performance Management Office (PMO) and a comprehensive performance management system, SBA also applied other key transformation practices that we recommended the agency adopt, including finalizing a transformation plan, identifying who would lead transformation efforts, and developing a performance management system that reflected agency goals and employees’ roles. As discussed above, SBA came under new leadership in 2006, and with the appointment of a new Administrator, SBA’s existing draft transformation strategy was never approved. Instead, the Administrator introduced a strategy to carry out remaining transformation objectives in a manner that addressed deficiencies in past implementation efforts and focused on SBA’s mission and priorities. The new strategy, which shifted SBA from organizational restructuring to improving performance, emphasized four specific priorities that made up the Administrator’s reform agenda: (1) meeting all compliance requirements, (2) ensuring that all SBA programs operated efficiently and effectively, (3) improving communication, and (4) providing effective employee training. This agenda is generally consistent with key transformation practices that we identified. Several SBA officials with whom we spoke pointed out that some of the negative impacts of past transformation efforts—including a significant reduction in SBA’s overall workforce and directed reassignments of district office employees to a new centralized loan center—had caused transformation to take on a negative tone across the agency. Our discussions with employees in many of the district offices confirmed this. Thus, the Administrator replaced “transformation” with the term “reform agenda” to describe SBA’s efforts to move forward, although it still aimed at accomplishing important transformation objectives. For example, transformation objectives included (1) focusing on SBA’s customers, (2) empowering employees, and (3) becoming more results-driven and performance-based. Similarly, the reform agenda was created on the foundation of SBA being (1) outcomes-driven; (2) customer-focused; (3) employee-enabled; and (4) accountable, efficient, and transparent. While we found that SBA’s earlier transformation strategy remained a draft plan, the new strategy, which has been presented as its performance management framework (PMF) (discussed below), has been implemented. To provide the infrastructure for SBA’s performance management efforts, including overseeing SBA’s strategic plan and goals, in January 2007, the Administrator established a new office, the Performance Management Office (PMO) and a new position, the Associate Administrator for Performance Management. At SBA, the Chief Financial Officer (CFO) serves concurrently in this new position. The PMO serves as an internal resource for SBA’s program offices, assisting them with data analysis, project management, and operational improvement. It reports on agency performance internally and externally (for example, performance and accountability and Program Assessment Rating Tool (PART) reporting). It also oversees and tracks SBA’s implementation and progress on the President’s Management Agenda initiatives and recommendations. In a report discussing action steps proposed by selected performance management experts for improving performance management in federal agencies, such experts stated that creating an office such as the PMO increases the focus on performance as an organizational priority. Additionally, the report lists “performance tracking and dialogue,” for which the PMO is responsible, as one of eight management practices that contribute to a performance culture. Moreover, the PMO is in line with recent efforts by the Office of Management and Budget (OMB) to promote program performance across the federal government. SBA further encouraged a collaborative approach to performance accountability by designating managers at various levels to lead initiative teams as “champions” or “project owners” of specific activities, thereby allowing them the opportunity to participate in leadership’s vision for SBA and addressing employees’ concerns about being uninformed and uninvolved in past transformation efforts. In our report on practices that empowered employees, we also note that using employee teams to help accomplish agency missions and involving employees in planning and sharing performance information with them are effective tools. In some instances, SBA designated employee champions to oversee improvements in specific programs or agency functions, and some initiatives had champions in headquarters and the field. For example, for initiatives involving staff support and training and SBA’s work with small businesses in underserved markets, program managers served as headquarters champions, while regional Administrators or district directors served as field champions. Following this approach, at the district office level, district directors also designated individual employees as champions of local initiatives related to the district offices’ mission and goals. District office initiatives are separate from output-related goals that SBA establishes for the district offices, such as loan volume. Instead, SBA considers initiatives to be “value-added” activities, such as leveraging resource partner services to serve small businesses and outreach to faith-based and community organizations, women, veterans, and other groups. District directors told us that SBA requires each office to submit a plan each fiscal year to accomplish the initiatives that are key to its local market and allocate resources from its existing budget. In some cases, district office employees had responsibility for initiatives that were related to their primary roles. SBA’s new PMF is the core component of its efforts to improve agency performance. According to SBA, the objectives of the PMF are to (1) prioritize objectives, (2) manage existing resources to meet priorities, (3) target items that produce measurable impacts, (4) discontinue activities not critical to success, (4) improve accountability, and (5) foster agency collaboration to achieve collective priorities. SBA documents state that the PMF allows SBA and its program offices to be more accountable, performance-oriented, transparent, and customer-focused—in line with the reform agenda—and provides the agency with flexibility by allowing prioritization of limited resources to best fit the needs of SBA’s customers. In the PMF, SBA’s priorities—including the reform agenda and other Administrator priorities, program office priorities, and goals and initiatives—form the basis for major area (program and district office) operating plans (see fig. 1). In developing its operating plans, SBA looks at the resources available to accomplish the metrics or milestones for each activity (compliance requirement, goal, or value-added initiative) performed in a given area. For example, district office operating plans include metrics for loans made to certain groups or communities that SBA targets, as well as the staff and budget allocation for achieving the loan goal. In turn, the agency assesses operations by using performance scorecards that include metrics and other measures for tracking the progress of goals and other activities. The scorecard then provides detailed information on the progress made toward achieving each metric within the operating plan. For fiscal year 2007, SBA developed new operating plans and streamlined scorecards to show fewer, higher-impact items, making 100 percent compliance with statutory requirements a priority. For the district offices, the number of scorecard items decreased from about 40 to less than 10. District office employees said that they preferred having fewer scorecard goals that they could work toward and that they had greater autonomy to achieve. In the past, district office scorecard items included some goals for which the district offices depended on local resource partners such as SCORE (formerly the Service Corps of Retired Executives), Small Business Development Centers (SBDC), and Women’s Business Centers (WBC) to achieve. For example, while the district offices have goals for training and counseling that they provide to small businesses, they previously also had goals for training and counseling that their resource partners provided to small businesses. Although some employees viewed the new scorecard as an improvement, others noted that some of their current responsibilities were not reflected in the new scorecard and that the district office did not receive credit for work on these additional responsibilities. For example, since SBA centralized its loan functions, the scorecard does not account for any of the assistance that district offices provide to the loan centers. Because district office staff resources were limited, the employees felt that they should receive credit for all the work they performed. As figure 1 indicates, the priorities set within the PMF also are reflected in SBA’s staffing decisions, employee roles and responsibilities, employee training, and personal business commitment plans that serve as individual performance contracts for employees. Scorecards then help measure performance on goals and objectives related to these functions and documents. For example, the personal business commitment plans of senior officials link their performance to goals in the scorecards for their respective areas. Finally, accountability for achieving results related to agency priorities is a key aspect of the PMF. SBA senior officials track the progress of PMF priorities monthly, and program managers are required to prepare monthly reports on scorecard metrics and meet with the Administrator and Deputy Administrator to discuss targets that have been met and action plans for addressing outstanding targets. (We provide an example of district office scorecard compliance requirements and goals in our discussion of the district offices’ roles and responsibilities.) Additionally, SBA holds regular progress review meetings at various levels in the agency, in part to create a culture of peer-based accountability and facilitate performance improvements agencywide. Several SBA employees with whom we spoke said that they understood the PMF, and others noted that it clarified the agency’s goals. One program office official added that SBA had made progress with the PMF by helping managers to make the connection between agency goals and their individual program offices. SBA’s current approach to organizational change offers some avenues for employee involvement, but some employees, including union representatives, expressed concerns about limited involvement or about providing input and seeing results. SBA has taken steps to involve district offices and solicit their feedback and ideas, but most of these initiatives involve district directors who may not always reflect district employees’ concerns or ideas. As noted above, district directors have an opportunity to serve on the FAC and communicate field issues and concerns to SBA headquarters. Additionally, in 2007, SBA implemented an informal “district director in residence” effort that allows district directors to work on temporary assignments in headquarters, where they can be exposed to headquarters operations and provide a field perspective on initiatives and activities that affect the field. Similarly, some headquarters managers have served in temporary field positions, such as acting district director, where they can manage field employees and provide feedback on the impact of headquarters directives on field personnel. For example, we met with a district director and an assistant district director who had worked in headquarters on initiatives that incorporated technology to make improvements to the 8(a) Business Development program. In both instances, the officials were able to provide practical input for these initiatives because the district offices are responsible for monitoring and assisting more than 9,600 firms in SBA’s 8(a) program and for communicating with other federal agencies that contract with these firms. Some district office employees with whom we met acknowledged that headquarters generally sought more input from the field regarding policies, procedures, and new program initiatives than in the past. SBA’s intranet also has a comments and suggestions tool that all employees can use to submit their ideas to the agency. In one location, the district director pointed out that two district office employees had submitted ideas that headquarters accepted for implementation—one for pursuing an initiative with the state to coguarantee SBA loans and encourage local lenders to make more SBA loans. However, district office employees remained concerned that they had little input in decisions that directly affected their work, and employees generally agreed that they were not certain that management incorporated their input when final decisions were made. For example, several recalled providing input for procedure revisions and developing new loan programs but said that the programs that headquarters had ultimately implemented were less attractive to (that is, cost-effective for) local lenders, making it challenging for district offices to market the programs and meet their loan goals. In addition, results from an OPM survey of SBA’s employees reflect an ongoing concern employees have about their involvement in decisions that affected their work. As shown in figure 2, about 32 percent of employees surveyed in 2004, 2006, and again in 2007 had negative responses regarding their involvement in decisions that affected their work. Similarly, the rate of positive employee responses has shown little change over the years, and SBA’s 2007 positive employee response of 45 percent was notably lower than the 2006 governmentwide positive response of 54 percent. Representatives from SBA’s employee union reported insufficient interaction with agency leadership and limited involvement in management decisions. For example, the representatives said that changing leadership within SBA’s human capital office made it difficult to establish relationships and that SBA’s senior leadership had not been responsive to the union’s efforts to collaborate. They also said that the union had not had opportunities to provide input for and negotiate on agency initiatives that affected employees, such as the new field staffing model, and they felt that SBA unilaterally changed agreements without seeking the union’s input. However, SBA officials said that federal union contracts typically allow management some discretion with regard to union involvement. Specifically, they told us that decisions affecting the agency’s mission, budget, and staff resources were made at the management level and did not involve the union. For example, they stated that SBA management would not negotiate with the union on SBA’s development of its field staffing model. In addition, they said that there was some misunderstanding among union representatives in this regard. We have previously reported that employee involvement strengthens the transformation process by including frontline perspectives and experiences. Specifically, employee involvement helps to create the opportunity to establish new networks and break down existing organizational silos, increase employees’ understanding and acceptance of organizational goals and objectives, and gain ownership for new policies and procedures. More specifically, in relation to eliciting feedback, OPM suggests that federal agencies could increase the usefulness of employee surveys by combining survey results with other personnel information or with additional feedback tools such as focus groups. In doing so, agencies could better determine the rationale behind employee survey responses and more accurately assess their human capital environment. SBA’s employee survey results and comments from employees with whom we met suggest that some employees were still not satisfied with the extent of their involvement in work-related decisions. In line with OPM’s suggestion, SBA officials said that they had begun efforts to examine several employee issues that they identified from human capital surveys for fiscal years 2005 through 2007. The officials said that they had followed up on questions that had the highest negative responses, including the question on employees’ satisfaction with their level of involvement in decisions that affected their work. SBA employees in headquarters and the field participated in focus groups in April and June 2008 and SBA plans to issue a report summarizing the focus group results on its intranet and develop pilot initiatives to address the issues it determines to be most critical. In June 2008, SBA’s Administrator, Steve Preston, was sworn in as Secretary of Housing and Urban Development. Prior to his nomination in April 2008, SBA managers and staff had expressed uncertainty about whether positive actions would continue under the next Administrator. When we spoke with managers and staff at SBA prior to the nomination, many said that under Administrator Preston, SBA’s approach to leadership had improved significantly, and some expressed concerns about the commitment of future leadership to progress at SBA. In particular, some employees were uncertain about whether positive actions would continue under the next Administrator and anticipated that a new Administrator might take the agency in a different direction, which could lead to the dismantling of many improvements. Some who were optimistic said that successful initiatives currently in place would continue if other SBA leaders shared the Administrator’s vision for improving the agency. In discussing efforts to institutionalize the positive initiatives that Administrator Preston had begun at SBA, the Associate Administrator for Performance Management (CFO) told us that she and other senior career officials had met to discuss how they could ensure that the positive changes and practices that had been put in place in the previous 2 years could be continued under a new Administrator. She said that while a new Administrator would have new ideas, she felt that it would be important to sustain practices and initiatives that had led to noticeable improvements in the agency’s capacity to implement a new vision or priorities. In particular, she pointed out that SBA’s efforts, such as the PMF, and improved communication, had improved its capacity to operate more effectively and efficiently, which would help any new programs or policies under a new Administrator. Additionally, the Chief of Staff told us that as the officials develop a transition plan for the next administration, they would include support for the positive actions that Administrator Preston implemented. Our previous work on key practices that support transformations noted that frequent turnover of political leadership in the federal government often made it difficult to sustain and inspire attention to make needed changes. Therefore, it is important that SBA’s leadership—career and political—remain committed to sustaining these efforts under existing and future administrations. After the centralization of loan liquidation and purchase guaranty functions in 2003, employee morale at SBA declined significantly, as indicated by employee survey results in 2004 and 2006. The SBA Administrator’s actions to improve communication and transparency were meant, in part, to assure employees that they would be better informed about what the agency was doing. In 2007, SBA administered another employee survey to measure whether these actions had improved employee morale and found that employee perceptions of leadership improved considerably. Employees also told us that they thought they were getting better information, which was corroborated in the 2007 employee survey. SBA also embarked on a major training initiative in 2007—SBA University—to provide core training to field employees and address concerns about training expressed in employee surveys. Employees told us that they appreciated SBA University because it showed that the agency recognized that employees needed training for their new roles and responsibilities. However, SBA has not yet developed a comprehensive and strategic training plan that includes specific goals, strategies, and milestones for developing and implementing core courses. Without such a plan, SBA lacks some accountability for ensuring that the training efforts would continue to be implemented. Following the notice of directed reassignments associated with the 2003 centralization of the 7(a) liquidation and purchase guaranty functions, employee morale declined significantly, as indicated in SBA’s 2004 employee survey results. Around the same time a new Administrator was appointed in 2006, SBA received the results of its 2006 employee survey, which continued to indicate low morale. The employee survey results in 2004 and 2006 showed a lack of respect for and trust in SBA leadership and a concern about training opportunities. According to SBA officials, the new Administrator made improving employee morale a priority. As discussed earlier, SBA leadership sought to improve communication and transparency across the agency. The new SBA Administrator’s efforts to communicate with employees included soliciting information from employees and visiting field locations to obtain their input on how to improve agency operations and morale. For example, during his first year at SBA, the Administrator attended 32 employee meetings and visited 23 district offices. At these meetings, he would ask managers to leave the room and tell employees that the discussions would remain confidential and individual employees’ comments would not be shared outside the meetings. According to employees with whom we met, the new Administrator’s positive and open management style generally improved morale within the agency. Many employees told us that they believed they were provided with more information on agency initiatives and the Administrator’s priorities. The results of the 2007 survey of employees showed an improvement in positive perceptions of leadership compared with 2004 and 2006 survey responses. For example, as shown in figure 3, there was a 21 percentage point increase between 2007 and previous years’ surveys in SBA employees agreeing with the statement, “I have a high level of respect for my organization’s senior leadership.” Furthermore, SBA’s 54 percent positive response for this question was higher than the governmentwide positive response level of 49 percent in 2006. Likewise, in 2007, 52 percent of SBA employees provided a positive response to a question about satisfaction with the information received from management. This response was an increase of nearly 20 percentage points from 2004 and at a level that also was higher than the governmentwide positive response of 47 percent in 2006. Other responses from the 2007 survey showed that employees still had concerns. As noted previously, a large percentage of employees felt that they were not sufficiently involved in decisions that affected their work. Also, 39 percent of SBA employees responded negatively when asked how satisfied they were with their opportunities to get a better job in the organization. While this level of negative response was an improvement from 2006, when 47 percent of SBA employees responded negatively, it was still notably worse than the 31 percent of governmentwide employees in 2006, who provided a negative response to this item on the 2006 OPM survey. SBA officials told us that the 2006 employee survey results clearly showed that employees wanted more and better training. We have noted in previous work that training and development can play a key role in helping agencies address the challenge of transformation and cultural change and help ensure that their workforces possess the knowledge, skills, and competencies needed to work effectively in a rapidly changing and complex environment. Prior to the arrival of SBA’s new Administrator in July 2006, the range of training available to SBA staff was limited, in part because SBA had not invested in training due to budget constraints. Most training was on the job or offered online. Recognizing that many employees had taken on new roles and responsibilities, the Administrator supported the need to develop a training program. This program—SBA University—started in 2007. The first sessions were for field office employees, with the goal of training them in the new roles and responsibilities incurred as part of transformation. SBA held three sessions of week-long training during late July and August 2007, with more than 1,300 employees from field offices across the country. The agency organized the sessions around seven modules of instructor-led training—(1) lender relations, (2) entrepreneurial development, (3) government contracting and business development, (4) public communications, (5) administration, (6) servicing centers, and (7) government contracting representatives. Within each module, some courses were program-specific, such as 8(a) annual reviews, while other courses were skill-specific, such as customer service. The agency offered another round of training under the SBA University in April 2008 for SBA managers (including managers in district offices and headquarters). Additional training is scheduled in 2008 for managers and headquarters employees. District office employees and management officials with whom we met generally were positive about SBA University. In addition, the positive reaction to the new emphasis on training was reflected in the 2007 employee survey results. For example, as shown in figure 4, employee response to an item asking about satisfaction with training shows a 13 percentage point increase in positive responses in 2007 compared to the 2006 survey. However, SBA’s positive response of 49 percent was lower than the 2006 governmentwide positive response of 54 percent and about a quarter of employees continue to express dissatisfaction with their training. Some employees told us that though they believed SBA University was a positive gesture to show that the agency would invest in training, they felt that many of the courses did not adequately prepare them to carry out their job responsibilities. Other employees told us that they were overqualified for some of the courses and would have benefited from more advanced training. SBA officials acknowledged that SBA University was a first step in developing a training program. They told us that SBA had not developed a training curriculum in recent years because of limited budgetary resources but that the agency was in the process of developing a curriculum focusing on core competencies needed for SBA’s mission-critical positions. As noted in SBA’s 2007 Human Capital Plan, SBA has identified six mission- critical occupations including (1) leadership (such as members of the Senior Executive Service and district directors); (2) lender relations and business development specialists; (3) contracts specialists, procurement analysts, and contracting officers; (4) loan specialists; (5) human resources specialists; and (6) information technology specialists. According to SBA officials, SBA has begun to develop core courses for the leadership and acquisition-related occupations. For example, SBA has developed a plan for leadership succession, including strategies for developing leaders, an implementation schedule, and an evaluation monitoring plan. They said that they started developing curriculums for these two occupations first partly because courses were available through external training resources. SBA officials stated that they planned to develop core curriculums to address the skills for the other mission- critical occupations. SBA officials told us that they have been working with the Office of Field Operations and subject matter experts in the district offices to identify the core competencies for the mission-critical occupations in the district offices. In addition, they plan to review individual development plans prepared by employees to identify training needs. Officials said that one of the challenges of developing and delivering training is ensuring that competent instructors with the right skills are available to teach courses—especially because vendors with the appropriate subject matter expertise may not be available to deliver the training. They added that while they started with acquisition-related and leadership curriculums, the next core courses probably would be for business development specialists. SBA also has created an Executive Development Council that provides direction, oversight and support for the development of leaders within SBA and development of employees at all levels of the organization. According to the council’s February 2008 charter, it will create a vision of development for SBA and design a developmental strategy that supports SBA’s mission and strategic goals. Though SBA has taken several positive steps related to training and development, SBA has not yet documented a comprehensive and strategic approach to training in a training plan. SBA’s Human Capital Plan describes the agency’s training and development initiatives—specifically, it describes the SBA University initiative as designed to advance and integrate employee learning with other critical business functions—but it does not provide any details on how and when training would be delivered. And while SBA’s Fiscal Year 2009 Performance Plan notes that the agency will design and institutionalize its long-term plan for the SBA University, this effort is still under way. Our previous work has shown that adequate planning allows agencies to establish priorities and determine the best ways to leverage investments to improve performance. For instance, a documented training plan would link the core competencies and mission-critical positions with specific goals, strategies, and milestones for developing and implementing core courses. A training plan can also present a business case for proposed training and development investments, including linkages with the agency’s strategic objectives, anticipated benefits, and projected costs. Without a plan for the rest of SBA’s core curriculum, SBA lacks some accountability for ensuring that the training efforts would continue to be implemented. In addition, without a plan, employees may not be able to anticipate future training opportunities and consequently judge agency efforts negatively. Since 2003, the roles and responsibilities of SBA district offices have shifted from processing loans and performing other loan-related functions to marketing SBA programs and services; conducting outreach to lenders, small businesses, and other organizations; and ensuring compliance with federal laws and regulations. The roles and responsibilities of the district offices have continued to evolve as SBA further refines what they should be. For example, select district office employees have begun assisting the loan centers with their loan and customer service functions, such as helping lenders prepare guaranty purchase requests. District offices also now have a formal role in providing disaster assistance. SBA defines district office roles and responsibilities through the annual compliance goals and requirements on the district office scorecards, but district offices can determine what actions they need to take to meet their goals, particularly in marketing and outreach. However, the district office employees still are adapting to their new roles, particularly those who used to process loans. Employees we interviewed said some aspects of their work had changed following the centralization of loan functions. For example, employees found that their work had become less structured and that it resulted in fewer tangible accomplishments. Due to decreased staffing levels, they also find themselves taking on multiple responsibilities, which affected their ability to conduct some activities. Since our 2003 report, the roles and responsibilities of SBA district offices have shifted from performing loan functions to conducting marketing and outreach for SBA programs and services and ensuring compliance with laws and regulations; these roles and responsibilities continue to evolve. Specifically, the loan processing, liquidation, and guaranty purchase functions that district offices used to perform were transferred to centralized facilities, leaving district offices to focus on promoting SBA programs and services through conducting marketing training lenders, providing support to and coordinating with SBA resource partners, identifying contracting opportunities for small businesses and developing them, and conducting annual eligibility reviews under the 8(a) business development program and other compliance functions. While a key component of centralization has been the removal of loan functions from the district offices, recent agency actions indicate that SBA continues to further amend the role of district offices vis-à-vis the loan centers. More specifically, SBA has created discrete roles for district offices to assist the loan centers in working with lenders. For example, as part of an SBA initiative to address problems arising from its centralization efforts at the NGPC, in February 2008, designated district office employees received training on the guaranty purchase process so that they could help lenders prepare complete and high-quality packages. (We discuss the changes related to centralized loan processing in more detail later in this report.) SBA also has established a joint marketing and customer service initiative between the Office of Capital Access and the district offices to develop a targeted and focused approach to marketing and outreach to the top lenders nationwide and in each district. According to SBA, the purpose of this initiative is to engage the top lenders, ensure that their purchase request packages and other loan paperwork are being processed in a timely manner and that SBA is responsive to their concerns, and identify opportunities to increase SBA lending activities. According to SBA, the guaranty purchase training and joint marketing and customer service initiatives are not new roles but are in line with the district office’s current roles and responsibilities. These initiatives have provided district offices with regular and concrete interactions with lenders, which the district office employees we interviewed said they had lost as a result of centralization. District offices also now have a more formalized support role in providing disaster assistance than they did previously. Under SBA’s 2007 Disaster Recovery Plan, district offices will provide “surge capacity” for large disasters (on scales approaching or exceeding the 2005 Gulf Coast hurricanes, which was the highest level of catastrophic disaster activity that SBA has faced to date) by helping process loan applications. According to the plan, district offices also would be responsible for conducting local media outreach and coordinating local resources. All of the district offices we visited had employees who received disaster assistance training. Some of the employees in these offices found this training informative. However, some employees expressed concern about the availability of future training and refresher courses because it might be several years before the next large-scale disaster occurred. Prior to the creation of the plan, district office employees volunteered to help process disaster loans on an ad hoc basis, and some employees we interviewed indicated that they had volunteered to work on the backlog of applications from victims of the Gulf Coast hurricanes Katrina, Rita and Wilma. However, we previously found that during the Gulf Coast hurricanes, SBA did not have a centralized and coordinated strategy for drawing upon district office staff resources, which contributed to the large backlog of disaster loan applications the agency experienced. As the overall function of the district offices has changed, the roles of individual employees also have changed and some employees continue to face challenges adjusting to their revised responsibilities. In the shift from loan-related functions to marketing and outreach, SBA removed the loan officer and loan servicing specialist positions from the district offices. The agency reclassified these employees primarily into one of two positions: lender relations specialist or business development specialist (see fig. 5). Lender relations specialists are responsible for conducting outreach to lenders through visits and training on SBA’s programs. Business development specialists are responsible for marketing SBA’s programs; conducting outreach to businesses and other economic development and community organizations, such as local chambers of commerce; conducting training, counseling, and technical assistance; and performing compliance activities. As part of their primary responsibilities, business development specialists also might have specific collateral duties in support of the office’s oversight, such as conducting annual reviews of 8(a) firms, and coordination of SBA’s programs and resource partners. The position descriptions contain a checklist of these duties, such as lead contact positions for SBA’s resource partners. For example, in one office we visited, one business development specialist also served as the 8(a) business development and point of contact for the Historically Underutilized Business Zone (HUBZone) program. Another business development specialist was the point of contact for the WBC and was also responsible for lender training. In some of the district offices we visited, employees who had been public information or information technology officers prior to transformation had been reclassified as business development specialists, even though they still retained these previous functions. The employees in the district offices we visited, particularly those who used to process loans, indicated that the nature of their work changed substantially as a result of transformation and that they were still adapting to these changes. Some of the employees said that they had performed marketing and outreach to some degree prior to centralization but that these functions were now their primary focus. Employees we interviewed indicated that since the loan functions had been centralized, their work had become less structured, provided fewer tangible accomplishments, and provided less regular interaction with lenders. Their new responsibilities also required them to travel more than they had previously. These changes have challenged some employees more than others, particularly those employees who used to process loans. However, SBA officials in the Office of Field Operations indicated that the new roles were meant to encompass more than one function and some district office employees concurred that they were now expected to be generalists rather than specialists. Moreover, SBA’s recent initiatives to refine the district office roles might provide district office staff with more tangible accomplishments and interaction with lenders. However, the business development specialists who conduct the 8(a) annual reviews said that their roles had not changed as a result of transformation, but that for some, the number of firms for which they were responsible had increased. Moreover, because of the emphasis on compliance requirements in the PMF, the significant amounts of time they spent meeting goals for annual reviews left them little time to work with clients to develop their businesses. However, SBA indicated that it has an initiative under way to automate the annual review process to help free up time for business development. SBA defines district office roles and responsibilities, including those for marketing and outreach, through the goal scorecards for district offices and other mechanisms. However, SBA still gives the district offices flexibility to determine how they would meet their goals. District office scorecards outline key annual compliance requirements and loan volume goals (see fig. 6), but as we noted earlier, meeting compliance requirements (such as completing 8(a) annual reviews) that are required by statute is a priority in SBA’s PMF. However, the district directors in the district offices we visited generally indicated that SBA gives them the flexibility to determine the actions they need to take to achieve their scorecard goals and tailor activities to their local markets. For example, district directors indicated that they can determine the number and types of events their employees should attend to market SBA’s programs. SBA also provides a range of marketing materials, including program literature, templates for marketing materials, and talking points for new loan programs on its intranet site that district offices can customize with local information. Some district office employees we interviewed found the agency’s marketing materials to be helpful in providing professional and consistent information to lenders on SBA programs, while still giving them the flexibility to tailor the information to their local market conditions. SBA further defined the roles and responsibilities of district office employees through the development of position descriptions and personal business commitment plans. The position descriptions outline current responsibilities as well as collateral duties for district office employees. While the district offices we visited have flexibility over how they carry out their responsibilities, some offices told us that they faced challenges in meeting their goals due to decreased staffing levels. Most of the offices we visited indicated that their staffing levels had decreased substantially after the transformation began due to factors such as reassignments, attrition, buyouts, and retirements and that these decreased staffing levels, at times, strained their ability to carry out their mission. According to SBA data on staff, by 2009, 34 percent of staff will be eligible to retire in fiscal year 2009, and according to an SBA official, half of those are in the district offices. However, the views on whether office staffing levels were adequate enough to meet district office goals and other responsibilities were mixed. Some district directors believed that they had enough staff resources to carry out their responsibilities, while others believed they could use additional employees. SBA uses a field staffing model that is based on factors such as market size and the number of 8(a) firms in a district to staff district offices. District directors in several offices appreciated the greater transparency the current model provided over the old one, in which a board met in closed sessions to make staffing decisions. However, some district directors believed that the model was not fully in tune with their resource needs. For example, several district directors said that while they had some flexibility in assigning specific responsibilities to staff to meet local market needs, they could not control the actual staffing process or ensure that positions were staffed effectively. In several locations, employees we interviewed noted that the agency did not have a succession plan for the district offices. As a result of operating with fewer staff, individual employees at district offices indicated that their workloads had increased and that they often found themselves taking on multiple responsibilities for which they did not always receive training because there were not enough individuals available to do any one function. In addition, fewer individuals were available to travel to fulfill the office’s marketing and outreach responsibilities in some of the geographically larger districts or single- district states. Since our 2003 report, SBA has completed the centralization of its loan functions. Based on district office and lender input, the transparency and communication surrounding SBA’s centralization activities had improved. This was evident in the most recent centralization of 7(a) loan guaranty processing for standard 7(a) loans and ongoing reforms at the National Guaranty Purchase Center (NGPC), the first centralization that took place under the transformation. SBA is now focused on re-engineering the NGPC to streamline and standardize processes and procedures. It also established a campaign to improve SBA’s image among lenders by decreasing turnaround times on purchase requests and improving overall customer service. SBA has established process-oriented measures to improve the timeliness of purchase reviews, as well as measures to assess the completeness of files that the centers receive from lenders. SBA has been developing a new quality assurance review process to assess the quality of its purchase decisions and ensure that it is following its own policies and procedures. However, the SBA OIG has found in past audits that inadequate purchase reviews by SBA employees have resulted in improper payments and continues to express concern about the quality of SBA purchase reviews. Moreover, though SBA has measures in place to track the timeliness of its purchase decisions, it has not yet developed performance measures to track quality assurance that would help to ensure that the focus on process efficiency does not come at the expense of the quality of the purchase reviews. SBA’s centralization efforts under transformation began with the centralization of the 7(a) loan guaranty purchase and liquidation functions at the NGPC in Herndon, Virginia, in 2003 to 2004. SBA’s centralization of 7(a) loan processing for standard 7(a) loan guaranty applications starting in September 2006 in Citrus Heights, California, completed SBA’s centralization of its loan functions. The circumstances at the time of the NGPC and 7(a) loan guaranty processing centralizations, including budgetary pressures and leadership, were dissimilar in some respects. However, based on our meetings with SBA and review of documentation, it appears that SBA’s standard 7(a) loan guaranty processing centralization was better planned and implemented than the NGPC centralization. First, there was greater transparency and communication about the 7(a) loan guaranty processing centralization across the agency. SBA sent all employees a notice of this centralization, as well as an implementation schedule. In contrast, district office staff in some offices we visited indicated that there had been limited communication regarding the implementation of the NGPC centralization. Second, the 7(a) loan guaranty processing centralization took place in phases—with the transfer of loan files from district offices to the center occurring between December 2006 and March 2007—unlike the NGPC centralization, in which all district offices sent their files to NGPC at once. Third, SBA appears to have based staffing levels for the 7(a) loan guaranty processing center on workload analyses and did not use directed reassignments to staff the center, as it did with the NGPC, which helped mitigate negative employee morale issues. Both lenders and district office employees have indicated that they have had positive experiences with the Standard 7(a) Loan Guaranty Processing Center. While the initial implementation of the NGPC lacked transparency, communication regarding the activities at the NGPC—including the ongoing re-engineering reforms (to be discussed later in this section)—has improved since the center was first formed. The Office of Financial Assistance, which manages SBA’s loan programs, holds monthly conference calls with employees across the agency to discuss the NGPC re-engineering efforts, participates in Office of Field Operations conference calls, briefs field managers at agencywide management conferences, and reports weekly to the Administrator on the center’s progress. The NGPC also has a Campaign Advisory Board that includes district office and regional representatives to provide input and feedback on the re-engineering efforts. For example, according to SBA, the field representatives on the Campaign Advisory Board suggested the customer support role, in which employees of district offices now help lenders prepare their guaranty purchase applications. While some district office employees and lenders continue to experience communication issues with NGPC in terms of responsiveness, they indicated that overall communication with NGPC has improved. Since SBA completed its centralization efforts, the agency has shifted its focus to re-engineering the operations of its centers, beginning with the NGPC. Through its re-engineering efforts—which have been designed to enhance workflow, redesign processes, and streamline procedures—SBA has taken actions to improve many of the problems that occurred as a result of the implementation of the NGPC centralization in 2004, such as the backlogs of files and inadequate staffing levels for loan specialist positions, which delayed the NGPC decisions on lender’s requests for guaranty purchases on defaulted loans. One key aspect of the re- engineering effort is the Brand Promise Restored (BPR) Campaign, which seeks to restore SBA’s reputation with lenders who have been frustrated by these delays and a lack of responsiveness on the part of NGPC staff to their concerns. The BPR’s key objectives, based on the recommendations of an outside consulting firm and input from various stakeholders within SBA, including district offices, are to reduce the center’s backlog of more than 4,000 loans; improve purchase review and decision turnaround times; and reduce lender errors in submitting applications (thereby improving the completeness of loan packages so that files are not sent back to lenders due to errors); and improve customer service. Furthermore, SBA has established performance measures to assess the progress of its re-engineering effort at the NGPC. In a weekly report to the Administrator and senior management officials, the NGPC provides information on measures such as reductions in backlogs, production targets for new purchase requests, and turnaround times. The NGPC also tracks measures related to district office assistance to lenders in preparing purchase requests and improving the completeness of files that the center receives, including the rate at which purchase applications can be reviewed without having to be sent back to lenders for revisions. SBA said that its other loan centers have similar performance measures, but the performance measures may differ in certain respects due to differing functions. For example, the Standard 7(a) Loan Guaranty Processing Center has a separate set of measures that are similar, including cycle times for completing processing functions and monthly production outputs, but that have different targets, such as a processing turnaround goal of 6 days rather than 45 days for the NGPC. Finally, because reforming NGPC was such a key priority within SBA, the center established weekly milestones that included steps such as pilot testing the new guaranty purchase manual, which were reported to the Administrator. The SBA OIG indicated that the NGPC has improved the efficiency of its guaranty purchase process as a result of its re-engineering and BPR effort. In its 2007 audit of the guaranty purchase process, the OIG made six recommendations, including that the center develop a plan to improve the quality of the purchase reviews and ensure adequate supervisory review and require loan officers to document their purchase decisions. According to the OIG, SBA has closed all but one recommendation, which relates to the recovery of certain improper purchase payments. Moreover, SBA indicated that it had increased the staffing levels for the NGPC from about 40 to about 100 (including contractors) based on a staffing model the agency developed to address the OIG’s recommendation that the agency fully staff the NGPC at the appropriate level. Finally, SBA indicated that is has cleared the backlog or pre-purchase reviews and most of its backlog of secondary market purchases since the BPR campaign began. While SBA has established process-oriented measures, it is still developing its quality assurance process and has not established measures management routinely track to assess the effectiveness of its guaranty purchase process and resulting decisions. The measures the center does use relate to the front-end completeness of files lenders submit to the NGPC, rather than the quality of purchase decisions. SBA officials told us they have a quality assurance initiative under way that should address some of the deficiencies in their quality assurance process but that this effort is still in its initial stages. The NGPC currently controls for quality by using the “rule of 3,” in which the recommendation for each purchase decision made by a loan specialist is reviewed by an “approver” and legal counsel before it is finalized. To encourage quality purchase reviews, SBA officials said its performance review process for loan specialists includes standards that measure how well loan specialists conduct reviews, with respect to accuracy, for example, by tracking how often their recommendations have to be substantially revised. The center also uses a “triage process” to ensure that the most complex cases are handled by the most experienced loan specialists. According to SBA, the goal of its quality assurance review initiative is to create a system that routinely examines all parts of the loan process to identify where in the loan process most errors occur. SBA officials also said they hope to remove the variability out of the purchase process, though there will always be some level of employee judgment involved. SBA officials said that its efforts under the Improper Payments Information Act of 2002 established a baseline on its error rate for improper payments for 7(a) guaranty purchases and that this information is used to measure and improve the quality of guaranty purchase reviews. While the OIG has noted improvements in the NGPC’s operations, it also has expressed to us ongoing concerns about the quality of purchase decisions at the NGPC, particularly with respect to the incidence of improper purchase payments. SBA’s review of lender requests for guaranty purchases on defaulted loans is an important tool for assessing lender compliance on individual loans and protecting SBA from making erroneous purchase payments. According to the OIG’s 2008 management challenges report, past OIG audits of early defaulted loans and SBA’s guaranty purchase process have shown that SBA has not consistently been able to detect lenders’ failures to administer SBA loans in compliance with SBA requirements and prudent lending practices, resulting in improper payments. For example, in a May 2007 audit of the guaranty purchase process, the OIG estimated that the rate of improper payments for all 7(a) loans that received a purchase review at the NGPC from October 1, 2004, through May 31, 2005, could be 17 percent (approximately $36 million). This estimated rate is substantially higher than the 1.56 percent that SBA reported for the 7(a) program in fiscal year 2006. SBA disagreed with the OIG’s estimate because it covered a different time period and because it did not believe that the sample used by the OIG was representative of SBA’s entire 7(a) loan portfolio. In general, both estimates are difficult to compare because they covered different time periods and employed different sampling methods. Additionally, neither estimate incorporates reviews of purchase decisions completed under the recently re-engineered guaranty purchase process. The OIG has an ongoing audit of SBA’s compliance with the Improper Payment Information Act that it expects to complete in February 2009 and continues to express concern that SBA may underreport its improper payment rate. In 2006, the OIG surveyed the NGPC’s existing quality assurance review process, which was designed to provide SBA with valid reviews, independent of the purchase approval process, to ensure the proper interpretation and consistent implementation of policies and procedures and consistent and accurate guaranty purchase decisions, thereby minimizing losses from improper payments. The OIG found that under SBA’s quality assurance review process at the time, SBA reviewed a number of loan purchase decisions, that may have been sufficient to assess compliance with the Improper Payments Information Act of 2002 but did not adequately cover the high- risk loan categories identified in their former quality assurance review plan. As a result, the OIG found that the NGPC was unable to ensure with any degree of reliability that guaranty purchase decisions made by the NGPC staff were complete, consistent, and accurate. We also previously found that this type of management challenge could increase the risk of improper payments. The NGPC does not have this quality assurance review process at this time because of SBA’s ongoing effort to replace and improve its quality assurance review process for all of its centers. SBA has efforts under way to improve its quality assurance review process but without measures focused on the quality of its purchase review process that management tracks regularly, SBA will not be able to assess whether its re-engineered process is also successful in producing appropriate purchase decisions that are in compliance with laws and regulations. According to the Government Performance and Results Act of 1993, agencies should develop a range of performance indicators, such as quantity, quality, timeliness, cost, and outcome, to help managers balance their priorities among several agency goals. Moreover, the creation and review of performance indicators help strengthen an agency’s internal control standards, which help to ensure both the effectiveness and efficiency of an agency’s operations. Reliance on any single type of measure could create a perverse incentive for managers to achieve one subgoal at the expense of the others. For example, without a measure that reflects the quality assurance, managers and employees at the NGPC could focus more on the turnaround time measures, and consequently, the quality of the purchase reviews could suffer leading to improper payments. SBA’s earlier efforts to restructure or transform agency operations have had some associated adverse effects on SBA employee morale and operations. However, in the past 2 years SBA has made progress in addressing these negative effects—by applying key practices that support successful transformations to improve agency operations and taking actions to address the recommendations we made in our 2003 report. The Administrator appointed in 2006 emphasized the importance of transparency in his reform agenda, made a concerted effort to improve communications across the agency, and engaged employees in improving the agency. Our meetings with employees affirmed that these efforts had a positive effect on many employees. We also found that though SBA had also taken some actions to obtain feedback and ideas from employees, many employees we interviewed, including union officials, continued to feel that SBA’s management did not sufficiently consider their ideas and concerns. SBA recently conducted focus groups to better understand employee concerns, specifically explored the concerns about employee involvement, and plans to implement pilot initiatives to address them. Such actions are important because a feeling of lack of involvement can affect employee productivity and job satisfaction and could hinder SBA’s ability to meet goals and improve its operations. We also note that many employees were concerned about the continuation of key efforts undertaken under Administrator Preston. Senior SBA career officials stated that continuing positive efforts was important to the agency and have taken steps to identify such efforts and institutionalize them. Such long-term commitments by senior leaders are important for ensuring that transformation and agency reforms are successful. SBA’s 2007 employee survey results suggest that recent actions, such as improving communication and training, have had a positive impact on employees, but also point to continuing concerns about these issues. SBA has recognized the need to provide better training—another area of continued concern for employees—to support employees whose roles and responsibilities changed as a result of transformation. Specifically, the creation of SBA University in 2007 was an important action not only in terms of providing training and demonstrating that the agency was willing to commit resources to invest in the development of employees. SBA officials also have been developing a training program that includes core curriculums for different staff positions. However, SBA has not developed a strategic training plan that lays out goals, strategies, and milestones. In an environment of constrained budgets and reduced staff resources, training is a key means to maximize or leverage the talents and skills of employees. A strategic plan would help to establish priorities and determine the best ways to leverage training investments to improve performance. In addition, a plan that is shared with employees could assure them that SBA continues to be committed to developing its employees. The most dramatic effects of transformation occurred in SBA’s district offices. In some ways, transformation still is occurring in the district offices as their roles and responsibilities continue to evolve. For instance, SBA recently determined that district offices still could have a role in the guaranty purchase review process, despite the centralization of this function. But the recent changes have introduced other concerns. In particular, district office employees continue to experience some frustration as they adjust to their new responsibilities and work toward achieving their goals with fewer resources. As SBA continues to adjust priorities and operations to achieve its goals, employing key practices relating to communications and employee involvement will position the agency in good stead in this area, as well. The centralization of the remaining loan functions previously performed by district offices was another key aspect of SBA’s transformation and those initiatives have been completed. But the functioning of the centers— particularly the NGPC—represents a work in progress. SBA recently has re-engineered the guaranty purchase process and center operations and developed measures to track progress. The emphasis has been on improving the efficiency of the guaranty process because of the need to reduce significant backlogs and the desire to provide good customer service to lenders. SBA said that it has been developing a new quality assurance review process and that its performance reviews of loan specialists also rates them on the accuracy of their reviews, but it has not yet developed any performance measures that management uses regularly to track the quality of its purchase reviews. However, performance measures focused on timeliness and volume, without a similar focus on quality in the purchase review process, could provide the wrong incentives for managers and employees and ultimately could result in improper payments and increased costs to the program. The OIG also has noted concerns about SBA’s purchase reviews and believes that SBA’s improper payment rate for its purchased guaranties may be much higher than what SBA currently reports. By focusing on the quality of purchase reviews and developing measures that track the quality of purchase decisions, SBA could help better ensure that the process produces decisions that are well supported, defensible, and balance the needs of lenders with the need to enforce requirements that protect the integrity of the loan guaranty process and prevent improper payments. We are making two recommendations to the Administrator of SBA: To support SBA’s development of a core curriculum for SBA University and communicate SBA’s training development efforts to employees, SBA should develop a strategic training plan with specific goals, strategies, and milestones. To ensure that its streamlined purchase guaranty process also provides an incentive to focus on quality and thereby results in good purchase decisions, SBA should develop performance measures that will provide information on the quality of its guaranty purchase reviews. We provided a draft of this report to SBA for review and comment. In written comments (see app. II), SBA generally agreed with the report and recommendations. SBA stated that the report recognized the agency’s efforts and initiatives leading to improvements in agency operations and employee morale and that it is committed to continuing to build upon the success achieved. In addition to agreeing to our recommendation to develop measures that focus on the quality of purchase reviews, SBA stated that the NGPC will continue to build on the quality processes in place and is developing new quality assurance review plans for NGPC and other processing centers that will be implemented during fiscal year 2009. Specifically, SBA provided additional information on how it measures quality in its processes for reviewing, recommending, and approving guaranty purchases. For example, NGPC tracks the number of work items of its loan specialists returned for substantial revision and randomly samples processed guaranty purchase disbursements to test them for accuracy. SBA stated that this information helps to identify areas where training is needed and ultimately contributes to improving the quality of its reviews. Though these actions were not reflected in the NGPC weekly management reports provided to us and therefore it is unclear to what extent these measures are regularly tracked by SBA management, we agree that these actions help SBA improve the quality of its purchase reviews. Additionally, SBA stated it will continue to ensure employees have access to the training and tools provided through the development of SBA University and has begun developing a draft plan for future training. SBA also provided technical comments that we incorporated as appropriate. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 dyas from the report date. At that time, we will send copies of this report to the Ranking Member, Senate Committee on Small Business and Entrepreneurship; the Chair and Ranking Member, House Small Business Committee; other interested congressional committees; and the Administrator of SBA. We will make copies available to others upon request. In addition, the report will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff have any questions concerning this report, please contact me at (202) 512-8678 or shearw@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix III. Our objectives were to review the status of the Small Business Administration’s (SBA) transformation efforts and related initiatives since our 2003 report. Specifically, we focused on (1) SBA’s progress in transforming agency operations and addressing GAO’s recommendations on key practices for successful transformation, (2) how SBA has addressed the impacts of transformation on its employees, (3) how SBA has defined the roles and responsibilities of its district offices, and (4) SBA’s efforts to assess its initiatives for centralization of its loan functions and how its approach for implementing centralization has changed since transformation began. To respond to these objectives overall, we reviewed agency planning and implementation documents related to SBA’s transformation and interviewed key officials at SBA headquarters in Washington, D.C., to obtain information about their roles and experiences with SBA transformation initiatives. SBA officials we met with included the SBA Acting Administrator (formerly Deputy Administrator), Chief of Staff, and senior officials representing the Office of Human Capital Management, the Office of Field Operations, the Office of Capital Access, and the Performance Management Office. In conducting our work, we focused on the transformation process itself and its impacts, rather than on making assessments of SBA’s administration of its various programs. To obtain documentary and testimonial information about district office involvement and experiences with various SBA transformation initiatives, we interviewed district directors and groups of employees from 10 of SBA’s 68 district offices. To provide national coverage, one district office was selected from each of SBA’s 10 geographical regions. These district offices were located in California, Georgia, Indiana, Iowa, Montana, New Jersey, Oregon, Texas, Vermont, and Virginia. In addition, we conducted site visits to the National Guaranty Purchase Center in Herndon, Virginia, and the Standard 7(a) Loan Processing Center in Citrus Heights, California, to observe and collect information and experiences about SBA’s centralization initiatives. We visited the National Guaranty Purchase Center because it was the first centralization effort completed under transformation and the Standard 7(a) Loan Processing Center because it was SBA’s most recent (and last) centralization completed under transformation to be able to compare and contrast SBA’s approaches to the two efforts. During both our district office and loan center site visits we held two meetings—-one with the offices’ directors followed by a group interview with office employees. All office employees were eligible to participate in the meetings. We requested that the employee meeting be open to all staff based on availability and also that the office director could recommend staff for participation. We further requested that the group interview with employees be conducted without the presence of management. We also met with officials from SBA’s Office of Inspector General (OIG) to discuss any audit results or major challenges related to the agency’s transformation initiatives. Finally, we discussed the effect of transformation on lenders and staff with, respectively, representatives of a lenders’ organization and SBA union officials. Questions prepared for the interviews, as well as the information obtained from them, were used to address issues related to each of our reporting objectives. To determine SBA’s progress on its transformation initiatives since 2002, its actions to implement key practices that are important to successful organizational transformation, and how the agency’s emphasis evolved over time, we reviewed and analyzed past GAO reports, SBA’s responses to GAO report recommendations, agency documents, and SBA OIG reports. In particular, we looked at differences and similarities between the new Administrator’s reform agenda and the prior Administrator’s transformation initiatives. In addition to examining district offices’ involvement and implementation of transformation activities, we reviewed the approaches SBA took on its transformation, management’s leadership of transformation efforts, positive and negative consequences of transformation, internal coordination, communication, performance management, and training. We reviewed documentation such as SBA rosters identifying the transformation leadership and implementation team members and their specific responsibilities; relevant SBA transformation policy and procedures; examples of communication modes used (type and frequency); and employee surveys. In particular, we reviewed agency documents and interviewed SBA management officials and employees to determine whether SBA has implemented the recommendations made in GAO’s October 2003 transformation report noted above, which was based on best practices for mergers and transformation identified by GAO. In addition, we obtained and analyzed SBA documents related to its new performance management system; current and previous scorecards that track the performance of agency components relative to the specific goals on which they are measured; operating plans that lay out strategies to achieve goals; information illustrating lines of communication and areas covered (for example, press, marketing, Web development, speechwriting, strategic alliances with external parties such as trade groups, and multimedia); and the agency’s communications plan that discusses its approach for addressing internal and external communication, along with descriptions of the responsibilities of its Office of Communication and Public Liaison, Office of Management and Administration, the Office of the Congressional Liaison, and the intergovernmental group. We analyzed the information we obtained from these document reviews and interviews to determine what actions SBA has taken to incorporate key practices and implementation steps for organizational transformations, whether SBA’s transformation laid out goals for the transformation, what communication strategies SBA used to promote frequent two-way communication at all levels of the agency and with stakeholders, what actions SBA has taken to involve employees and ensure that their concerns and ideas are considered, how SBA has aligned its transformation activities to achieve agency performance and strategic goals, and how SBA’s performance management system linked overall agency goals to individual goals. To assess how SBA has addressed the impacts of its transformation activities on employees, we discussed this matter in interviews with officials from the various headquarters officers, the district office directors, and groups of employees at the 10 district offices we selected. The number of participants in our employee group interviews ranged from 5 to 12 (and represented from about 21 percent to 100 percent of the employees assigned to the various locations). While the limited number of offices we visited is too small for generalizing the information obtained to all district offices, the recurring nature of the observations and perspectives expressed across these offices, along with other material we examined, suggest that these views were not limited to the offices we visited. To distinguish specific activities affecting employees as a result of transformation, we reviewed materials that included SBA’s fiscal year 2007 Human Capital Plan, Information and Technology Strategic Plan, fiscal year 2007 Annual Performance Report, fiscal year 2009 congressional submission (for budget), the performance management framework, and the SBA OIG’s Most Serious Management Issues Reports for fiscal years 2005, 2006, and 2007. We also analyzed results from the Office of Personnel Management’s (OPM) 2004 and 2006 Federal Human Capital Surveys of SBA staff and SBA’s 2007 Annual Employee Survey to help identify specific impacts of transformation on employees and the effect of actions SBA had taken to address those impacts. In particular, we examined such areas as employee morale, training, communication, and employee involvement in decisionmaking—to the extent they were relevant to our prior 2003 report’s recommendations. To assess the reliability of the OPM survey data and SBA’s 2007 survey, we discussed the conduct of these surveys with knowledgeable OPM officials. We also obtained a copy of SBA’s 2007 survey data set and replicated selected survey results. We determined that these survey data were sufficiently reliable for purposes of this report. For determining how and to what extent SBA has defined the roles and responsibilities of district offices, we used information from the interviews we conducted. We also obtained and reviewed documentation related to the roles and responsibilities of district office staff before and after transformation and reform agenda activities. This included material on job descriptions and district office staffing information, such as the Field Staffing Model and office structure by position. We also examined relevant parts of training materials that related to district office responsibilities; implementation plans specifying what individual staff responsibilities should be; SBA’s district office scorecard and other performance management materials that discuss the functions and roles on which district offices and staff are evaluated; strategic plan, human capital plans, and transformation plans; SBA memorandums, policies, and procedures related to performance management assessments and measures. To determine how SBA has assessed the progress of its centralization efforts and how its approach for implementing centralization has changed since transformation began, we used information from our interviews with officials from SBA headquarters and our site visits. We also obtained and analyzed documentation from SBA regarding its centralization efforts, including decision memorandums and implementation schedules, centralization performance and progress reports, documentation on specific initiatives related to centralization. We also relied on recent OIG audit reports and its ongoing audit work addressing both the progress and deficiencies of SBA’s centralization initiatives. In particular, we met with representatives from that office to discuss their audit results, including recommendations they made to the agency, as well as management challenges faced by SBA. We also compared SBA’s efforts to assess the progress of its centralization efforts against Government Performance and Results Act guidance and GAO’s Standards for Internal Control in the Federal Government. We conducted this performance audit from August 2007 to September 2008 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. In addition to the contact named above, Kay Kuhlman (Assistant Director), Meghana Acharya, Johnnie Barnes, Thomas Beall, Bernice Benta, Tania Calhoun, Alexandra Martin-Arseneau, Marc Molino, Barbara Roesmann, and Sarah Veale made key contributions to this report. | Over the past 6 years, the Small Business Administration (SBA) has sought to transform the agency and improve its operations. A major focus of transformation was to centralize the remaining loan functions performed by 68 district offices. SBA's implementation of early transformation efforts did not reflect key practices GAO recommended in a 2003 report as important for successful transformations. Consequently, its centralization of the guaranty purchase process for one of its loan programs resulted in backlogs and other problems reported by SBA's Inspector General. Some of SBA's actions also led to a reduction in staff at district offices and a decline in employee morale. GAO was asked to assess how SBA has (1) responded to GAO's 2003 recommendations, (2) addressed the impacts of transformation on employees, (3) defined the roles and responsibilities of district offices, and (4) assessed the centralization of loan functions. GAO reviewed documents related to SBA's transformation and reform efforts, interviewed SBA officials, and analyzed SBA employee survey data. GAO also visited 10 district offices and two centers and interviewed groups of employees. In the past 2 years, SBA has applied key practices that support successful transformations to improve agency operations and, thereby, has taken actions to address recommendations GAO made in its 2003 report, including improving communication, performance management, and employee involvement. The Administrator appointed in 2006 emphasized the importance of transparency in his reform agenda and took actions to improve communications across the agency. The development of a performance management framework was a key step in linking the agency's reforms with strategic goals and employee roles. The Administrator also made a concerted effort to engage SBA's employees in improving the agency, and meetings with employees affirmed that these efforts had a positive effect. Some employees continued to feel that management does not consider their ideas and concerns. SBA recently conducted focus groups to understand these concerns and plans to implement initiatives to address them. SBA senior officials also said that they are taking steps to institutionalize these improvements. SBA leadership's commitment will be important to ensure that the agency's transformation and reforms are successful. SBA took some actions to address its low employee morale, which had declined significantly following the centralization efforts, as shown in the 2004 and 2006 Federal Human Capital Surveys. SBA's 2007 survey results suggest that these recent actions, such as improving communication and training, have had a positive impact on employees. The creation of SBA University in 2007 was an important action since it provided training and also showed that the agency was willing to invest resources in the development of employees. SBA officials said they are developing a core training program. However, SBA has not developed a training plan that lays out goals, strategies, and milestones. Such a plan would help to establish priorities and could assure employees that SBA remains committed to developing its employees. SBA continues to define the roles and responsibilities of the district offices, as evidenced by its recent determination that district offices should retain a role in loan processes that have been centralized. District directors and employees made positive comments about the flexibility they had in using resources to meet office goals. But they also said that they still were adjusting to new responsibilities and the reduction of staff in their offices. SBA recently re-engineered its guaranty purchase process. Its measures to track progress have emphasized the timeliness of the process, completeness of packages lenders submit, and customer service. SBA reviews each purchase decision and is developing a new quality assurance review process, but has not yet developed performance measures to track the quality of its purchase reviews. SBA Inspector General audits have noted concerns about the quality of purchase reviews and found that the center's purchase reviews do not adequately prevent improper payments. Performance measures could provide more attention to the quality of reviews. |
As we reported in 2000, commercial activities in school can generally be classified in four categories—product sales, direct advertising, indirect advertising, and market research—although each category encompasses a wide range of activities. For example, advertising activities could range from selling advertisements for a high school football game to selling naming rights to a school. Although this report synthesizes statutes, regulations, and proposed legislation addressing all four categories, our discussions of school district policies and Education’s activities focus on the fourth category, market research, because of the amendments made by NCLBA that place requirements on districts that deal with the collection, disclosure, and use of student data for marketing and selling. (See table 1.) In recent years, the growth of the Internet has had a large impact on commercial activities, particularly market research, by enabling marketers to elicit aggregated and personally identifiable information directly from large numbers of students. For example, some Web filtering systems used in schools that block student access to certain Web sites also allow the company that maintains that software to measure and analyze how children use the Internet by tracking which Web sites they visit and how long they stay there. Although this information is aggregated and does not identify particular children, this information, especially when used with demographic data, can help businesses develop advertising plans that target particular audiences if districts allow the installation of the software. Also, Web sites directly elicit the participation of students in market research panels by offering them cash or prizes in exchange for information about themselves and their preferences. This makes it possible for companies to engage large-scale customized panels of students to test out marketing strategies and provide data to develop product lines and product loyalty without relying on schools. NCLBA addresses some concerns about commercial activities and student data by amending and expanding certain student data safeguards that were established in PPRA. Prior to NCLBA, PPRA generally prohibited requiring students to submit to a survey concerning certain personal issues without prior written parental consent. As amended, PPRA for the first time requires districts to develop and adopt new policies, in consultation with parents, for collecting, disclosing, and using student data for marketing or selling purposes. Districts are also required to directly notify parents of these policies and provide parents an opportunity to opt their child out of participation in such activities. Furthermore, districts are required to notify parents of specific activities involving the collection, disclosure, and use of student information for marketing or selling purposes and to provide parents with an opportunity to review the collection instruments. PPRA did not contain deadlines for districts to develop policies. Also, PPRA requires Education to annually inform each state education agency and local school districts of their new obligations under PPRA. Finally, PPRA continues to require Education to investigate, process, and adjudicate violations of the section. For the past 30 years, student and parent privacy rights related to students’ education records have been protected primarily under the Family Educational Rights and Privacy Act (FERPA), which was passed in 1974. FERPA protects the privacy of students’ education records by generally requiring written permission from parents before records are released. FERPA also allows districts to classify categories of information as publicly releasable directory information so long as the district has provided public notice of what will constitute directory information items and has allowed parents a reasonable period of time to advise the district that directory information pertaining to their child cannot be released without consent. Under FERPA, directory information may include a student’s name, telephone number, place and date of birth, honors and awards, and athletic statistics. Unlike PPRA, FERPA does not address the participation of students in surveys or the collection, disclosure, or use of student data for marketing or selling purposes. (See table 2.) As a result of the NCLBA amendments, Education is required to annually inform each state and local education agency of the educational agency’s obligations on PPRA and FERPA. State education laws are enacted by state legislatures and administered by each state’s department of education, which is led by the state’s chief state school officer. The Council of Chief State School Officers represents states’ education interests in Washington, D.C., and acts as a conduit of information between the federal government and the states regarding federal education laws. Each state department of education provides guidance and regulations on state education laws to each school district. School district policies are generally set by local school boards according to the authority granted to them by state legislatures. The policies are then administered by the school district’s superintendent and other school district staff. Local school boards in each state have come together to form a state school boards association. They provide a variety of services to their members including help on keeping their local school board policies current. For example, a partial list of services offered by one school board association includes policy development services, advocacy, legislative updates, legal services, executive search services, conferences and training, and business and risk management services. Since 2000, 13 states have established statutes, regulations, or both that address one or several categories of commercial activities in schools. Six of these states established provisions addressing market research by restricting the use of student data for commercial activities and for surveys. Other states passed statutes or issued regulations addressing product sales and advertising. In addition, as of February 2004, at least 25 states are considering proposed legislation that would affect commercial activities. Most of these proposals would affect product sales, particularly the sale of food and beverages. Prior to 2000, 28 states had passed provisions addressing commercial activities. At that time, most provisions addressed direct advertising and product sales. The seven districts we visited in 2000 continued to conduct a variety of commercial activities, particularly product sales, and three districts reported that they have increased the level of activities with local businesses. However, the types of activities in these districts have not substantially changed since our visit. Since our previous report in 2000, 13 states have enacted 15 statutory provisions and issued 3 regulatory provisions addressing one or more types of commercial activities in schools. Six states passed legislation affecting marketing research. Three of these 6 passed laws restricting the disclosure or use of student data for commercial purposes, and another 3 placed restrictions on students’ participation in surveys. For example, an Illinois statute prohibited the disclosure of student data to businesses issuing credit or debit cards, and a New Mexico regulation prohibited the sale of student data for commercial reasons without the consent of the student’s parent. Laws in Arizona, Arkansas, and Colorado prohibited student participation in surveys without the consent of their parents. Five states passed new provisions affecting product sales. In most cases, these laws targeted the sale of soft drinks and snack food. Other new provisions addressed direct and indirect advertising. Prior to 2000, 28 states had established one or more statutes or regulations that affected commercial activities in schools. Twenty-five states established provisions addressing advertising—in 19 states, measures affected direct advertising and in 6, indirect advertising. Sixteen states established provisions addressing product sales. Only 1 state established a measure that addressed market research. See appendix II for a state-by- state listing of provisions addressing commercial activities. Legislatures in 25 states have recently considered one or more bills that affect commercial activities in schools, with most having a particular focus on child nutrition. These bills are intended to improve child nutrition and reduce obesity, and to achieve this intention, place limitations, restrictions, or disincentives on the sale of beverages and food of limited nutritional value. Legislatures in 24 of the 25 states recently considered bills that restrict or ban the sale of beverages and food of limited nutritional value in schools. For example, a bill in New York would prohibit vending machines from selling food and drinks of minimal nutritional value. Additionally, legislatures in several states have considered bills that restrict the hours when students can buy products of limited nutritional value. For example, bills in Alaska and Ohio would restrict the sale of soft drinks during certain hours. Finally, pending legislation in Maryland would require schools to sell food of limited nutritional value at higher prices than nutritious food. Legislatures in seven states have recently proposed bills that focus on other aspects of commercial activities in schools. In three states— Connecticut, Minnesota, and North Carolina—bills would restrict the ability of schools to enter into exclusive contracts with beverage and food venders. In two statesNew Jersey and North Carolinabills would place limits on the ability of schools to release or collect personal information about students, such as prohibiting the release of data from the student- testing program to any marketing organization without the written permission of the parent or guardian. Other proposed bills addressed a variety of issues, such as allowing schools to sell advertising and accept supplies bearing logos or other corporate images or requiring school boards to disclose the portion of proceeds from fundraising activities that is contributed to the school activity fund. See appendix III for a state-by- state listing of legislative proposals. In updating the site visit information we collected in 2000, we found only slight changes in commercial activities in all seven school districts. All districts reported they continued to engage in product sales and display advertising. As we found earlier, most commercial activities, particularly product sales and advertising, occurred in high schools. All the high schools we visited in 2000 still sold soft drinks, and most sold snack or fast food. To varying degrees, all displayed corporate advertising. High schools continued to report the receipt of unsolicited samples, such as toiletries, gum, razors, and candy, that they did not distribute to students. In contrast, the elementary schools we contacted did not sell carbonated soft drinks to students or display corporate advertising. Grocery and department store rebate programs continued to operate in almost all schools, but coupon redemption programs were largely an elementary school enterprise. As we found before, none of the districts reported using corporate-sponsored educational materials or engaging in market research for commercial purposes. Officials did report some changes in commercial activities. Three of these districts reported stronger ties with local businesses, and three schools in two districts reported they now sell healthier soft drinks. One district reported a new relationship with a computer firm headquartered in its area that provided tutors as well as cash donations to schools in the district. Under this relationship, company employees tutored students who were at risk of failing, and the company donated $20 to schools for each 10 hours of tutoring that its employees provided. A principal in this district reported that many students in her school benefit substantially from this relationship and her school earned between $6,000 and $9,000 per year in donations. Another district reported it had entered into a new contract with a local advertising agency to raise revenue to renovate sport concession stands, and a third had organized a new effort to sell advertisements to fund construction on the district’s baseball field. Three principals told us that vending machines in their schools now offer a different mix of beverages—for example, more juice, milk, and water and fewer carbonated beverages—than they did when we visited in 2000. We estimate that about two-thirds of the districts in the nation were either developing or had developed policies addressing the new provisions on the use of student data for commercial purposes. However, only 19 of the 61 districts that provided us copies of their policies specifically addressed these provisions. Very few school districts reported releasing student data for marketing and selling, and all these releases were for student-related purposes. Of the seven districts we visited in 2000, three adopted new policies on the use of student data since our visit, and only one released data and that was for graduation pictures. Although districts reported they had developed policies, many of the policies they sent us did not fully address PPRA requirements. On the basis of the results of our surveys, we estimate about a third of districts were developing policies regarding the use of student data for commercial purposes; another third had developed policies; and about another third had not yet developed policies. However, when we analyzed policies that 61 districts sent to us, we found only 19 had policies that specifically addressed marketing and selling of student information. Of these, 11 policies addressed the collection, release, and use of student information for commercial purposes. Eight policies partially addressed the provisions by prohibiting the release of student data for these purposes. Policies in the 42 remaining districts did not address the new PPRA provisions. Many of these districts provided us policies concerning FERPA requirements. We telephoned all districts in our sample that reported they release data for commercial purposes and a subsample of districts that reported they had not. Of the 17 districts that released data for commercial purposes, all reported that they released data only for school-related purposes. For example, all 17 released students’ names to photographers for graduation or class pictures. Two of these districts also released student data to vendors who supplied graduation announcements, class rings, and other graduation-related products, and another two districts released student information to parent-teacher organization officials who produced school directories that they sold to students’ parents. Of the 16 districts that reported they did not release student data, one actually did release student data. As in the other cases, that district released it to a school photographer. Of the seven districts we visited in 2000, three adopted new policies on the use of student data. One of the districts we visited adopted new policies that incorporated PPRA provisions on the use of student data for commercial purposes. Two adopted policies with blanket prohibitions against some uses of student data for marketing and selling. In one of these districts, policies prohibited the release of students’ data for any survey, marketing activity, or solicitation, and policies in the other banned the use of students to support any commercial activity. Officials in all seven districts reported that their district did not collect student data for marketing or selling purposes, and several expressed surprise or disbelief that this practice did in fact occur. However, a high school in one district reported that it disclosed information on seniors to vendors selected by the district to sell senior pictures, school rings, and graduation announcements. As required by NCLBA, Education has developed guidance and notified every school district superintendent and chief state school officer in the country of the new required student information protections and policies, and has charged the Family Policy Compliance Office to hear complaints on PPRA. Education issued guidance about the collection, disclosure, and use of student data for commercial purposes as part of its general guidance on FERPA and PPRA in 2003 and 2004. In addition, although not required by statute to do so, Education provided superintendents with model notification information that districts could use to inform parents of their rights, included information about PPRA in some of its training activities, and posted its guidance and other PPRA-related material prominently on its Web site. Education has charged its Family Policy Compliance Office to hear complaints and otherwise help districts implement the new student data requirements. Although the office has received some complaints about other provisions related to student privacy, as of June 2004, officials from that office reported they have received no complaints regarding the commercial uses of student data. Many districts did not appear to understand the new requirements, as shown by our analysis of the 61 policies sent to us by districts in our sample. Although we asked districts to send us their policies that addressed these new provisions, only 11 districts sent policies that addressed these new provisions comprehensively, and 8 sent policies that covered these provisions partially. The 42 remaining districts sent policies that did not contain specific language addressing the collection, release, or use of student data for commercial purposes, although districts sent them to us as documentation that the districts had developed such policies. Most of these policies contained only general prohibitions about the release of student records and concerned FERPA. Although Education is not required to issue its guidance to state school boards associations, four districts in two states in our survey offered unsolicited information that they relied on state school boards associations to develop policies for their consideration and adoption. Two districts in a third state that sent us policies used policies developed by their state school boards association to address commercial activities in schools. However, Education did not distribute its guidance to these associations. Although state laws both limit and support commercial activities in schools, many state legislatures have chosen to pass laws addressing only specific activities such as permitting or restricting advertising on school buses. In addition, many states have not enacted legislation concerning commercial activities or have passed the authority to regulate these activities to local districts, thus allowing district school boards, superintendents, or principals to determine the nature and extent of commercial activities at the local level. Not only do commercial activitiesproduct sales, direct advertising, indirect advertising, and market researchencompass a broad spectrum of activities, but also the levels of these activities and the levels of controversy attached to them vary substantially. For example, few would equate selling advertisements for a high school football program with selling the naming rights to a school, although both are examples of direct advertising. Because of these differences, as well as philosophical differences among districts and communities, it is probably not surprising that states legislatures have taken various approaches toward the regulation of commercial activities. Perhaps because providing student information for commercial purposes may have serious implications, few districts do so. In fact, some school officials said they were skeptical that schools would allow the use of student data for this purpose. In the past, marketers may have approached schools to survey students about commercial products or services. Today, however, technology, particularly the proliferation and availability of the Internet, provides marketers with quick and inexpensive access to very large numbers of children without involving the cooperation of schools. As Internet users, children often submit information about themselves and their personal product preferences in exchange for cash or prizes. Because of the disinclination of school officials to sell student data and the ability of marketers to get data directly from students without involving schools, it may be understandable that relatively few districts as yet have actually adopted policies that specifically address the selling and marketing provisions of PPRA. On the other hand, few would argue against the need to protect students’ personal information. Many businesses, particularly local businesses catering to youth markets, might still profit from acquiring student information from schools. Although we found districts did not use student data for purposes generally viewed as offensive, this does not mean such use would not happen in the future in the absence of safeguards. It appears that some superintendents may not be aware of the new PPRA requirements or have not understood Education’s guidance because many thought their district’s policies reflected the latest federal requirements on use of student data when, in fact, they did not. Also, several districts told us that they relied on state school boards associations to develop policies. Unlike models or guidance that reflect only federal law, policies developed by these groups may be most useful to districts because they correspond to both federal and state requirements. These associations are not on Education’s guidance dissemination list. We recommend that the Secretary of Education take additional action to assist districts in understanding that they are required to have specific policies in place for the collection, disclosure, and use of student information for marketing and selling purposes by disseminating its guidance to state school boards associations. We provided a draft of this report to the Department of Education for review and comment. Education concurred with our recommendation. Education’s comments are reproduced in appendix V. Education also provided technical comments, which were incorporated as appropriate. Unless you publicly announce its contents earlier, we plan no further distribution until 30 days after the date of this letter. At that time, we will send copies of this report to the Secretary of Education, appropriate congressional committees, and others who are interested. We will also make copies available to others upon request. In addition, the report will be available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staff have any questions, or wish to discuss this material further, please call me on (202) 512-7215. Other contacts and staff acknowledgments are listed in appendix VI. To conduct our work, we reviewed state statutes, regulations, and proposed legislation; received mail questionnaires from 219 school districts selected on the basis of a national stratified probability sample design; conducted additional brief telephone interviews with 36 of these districts; analyzed policies voluntarily provided by 61 districts; interviewed officials at Education; and examined guidance issued by the department. In addition, we conducted telephone interviews with district and school officials in the 7 districts that we visited in 2000 for our previous study on commercial activities in schools to update our previous findings. To update our compilation of state statutes and regulation contained in our 2000 report, we researched legal databases, including Westlaw and Lexis, to identify laws passed between January 2000 and May 2004. To identify pending laws, we researched information available on databases maintained by state legislatures or followed links provided by these databases to identify bills introduced between January 2003 and February 2004. However, there are inherent limitations in any global legal search, particularly when—as is the case here—different states use different terms or classifications to refer to commercial activities in schools. We selected a national probability sample of districts, taken from school districts contained in the Department of Education’s Common Core of Data (CCD) Local Education Agency (LEA) file for the 2000-2001 school year. After removing districts from this list that were administered by state or federal authorities, we identified a population of 14,553 school districts. In the course of our study, we learned that some special education and other units in this list do not have legal authority to establish formal policies. As a result, we estimate that our study population consists of 13,866 districts in the 50 states and the District of Columbia. The sample design for the survey consisted of a stratified random probability sample design: 271 districts were drawn from the three strata shown in table 3. The strata were designed to draw relatively large numbers of districts from states likely to include districts that had engaged in or planned to engage in one or more specific activities involving the collection, disclosure, or use of student information for the purposes of marketing or selling or providing information to others for these purposes. Because we thought the activities of interest were low incidence activities, we wanted to maximize our ability to examine situations involving the use of student data for commercial purposes. The expected high-activity strata were defined as states that we identified as having laws that permitted commercial activities when we performed our work in 2000. As shown in table 4, the response rates were 76 percent, 83 percent, and 88 percent in the three sampling strata. The overall estimated response rate was 87 percent. Because we followed a probability procedure based on random selections, our sample is only one of a large number of samples that we might have drawn. Since each sample might have provided different estimates, we express our confidence in the precision of our particular sample’s results as a 95 percent confidence interval (for example, plus or minus 9 percentage points). This is the interval that would contain the actual population value for 95 percent of the samples we could have drawn. As a result, we are 95 percent confident that each of the confidence intervals in this report will include the true values in the study population. The practical difficulties of conducting any survey introduce nonsampling errors. For example, errors could be made in keying questionnaire data, some people may be more likely than others to respond, or questions may be misinterpreted. To minimize data-handling errors, data entry and programs were independently verified. To reduce the possibility of misinterpreting questions, we pretested the questionnaire in four districts. The full questionnaire is reproduced in appendix IV. We took additional steps to check answers for a subsample of respondents because of concerns about misinterpretation. We were concerned about possible misinterpretation of a question about implementing the law (question 1) because we discovered during our pretest that there was confusion between a narrow and (probably) little used portion of a student privacy law (PPRA) regarding selling and marketing of student data and a more familiar older law (FERPA) concerning student records. We were also concerned that there could be an underreporting of commercial activities because our question did not specify the activities (“During school year 2003-2004, did your LEA, or any school in your LEA, engage or plan to engage in one or more activity regarding the collection, disclosure, or use of student information for the purposes of marketing or selling or providing the information to others for these purposes?”) and because very few schools reported using student information for commercial purposes. Further, our pretesting indicated a potential problem with respondents suggesting they would be hesitant to report commercial activities. Moreover, because 86 percent of our respondents answered “no” to question 2, a higher rate than expected, there was some question in our minds whether our respondents who answered no had really considered all the possible ways in which student information could be used for commercial purposes when formulating their answers. We attempted to verify the answers to our question about commercial practices by telephoning 36 districts, the 20 districts that reported using student data for commercial purposes, and the 16 randomly selected districts from among the districts that reported not using student data for commercial purposes. Three of the 20 districts originally reporting the use of student data were found not to be using the data for commercial purposes because they were supplying the data to military organizations or for scholarships, allowed uses. For the 16 sampled districts reporting not using students’ data for commercial purposes, we asked the superintendent or other knowledgeable person in the district detailed questions about 20 possible commercial activities. We asked whether the school provides student addresses or phone numbers for specific activities (student pictures, letter jackets, any types of school uniforms, yearbooks, class rings, tuxedo rentals for prom, corsages for prom, musical instrument rentals, caps and gowns for commencement, preparation for Scholastic Aptitude Test or other tests, any other type of tutoring, transportation for school field trips, or travel for other trips such as spring break or ski trips), or to outside organizations for students to serve on an Internet or study panel, answer questionnaires, test or try out a product, or receive mailings from or talk with representatives of outside organizations who are selling services or products. One of these 16 districts was found to be using the data for commercial purposes in connection with photographers for school pictures. As a result of these telephone calls, we corrected the three incorrect records but did not make further adjustments for districts that were not contacted. We attempted to verify the answers about the development of policies by examining policies that were voluntarily submitted by districts in our sample. Sixty-one districts complied with our request to submit a copy of their policy. We analyzed these policies to determine if they specifically referred to marketing or commercial activities. We found 11 districts submitted policies that addressed these provisions and an additional 8 submitted policies that partially addressed the provisions in that they prohibited the release of student data for commercial purposes but did not address the collection or use of such data. Therefore, our questionnaire gathered relevant data about districts’ perceptions of the extent to which they thought they were implementing the provision, rather than the extent to which these policies actually did so. This probably reflects confusion in interpreting the provision or lack of awareness of its existence. We interviewed officials at Education’s Family Policy Compliance Office and the Office of the General Counsel. We examined in detail the guidance issued by the department to assist schools in implementing PPRA provisions on use of student data for marketing and selling purposes. We conducted telephone interviews with district and school officials in the seven districts we visited to collect information for Public Education: Commercial Activities in Schools (GAO-00-156), a report we issued in 2000, to discern changes in commercial activities in these districts. (See table 5.) We selected these districts because they engaged in a variety of commercial activities, served diverse populations—ranging from large numbers of poor students to children from affluent families—and varied in terms of geography and urbanicity. In updating that information for this report, we interviewed district-level officials, including superintendents and business managers, and elementary, middle school, and high school principals. This appendix lists state statutory and regulatory provisions relating to commercial activities as of May 2004. Shaded entries were enacted since 2000. This updated table identifies 15 new statutes and 3 new regulations and also includes state laws pertaining to the sale of competitive food in schools. In addition, the table includes several laws that were not identified in our previous report. Requires school committee to establish a travel policy that addresses expectations for fundraising by students . This appendix lists proposals addressing commercial activities in schools that have been introduced by some state legislatures between January 2003 and February 2004. The data are taken from the Web sites maintained by state legislatures. Many of these sites are revised only periodically, and information on some is limited to the current legislative session. Therefore this information should be viewed as a rough snapshot, rather than a comprehensive analysis. Would establish nutrition standards for beverages sold to students in public schools Would require food and drink items in public school vending machines to comply with board of education standards standards for vending machine food and drink items vending machines that are not within the cafeteria to be used exclusively for physical education and nutrition education and candy from being dispensed to students by school vending machines. In addition to those names above, the following people made significant contributions to this report: Susan Bernstein, Carolyn Blocker, Richard Burkard, Jim Fields, Behn Kelly, James Rebbe, and Jay Smale. | Congress has continuing interest in commercial activities in U.S. public schools. These include product sales, advertising, market research, and the commercial use of personal data about students (such as names, addresses, and telephone numbers) by schools. To update information about commercial activities in schools, Congress asked us to answer the following questions: (1) Since 2000, what statutes and regulations have states enacted and proposed to govern commercial activities in schools? (2) To what extent have districts developed policies implementing amended provisions of the Protection of Pupil Rights Amendment (PPRA) in the No Child Left Behind Act on the use of student data for commercial purposes? (3) What guidance has the Department of Education (Education) disseminated? To answer these questions, we researched state laws, surveyed a national sample of school districts, analyzed policies provided by districts, interviewed officials at Education, and examined its guidance. In addition, we updated findings from the districts we visited in 2000. Since we reported on commercial activities in 2000, 13 states have established laws addressing commercial activities in public schools, and at least 25 states are considering such legislation. Of the states establishing new laws, 6 established laws affecting market research by addressing the use of student data for commercial activities. Almost all of the proposed bills target the sale of food and beverages. Prior to 2000, 28 states established laws addressing commercial activities, particularly product sales and advertising. At that time, only 1 state passed a provision affecting market research. PPRA provisions required districts to implement policies on the collection, disclosure, or use of student data for marketing and selling purposes, and we estimate that about two-thirds of the districts in the nation believe they are developing or have developed such policies. However, of the 61 districts that sent us policies, only 19 policies addressed these issues. No district reported having collected student data for commercial purposes. Only a few reported disclosing student information for these purposes, and all had done so for school-related purposes such as graduation pictures. Education has undertaken several activities, such as sending guidance to state education agencies and school district superintendents and posting information on its Web page, to inform districts about the student information provisions of PPRA, but many districts appear not to understand the new requirements. Some districts told us that they relied on their state school boards association to develop policies for them because state school boards associations address federal and state laws. School districts in one state sent us policies that addressed commercial activities that had been developed by their state school boards association. Education was not required to disseminate guidance to associations of local school boards in each state and has not done so. |
The PMAs were established from 1937 through 1977 to sell and transmit electricity generated mainly from federal hydropower facilities. Most of these facilities were constructed and continue to be owned and operated by the Bureau and Corps (operating agencies). The operating agencies constructed these facilities as part of a larger effort to develop multipurpose water projects that have functions in addition to power generation, such as flood control, irrigation, navigation, and recreation. As required by law, the PMAs give preference in the sale of power to public power customers. These preference customers include public utility and irrigation districts, customer-owned cooperatives, municipally-owned utilities, and, in some cases, state governments and the federal government. With the exception of Bonneville, the Congress appropriates money each year to the PMAs for power-related purposes and to the operating agencies for both power and nonpower purposes. The PMAs, other than Bonneville, generally receive appropriations annually to cover operations and maintenance expenses (O&M) and, if applicable, capital investments in their transmission assets. Since 1974, Bonneville has operated without annual appropriations from the Congress and has financed its activities through a revolving fund. Bonneville is, however, responsible for repaying its pre-1974 appropriations. The operating agencies receive appropriations for all aspects of their multipurpose water projects, including O&M expenses and capital expenditures. The portion of these appropriations expended for power-related purposes is allocated to the PMAs for repayment by power customers. Section 9(c) of the Reclamation Project Act of 1939 and section 5 of the Flood Control Act of 1944 generally require that the PMAs recover through power rates the costs of producing, transmitting, and marketing federal hydropower. Bonneville is covered additionally under the Northwest Power Act. In addition, the PMAs that market power from multipurpose projects also having irrigation as a purpose—Bonneville and Western—are responsible for repaying certain irrigation assistance costs. Costs are recovered by the PMAs through rates charged the customers who benefit from the federal power. The PMAs generally repay appropriations expended for O&M expenses in the same year that the expenses are incurred, but generally repay appropriations expended for capital investments and other debt, with interest, within the repayment period prescribed by law and/or DOE order. As shown in table 1, the PMAs differ substantially in annual revenues and the amount of outstanding debt to be repaid to the federal government. Table 1 and the notes thereto include information on (1) the outstanding appropriated debt and irrigation costs that are to be repaid by the PMAs, (2) Bonneville Power Administration Treasury bonds, and (3) certain nonfederal debt. DOE’s policy for implementing the cost recovery requirement is set forth in DOE Order RA 6120.2, which states that all costs of operating and maintaining the power system, as well as transmission and irrigation assistance costs, are generally to be included in the rates set by the PMAs. This order does not specifically identify and define all costs that must be recovered. To define the full costs associated with producing, transmitting, and marketing the federal hydropower, we referred to Office of Management and Budget (OMB) Circular A-25 and federal financial accounting standards recommended by the Federal Accounting Standards Advisory Board (FASAB) and adopted by GAO, OMB, and Treasury. OMB Circular A-25 defines the full costs of providing goods or services—in this case federal power marketed by the PMAs—as all direct and indirect costs of delivering those goods or services. The federal financial accounting standard defines the full costs of an entity’s outputs as “the sum of (1) the costs of resources consumed by the segment that directly or indirectly contribute to the output, and (2) the costs of identifiable supporting services provided by other responsibility segments within the reporting entity, and by other reporting entities.” As we reported in 1996, applying these definitions of full cost, the full cost of the power marketed by the PMAs includes all direct and indirect costs incurred by the operating agencies to produce the power, the PMAs to market and transmit the power, and any other agencies to support the operating agencies and PMAs. DOE Order RA 6120.2 requires the PMAs to annually conduct power repayment studies to evaluate whether power rates are sufficient to recover all costs that must be repaid within the rate-making period. These power repayment studies form the basis for setting the PMAs’ rates. Specifically, the power repayment studies must identify, among other things, estimated revenues, expenses, repayments of debt, and the total amount of debt to be repaid generally over the next 50 years. Under the Department of Energy Organization Act of 1977 (DOE Act), the Secretary of Energy is responsible for monitoring the PMAs to ensure that all applicable costs are recovered. The Secretary has delegated to the Deputy Secretary the primary responsibility for PMA-related issues. The Deputy Secretary’s office is to review the rate proposals of the three PMAs before they are sent to FERC for review. For Bonneville, in accordance with provisions of the Northwest Power Act, the administrator develops the rate proposals and submits them directly to FERC without review by the Deputy Secretary. FERC, an independent agency within DOE, reviews the three PMAs’ rates under authority delegated to it by the Secretary of Energy under DOE Delegation Order 0204-108. For Bonneville, FERC reviews rates under the requirements of the Northwest Power Act. The Secretary’s delegation order and legislation limit FERC’s review authority. The public also plays a role in the rate review process. Whenever a PMA proposes a new rate, a public hearing process takes place to obtain input on the proposed rate. This process ensures that members of the public, such as PMA customers, have an opportunity to provide input for the PMAs’ consideration before the rate becomes effective. According to DOE and FERC officials, most comments received during this public process are received from customers, who have an incentive to keep rates as low as possible. See appendix II for more information on the legal responsibilities and delegated authorities regarding PMA rate-making. Current monitoring activities do not ensure that all power-related costs are recovered and that cost recovery issues identified in audit reports are resolved in a timely manner. There is little monitoring performed at DOE’s departmental level and FERC’s monitoring efforts are limited in scope. Neither DOE nor FERC performs independent, detailed reviews of the power repayment studies that form the basis for the PMAs’ rates. Thus, there is little assurance that the power repayment studies provide for complete, accurate, and timely repayment of the PMAs’ power-related costs and debt. Treasury’s involvement in the repayment process includes receiving and recording the repayment transactions, but does not include monitoring the repayment for appropriateness. The Secretary of Energy has delegated to the Deputy Secretary the responsibility for monitoring the PMAs’ activities to ensure that power-related costs and debt are repaid. The Deputy Secretary does so through interaction with the PMA administrators in the field and with the two Washington, D.C., PMA liaison offices at DOE headquarters. The Deputy Secretary receives the rate proposal packages from the three PMAs, approves the rates established by the three PMAs on an interim basis, and sends the rate proposal packages to FERC for review. The sole staff person in the Office of the Secretary involved in doing this told us that he spends most of his time on national energy policy issues, such as the administration’s recent proposal for restructuring the electricity industry. He estimated that only about 30 percent of his time is spent on PMA activities, little of which is devoted to detailed review of rate-making and cost recovery issues. As a result, the Deputy Secretary’s office does not perform the monitoring activities necessary to ensure that all of the appropriate costs are included in the PMAs’ power repayment studies, on which rates are based. In addition, not all information that would be relevant to FERC’s consideration of rate proposals, such as audit reports, is routinely gathered and submitted to FERC. Bonneville’s rates are established by the administrator and sent directly to FERC for review without review by the Deputy Secretary’s office. Therefore, as with the current situation for the three PMAs, there is no assurance that FERC will routinely receive all information, such as audit reports, that would be relevant to FERC’s consideration of Bonneville’s rate proposals. The Secretary of Energy has delegated to FERC the responsibility for reviewing and approving the final rates of the three PMAs. According to FERC officials, their review of the PMAs’ rate proposals focuses on reviewing financial and other information provided to FERC by the PMAs. Although FERC has the authority to do so, FERC officials told us that when reviewing a rate proposal they generally do not obtain and review documentation related to the three PMAs’ costs, such as audit reports that have raised cost recovery issues, beyond that provided by the three PMAs. Under Department of Energy Delegation Order 0204-108, FERC’s review of the three PMAs’ rates includes assessing (1) whether the rates are the lowest possible to customers consistent with sound business principles, (2) whether the revenue levels generated by the rates are sufficient to recover the costs of producing and transmitting electric energy, and (3) the assumptions and projections used in developing the rate components. However, according to FERC officials, the delegation order allows FERC to reject a rate proposal only if it finds the proposal to be “arbitrary, capricious or in violation of law.” According to FERC officials, since this standard for rejection imposes a significant practical limitation on FERC’s review of the three PMAs’ rates, FERC rarely disapproves a rate request. FERC officials told us that they have rejected a rate proposal on only one occasion. This proposal was by Western for the Parker-Davis Project and FERC rejected it because it did not provide for the recovery of significant expected costs of future power facility additions and replacements. FERC’s review and final approval of Bonneville’s rates is authorized under the Northwest Power Act. This act requires FERC to assess whether Bonneville’s rates (1) are sufficient to ensure repayment of the federal investment in the Federal Columbia River Power System (FCRPS), (2) are based on total system costs, and (3) reflect equitable allocation of transmission system costs between federal and nonfederal users. However, FERC’s review of Bonneville’s rate proposals is limited by its interpretation of two court opinions issued by the Ninth Circuit Court of Appeals, which is statutorily charged with reviewing actions arising under the Northwest Power Act. FERC has interpreted the opinions to mean that it may not obtain and review documentation other than that provided by Bonneville and therefore FERC generally does not request additional documentation from Bonneville in its consideration of Bonneville’s rate proposals. In one case, the court held that under the Northwest Power Act, FERC’s review of rates is limited to the three previously mentioned standards specified in the act and that FERC’s limited review properly reflects congressional desire to limit its role to financial oversight rather than rate-making. In a second case, the court held that FERC should not seek new evidence in reviewing Bonneville’s rate proposals. Instead, the court held that FERC should evaluate the evidence developed by Bonneville, which is required to develop a “full and complete record” under section 7.(i) of the Northwest Power Act. As a result of its interpretation of these court opinions, FERC generally does not request additional documentation and limits its review to the documentation submitted with Bonneville’s rate proposals. For all four PMAs, FERC analyzes the PMA-provided information to determine whether the proposed rates appear to be sufficient to recover the costs the PMAs have included in the power repayment studies. FERC relies on the amounts and due dates for repayment reported by the PMAs in the power repayment studies as being complete, accurate, and timely. According to FERC officials, they rely on the financial and other data provided by the PMAs, and not on other evidence, such as audit reports. Based on its restricted analysis, FERC either approves or disapproves the PMAs’ rate proposals; it cannot modify a proposed rate. However, FERC officials told us if FERC disapproves a PMA’s rate proposal, the existing (and typically lower) rate remains in effect until the PMA submits a new one that is reviewed and approved by FERC. The Department of the Treasury’s role in the repayment process is minimal. Currently, Treasury is responsible for receiving and recording the repayments, but is not responsible for monitoring or assessing the appropriateness of the repayment amounts and timeliness. In fact, Treasury does not have available to it the information on amounts due, due dates, and interest rates that would be necessary for it to do so. In addition, since it has no role in rate-making, Treasury does not have staff with the knowledge and expertise necessary to assess the PMAs’ power repayment studies and rate proposals. Audits by external auditors and GAO have identified various unrecovered power-related costs that have resulted in financial loss to the federal government. Until an effective monitoring system is implemented, the federal government will continue to be exposed to financial loss due to the under-recovery of costs. Better monitoring is essential to protect the federal government’s right to recover the costs of its investments that are to be repaid through power revenues. The PMAs’ financial statements are audited by external auditors. The results of the financial statement audits are contained in auditor’s reports, reports on compliance with laws and regulations, reports on internal controls, and management letters. All of these documents can contain valuable information pertaining to cost recovery by the PMAs. For the three PMAs, the external auditors have repeatedly identified certain power-related costs that are not being recovered, even though they do not specifically evaluate the PMAs’ power repayment studies as part of the financial statement audits. Unrecovered costs that have been identified relate to federal employee CSRS pension, postretirement health benefits, life insurance benefits, workers’ compensation benefits, and certain interest expenses related to federal appropriations. Some of the same cost recovery issues also exist at Bonneville, but they have generally not been raised by Bonneville’s external auditors in their audit reports. These types of costs are recoverable under the definitions of full cost contained in OMB Circular A-25 and Statement of Federal Financial Accounting Standard No. 4. Examples of unrecovered costs identified by external and GAO auditors that are symptomatic of the lack of effective monitoring of repaymentinclude the following. PMAs historically have not recovered the full costs of providing pensions and postretirement health benefits for PMA employees and operating agency employees involved in power production and marketing. As part of their fiscal year 1994 financial statement audits, the three PMAs’ external auditors raised the lack of recovery of these costs as a compliance issue in reports to the PMA administrators. We determined that Bonneville also is not recovering the full costs of pensions and postretirement health benefits. In 1996, and again in 1997, we reported that the PMAs still were not recovering the full costs to the federal government of providing these benefits. We reported that the full costs of providing CSRS pensions were not being recovered because the combined contributions of federal employees and the PMAs do not cover the federal government’s full cost of providing these benefits, including payments by other federal agencies. The PMAs’ rates include only the costs actually paid by the PMAs, not the additional payments by the Office of Personnel Management (OPM) related to PMA and operating agency personnel involved in power-related activities. We also reported that the full costs of providing the federal government’s portion of postretirement health benefits were not being recovered and would eventually have to be paid by the general fund of the Treasury. For fiscal year 1996, we estimated that the net cost to the federal government of providing these benefits was about $37 million ($21 million for Bonneville and $16 million for the other three PMAs). Cumulatively, for fiscal years 1992 through 1996, we estimated that the net cost, in constant 1996 dollars, was about $192 million ($110 for Bonneville and $82 million for the other three PMAs). These estimates covered only current PMA and operating agency employees involved in power-related activities. All four PMAs have said that they plan to begin recovering the full costs of providing these benefits. Bonneville began recovering some of these costs in 1998 and plans to phase in full cost recovery over time, with full cost recovery beginning in 2002. Bonneville estimated the amounts related to operating agency personnel, since the operating agencies did not determine the appropriate amounts and allocate the costs to Bonneville. The three PMAs began recovering the full costs of CSRS pension and postretirement health benefits for their own employees in 1998; however, they have not begun to recover the costs for operating agency personnel. According to PMA officials, the three PMAs will begin recovering the costs for operating agency personnel if the operating agencies determine the amounts to be recovered and allocate the costs to the PMAs. However, PMA and operating agency officials told us that the operating agencies are still deciding how this will be done. None of the four PMAs plans to recover these costs retroactively by placing them in current power rates, according to PMA officials. Similarly, the PMAs have not recovered the full costs of providing life insurance benefits for PMA employees and operating agency employees involved in power production and marketing. As part of their fiscal year 1994 financial statement audits, the three PMAs’ external auditors reported to the respective administrators that the PMAs were not recovering these costs. Recovery of the costs is still under consideration by the three PMAs. We determined that Bonneville also had not been fully recovering the costs of life insurance. However, Bonneville and operating agency officials told us that Bonneville began recovering some of these costs in fiscal year 1998 by including an estimate of the amounts related to life insurance benefits in the estimates discussed above. As with CSRS pensions and postretirement health benefits, Bonneville plans to phase in full cost recovery in rates over time, with full cost recovery beginning in 2002. Southeastern’s external auditor reported as part of its fiscal year 1995 financial statement audit that Southeastern was not recovering the full costs to the federal government of providing workers’ compensation benefits. Specifically, the auditor reported that the Corps was not allocating any of its workers’ compensation costs to Southeastern and that Southeastern was therefore not recovering these costs through rates. Auditors have also identified instances of incorrect interest calculations and/or payments by the three PMAs. The errors related to interest on moveable equipment at Western, interest on deferred assets at the Harry S. Truman Dam and Reservoir (Truman Project) marketed by Southwestern, and calculations using incorrect interest rates at certain projects marketed by Southeastern and Southwestern. As part of its fiscal year 1993 financial statement audit, Western’s external auditor reported that Western was not recovering about $3 million annually in interest on moveable equipment. In fiscal years 1994 and 1995, Western developed a common approach for all projects designed to ensure that it recovered this interest in the future. However, Western’s external auditor reported that the problem had not yet been resolved as of the end of fiscal year 1995. Specifically, the auditor reported that certain area offices had not calculated interest on an allocated amount of movable equipment located at headquarters. Western did not address this issue until 1996, when it began charging interest on the balance of all moveable equipment. However, according to Western’s external auditor, Western did not retroactively charge interest on moveable equipment for years prior to 1993. We also found an error related to the calculation of interest on deferred assets at the Truman Project, marketed by Southwestern. Because of fish kills, the project has never operated at its 160,000 kilowatt capacity; instead, only 53,300 kilowatts have been declared to be in commercial operation. As a result, Southwestern has deferred from recovery the estimated costs—$31 million—of the nonoperational portion of the Truman Project. However, Southwestern’s stated policy is to recover the interest expense of the Truman deferred investment annually. Until 1994, the Corps calculated the interest expense for Truman and other projects marketed by Southwestern. Interest costs were to be based on the entire power-related construction costs of these projects, including the $31 million Truman deferral. In 1995, Southwestern began calculating the interest expense on the projects it markets. However, Southwestern’s fiscal year 1995 calculation of interest expense for the Truman project mistakenly excluded interest associated with the $31 million Truman deferral. As a result, about $930,000 in interest on the Truman deferral was not recovered. Southwestern officials acknowledged the mistake and said that the underpayment of interest would be corrected in fiscal year 1996. Southwestern did subsequently recover the $930,000 associated with the Truman deferral, along with approximately $71,000 in additional accrued interest. As a result of its fiscal year 1994 financial statement audit, Southeastern’s external auditor informed Southeastern’s management that, in several instances, the Corps had improperly calculated interest expense on the federal government’s outstanding power-related appropriations. According to the auditor, these errors resulted from the misinterpretation of the interest calculation guidance. For example, the external auditor reported that since 1983, three Corps district offices had not applied the current interest rate on plant additions, even though the policy has been to use current interest rates since 1983. Instead, the Corps had been using the original project interest rates, which for 1984 through 1994 ranged from a low of 2.5 percent to a high of 6.1 percent. These interest rates were considerably lower than the 7.1 to 12.4 percent interest rates that should have been applied to hydropower investments during this period. Under a methodology proposed by its auditor, Southeastern estimated that as of September 30, 1994, using the incorrect interest rates had understated interest expense by about $1.7 million at two of its four rate-making systems. Southeastern officials did not agree with the methodology proposed by its auditor and told us that the $1.7 million represents the maximum amount of the interest understatement. Southeastern subsequently took over the calculation of interest on federal investment that had previously been done by the Corps; however, it did not determine the total magnitude of the error. Southeastern officials told us that they do not plan to recover these costs because they considered them immaterial to Southeastern’s financial statements. We do not agree that materiality in relation to financial statements is the appropriate criteria for deciding whether to recover a known power-related cost. Similarly, as a result of its fiscal year 1994 financial statement audit, Southwestern’s external auditor informed Southwestern’s management that, for the years 1984 through 1988, the Corps had used interest rates lower than those that should have been used to calculate interest on additions to the federal power facilities marketed by Southwestern. The auditor also noted that, as of fiscal year 1994, three Corps districts had continued to use the lower interest rates to calculate interest. Southwestern adjusted its books to properly record interest expense for fiscal years 1989 through 1995 when it took over responsibility for calculating interest on investment from the Corps in fiscal year 1995. However, as of fiscal year 1996, Southwestern had not applied the appropriate interest rates to additions to the federal investment for fiscal years 1984 through 1988, and Southwestern officials told us that they do not plan to do so. A long delay in recovering the costs of a transmission line on which construction began in 1965, and which was later abandoned, is another example of the ineffectiveness of the current monitoring system. According to the Bureau, a transmission line, which was planned to be part of the Pacific Northwest-Pacific Southwest Intertie Project, was abandoned in fiscal year 1969 due to sporadic funding. Western and its external auditor agreed in 1996 that the total unrecovered amount for the abandoned transmission line was about $20 million. However, even though these costs were power-related, Western was able to delay making any principal or interest payments to the federal government until 1997, about 28 years after the project had been abandoned. The current monitoring efforts do not ensure that costs such as those described above are fully recovered, and therefore that the federal government does not incur unnecessary losses on power-related activities. It is important to note that although the aforementioned unrecovered costs have been documented, there is no assurance that all unrecovered costs have been identified. Other instances of under-recovery could exist which we did not identify as a result of the limited scope of our work. Thus, the full magnitude of the unrecovered costs is unknown. If the PMAs were to begin recovering all of these costs, in many cases they would first need to develop estimates of the amounts to be recovered in cooperation with the appropriate operating agency. These estimated costs would then need to be included in the PMAs’ power repayment studies and rates for recovery, as is commonly done now for other costs. If the PMAs were to begin including these unrecovered costs in power rates, the result could be upward pressure on rates. The current activities for monitoring the repayment of power-related costs and debt are less extensive than those performed in prior years. For several years prior to 1984, DOE’s Office of Power Marketing Coordination (OPMC) monitored the PMAs’ activities and reviewed their rate proposals. This provided some additional assurance that information provided to FERC on costs to be recovered by the PMAs was complete and accurate. In addition, before the Secretary of Energy’s order delegating to FERC the authority to review the PMAs’ rates was revised in 1983, FERC had more latitude in reviewing the PMAs’ rates. According to FERC officials, this latitude enabled FERC to more actively review rate proposals and challenge rates that they thought did not recover all relevant costs. FERC’s review of Bonneville’s rate proposals was similarly broader before the passage of the Northwest Power Act. The act limited FERC’s scope of review to the standards specified in the act. Prior monitoring practices used by OPMC, which was disbanded in 1984, were more extensive than those performed by DOE today. According to former OPMC officials, including the former director of that office, OPMC’s monitoring practices included reviewing the PMAs’ rate proposals and power repayment studies before they were sent to FERC. They told us that one of OPMC’s key responsibilities was to assess whether appropriate costs were included in rates, and the office employed a staff with expertise in PMA rate-making to help it carry out this responsibility. However, OPMC did not perform detailed reviews of the power repayment studies which form the basis for the PMAs’ rate proposals. According to DOE officials, when OPMC was disbanded, its monitoring activities were not assigned to another entity and have not been subsequently assumed by another entity. Although the two Washington PMA liaison offices began to perform some of the functions of the disbanded OPMC, they generally have not reviewed the PMAs’ rate proposals and power repayment studies. As a result, the assurance that the PMAs’ rate proposals provide for accurate, complete, and timely repayment was decreased when OPMC was disbanded. Because the Secretary of Energy’s 1978 delegation order did not specify the standard of review, FERC interpreted the order as allowing the same type of review that it had previously carried out as the Federal Power Commission (FPC) when it applied its independent expertise in evaluating the PMAs’ rate proposals. Subsequently, the December 14, 1983, revision to the delegation order, which is currently in effect, limited FERC’s review to the previously discussed three standards set forth in the order. In FERC’s view, under this order, rate proposals can only be rejected if found to be arbitrary, capricious, or in violation of law. In addition, the revision indicated that the operating agencies’ policy judgments and interpretations of laws and regulations were not reviewable by FERC. These limitations significantly reduced FERC’s ability to ensure that all power-related costs are included in the three PMAs’ rates. Prior to the passage of the Northwest Power Act in 1980, the DOE delegation order to FERC applied to all PMAs, including Bonneville. The current delegation order does not cover Bonneville due to provisions of the act. As previously discussed, the act limits the scope of FERC’s review of Bonneville’s rates. While Treasury is responsible for receiving and recording the repayment of the PMAs’ power-related costs and debt, it is not responsible for monitoring repayment and is not in a position to effectively do so. DOE is responsible under the law for monitoring the repayment of the PMAs’ power-related costs and debt and is in the best position to perform this function. However, until an effective monitoring system is established by DOE, including at FERC, the federal government will continue to be exposed to financial loss. We recommend that the Secretary of Energy move quickly to enhance the department’s oversight of the repayment of the PMAs’ power-related costs and debt and thereby increase the likelihood that the federal government will receive all money due it in a timely manner. To provide this additional assurance, the Secretary of Energy should do the following. Require independent, outside reviews by qualified parties of the power repayment studies prepared by the PMAs to increase assurance that all power-related costs are included in rates. These reviews should assess (1) the appropriateness of the assumptions and methodologies used in the power repayment studies, (2) whether all power-related costs are included, and (3) whether appropriate interest rates are used by the PMAs and operating agencies to calculate interest to be charged on completed projects or capitalized for projects being constructed. The results of the reviews should be summarized in written reports. These independent reviews should initially be done for each power repayment study the next time it is updated. Thereafter, the frequency of the reviews should be based on the results of prior review(s) and assessments of the risk of financial loss to the federal government. The independent review costs, as valid power-related costs, should be included in power rates. Include the full costs of CSRS pension and postretirement health benefits, life insurance, and workers’ compensation benefits in the PMAs’ rates. The costs should include not only those for PMA employees, but also those for operating agency employees involved in power-related activities either full-time or part-time, directly or indirectly. Amounts pertaining to operating agency personnel should be obtained from the operating agencies or, if necessary, estimated by the Department of Energy in cooperation with the appropriate operating agency. Incorporate and maintain updated cost recovery guidance in DOE Order RA 6120.2 that ensures full recovery of power-related costs, including the full costs of CSRS pension and postretirement health benefits, life insurance, and workers’ compensation benefits for all PMA employees as well as operating agency employees involved in power-related activities either full-time or part-time, directly or indirectly. Establish a process within DOE for tracking and resolving issues that affect the repayment of power-related costs and debt. Specifically, DOE should: Review the three PMAs’ (Southeastern, Southwestern, and Western) rate proposals before they are sent to FERC. For all four PMAs (Bonneville, Southeastern, Southwestern, and Western), review the reports summarizing the results of the independent, outside evaluations of the power repayment studies, for purposes of identifying any issues that require follow-up and resolution with the PMAs. Review audit reports (Auditor’s Reports, Reports on Compliance with Laws and Regulations, Reports on Internal Controls, and Management Letters) for all four PMAs and ensure the timely resolution of all identified management, cost recovery, and repayment issues. Ensure that all four PMAs pass onto FERC the reports referred to above for its use in reviewing PMA rate proposals. Revise Delegation Order 0204-108 to give FERC more authority to review and challenge the rate proposals of the three PMAs. Specifically, the Secretary of Energy should (1) clarify that the “arbitrary and capricious” standard does not preclude FERC from rejecting a rate proposal that it finds to be inconsistent with cost recovery guidance contained in DOE Order RA 6120.2, (2) allow FERC to go beyond the three specific standards of review specified in the order, as necessary, and (3) allow FERC to modify a rate proposal rather than merely accept or reject it. In addition, we recommend that the Federal Energy Regulatory Commission utilize this additional authority in its reviews of PMA rate proposals, including analysis and consideration of audit reports and the results of independent evaluations of the PMAs’ power repayment studies. We provided a draft of this report to FERC, DOI, Treasury, DOD, DOE, and the PMAs for comment. FERC, DOI, Treasury, and DOD generally agreed with the report. FERC and DOD provided informal technical comments which we incorporated into the report, as appropriate. DOE’s Office of the Secretary did not provide written comments, but told us that they concurred with the PMAs comments. The PMAs agreed with many of the facts contained in our report, but disagreed with our conclusions and recommendations for improving the monitoring. The PMAs’ comments are reprinted in appendix III. Their major comments are evaluated below; other comments are evaluated in appendix III. The PMAs also provided us with informal technical comments which we evaluated and incorporated, as appropriate. The PMAs stated that current monitoring activities are extensive and additional monitoring is not necessary. They cited 10 types of oversight, monitoring, and appeals that are currently available over PMA repayment practices, and stated that our recommendation for an independent, outside review of the PMAs’ power repayment studies would seem to be an unnecessary, expensive, and time-consuming duplication of existing reviews. We do not agree. Based on our review, the current monitoring activities are not designed to ensure that repayment is complete, accurate, and timely. For example, none of the activities delineated by the PMAs include detailed reviews of the PMAs’ power repayment studies, upon which rates are based. Therefore, there is little assurance that these repayment studies provide for complete, accurate, and timely repayment of the costs that the federal government is entitled to recover under current law. Our report includes several examples of (1) significant power-related costs that historically have not been recovered and (2) the lack of timeliness in resolving cost recovery issues once they were raised. It is important to note that our review was not designed to detect and quantify all losses, but rather to evaluate the effectiveness of the system for monitoring repayment and determine if there was the potential for financial loss to the federal government. Based on our assessment of the monitoring system, we concluded that there is the potential for loss to the federal government, beyond the identified examples in the report, and therefore that the system is ineffective and needs enhancing through closer monitoring of the PMAs’ repayment. However, in response to the PMAs’ comments, we have revised our report to clarify that we are not proposing multiple detailed reviews of the power repayment studies. Instead, the detailed, independent power repayment study reviews should be performed and summaries of the results of those reviews used by DOE to identify issues that require follow-up and resolution with the PMAs and by FERC in assessing the PMAs’ rate proposals. The PMAs also stated that the three PMAs’ respective financial statement auditors perform compliance testing which already accomplishes our recommendation that the power repayment studies be independently reviewed to determine whether they provide for complete, accurate, and timely repayment. We do not agree. As we state in the report, the objective of a financial statement audit is to express an opinion as to whether the statements are fairly stated and the audit would not be expected to detect all issues related to repayment or cost recovery. The financial statement audits—and other audits—have not included detailed reviews of the power repayment studies that assessed their completeness, assumptions, methodologies, and the reasonableness of estimates used. Since the existing reviews of the power repayment studies have not included such assessments to ensure that repayment is complete, accurate, and timely, it is not possible to determine whether all cost recovery issues have been identified. Moreover, the PMAs’ assertion relates to the compliance testing at the three PMAs rather than at Bonneville, which receives a different type of audit and less compliance testing, according to a Bonneville official. As we state in our report, some of the same cost recovery issues exist at Bonneville but generally have not been raised by its financial statement auditor in audit reports. Additionally, even when cost recovery issues have been identified as part of the audits, they have often not been addressed in a timely manner by the PMAs. For example, the unrecovered CSRS pensions and postretirement health benefits cost issue was first reported by Western’s financial statement auditor as part of its fiscal year 1993 audit and by Southeastern’s and Southwestern’s auditors in 1994, but still has not been fully resolved by PMA management. The PMAs stated that the failure to recover costs that we identified in this and previous reports largely involved differing interpretations of law or policy, rather than a failure to follow clear guidance. While differing interpretations of law or policy can lead to a failure to recover power-related costs, this is one reason closer monitoring is needed so that issues can be resolved in a more timely manner. For example, as discussed above, Western’s financial statement auditor first recommended that Western begin recovering CSRS pension and postretirement health benefits costs in 1993. This issue has only recently begun to be resolved. Some of the delay in resolving the issue could be attributable to differing interpretations of law or policy. However, a monitoring system at the DOE level would likely have identified this as an issue that might also exist elsewhere and would have facilitated quick resolution at all four PMAs to ensure that the federal government did not continue to suffer financial loss. As we state in our report, the amount of the four PMAs’ unrecovered CSRS pension and postretirement health benefits costs over a 5-year period (fiscal years 1992-1996) was an estimated $192 million (in constant 1996 dollars). This financial loss could have been mitigated by timely identification of the scope of the problem at the DOE level and resolution at each of the PMAs. The PMAs also maintain that OMB Circular A-25 and federal accounting standards do not apply when “the amount to be priced is provided for by statute or regulation.” This statement does not appear to be relevant to the unrecovered costs identified in our report, which the PMAs have now agreed should be recovered. Recovering these power-related costs is in accordance with laws and regulations. Moreover, statutes, regulations, and DOE Order RA 6120.2 require that the PMAs’ power-related costs be included in rates, but do not specifically identify and define all costs that must be recovered. Therefore, using guidance set forth by OMB Circular A-25 and federal accounting standards to help define the full costs of federal power is appropriate. The PMAs state that the description of unrecovered power-related costs included in our report is misleading in that it claims that four types of power-related costs are not being recovered. The PMAs further state that they generally agree that these costs should be recovered and have been taking steps to ensure that this occurs, beginning in fiscal year 1998. They also object to our characterization of progress towards resolving problems as slow or nonexistent. However, as our report demonstrates, progress toward resolving some of the cost recovery issues has been slow. For example, Western delayed making principal or interest payments on a transmission line for about 28 years after the project was abandoned. The total amount of the unrecovered costs during that time period was approximately $20 million. No progress has been made toward resolving certain other cost recovery issues. For example, our report discusses unrecovered workers compensation costs at Southeastern that were reported as a result of Southeastern’s fiscal year 1995 financial statement audit. The PMAs’ comments acknowledge that these costs should be included in rates; however, no progress has been made to do so. In addition, Western’s financial statement auditor reported as part of its fiscal year 1993 financial statement audit that Western was not recovering about $3 million annually in interest on moveable equipment. Western subsequently recovered interest for 1993 and later years; however, it has not taken any steps to recover interest for prior years, even though DOE policy has always been to recover interest on the federal government’s outstanding investment in power-related activities. Similarly, our report describes unrecovered interest costs at Southeastern and Southwestern related to the application of incorrect interest rates to the federal investment that remain unrecovered. The PMAs state that many of the findings reported by the external auditors and GAO were initially raised by PMA personnel. They also state that because the errors related to unrecovered interest on federal investments at certain projects were raised by the three PMAs’ external auditors, that “it is apparent that the present oversight and monitoring practices are working.” We do not agree. Many of the issues we discuss had been problems for a number of years before they were identified. Further, while financial audits are an important element of monitoring, they do not relieve management of its responsibility to establish an effective monitoring system within its own organization. The PMAs believe that FERC’s current authority is sufficient to allow FERC to provide effective monitoring and oversight of the three PMAs’ repayment. This is at odds with FERC’s position. FERC officials told us that, in their opinion, the “arbitrary and capricious” standard specified in the current delegation order is very difficult to meet and thus imposes a significant practical limitation on FERC’s review authority. As a result, FERC rarely disapproves a rate request. Our recommendation involves returning to a more general delegation of authority to FERC that does not impose unnecessary limits on its review authority, including the “arbitrary and capricious” standard that was not a part of the Secretary of Energy’s previous delegation order. We agree with the PMAs’ position that revising the delegation order to allow FERC to modify the three PMAs’ rate proposals, rather than merely accepting or rejecting them, is worthy of consideration. This change, along with the removal of the restrictions on FERC’s review, would better enable FERC to help ensure that costs are recovered. As a result, we have revised our recommendation pertaining to revising the delegation order. As agreed with your offices, unless you publicly announce its contents earlier, we will not distribute this report until 30 days from its date. At that time, we will send copies to appropriate House and Senate Committees; the Ranking Minority Members of the House Committee on the Budget and the House Committee on Resources’ Subcommittee on Water and Power Resources; interested Members of the Congress; the Secretary of Energy; the Secretary of the Interior; the Secretary of Defense; the Secretary of the Treasury; the Chairman of the Federal Energy Regulatory Commission; the Director of the Office of Management and Budget; and other interested parties. We will make copies available to others upon request. Please call me at (202) 512-8341 if you or your staff have any questions. Major contributors to this report are listed in appendix IV. We were asked by the Chairmen of the House Budget Committee and the Subcommittee on Water and Power Resources, House Resources Committee, to examine the monitoring of the repayment of the power-related costs and debt of the Department of Energy’s (DOE) Power Marketing Administrations (PMA). Specifically, the Chairmen asked us to determine (1) if DOE or the Department of the Treasury actively monitors the amounts of debt to be repaid and the appropriateness of the annual payments and (2) if there is a potential for financial loss to the federal government due to lack of such monitoring. We determined whether the current activities for monitoring repayment effectively ensure that repayment amounts are complete, accurate, and timely, but did not attempt to quantify the financial loss to the federal government resulting from any monitoring ineffectiveness. This report provides information on the Bonneville, Southeastern, Southwestern, and Western Area Power Administrations. It also includes the power-related activities of the Department of the Interior’s Bureau of Reclamation (Bureau) and U.S. Army Corps of Engineers (Corps), which own and operate virtually all of the multipurpose federal water projects that provide power to the PMAs’ customers. To determine if current monitoring activities ensure that all repayments were accurate, complete, and timely, we first determined who has the legal responsibility for ensuring that the PMAs recover all power-related costs and repay debt and then obtained an understanding of the current system for monitoring the repayment. We did this by researching relevant laws, regulations, and court cases and interviewing senior officials from DOE headquarters; the two PMA liaison offices in Washington, D.C.; the four PMAs; the Department of the Treasury; the Federal Energy Regulatory Commission (FERC); and the operating agencies (the Department of the Interior’s Bureau of Reclamation and U.S. Army Corps of Engineers). We then assessed the potential for financial loss to the federal government. In doing so, we relied extensively on audited financial information for each of the four PMAs. Specifically, we obtained and analyzed the external auditors’ reports of the results of their financial statement audits, reports on the PMAs’ compliance with laws and regulations, reports on the effectiveness of the PMAs’ systems of internal controls, and letters to management highlighting cost recovery and other issues. Many of these reports contained information on cost recovery issues. For each PMA, we reviewed the most recently available reports as well as those for several previous years. In addition, we reviewed and synthesized recent GAO reports that assessed cost recovery by the PMAs. In performing the reviews that resulted in these previous reports, we had developed a good understanding of the PMAs’ requirement for recovering power-related costs and for several of the costs that were not being fully recovered. We also reviewed reports by the DOE, DOI, and DOD Inspector General offices. After developing an understanding of cost recovery and repayment issues at each of the four PMAs, we corroborated our understanding of those issues by interviewing appropriate officials. Specifically, we corroborated the results of our analyses by interviewing DOE officials and the PMAs’ external auditors. We identified how current activities for monitoring the repayment of the PMAs’ power-related costs and debt differ from prior ones by obtaining and reviewing documentation on organizational changes within DOE pertaining to the monitoring of the PMAs. Specifically, we obtained and reviewed organization charts and delegation orders spanning more than a decade. In addition, we interviewed officials from DOE, the PMA liaison offices, the PMAs, the Department of the Treasury, and FERC. To follow up on these discussions, we also interviewed former officials of DOE’s disbanded Office of Power Marketing Coordination (OPMC), which previously monitored the PMAs’ activities in the late 1970s and early 1980s. Among the officials we interviewed was the former director of OPMC. We then assessed whether these changes could affect the potential for financial losses to the federal government. We conducted our review from December 1997 through June 1998 in accordance with generally accepted government auditing standards. The following are GAO’s comments on the Department of Energy’s letter dated June 5, 1998. 1. Discussed in the “Agency Comments and Our Evaluation” section of the report. 2. Our involvement in raising cost recovery issues to the Congress should not be considered a part of the monitoring process designed by the Secretary of Energy to ensure that he or she fulfills his or her responsibility to recover these costs. Our audit work is generally performed at the specific request of the Congress and cannot be considered to be routine, ongoing monitoring of PMA activity. 3. We believe that the return of the federal government’s investment in the PMAs’ power-related activities to the Treasury through scheduled payments is an expectation of the federal government and the taxpayers, and should not necessarily be thought of as a record of achievement. While we agree that the PMAs have made substantial repayments to Treasury on a timely basis, this report and our previous reports clearly point out that the PMAs have not recovered all of the costs that the federal government has incurred to produce, market, and transmit federal power. 4. We disagree. There should not be diversity among the PMAs in recovering the costs discussed, or in recovering other costs that are power-related. Clarifying the requirement that these costs be recovered in DOE Order RA 6120.2 would help ensure the necessary consistency among the PMAs and, as discussed previously, would avoid any confusion which might arise from differing interpretations regarding these costs in the future. Moreover, significant unrecovered cost categories identified in future monitoring efforts, such as detailed reviews of the PMAs’ power repayment studies, should be specified as recovery requirements in the DOE Order. In addition, since Bonneville is not subject to the delegation order under which DOE reviews the other three PMAs’ rate proposals before they are sent to FERC, monitoring of cost recovery issues at Bonneville at the DOE level is curtailed; clarifying cost recovery guidance in DOE Order RA 6120.2 is especially important to ensure power-related costs are treated alike among all the PMAs. 5. The PMAs’ willingness to provide audit reports, including management letters, to FERC will enhance FERC’s ability to effectively evaluate the PMAs’ rate proposals. However, as noted in our response to comment 4, it is important to include this requirement in RA 6120.2 because Bonneville is not subject to the delegation order under which DOE transmits case materials to FERC. David W. Bogdon, Senior Evaluator Larry L. Feltz, Senior Evaluator Robert J. Bresky, Jr., Evaluator The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO reviewed the monitoring of the repayment of the power-related costs and debt of four of the Department of Energy's (DOE) power marketing administrations (PMA), focusing on determining whether: (1) DOE or the Department of the Treasury actively monitors the amount of debt to be repaid and the appropriateness of the annual payments; and (2) there is a potential for financial loss to the federal government as a result of any lack of such monitoring of the repayment. GAO noted that: (1) current monitoring activities do not ensure that the federal government recovers the full cost of its power-related activities from the beneficiaries of federal power; (2) the full cost of the power-related activities includes all direct and indirect costs incurred by the federal government in producing, transmitting, and marketing federal power; (3) audits by external auditors and GAO's own work have identified various unrecovered power-related costs that resulted in financial loss; (4) progress toward resolving cost recovery issues has been slow or nonexistent; (5) unrecovered power-related costs relate to: (a) Civil Service Retirement System (CSRS) pensions and post-retirement health benefits; (b) life insurance benefits; (c) workers' compensation benefits; and (d) interest on some of the federal appropriations used to construct certain projects; (6) GAO estimated that the federal government's unrecovered costs for CSRS pensions and post-retirement health benefits were about $37 million for fiscal year (FY) 1996 and about $192 million for FY 1992 through FY 1996; (7) the full magnitude of the under-recovery of power-related costs is unknown; (8) until an effective monitoring system is implemented, the federal government will continue to be exposed to financial loss to the under-recovery of power-related costs; (9) the current activities for monitoring the repayment of power-related costs and debt are less extensive than those undertaken in prior years; (10) previously, DOE's Office of Power Marketing Coordination (OPMC) monitored repayment and reviewed rate proposals before they were sent to the Federal Energy Regulatory Commission (FERC) for review; however, DOE disbanded OPMC in 1984 and its monitoring duties generally were not assigned to another entity; (11) OPMC assessed whether appropriate costs were included in rates, but did not review the PMAs' power repayment studies in detail; (12) the scope of FERC's review of the three PMAs' rates was limited by the Secretary of Energy's 1983 revision to the delegation order under which FERC carries out that function; (13) the scope of FERC's review of Bonneville's rates was limited by the passage of the Pacific Northwest Electric Power Planning and Conservation Act; and (14) the review procedures previously performed by DOE's OPMC and FERC provided greater assurance that repayment amounts were accurate, complete, and timely. |
The FAA, an agency of the Department of Transportation, is primarily responsible for the advancement, safety, and regulation of civil aviation, as well as overseeing the development of the air traffic control system. Its stated mission is to provide the safest, most efficient aerospace system in the world. This system, known as the National Airspace System (NAS), includes air traffic control systems, ATC procedures, operational facilities, aircraft, and the people who certify, operate, and maintain them. According to FAA, the system includes more than 19,000 airports, nearly 600 air traffic control facilities, and approximately 65,000 other facilities, including radar, communications nodes, ground-based navigation aids, computer displays, and radios, intended to provide safe and efficient flight services for the public. Over 46,000 FAA personnel and approximately 608,000 pilots operate about 228,000 aircraft within the NAS, including up to 2,850 flights at any given moment. The system operates on a continuous basis, 24 hours a day, every day of the year. As aircraft move across the NAS, controllers at several types of air traffic control facilities manage their movements during each phase of flight. According to FAA, these facilities include the following: More than 500 air traffic control towers supervise flights within about 5 miles from the airport runway. They give pilots taxiing and take off instructions, air traffic clearance, and provide separation between landing and departing aircraft. One hundred sixty Terminal Radar Approach Control facilities provide air traffic control services for airspace that is located within approximately 40 miles of an airport and generally up to 10,000 feet above the airport. These facilities handle sequencing and separation of aircraft as they approach major metropolitan areas. Twenty-two Air Route Traffic Control Centers (ARTCC) control and monitor airplanes over the continental United States and between airports. Controlling traffic usually at or above 17,000 feet, the typical center has responsibility for more than 100,000 square miles of airspace generally extending over a number of states. Three of the centers also control air traffic over the oceans. Controllers at these facilities work with pilots to ensure the flight path is smooth and free of other traffic. The Air Traffic Control System Command Center manages the flow of air traffic within the United States. This facility regulates air traffic when weather, equipment, runway closures, or other conditions place stress on the national airspace system. In these instances, traffic management specialists at the command center take action to modify traffic demands in order to keep traffic within system capacity. Figure 1 provides a visual summary of air traffic control across the different phases of flight. The FAA’s ability to fulfill its mission depends on the adequacy and reliability of its air traffic control systems, a vast network of computer hardware, software, and communications equipment. The agency relies on more than 100 air traffic control systems to process and track flights around the world. These complex and highly automated systems process a wide range of information, including radar, weather, flight plans, surveillance, navigation/landing guidance, traffic management, air-to- ground communication, voice, network management, and other information—such as airspace restrictions—that is required to support the agency’s mission. In order to successfully carry out air traffic control operations, it is essential that these systems interoperate, functioning both within and across facilities as one integrated system of systems. According to FAA data available at the time of our review, about one-third of air traffic control systems rely on Internet Protocol (IP)-based networking technologies for communication. FAA’s ongoing effort to modernize the ATC system is referred to as the Next Generation Air Transportation System (NextGen). NextGen involves changes to many aspects of air transportation, including the acquisition of new integrated systems (both software and hardware), flight procedures, aircraft performance capabilities, and supporting infrastructure to transform the current air transportation system into one that uses satellite- based surveillance and navigation operations instead of ground-based radar. These changes are intended to increase the efficiency and capacity of the air transportation system while maintaining safety and accommodating anticipated future growth. As demand for the nation’s increasingly congested airspace continues to grow, NextGen improvements are intended to enable the FAA to guide and track aircraft more precisely on more direct routes, reduce delays, save fuel, and reduce aircraft exhaust emissions. The following five key air traffic control systems perform critical air traffic control functions: FAA Telecommunications Infrastructure (FTI) forms the basic telecommunications infrastructure for NextGen, replacing the agency’s legacy networks to provide consolidated telecom services for the NAS. The FTI is also intended to reduce costs, improve bandwidth, and offer improved information security services, such as encryption, so that an enterprise-wide approach to information security assurance can be achieved. Surveillance and Broadcast Service System (SBSS) provides surveillance services to FAA and the aviation community. Most notably, SBSS provides the Automatic Dependent Surveillance- Broadcast service—one of the six NextGen transformational programs—which is FAA’s satellite-based successor to radar. This service makes use of Global Positioning System technology to determine and share precise aircraft location information, and streams additional flight information to the cockpits of properly equipped aircraft. En Route Automation Modernization (ERAM) replaces the legacy en route Host computer and backup system, developed more than 40 years ago. According to FAA, much of the system has already been deployed. ERAM is designed to be at the heart of NextGen and is key to advancing FAA’s transition from a ground-based system of air traffic control to a satellite-based system of air traffic management. As ERAM evolves, it is intended to provide benefits for users and the flying public by increasing air traffic flow and improving automated navigation and conflict detection services, both of which are vital to meeting future demand and preventing gridlock and delays. En Route Communications Gateway (ECG) is a communications system that receives data from sources outside the ARTCC, such as flight plan information and weather data, and passes it to other systems, including ERAM. Traffic Flow Management-Infrastructure (TFM-I) provides information processing support for FAA traffic management personnel as they coordinate the use of the NAS and respond to conditions of excess demand. TFM-I receives information on planned and active flights, generates forecasts of demand up to several hours ahead, presents this information to Traffic Management Personnel, and provides automation support for traffic management initiatives to resolve or ameliorate congestion. Each of these systems, in conjunction with many others that make up the NAS computing infrastructure, works to ensure safe flight passageways for aircraft in U.S. airspace from takeoff to landing. Safeguarding federal computer systems and the systems supporting the nation’s critical infrastructures, including the NAS, is essential to protecting national and economic security, and public health and safety. For government organizations information security is also a key element in maintaining the public trust. Inadequately protected systems may be vulnerable to insider threats as well as the risk of intrusion by individuals or groups with malicious intent who could use their illegitimate access to obtain sensitive information, disrupt operations, or launch attacks against other computer systems and networks. Accordingly, since 1997, we have designated information security as a government-wide high-risk area. In 2003, we expanded this high-risk area to include protecting systems supporting our nation’s critical infrastructure.those of agency inspectors general, describe persistent information security weaknesses that place a variety of federal operations at risk of disruption, fraud, and inappropriate disclosure. Our previous reports, and Recent federal guidance demonstrates that securing critical infrastructures from internal and external threats is a national priority. For example, in February 2013, the President signed Executive Order 13636, “Improving Critical Infrastructure Cybersecurity,” to address concerns about better securing critical infrastructure from cyber threats. This order, among other things, directed executive branch agencies to promote the adoption of cybersecurity practices; increase the volume, timeliness and quality of cyber threat information sharing; incorporate privacy and civil liberties protections into every initiative to secure critical infrastructure; and explore the use of existing regulation to promote cybersecurity. The order also directed the National Institute of Standards and Technology (NIST) to develop a technology-neutral voluntary framework for improving critical infrastructure cybersecurity. The framework, issued in February 2014, with business requirements, risk tolerances, and resources. is designed to help organizations align their cybersecurity activities Although many legacy air traffic control systems continue to rely on point- to-point communications, NAS systems, including NextGen systems, increasingly use IP technologies to communicate over interconnected computer networks. With the increased use of such technologies, however, comes increased risk: integrating critical infrastructure systems with information technology networks provides significantly less isolation from the outside world than predecessor systems, creating a greater need to secure these systems from remote, external threats. NIST, Framework for Improving Critical Infrastructure Cybersecurity, Version 1.0 (Gaithersburg, Md.: Feb. 12, 2014). The operational arm of the FAA is the Air Traffic Organization (ATO), which is responsible for providing safe and efficient air navigation services to the nation’s airspace. ATO is led by a Chief Operating Officer who reports directly to the FAA Administrator. The ATO includes seven service units: Air Traffic, Management, Mission Support, Program Management Organization, Safety and Technical Training, Systems Operations, and Technical Operations. Several entities within ATO share responsibility for information security- related activities over air traffic control systems. The NAS Security Risk Executive is the individual with overall responsibility for overseeing all information security-related activities across ATO, including the security of air traffic control systems. ATO’s Information Security Systems Group, within Technical Operations Services, includes NAS Cyber Operations (NCO), ISS Engineering, and an Authorization Team. Ensuring that each air traffic control system is properly secured and protected is the responsibility of an Information Systems Security Officer (ISSO), and the system owner is responsible for planning, directing, and managing resources for the system, including ensuring that the system meets all security requirements throughout its life cycle. The Security Risk Executive, Authorization Team, NCO, system owners, and ISSOs are jointly responsible for carrying out security program activities for NAS systems, including conducting system risk assessments, documenting system security plans, testing security controls, managing remedial action plans, monitoring and responding to incidents, and planning for contingencies. The FAA Office of Information and Technology (AIT) is responsible for the security of FAA’s non-NAS information systems. The office resides within FAA’s Office of Finance and Management and is headed by the Chief Information Officer (CIO), who has overall responsibility to oversee the security of the agency’s information and information systems. The FAA Chief Information Security Officer is responsible for developing, documenting, and implementing an agency-wide information security program and for assisting the CIO in ensuring compliance with federal law pertaining to information security and other applicable policies, standards, requirements and guidelines. Federal law and guidance specify requirements for protecting federal information and information systems. The Federal Information Security Management Act of 2002 (FISMA) requires each agency to develop, document, and implement an agency-wide information security program to provide security for the information and information systems that support operations and assets of the agency, including those provided or managed by another agency, contractor, or another organization on behalf of an agency. FISMA assigns certain responsibilities to NIST, which is tasked with developing, for systems other than national security systems, standards and guidelines that must include, at a minimum, (1) standards to be used by all agencies to categorize all of their information and information systems based on the objectives of providing appropriate levels of information security, according to a range of risk levels; (2) guidelines recommending the types of information and information systems to be included in each category; and (3) minimum information security requirements for information and information systems in each category. NIST, Contingency Planning Guide for Federal Information Systems, SP 800-34, Revision 1 (Gaithersburg, Md.: May 2010). NIST, Managing Information Security Risk: Organization, Mission, and Information System View, SP 800-39 (Gaithersburg, Md.: March 2011). NIST, Information Security Handbook: A Guide for Managers, SP 800-100 (Gaithersburg, Md.: October 2006). provides guidance for facilitating a more consistent approach to information security programs across the federal government. Although FAA has taken steps to safeguard its air traffic control systems, significant security control weaknesses remain in NAS systems and networks, threatening the agency’s ability to adequately fulfill its mission. FAA established policies and procedures for controlling access to NAS systems and for configuring its systems securely, and it implemented firewalls and other boundary protection controls to protect the operational NAS environment. However, a significant number of weaknesses remain in the technical controls—including access controls, change controls, and patch management—that protect the confidentiality, integrity, and availability of its air traffic control systems. Additionally, significant interconnectivity exists between non-NAS systems and the NAS operational environment, increasing the risk from these weaknesses. Further, the agency had not yet fully implemented an agency-wide information security program to ensure that controls are appropriately designed and operating effectively. A key reason for both the technical control weaknesses and the security management weaknesses is that FAA had not fully established an integrated, organization-wide approach to managing information security risk that is aligned with its mission. These shortcomings put NAS systems at increased and unnecessary risk of unauthorized access, use, or modification that could disrupt air traffic control operations. A basic management objective for any agency is to protect the resources that support its critical operations and assets from unauthorized access. An agency can accomplish this by designing and implementing controls that are intended to prevent, limit, and detect unauthorized access to computer resources (e.g., data, programs, equipment, and facilities), thereby protecting them from unauthorized disclosure, modification, and loss. Specific access controls include boundary protection, identification and authentication of users, authorization restrictions, cryptography, and audit and monitoring procedures. Without adequate access controls, unauthorized users, including intruders and former employees, can surreptitiously read and copy sensitive data and make undetected changes or deletions for malicious purposes or for personal gain. In addition, authorized users could intentionally or unintentionally modify or delete data or execute changes that are outside of their authority. Although FAA had issued security policies, it did not consistently protect its network boundary from possible intrusions; identify and authenticate users; authorize access to resources; ensure that sensitive data are encrypted; or audit and monitor actions taken on NAS systems and networks. Boundary protection controls are used to restrict connections into and out of networks and to control connections between network-connected devices. Implementing multiple layers of security to protect an information system’s internal and external boundaries can reduce the risk of a successful cyber attack. For example, multiple firewalls can be deployed to prevent both outsiders and insiders from gaining unauthorized access to systems. NIST Special Publication 800-53 recommends organizations monitor and control communications both at the external boundary of the system and at key internal boundaries within the system. NIST also recommends information systems connect to external networks or systems only through managed interfaces such as gateways, routers, or firewalls. FAA policy states that connections between NAS systems/networks and any non-NAS network or non-NAS information systems must occur through the NAS Enterprise Security Gateway. While FAA implemented numerous controls to separate NAS systems from non-NAS systems, it did not always sufficiently protect connections between external partners and NAS operational systems or limit interconnectivity between operational and mission support environments. The excessive interconnectivity between NAS and non-NAS environments increased the risk that FAA’s mission-critical air traffic control systems could be compromised. Information systems need to be managed to effectively control user accounts and identify and authenticate users. Users and devices should be appropriately identified and authenticated through the implementation of adequate logical access controls. Users can be authenticated using mechanisms such as a password and user ID combination. NIST SP 800- 53 recommends, and FAA policy requires, strong password controls for authentication, such as passwords that are at least eight alphanumeric characters in length, contain at least one upper- and one lower-case letter, contain numbers and special characters, and expire after a predetermined period of time. However, FAA did not consistently implement identification and authentication controls in accordance with its security policies and NIST guidance. For example, certain servers and applications supporting NAS systems did not implement sufficiently strong password controls. As a result, FAA is at increased risk that accounts could be compromised and used by unauthorized individuals to access sensitive information or systems. Authorization encompasses access privileges granted to a user, program, or process. It is used to allow or prevent actions by that user based on predefined rules. Authorization includes the principles of legitimate use and least privilege.implement controls to ensure that only authorized users can access the system. This includes, but is not limited to, uniquely identifying all users, periodically reviewing access to the system, disabling accounts that no longer need access to the system, and assigning the lowest level of permission necessary for a task. NIST also recommends that systems and devices be configured so that only the functionality necessary to support organizational operations is enabled in order to prevent unauthorized connection of devices, unauthorized transfer of information, or unauthorized network connectivity. NIST guidance recommends that organizations FAA did not always ensure that users’ access to key air traffic control systems was authorized in accordance with FAA policies and NIST guidance. In several cases, FAA and its contractors did not properly document that system users were authorized to access NAS systems. Additionally, FAA did not always ensure that periodic reviews of user access to NAS systems were performed in accordance with FAA policies and NIST guidance. As a result, users of these air traffic control systems may have greater access than they need to fulfill their responsibilities, increasing the risk that these systems could be compromised, either inadvertently or deliberately. Cryptographic controls can be used to help protect the integrity and confidentiality of data and computer programs by rendering data unintelligible to unauthorized users and/or protecting the integrity of transmitted or stored data. Cryptography involves the use of mathematical functions called algorithms and strings of seemingly random bits called keys to (1) encrypt a message or file so that it is unintelligible to those who do not have the secret key needed to decrypt it, thus keeping the contents of the message or file confidential; (2) provide an electronic signature that can be used to determine if any changes have been made to the related file, thus ensuring the file’s integrity; and (3) link a message or document to a specific individual’s or group’s key, thus ensuring that the “signer” of the file can be identified. NIST guidance states that the use of encryption by organizations can reduce the probability of unauthorized disclosure of information. NIST Special Publication 800-53 recommends that organizations employ cryptographic mechanisms to prevent unauthorized disclosure of information during transmission, encrypt passwords while being stored and transmitted, and establish a trusted communications path between users and security functions of information systems. Additionally, when federal agencies employ cryptography, NIST standards require them to use Federal Information Processing Standard (FIPS) 140-2-validated algorithms. FAA did not always ensure that sensitive data were encrypted when transmitted or stored, as called for by its policies and NIST guidance. For example, certain network devices supporting NAS systems did not always encrypt authentication data when transmitting them across the network, and other systems did not always encrypt stored passwords using sufficiently strong encryption algorithms in compliance with FIPS 140-2. Due to these weaknesses, FAA faces an increased risk that attackers could compromise accounts or intercept, view, and modify transmitted data, thereby threatening the confidentiality, integrity, and availability of the NAS. Audit and monitoring involves the regular collection, review, and analysis of auditable events for indications of inappropriate or unusual activity, and the appropriate investigation and reporting of such activity. Automated mechanisms may be used to integrate audit monitoring, analysis, and reporting into an overall process for investigation of and response to suspicious activities. Audit and monitoring controls can help security professionals routinely assess computer security, perform investigations during and after an attack, and even recognize an ongoing attack. Audit and monitoring technologies include network- and host-based intrusion detection systems, audit logging, security event correlation tools, and computer forensics. Network-based intrusion detection systems capture or “sniff” and analyze network traffic in various parts of a network. FISMA requires that each federal agency implement an information security program that includes procedures for detecting, reporting, and responding to security incidents. FAA did not consistently implement sufficient audit and monitoring controls. For example, FAA did not always have sufficient capability to monitor network traffic or ensure that NAS systems were sufficiently logging security-relevant events. As a result of these weaknesses, FAA faces an increased risk that it will be unable to detect and respond to unauthorized or malicious activities on its systems. In addition to access controls, other important controls should be in place to ensure the confidentiality, integrity, and availability of an agency’s information. These controls include policies, procedures, and techniques for implementing personnel security and securely configuring information systems. While FAA conducted background investigations in accordance with its policy, weaknesses in its configuration management processes increase the risk of unauthorized use, disclosure, modification, or loss of sensitive information and information systems supporting FAA’s mission. Policies related to personnel actions, such as hiring, termination, and maintaining employee expertise, are important considerations in securing information systems. If personnel policies are not adequate, an entity runs the risk of (1) hiring unqualified or untrustworthy individuals; (2) providing terminated employees opportunities to sabotage or otherwise impair entity operations or assets; (3) failing to detect continuing unauthorized employee actions; (4) lowering employee morale, which may in turn diminish employee compliance with controls; and (5) allowing staff expertise to decline. Hiring procedures should include contacting references, performing background investigations, and ensuring that periodic reinvestigations are consistent with the sensitivity of the position, in accordance with criteria from the Office of Personnel Management. FAA policy requires positions to be designated by sensitivity and risk level, and describes requirements for conducting background investigations for employees and contractors, including periodic reinvestigations of individuals in positions of higher risk or sensitivity. FAA ensured that the employees and contractors we sampled on the TFM-I, ERAM, SBSS, and FTI programs had appropriate background investigations. Specifically, all of the employees and contractors we sampled had up-to-date background investigations that were consistent with the risk designation of their positions. As a result, FAA reduced its risk that it has employed or contracted for unqualified or untrustworthy individuals on these programs. Configuration management is an important control that involves the identification and management of security features for all hardware and software components of an information system at a given point and systematically controls changes to that configuration during the system’s life cycle. Configuration management involves, among other things, (1) verifying the correctness of the security settings in the operating systems, applications, or computing and network devices and (2) obtaining reasonable assurance that systems are configured and operating securely and as intended. In addition, establishing controls over the modification of information system components and related documentation helps to prevent unauthorized changes and ensure that only authorized systems and related program modifications are implemented. This is accomplished by instituting policies, procedures, and techniques that help make sure that all hardware, software, and firmware programs and program modifications have been properly authorized, tested, and approved. Patch management, a component of configuration management, is important for mitigating the risks associated with software vulnerabilities. When a software vulnerability is discovered, the software vendor may develop and distribute a patch or work-around to mitigate the vulnerability. Without the patch, an attacker can exploit the vulnerability to read, modify, or delete sensitive information; disrupt operations; or launch attacks against other systems. Outdated and unsupported software is more vulnerable to attack and exploitation because vendors may no longer provide updates, including security updates. According to NIST SP 800-53, configuration management activities should include documenting approved configuration-controlled changes to information systems, retaining and reviewing records of the changes, auditing those records, and coordinating and providing oversight for configuration change control activities through a mechanism such as a change control board. Additionally, NIST Special Publication 800-128states that patch management procedures should define how the organization’s patch management process is integrated into configuration management processes, how patches are prioritized and approved through the configuration change control process, and how patches are tested for their impact on existing secure configurations. FAA policy describes detailed requirements for controlling changes to NAS systems. FAA did not always ensure that changes to network devices supporting air traffic control systems were managed in accordance with FAA policies for configuration change control. Specifically, significant changes were made to a key network device on one NAS system without following the system’s defined change control process, which requires that changes be documented, analyzed for potential security impacts, tested, and approved before being implemented. Without adequately controlling configuration changes to network devices, an increased risk exists that changes could be unnecessary, may not work as intended, or may result in unintentional side effects that could impact mission-critical operations. Additionally, the agency did not always ensure that security patches were applied in a timely manner to servers and network devices supporting air traffic control systems, or that servers were using software that was up-to- date. For example, certain systems were missing patches dating back more than 3 years. Additionally, certain key servers had reached end-of- life and were no longer supported by the vendor. As a result, FAA is at an increased risk that unpatched vulnerabilities could allow its information and information systems to be compromised. An entity-wide information security program is the foundation of a security control structure and a reflection of senior management’s commitment to addressing security risks. The security program should establish a framework and continuous cycle of activity for assessing risk, developing and implementing effective security procedures, and monitoring the effectiveness of these procedures. Without a well-designed program, security controls may be inadequate; responsibilities may be unclear, misunderstood, or improperly implemented; and controls may be inconsistently applied. FISMA requires each agency to develop, document, and implement an information security program that, among other things, includes policies and procedures that (1) are based on risk assessments, (2) cost-effectively reduce information security risks to an acceptable level, (3) ensure that information security is addressed throughout the life cycle of each system, and (4) ensure compliance with applicable requirements; security awareness training to inform personnel of information security risks and of their responsibilities in complying with agency policies and procedures, as well as training personnel with significant security responsibilities for information security; periodic testing and evaluation of the effectiveness of information security policies, procedures, and practices, to be performed with a frequency depending on risk, but no less than annually, and that includes testing of management, operational, and technical controls for every system identified in the agency’s required inventory of major information systems; a process for planning, implementing, evaluating, and documenting remedial actions to address any deficiencies in information security policies, procedures, or practices; procedures for detecting, reporting, and responding to security plans and procedures to ensure continuity of operations for information systems that support the operations and assets of the agency. To its credit, FAA has taken steps to implement an information security program and manage information security risks for its air traffic control systems. FAA produced system security plans for the systems we reviewed which specified the systems’ operational contexts, relationships with others systems, general security requirements, and security controls. Additionally, FAA policy requires the periodic testing and evaluation of the controls on agency systems, and the agency has a process for planning, implementing, evaluating and documenting remedial actions to address deficiencies in those controls. FAA also documented a risk assessment policy and conducted risk assessments on the major systems we reviewed. However, FAA did not always consistently document incident response policies, ensure that contractors took required training, adequately test security controls, mitigate security weaknesses in a timely manner, ensure that incident response capabilities for NAS systems were adequate, or fully document and test contingency plans for its air traffic control systems. A key element of an effective information security program is to develop, document, and implement risk-based policies, procedures, and technical standards that govern the security over an agency’s computing environment. Regarding incident response activities, NIST Special Publication 800-53 recommends agencies create an incident response policy and procedures to facilitate the implementation of the policy and associated incident response controls. Although FAA had developed and documented many information security policies and procedures, incident response policies and procedures were not always complete or approved. For example, although the NCO security group operates as the focal point for NAS incident response activities, ATO’s incident response policy establishing NCO as the incident response focal point was still in draft at the time of our review. Additionally, although system-level incident response policies had been finalized for four of the NAS systems we reviewed, they did not always specify incident reporting timeframes or the need for all incidents to be reported. Without finalized and harmonized incident response policies, FAA faces an increased risk that incident response authorities and responsibilities will not be clearly understood by all stakeholders, impeding the agency’s ability to efficiently and effectively respond to incidents or to do so in a timely manner. According to FISMA, an agency-wide information security program must include security awareness training for agency personnel, contractors, and other users of information systems that support the agency’s operations and assets. This training must cover (1) information security risks associated with users’ activities and (2) users’ responsibilities in complying with agency policies and procedures designed to reduce these risks. FISMA also includes requirements for training personnel who have significant responsibilities for information security. Additionally, NIST Special Publication 800-53 recommends that agencies provide incident response training to information system users consistent with their assigned roles and responsibilities. FAA officials were unable to tell us when the policy would be finalized. individuals with significant information system security responsibilities must receive role-based training specific to their responsibilities. For both security awareness and role-based training, FAA policy also states that records must be kept documenting the name and title of the individual receiving training, their security responsibilities, the type of training received, and the training date. However, FAA did not always ensure that its employees and contractors took required information security training, including specialized security training and system-specific training, in a timely manner. For example: FAA contractors supporting certain NAS systems did not take the required security awareness training. Additionally, FAA did not provide periodic refresher security training to individuals with significant security responsibilities on two NAS systems. Although FAA stated that the Department of Transportation security awareness training, combined with on-the-job training, was sufficient, the Department of Transportation training does not cover security topics specific to these systems, and FAA did not define required content for on-the-job training or keep training records, as required by FAA policy. FAA had also not sufficiently documented that persons with incident response roles and responsibilities for one of the NAS systems we reviewed had taken required training, and did not provide formal incident response training for personnel on another NAS system. Without adequately ensuring that personnel take required security training, FAA faces an increased risk that employees may not recognize and respond appropriately to potential security threats and vulnerabilities. Further, without sufficiently documenting that incident responders have taken required training, FAA faces an increased risk that employees will not receive training on performing their roles and responsibilities on a regular basis. FAA policy, in accordance with NIST guidance, states that security control assessments are to determine the extent to which controls are implemented correctly, operate as intended, and produce the desired outcome with respect to meeting the security requirements of the system. NIST Special Publication 800-53A notes that while a high-level examination of a limited body of evidence can support an assessor’s determination that a control is implemented and free of obvious errors, determining whether a control is implemented correctly and operating as intended requires an in-depth analysis of a substantial body of relevant evidence. While FAA prepared assessments of the security controls for the NAS systems we reviewed, its control assessments were not always comprehensive enough to identify weaknesses that we found in user account management, configuration management, and security awareness training controls. For example, FAA concluded that the account review control for one system was implemented based on examining the system security plan and interviewing officials, but the testers did not examine artifacts such as audit reports to determine if reviews were being conducted as described in the security plan. FAA also concluded that configuration change control had been implemented for another system after examining the system security plan and interviewing personnel; however, artifacts such as change tickets and approval documents were not examined to determine if system changes were controlled in accordance with procedures. Further, FAA’s test results indicated that the security awareness training common control was defined, but the testers did not examine training records to verify that personnel on the systems that rely on the control were taking the training as required. ATO officials stated that control assessments were often supported by additional documentation not described in the security control assessment reports; however, the agency was unable to provide evidence to corroborate this statement. By not conducting more comprehensive tests of security controls, FAA has decreased assurance that controls are implemented correctly and operating as intended. Additionally, because security control assessments are used by agencies to evaluate whether contractors are implementing information security controls effectively, FAA had reduced assurance that its contractors were adequately securing and protecting air traffic control systems. FISMA requires that agency-wide information security programs include a process for planning, implementing, evaluating, and documenting remedial actions to address any deficiencies in the information security policies, procedures, and practices of the agency. Agencies must establish procedures to reasonably ensure that all information security control weaknesses, regardless of how or by whom they are identified, are addressed through the agency’s remediation processes. For each identified control weakness, the agency is required to develop and implement a plan of actions and milestones (POA&M) based on findings from security control assessments, security impact analyses, continuous monitoring of activities, audit reports, and other sources. When considering appropriate corrective actions to be taken, the agency should, to the extent possible, consider the potential agency-wide implications and design appropriate corrective actions to systemically address the deficiency. While FAA established POA&Ms for addressing identified security control weaknesses, it did not always complete remedial actions in accordance with established deadlines. For example, of the 147 POA&Ms we reviewed on 4 NAS systems, 58 were not completed by their planned completion dates, and the planned completion dates for 50 had been extended from between 8 months to more than 3 years past the dates that they were originally scheduled to be completed. According to ATO officials, one reason that original deadlines are often missed is that the programs lack sufficient resources and funding to address weaknesses by their original due dates. Without resolving identified vulnerabilities in a timely manner, FAA faces an increased risk, as continuing opportunities exist for unauthorized individuals to exploit these weaknesses and gain access to sensitive information and systems. Comprehensive monitoring and incident response controls are necessary for rapidly detecting incidents, minimizing loss and destruction, mitigating the weaknesses that were exploited, and restoring computing services. While strong controls may not prevent all incidents, agencies can reduce the risks associated with these events by detecting and promptly responding before significant damage is done. NIST Special Publication 800-53 recommends that agencies test incident response capabilities for effectiveness. NIST guidance also notes that the ability to identify incidents using appropriate audit and monitoring techniques enables an agency to initiate its incident response plan in a timely manner. Further, NIST guidance also recommends that once an incident has been identified, an agency’s incident response processes and procedures should provide the capability to correctly log the incident, properly analyze it, and take appropriate action. NIST guidance also recommends that agencies test their information technology plans, including incident response plans, and document the test results in an after action report. FAA’s ATO assigned responsibility for incident handling on NAS systems to NCO, although NAS system owners and operators also have a responsibility to coordinate with NCO in responding to incidents affecting their systems. FAA has notable shortcomings in security monitoring and incident detection over air traffic control systems. For example: NCO does not have sufficient access to effectively monitor the NAS operational environment. For example, there was no full network packet capture and anomaly detection capability for network traffic at major network interface points at FAA operational facilities. Additionally, network traffic flow session data was not integrated into the ad-hoc query systems used by the NCO. Further, NCO lacked access to data from sensors on key network gateways, including intrusion detection, network packet capture, and network flow data, and so cannot adequately monitor the gateways for security-relevant events. Although NCO has been given responsibility for incident response for the NAS environment, 26 of the 35 IP-connected NAS systems did not provide security event logs to NCO, including 3 of the systems we reviewed, severely limiting the ability of NCO to effectively monitor the NAS environment. The NCO database system containing centralized security logs collected from various NAS systems was ineffective due to weaknesses in its searching function. Specifically, the system could not search past any gaps in the log data. To compensate, NCO personnel would manually parse the data from multiple queries together, but there was not sufficient assurance that all data needed for incident investigations had been retrieved. NCO did not have a formal process in place to review and document the potential impact to NAS operations from significant incidents identified internally or by FAA’s Cyber Security Management Center. Specifically, NCO did not formally assess the potential risks to the NAS for any of the incidents we reviewed. Testing of the NAS incident response capability has been limited. Specifically, while one system had conducted and documented tests of its incident response capability, NCO has not developed after- action reports for all phases of its incident response capability tests, and officials with three systems all indicated they do not test their incident response capabilities. As a result, there is an increased risk that FAA will not be able to adequately detect, contain, eradicate, or recover from incidents affecting air traffic control systems. Losing the capability to process, retrieve, and protect electronically maintained information can significantly affect an agency’s ability to accomplish its mission. If contingency planning controls are inadequate, even relatively minor interruptions can result in lost or incorrectly processed data, which can cause financial losses, expensive recovery efforts, and inaccurate or incomplete information. NIST Special Publication 800-34 recommends that contingency plans include procedures for diagnosing and addressing problems, notifying recovery personnel when the plan needs to be activated, and procedures to be followed in the event that specific personnel cannot be contacted. NIST also recommends that contingency plans include procedures for notifying users when a system has been reconstituted and normal operations have resumed. Additionally, NIST Special Publication 800-53 notes that contingency plans should be tested to determine the plan’s effectiveness and the organization’s readiness to execute the plan. FAA did not always ensure that contingency plans for air traffic control systems were complete or that tests of the plans were adequate. Although FAA documented contingency plans for three systems, it did not always include important information in these plans. For example, the contingency plans for two systems did not sufficiently document the means by which key personnel were to be contacted in the event of a disaster or define procedures to follow in the event that specific personnel could not be contacted. Although separate notification procedures were developed for one of the systems, they were not included in the contingency plan either explicitly or by reference. Also, although procedures had been established for notifying users when two of the systems had been reconstituted and normal operations had resumed, the contingency plans for those systems did not include or reference the procedures. Further, the contingency plan for another system did not contain the actual assessment and recovery procedures for the system. Also, FAA tested the contingency plans for three NAS systems, but the tests did not always address key elements of the plans, including notification procedures, recovering the system on an alternate platform, and system performance on alternate equipment. Without including important information in its contingency plans for air traffic control systems or sufficiently testing its contingency plans, FAA is at an increased risk of employees or contractors not following the correct procedures to appropriately recover systems in a timely manner from service disruptions. One important reason for many of the weaknesses in security controls as well as the security program shortcomings identified in our review is that FAA has not yet fully established an integrated, organization-wide approach to managing information security risk. According to NIST, effective risk management requires organizations such as the FAA to operate in highly complex, interconnected environments using state-of- the-art and legacy information systems—systems that organizations depend on to accomplish their missions and to conduct important business-related functions. The complex relationships among missions, mission/business processes, and the information systems supporting those missions and processes require an integrated, organization-wide view for managing risk. Effective management of information security risk is also critical to the success of organizations in achieving their strategic goals and objectives. FISMA requires the head of each federal agency to ensure that information security management processes are integrated with agency strategic and operational planning processes. NIST SP 800-39 provides agencies with guidance for developing and implementing an integrated, organization-wide program for managing information security risk to agency operations (i.e., mission, functions, image, and reputation), organizational assets, individuals, other organizations, and the nation resulting from the operation and use of federal information systems. It describes an integrated approach for addressing information security risk at the organization level, the mission/business process level, and the information system level. NIST SP 800-39 states that, in managing information security risk at the organizational level, agencies should establish and implement information security governance, a risk executive function, and risk management strategy in order to ensure that risk management decisions are aligned strategically with the agency’s missions and business functions consistent with the organizational goals and objectives. The publication also states that an organization’s mission/business processes should be designed to manage risk in accordance with the organizational information security risk management strategy. Further, NIST SP 800-100 notes that agencies should have a strategic plan for information security, which identifies goals and objectives related to the agency’s mission, specifies a plan for achieving those goals, and establishes short- and mid-term performance targets and measures that allow the agency to track manage and monitor its progress towards those goals and objectives. Also, in February 2014, NIST established its Framework for Improving Critical Infrastructure Cybersecurity, which presents a set of industry standards and best practices to help organizations manage cybersecurity risks to critical infrastructure systems, and contains a methodology for organizations to evaluate and strengthen information security programs for critical infrastructure. While it has taken initial steps, the FAA has not yet implemented an effective, organization-wide program for managing information security risk to its mission and operations. Specifically, the FAA Chief Information Security Officer stated that in November 2013, the agency established a risk executive function at the agency level in the form of a Cyber Security Steering Committee. The committee includes representatives from across FAA, including ATO, NextGen, and FAA’s office of security. However, FAA has not yet fully established the governance structures and practices necessary for ensuring that its information security risk management decisions are aligned strategically with its mission. Specifically: FAA has not clearly and consistently established roles and responsibilities for information security for NAS systems. FISMA requires the head of an agency to (1) ensure that senior agency officials provide security for information and information systems that support operations and assets under their control, and (2) delegate to the agency CIO the authority to ensure compliance with FISMA requirements. Further, according to NIST Special Publication 800-39, one of the tasks of the risk executive is to establish risk management roles and responsibilities. However, existing FAA practices, policies, and documentation are inconsistent in establishing responsibility for NAS information security. While FAA’s security management program policy states that primary responsibility for NAS information security rests with the CIO, ATO officials stated that primary responsibility for NAS information security lies with ATO rather than the CIO. FAA’s portion of the President’s Fiscal Year 2014 Budget Submission states only that the CIO is responsible for non-NAS systems. The steering committee has defined updated roles and responsibilities for information security, which state that executive-level responsibility for the organization-wide information security program lies with the CIO. However, these roles and responsibilities have not yet been implemented. Additionally, AIT officials and ATO officials disagreed about whether the roles and responsibilities had been approved. Better defining roles and responsibilities is important because FAA officials in AIT and ATO expressed diverging opinions about how information security controls should be implemented in the NAS environment. For example, AIT officials stated that ATO and AIT should collaborate to identify and reduce duplicative incident response activities between NCO and FAA’s existing Cyber Security Management Center, but ATO officials stated that the NCO capability should remain separate from the Cyber Security Management Center because of the unique security requirements of the NAS critical infrastructure. FAA does not have a strategic plan for information security that is up to date and reflects current conditions. FAA’s Information Systems Security Program Policy requires a multiyear information security strategic plan to be developed, maintained, and updated annually. Additionally, NIST SP 800-100 states that agencies should revisit the information security strategic plan when a major change in the agency information security environment occurs. However, the FAA information security strategic plan has not been updated since 2010. Significant changes in the NAS environment, such as the increased reliance on IP networks, increased connectivity between systems, the introduction of NextGen systems, and the designation of the NAS as part of the nation’s critical infrastructure, have changed the level and nature of the information security risks facing air traffic control systems. The evolution of connectivity for NAS systems substantially increases risks of Internet- based intrusions and disruptions. AIT officials told us that the Cyber Security Steering Committee plans to revise the information security strategic plan during fiscal year 2015. Because FAA lacks an up-to-date information security strategic plan, the ATO organization does not have a clear set of goals, objectives, and performance measures around which it can organize its information security program. Responsibility for NAS information security in ATO is distributed across different entities and programs, and ATO officials stated that variation in emphasis on security goals and priorities across the organization makes it challenging to manage information security activities in the NAS environment. For example, according to ATO officials, the NAS incident response organization, NCO, has limited capabilities and available staff because it is required to obtain funding from other program units within ATO, which have different priorities. Additionally, ATO officials stated that remedial actions are often delayed because the program units do not hold system owners accountable for addressing information security weaknesses in a timely manner. In the absence of clearly defined roles and responsibilities or an updated strategic plan, ATO has begun moving forward with strategic planning activities separately from the steering committee’s risk management responsibilities. During our review, ATO evaluated its information security processes and capabilities against the guidance in the NIST Framework for Improving Critical Infrastructure Cybersecurity, with plans to use the results of the evaluation to develop an information security strategic plan for ATO. However, other organizations represented on the steering committee are not involved in this process. Although ATO officials told us that they plan to inform the members of the committee about the results of the review, it is not clear whether the committee intends to use the results in its efforts, or whether ATO’s planned information security strategic plan will reflect the priorities of FAA as a whole. Until it fully establishes an integrated, organization-wide approach to managing information security risk and ensures that federal guidance for securing critical infrastructure is incorporated into its risk management processes, FAA is likely to continue to face challenges in ensuring that risk management decisions are aligned strategically with the its mission and effectively implementing information security controls for air traffic control systems. As a result, the weaknesses we identified are likely to persist. A large, complex, interconnected system like the NAS inherently faces many security risks. Although FAA took many steps to address these risks, weaknesses remain that challenge the FAA in fulfilling its mission of ensuring the safety and efficiency of the nation’s airspace operations. Many weaknesses in access controls and configuration management pose risks to the security of the NAS. The effect of these weaknesses is increased by the significant interconnectivity that exists between the FTI NAS operational environment and the FTI mission support network. Additionally, significant shortcomings limit NCO’s ability to detect and respond to security incidents across NAS systems. These weak controls are mirrored in weak security management processes, such as incomplete policies and procedures for incident response and insufficient testing of security controls. Additionally, actions to mitigate identified security weaknesses are often delayed—sometimes for years. All of these weaknesses combine to pose increased risks to the confidentiality, integrity, and availability of NAS systems and thus put the safe and uninterrupted operation of the nation’s air traffic control system at risk. A fundamental cause for these various weaknesses is that FAA has not yet implemented an effective program for managing organizational information security risk to its mission. Although FAA established a cyber security steering committee, roles and responsibilities remain unclear, and AIT and ATO officials continue to disagree on who should be responsible for the security of NAS systems. Likewise, an out-of-date information security strategic plan contributes to the lack of an adequate risk-based structure to guide implementation of security controls. Further, due in part to the lack of an up-to-date strategic plan for the agency, ATO lacks a clear set of goals, objectives, and performance measures around which it can organize its information security program for NAS systems, making it challenging to manage information security activities such as ensuring that controls are effectively implemented and that system owners address identified security weaknesses in a timely manner. Until FAA establishes stronger agency-wide information security risk management processes, fully develops its NAS information security program, and ensures that remedial actions are addressed in a timely manner, the weaknesses that we identified are likely to continue, placing the safe and uninterrupted operation of the nation’s air traffic control system at increased and unnecessary risk. To fully implement its information security program and ensure that unnecessary risks to the security of NAS systems are mitigated, we recommend that the Secretary of Transportation direct the Administrator of FAA to implement the following 14 recommendations: Finalize the incident response policy for ATO and ensure that NAS system-level incident response policies specify incident reporting timeframes and the need for all incidents to be reported in accordance with FAA guidance. Establish a mechanism to ensure that all contractor staff complete annual security awareness training as required by federal law and FAA policy. Establish a mechanism to ensure that all staff with significant security responsibilities receive appropriate role-based training. Establish a mechanism to ensure that personnel with incident response roles and responsibilities take appropriate training, and that training records are retained. Take steps to ensure that testing of security controls is comprehensive enough to determine whether security controls are in place and operating effectively, by, for example, examining artifacts such as audit reports, change tickets, and approval documents. Take steps to ensure that identified corrective actions for security weaknesses are implemented within prescribed timeframes. Provide NCO with full network packet capture capability for analyzing Provide NCO with access to network sensors on key network network traffic and detecting anomalies at major network interface points at FAA operational facilities. Integrate network traffic flow data into NCO’s ad-hoc query systems. gateways for reviewing intrusion detection, network traffic, and network session data. Provide NCO with security event log data for all IP-connected NAS systems. Address identified weaknesses in the search function of the NCO database event query system to eliminate the need for manual workarounds and ensure that all data relevant for security investigations can be retrieved. Develop a formal process for NCO to assess significant identified incidents for potential impact to NAS operations. Ensure that NAS incident response capabilities are adequately tested, and that test results are sufficiently documented. Ensure that contingency plans for NAS systems are sufficiently documented, and that tests of contingency plans address key elements of the contingency plans, including notification procedures, recovering the system on an alternate platform, and system performance on alternate equipment. Further, to establish an integrated organization-wide approach to managing information security risk and to ensure that risk management decisions are aligned strategically with the FAA’s mission, we recommend that the Secretary of Transportation direct the Administrator of FAA to take the following three actions: Clearly define organizational responsibilities for information security for NAS systems, and ensure that all relevant organizations, including AIT and ATO, are in agreement with them. Update the FAA information security strategic plan to reflect current conditions, including the increased reliance on IP networking and the designation of the NAS as one of the nation’s critical infrastructures. Create an agency-wide commitment to strategic planning for information security by ensuring that planning activities are coordinated with all relevant organizations represented on the Cyber Security Steering Committee. We are also making 168 recommendations to address 60 findings in a separate report with limited distribution. These recommendations consist of actions to implement and correct specific information security weaknesses related to access controls and configuration management. In written comments (reprinted in appendix II) on a draft of this report, the Department of Transportation stated that FAA concurred with our recommendations. The department also stated that FAA recognizes the need to secure the NAS environment as part of the nation’s critical infrastructure, and that FAA has taken several steps to improve NAS information security. Additionally, the department stated that FAA recognizes that mission assurance requires the integration of all agency cyber capabilities to support operations and is continuing its efforts to establish an integrated organization-wide approach to managing information security risk. We agree that these actions are important steps for FAA to take, and we also believe that addressing our recommendations will result in valuable improvements in information security over air traffic control systems. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies of this report to appropriate congressional committees, the Secretary of Transportation, and other interested parties. In addition, this report will be available at no charge on the GAO website at http://www.gao.gov. Should you or your staffs have questions on matters discussed in this report, please contact Gregory C. Wilshusen at (202) 512-6244, Dr. Nabajyoti Barkakati, Ph.D., at (202) 512-4499, or Gerald L. Dillingham, Ph.D., at (202) 512-2834. We can also be reached by e-mail at wilshuseng@gao.gov, barkakatin@gao.gov and dillinghamg@gao.gov. GAO staff who made major contributions to this report are listed in appendix III. The objective of our review was to evaluate the extent to which the Federal Aviation Administration (FAA) has effectively implemented appropriate information security controls to protect the confidentiality, integrity, and availability of its existing air traffic control (ATC) systems. To determine the effectiveness of the FAA’s security controls, we gained an understanding of the overall National Airspace System (NAS) control environment and examined controls for the agency’s systems and facilities. Specifically, we reviewed controls over the network infrastructure and systems that support the FAA’s mission to provide the safest, most efficient aerospace system in the world. We performed our work at FAA headquarters in Washington, D.C. and other locations supporting key systems, specifically: the Air Traffic Control System Command Center in Warrenton, Virginia; the FAA’s William J. Hughes Technical Center in Egg Harbor Township, New Jersey; and at contractor facilities in Herndon, Virginia; Egg Harbor Township, New Jersey; and Melbourne, Florida. To select the information systems for our audit, we evaluated each NAS information system based on several factors, including its relative importance in supporting FAA’s mission, expected lifetime, and how widely it is used. Further, we selected only systems that use Internet Protocol (IP)-based communications. Based on this evaluation, we for review: FAA selected a non-generalizable sample of five systemsTelecommunications Infrastructure (FTI), Surveillance and Broadcast Service System (SBSS), Traffic Flow Management-Infrastructure (TFM-I), En Route Automation Modernization (ERAM), and En Route Communications Gateway (ECG). GAO, Federal Information System Controls Audit Manual (FISCAM), GAO-09-232G (Washington, D.C.: February 2009). and procedures; and standards and guidelines from relevant security and IT security organizations, such as the National Security Agency and the Center for Internet Security. reviewed network access paths to determine if boundaries had been adequately protected; reviewed the complexity and expiration of password settings to determine if password management was being enforced; analyzed users’ access authorizations for four systems to determine whether system access had been approved and properly documented; observed configurations for transmitting data across the network to determine whether sensitive data were being encrypted. For the ERAM and ECG systems, we reviewed configuration settings for network devices by reviewing the network device builds that are distributed to Air Route Traffic Control Centers (ARTCC) by the William J. Hughes Technical Center; reviewed system security settings to determine if sufficient audit and monitoring controls had been implemented; evaluated configuration change control processes for selected systems to determine whether changes were sufficiently documented, tested, and approved in accordance with system procedures; and inspected key network devices and servers to determine if critical patches had been installed and/or were up to date. Additionally, our review of boundary protection controls focused on interconnections between NAS and non-NAS systems. Using the requirements identified by the Federal Information Security Management Act of 2002, which establishes key elements for an effective agency-wide information security program, and associated NIST guidelines and agency requirements, we evaluated the FAA’s information security program by analyzing processes and documentation that were part of FAA’s information security risk management processes to determine the extent to which the process sufficiently supported the agency’s mission and operations; examining system security plans to determine whether they described the security controls in place or planned for meeting the security requirements of the system; examining security awareness training records to determine whether employees and contractors had received training according to FAA policy and federal guidance; analyzing security testing and evaluation results for four systems to determine whether testing of management, operational, and technical controls was sufficient to conclude that controls were in place and operating effectively; examining remedial action plans for four systems to determine whether FAA addressed identified vulnerabilities in a timely manner; examining contingency plans and contingency test results for selected systems to determine whether those plans were appropriately documented and had been sufficiently tested; and reviewing FAA’s processes for identifying and responding to information security incidents to determine whether FAA has implemented an effective incident response capability for its air traffic control systems. As part of our review of the FAA’s information security program, we reviewed several sources of computer-generated data. These included FAA’s employee background investigation data. inventory of NAS information systems, IT security training completion data, and To verify the reliability of these data, we examined them for obvious outliers, omissions, errors, and consulted with FAA officials to resolve any identified anomalies. We also interviewed knowledgeable officials regarding controls on the security training and employee background investigation data, including how and by whom it is input and used. We determined that these sources of data were sufficiently reliable for our purposes. We conducted this performance audit from August 2013 to January 2015 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objective. In addition to the contacts named above, Gary Austin, Lon Chin, West Coile, John de Ferrari, Wilfred Holloway, Nick Marinos, and Chris Warweg (assistant directors); Sher’rie Bacon; Chris Businsky; William Cook; Saar Dagani; Jennifer Franks; Lee McCracken; John Ockay; Justin Palk; Krzysztof Pasternak; Monica Perez-Nelson; Eugene Stevens; Michael Stevens; and Adam Vodraska made key contributions to this report. | In support of its mission, FAA relies on the NAS—one of the nation's critical infrastructures—which is comprised of air traffic control systems, procedures, facilities, aircraft, and people who operate and maintain them. Given the critical role of the NAS and the increasing connectivity of FAA's systems, it is essential that the agency implement effective information security controls to protect its air traffic control systems from internal and external threats. GAO was asked to review FAA's information security program. Specifically, the objective of this review was to evaluate the extent to which FAA had effectively implemented information security controls to protect its air traffic control systems. To do this, GAO reviewed FAA policies, procedures, and practices and compared them to the relevant federal law and guidance; assessed the implementation of security controls over FAA systems; and interviewed officials. This is a public version of a report containing sensitive security information. Information deemed sensitive has been redacted. While the Federal Aviation Administration (FAA) has taken steps to protect its air traffic control systems from cyber-based and other threats, significant security control weaknesses remain, threatening the agency's ability to ensure the safe and uninterrupted operation of the national airspace system (NAS). These include weaknesses in controls intended to prevent, limit, and detect unauthorized access to computer resources, such as controls for protecting system boundaries, identifying and authenticating users, authorizing users to access systems, encrypting sensitive data, and auditing and monitoring activity on FAA's systems. Additionally, shortcomings in boundary protection controls between less-secure systems and the operational NAS environment increase the risk from these weaknesses. FAA also did not fully implement its agency-wide information security program. As required by the Federal Information Security Management Act of 2002, federal agencies should implement a security program that provides a framework for implementing controls at the agency. However, FAA's implementation of its security program was incomplete. For example, it did not always sufficiently test security controls to determine that they were operating as intended; resolve identified security weaknesses in a timely fashion; or complete or adequately test plans for restoring system operations in the event of a disruption or disaster. Additionally, the group responsible for incident detection and response for NAS systems did not have sufficient access to security logs or network sensors on the operational network, limiting FAA's ability to detect and respond to security incidents affecting its mission-critical systems. The weaknesses in FAA's security controls and implementation of its security program existed, in part, because FAA had not fully established an integrated, organization-wide approach to managing information security risk that is aligned with its mission. National Institute of Standards and Technology guidance calls for agencies to establish and implement a security governance structure, an executive-level risk management function, and a risk management strategy in order to manage risk to their systems and information. FAA has established a Cyber Security Steering Committee to provide an agency-wide risk management function. However, it has not fully established the governance structure and practices to ensure that its information security decisions are aligned with its mission. For example, it has not (1) clearly established roles and responsibilities for information security for the NAS or (2) updated its information security strategic plan to reflect significant changes in the NAS environment, such as increased reliance on computer networks. Until FAA effectively implements security controls, establishes stronger agency-wide information security risk management processes, fully implements its NAS information security program, and ensures that remedial actions are addressed in a timely manner, the weaknesses GAO identified are likely to continue, placing the safe and uninterrupted operation of the nation's air traffic control system at increased and unnecessary risk. GAO is making 17 recommendations to FAA to fully implement its information security program and establish an integrated approach to managing information security risk. In a separate report with limited distribution, GAO is recommending that FAA take 168 specific actions to address weaknesses in security controls. In commenting on a draft of this report, FAA concurred with GAO's recommendations. |
The United States, which possesses the largest, most extensive aviation system in the world, has more than 18,000 airports. U.S. airports range from large commercial transportation centers enplaning more than 30 million passengers annually to small grass strips serving only a few aircraft each year. Of these, 3,304 are designated as part of the national airport system and are therefore eligible for federal assistance. The federal interest in capital investment for airports has been guided by several objectives, most notably ensuring safety and security, preserving and enlarging the system’s capacity, helping small commercial and general aviation airports, funding noise mitigation, and environmental protection. National system airports are of two types—commercial service airports, of which there are 540, and general aviation airports, of which there are 2,764. The Federal Aviation Administration further divides commercial service airports, defined as those publicly owned airports that enplane 2,500 or more passengers and have scheduled service—into primary airports (enplaning more than 10,000 passengers annually) and other commercial service airports. The 413 designated primary airports are arranged into various classes of hub airports—large, medium, small, and nonhub—as explained in figure 1. Statutorily, large and medium hub airports are designated as large primary airports and must contribute a larger share to projects funded under the Airport Improvement Program (AIP) as well as forgo a portion of their AIP grants if they collect passenger facility charges (PFC). This report follows that convention in grouping large and medium hub airports together separate from all other national system airports in considering airports’ financial capabilities. In addition, financial information on each category of airport is presented in appendix II. Airports, airlines, FAA, and a congressionally commissioned study have made various estimates of the future financing requirements needed to meet airports’ future development plans. These estimated requirements have varied widely, ranging from $4 billion to $10 billion annually over the next several years. In our April 1997 report, we concluded that these estimates varied so widely because of differing views about what kinds of projects and airports to include as part of an estimate. In that report, we provided four estimates of future development, varying upon how many categories of projects are included. Our estimates ranged from $1.4 billion per year to fund safety, security, noise mitigation, and reconstruction projects; $2.8 billion if other high-priority projects, primarily capacity-related projects, are added; $6.1 billion for all AIP-eligible projects; and $10.1 billion per year to fund all types of projects, including those not eligible for AIP funding. Airports rely on a variety of funding sources, some public and some private, to finance their capital development. The major funding sources, listed in further detail in table 1, are federal and state grants, PFCs airport and special facility bonds, and airport-generated income. In 1996, U.S. airports raised an estimated $7 billion from these sources. Additional information on each of these funding sources and their distribution among various categories of airports is contained in appendix I. As figure 2 shows, total airport funding varies year to year. For example, funding in 1993 declined by 45 percent from 1992. This variability results primarily from year-to-year changes in the amount of funding from bonds, which in turn is affected by changing interest rates, the demand for air travel, and airlines’ agreements with airports. The amount and type of funding varies considerably by the type of airport. The 71 large and medium hub airports, which accounted for almost 90 percent of all passenger traffic, also obtained 79 percent of all funding in 1996, while the 3,233 other national system airports accounted for the remaining 21 percent of the funding. In addition, as shown in figure 3, large and medium hub airports rely most heavily on private airport bonds, which constitute roughly 62 percent of their total funding, while the other airports rely on federal grants for about half of their funding. Additional information on each of these funding sources and their distribution among various categories of airports is contained in appendix I. To test the strength of the relationship between the size of an airport and its reliance on the various funding sources, we analyzed the correlation between size, as measured by passenger enplanements, and funding sources, as measured by the proportion that each source contributes to the total funding for the 300 commercial airports for which we had complete financial data. We found that as an airport’s size increases, both within and across airport categories, an airport’s reliance on AIP diminishes, while the use of other funding sources, such as bonds and PFCs, increases. Airports’ planned capital development over the next 5 years may total as much as $10 billion per year, or $3 billion more per year than their 1996 funding. Figure 4 compares airports’ total capital development funding in 1996 to their annual planned development over the next 5 years. Funding for 1996 is shown by source; planned spending is shown by the relative priority of the projects. FAA’s highest priorities are for projects to meet safety, security, and environmental requirements, including noise mitigation, and for projects that maintain existing infrastructure (reconstruction). Other high-priority projects are primarily for adding capacity, while other AIP-eligible projects are a relatively lower priority, such as projects aimed at helping airports better meet FAA’s design standards. Some projects, such as expanding commercial space in terminals and parking garages, are typically not eligible for funding from FAA. Although a difference may exist between funding and planned spending in total, there is a much closer match between funding from AIP and planned spending on FAA’s highest-priority projects (reconstruction and mandates). In the aggregate, the $1.372 billion in AIP grants in 1996 roughly equates to the $1.414 billion in estimated development planned for the highest-priority projects. However, because about one-third of AIP grants are awarded to airports on the basis of the number of passengers enplaned and not necessarily projects’ priority, the full amount of AIP grants may not be going to the highest-priority projects. The difference between current funding and planned development for smaller airports represents a greater proportion of their total planned development than for large and medium hub airports. Current funding at the 3,233 small, nonhub, other commercial service, and general aviation airports is a little over half of the estimated cost of their total planned development, producing a difference of more than $1.4 billion (see fig. 5). The difference may actually be even greater if it were not for $250 million in special facility bonding for a single cargo/general aviation airport. For this group of airports, the $782 million in 1996 AIP grants surpasses the annual estimate of $750 million for reconstruction, noise mitigation, and federally mandated projects. As a portion of total funding, the potential funding difference for the 71 large and medium hub airports is comparatively less than for their smaller counterparts (see fig. 6). However, because total expenditures for capital projects are so much greater for these airports, this smaller portion represents a potential shortfall of $1.5 billion, or $87 million greater than smaller airports’ collective shortfall. Figure 6 also indicates that $590 million in AIP grants falls $74 million short of the estimated cost to meet FAA’s highest-priority development—meeting federal mandates and maintaining current infrastructure. For a more detailed analysis that shows the difference between current funding and planned development for each of the six categories of airports, see appendix II. Evaluating the various proposals to provide additional funding for airport development involves the consideration of the trade-offs among the various funding types as well as the potential effect each proposal would have on airports. Initiatives to increase funding for airport development include increasing AIP funding, raising the ceiling on PFCs, and other initiatives, such as FAA’s pilot programs for innovative financing and privatization. In addition, we examined the potential benefits of state-administered revolving funds. Choosing to increase one source of funding instead of another involves making trade-offs because the current funding sources differ in several key characteristics, as shown in table 2. For example, increasing funding through grant programs will increase the extent to which the government can specify the recipient, the project, and the amount of funds that will be awarded because grant programs facilitate such targeting better than other funding mechanisms. However, because grant programs in general are relatively costly to administer, increasing funding through grants would increase administrative costs more than a similar amount from bonds or airport revenues. The funding mechanisms also differ with respect to who bears the cost of airport financing. These differences affect the extent to which beneficiaries pay in proportion to the benefits they receive—a measure of economic efficiency and equity. In choosing, for example, between bonds and grants, it is useful to consider that they may have different efficiency and equity effects because of differences in the share of costs borne by users and nonusers of airports under each funding mechanism. Grants are funded through AIP, which is, in turn, funded primarily by the ticket tax. Thus, users pay for grants to airports. In contrast, part of the cost of tax-exempt bonds is borne by nonusers of airports because the interest earned by bondholders is exempt from federal income taxation. As a result, more of the cost of bond financing is borne by nonusers of airports than in the case of grants. Even so, it is uncertain whether using bonds to increase funding would improve or worsen the overall efficiency and equity of airport financing. This uncertainty arises because of the uncertainty in determining how much nonusers benefit from airport development that may stimulate economic development in the community surrounding any given airport. As a result, it is difficult to compare such benefits with the costs that nonusers currently bear. If, as some believe, these benefits are small, then increasing the use of bonds could reduce the overall efficiency and equity of airport financing. But this decrease in equity and efficiency might be justified because bonds have lower administrative costs than grants. Increasing the total AIP funding would benefit smaller airports more than hub airports under the existing distribution formula. Increasing the level of AIP under the existing distribution formula provides a slightly increasing share of AIP funds to the smaller airports, with a concomitant decrease for the large and medium hub airports. The Congress increased AIP funding for fiscal year 1998 by $240 million to $1.7 billion, but $647 million less than the 1998 authorized level of $2.347 billion, a funding level supported by the airport groups—the American Association of Airport Executives (AAAE) and Airports Council International-North America (ACI-NA). Both the National Civil Aviation Review Commission and the Air Transport Association (ATA), the commercial airline trade association, have recommended that future AIP funding levels be stabilized at a minimum of $2 billion annually. Table 3 provides the amount and share of funds that may go to hub airports and other smaller airports if AIP funding were made available at the current ($1.7 billion), proposed ($2.0 billion), and authorized ($2.347 billion) levels under the existing distribution formula. Smaller airports’ increasing share of AIP under higher funding levels is due primarily to higher levels of state apportionment funds and higher levels of discretionary funding if AIP funding increases. State apportionment funds constitute 18.5 percent of the total program funding, and those funds are for general aviation airports. As AIP increases, state entitlement funds increase from $314.5 million (at a $1.7 billion level of AIP), to $370 million (at $2.0 billion of AIP), and finally to $434.2 million (at $2.347 billion of AIP). Under the current formula, the amount of apportionment funds for primary airports remains constant and, therefore, increasing total funding causes discretionary funding to account for an increasing proportion of the total funds. Greater discretionary funding, in turn, means more funding for smaller airports because one-third of discretionary funding in excess of $300 million is directed to general aviation and other commercial service airports. For example, the one-third set-aside for general aviation and other commercial service airports increases from $29.9 million at the $1.7 billion level, to $81.3 million at the $2.0 billion level, and finally to $140.7 million at the $2.347 billion level. While the National Civil Aviation Review Commission and the ATA have recommended a minimum funding level of $2.0 billion for AIP, the ATA also has recommended redefining airport categories and the distribution formula for AIP. The ATA proposes that national system airports be grouped into four categories and that a specified portion of AIP funds be distributed to airports in each category. As table 4 shows, a slightly higher portion of a $2.0 billion AIP would go to the larger airports and a slightly smaller portion to the smaller airports under the ATA’s proposal than under the current approach. Under the ATA’s proposal, the 55/45 percent distribution would remain constant at higher levels of AIP funding. With the increasing share of AIP funds that might go to the smaller airports at higher funding levels, the disparity between the ATA’s proposal and the current distribution would increase. Increasing PFC-based funding would help larger airports that have a large passenger base, while only minimally aiding smaller airports. Airport groups have actively supported increasing the amount of funding airports can raise through PFCs by eliminating the current $3 per passenger ceiling. Meanwhile, the ATA actively opposes any increase in PFCs because it would increase passenger costs and, the association believes, reduce passenger traffic. The National Civil Aviation Review Commission stated that the PFC ceiling will need to be raised if AIP funding is not substantially greater than $2 billion per year, but the Commission has also recommended that airlines have a greater voice than they currently do in deciding whether an airport needs a PFC. According to airport groups, airports require more PFC funding to reduce congestion at airports, especially for passengers trying to access the airport and moving through the terminal. For some airports, roadside and terminal congestion may be more severe than that on the airfield and harder to finance, according to airport groups, because airlines are not as supportive of nonairfield projects, and these projects are ineligible for or a low priority for AIP grants. As a result, a majority of past and future PFC collections are dedicated to terminal and airport access projects and interest payments on debt. As of January 1, 1998, 264 commercial service airports—almost half of all such airports—imposed a PFC. The larger the airport, the greater the likelihood that a PFC is in place (see fig. 7). About three-quarters of large and medium hub airports impose a PFC, while only 45 percent of nonhub and less than 10 percent of other commercial service airports impose a PFC. If the airports currently charging PFCs were to increase them to $4, $5, or $6 per passenger instead of the current $3 limit, total collections would increase to $1.5 billion, $1.9 billion, and $2.2 billion, respectively, on the basis of 1996 enplanements and collection rates. On the basis of 1996 passenger levels, PFC collections could increase to $2.9 billion, but only if all commercial airports imposed a $6 PFC. Figure 8 shows the estimates by airport category. If enplanements continue to grow as expected, then future collections will be proportionally greater. Increasing the PFC ceiling would not substantially benefit smaller commercial airports. Because smaller airports have relatively few enplanements, PFCs do not generate much funding. In addition, while the PFC program requires large and medium hub airports that impose a PFC to forgo a portion of their AIP grants so that these funds can be redirected to smaller airports, most of these larger airports are already returning their maximum amount, according to FAA officials. When PFCs were first introduced, airport groups hoped that PFCs would provide airports with additional cash flow that could be used to support more airport bonds and, therefore, capital development. However, the issuance of PFC-backed bonds has been limited, in part, by the Department of Transportation’s authority to terminate a PFC if the airport does not use its collections as agreed or if it violates the Airport Noise and Capacity Act. Since 1995, FAA has worked to lessen the likelihood of termination by instituting a lengthy review and termination process, and, as a result, in 1996, two airports issued bonds secured by future PFC collections, and other airports are currently considering following suit. Airports are also using their PFC collections as additional security in bonding arrangements, thereby expanding their overall debt capacity. In recent years, FAA, with congressional urging and direction, has sought to expand airports’ available capital funding through more innovative methods, including more flexibly applying AIP grants and attracting more private capital. The 1996 Federal Aviation Reauthorization Act gave FAA the authority to test three innovative approaches to financing airport development. In addition, the act authorized a pilot to privatize a limited number of airports. Thus far, these two innovative methods have attracted only limited interest among airports. Finally, another innovative alternative—funding state-administered loan funds for smaller airports—while not currently permitted, may hold some promise for increasing funding for smaller airports. Interest in alternative financing approaches was initially spurred by declining AIP funding and progress in establishing innovative approaches to finance surface transportation and other infrastructure improvements. In the FAA Authorization Act of 1994, the Congress directed FAA to study innovative approaches to using federal funds to finance airport development. FAA’s study, released in March 1996, determined that investment at large and medium hub airports has kept pace with aviation growth and that therefore these airports generally do not face systemic financial constraints. However, the study also found that small hub airports may be financially constrained, particularly in connection with terminal and other nonairfield projects. FAA’s study examined four alternatives—a federal guarantee for airport loans; using AIP to fund reserve accounts that act as a safety margin for future interest and principal payments; using AIP to pay for bond insurance; and using AIP to capitalize an airport loan fund—as innovative means to increase airport investment. The study concluded that these options offered modest potential gains but possibly greater benefits in certain circumstances if directed to smaller airports and properly targeted to avoid crowding out current investment. The 1996 FAA Reauthorization Act gave the FAA the authority to test three innovative uses for AIP grants—(1) permitting greater percentages of local matching for AIP grants, (2) paying interest costs on debt, and (3) purchasing bond insurance—for up to 10 projects. Thus far, FAA has received 30 applications and approved 5 projects with grants totaling $15.36 million. All five projects test the first innovative use of grants—allowing local contributions in excess of standard grant matching amounts, which for most airports and projects are otherwise fixed at 10 percent. FAA and state aviation representatives generally support the concept of flexible matching because it means that projects that otherwise might not get under way because of a lack of funding from FAA, can get started sooner; in addition, flexible funding may ultimately increase funding to airports. Applicants have shown less interest in the other two options, which, according to FAA and investment banking officials, do not offer new or substantial benefits for airports. Declining airport grants and broader government privatization efforts spurred interest in airport privatization as another innovative means to bring more capital to airport development, but thus far efforts have shown only limited results. As we previously reported, the sale or lease of airports in the United States faces many hurdles, including legal and economic constraints. As a way to test privatization’s potential, the Congress directed FAA to establish a limited pilot program under which some of these constraints would be eased. Starting December 1, 1997, FAA began accepting applications from airports to participate in the pilot program on a first-come, first-served basis for up to five airports. Thus far, two airports have applied to be part of the program. Allowing FAA to capitalize states’ revolving airport loan funds with AIP grants is an innovative concept that some federal transportation, state aviation, and airport bond rating and underwriting officials believe would help smaller airports obtain additional financing. State revolving loan funds have been successfully employed to finance other types of infrastructure projects, such as waste water projects and, more recently, drinking water and surface transportation projects. While loan funds can be structured in various ways, basically they use federal and state moneys to capitalize the funds, from which loans are then made. Interest and principal payments are recycled to provide additional loans. Once established, loan funds can more quickly expand by issuing bonds using the funds’ capital and loan portfolio as collateral. These revolving funds would not create any contingent liability for the U.S. government because they would be under state control. Some officials of bond rating agencies, underwriting firms, and federal and state transportation agencies believe that revolving loan funds would help smaller airports that have trouble obtaining affordable debt financing. Loan funds could also help speed construction and lower construction costs by providing financing up front, instead of incrementally, as is the case with AIP grants, according to a rating agency official. FAA cannot use AIP grants to capitalize state loan funds because AIP construction grants can go only to a designated airport and project. Recently, FAA received an application from a state for an AIP grant to help establish a revolving fund as part of FAA’s pilot program for innovative financing. However, the application was denied because FAA officials determined that such a grant could be construed as a guarantee of airport debt, which is expressly prohibited under the program. Currently, Florida is the only state with an established revolving loan program. Since 1985, the state has provided $75 million in loans to airports for land acquisition and capital projects. While some of the loans are later reimbursed through AIP grants for eligible projects, the state funds the loan program itself. In addition, 39 states have established state infrastructure banks (SIB) using federal and state grant money to fund surface transportation projects.This same SIB structure could also be used to fund aviation projects, and at least one state—Ohio—has already authorized its SIB to fund aviation projects using state funds. The total funding for airport development peaked in 1992 in real terms and declined to about $7 billion in 1996. Meanwhile, planned development at airports may total as much as $10 billion per year over the next 5 years. Most of this amount—perhaps as much as $7 billion per year—is attributable to the potential costs of development at large and medium hub airports, which enplane 9 of every 10 passengers. Continued funding for these airports will be critical to ensuring adequate capacity for the national airport system and avoiding congestion and delays. However, while the need for funding at hub airports may be considerable, these airports also have access to many funding sources, particularly tax-exempt bonds. The more difficult problem may rest with meeting the funding demands of smaller airports. Smaller airports, especially small, nonhub, other commercial service, and general aviation airports, confront a potential funding shortfall that in percentage terms is far greater than for larger airports. These airports have the fewest funding options, relying on federal grants for half of their funding, which is sufficient to fund FAA’s highest priorities but little else. Protecting the financial viability of these smaller airports will require adequate funding from existing federal and state grant programs, but also more innovative applications of existing funding. FAA has been testing several innovative approaches authorized by the Congress and expects to report to the Congress on the results of this testing later this year. However, a state revolving loan fund was not among those tested, and while not a panacea, state loan funds may offer some potential for helping smaller airports fund their development. FAA has determined that it does not currently have the legal authority to use Airport Improvement Program grants to fund state revolving loan funds. Nevertheless, state revolving loan funds have the support of some federal transportation, state aviation, and airport bond rating and underwriting officials because they believe that the funds could provide loans to airports that otherwise might not have access to debt financing. In addition to expanding available financing, these loans could also speed construction and lower costs by providing funding up front. To help smaller airports fund some of the cost of their capital development, but to avoid undermining the level of federal support for larger airports, we recommend that the Secretary of Transportation seek authority from the Congress to use Airport Improvement Program grants to capitalize state revolving funds in those circumstances where states have a demonstrated capability and desire to manage a revolving fund. We provided the Department of Transportation, FAA, the American Association of Airport Executives, the Airports Council International-North America, the Air Transport Association, and the National Association of State Aviation Officials with a copy of our draft report for review and comment. We met with agency and association officials, including the Director of FAA’s Office of Airport Planning and Programming; the Senior Vice President, Federal Affairs, of the American Association of Airport Executives; the Executive Vice President of the Airports Council International-North America; the Senior Vice President, Federal Affairs and Airports, of the Air Transport Association; and the Vice President of the National Association of State Aviation Officials. The agency and associations generally agreed with the facts presented and provided some clarifying comments and information, which we included in the report as appropriate. Regarding our recommendation for state revolving loan funds, FAA stated that there is strong interest in revolving loan funds among the states and noted that FAA has received inquiries from states interested in pursuing creation of such funds. To develop estimates of various airport funding, we analyzed five different databases maintained by FAA, industry organizations, and private data vendors. We did not audit the accuracy of the databases but did perform some limited cross-checking of information to assess their reasonableness. We then compared these estimates to planned development as reported in our April 1997 report. A more detailed discussion of our data sources and analytical methodology is contained in appendix III. We conducted our work from June 1997 through February 1998 in accordance with generally accepted government auditing standards. As agreed with your offices, unless you publicly announce its contents earlier, we plan no further distribution of this report until 7 days from the date of this letter. At that time, we will send copies of this report to the Secretary of Transportation and the Administrator, Federal Aviation Administration. We will also make copies available to others on request. Please call me at (202) 512-3650 if you have any questions about this report. Major contributors to this report are listed in appendix IV. Funding for airport development comes from five primary sources: federal Airport Improvement Program (AIP) grants, passenger facility charges (PFC), airport and special facility bonds, state grants, and airport revenue. Airports vary in their reliance on these sources of funds. AIP grants are made available from the Airport and Airway Trust Fund.The Federal Aviation Administration (FAA) allocates most AIP grants on the basis of (1) a legislated apportionment formula, tied to the number of passengers an airport enplanes, and (2) set-aside categories earmarked for specific types of airports and projects. FAA has discretionary authority to allocate the remaining funds—about $300 million out of the $1.46 billion made available for fiscal year 1997—on the basis of needs identified by airports. AIP grants peaked in 1992 at $2.264 billion, as measured in 1996 dollars, and declined to $1.372 billion in 1996, as shown in figure I.1. 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 While total AIP funding grew during the program’s first decade and then declined since 1992, the allocation of funds among the various airport categories has been fairly consistent. Over the 15-year period, large hub airports garnered 27 percent of the grants, followed closely by general aviation airports (26 percent) and, to a lesser extent, medium hubs (17 percent), small hubs (14 percent), nonhubs (11 percent), and other commercial service airports (4 percent). Figure I.2 shows the percentage of AIP funding that each category of airport received in each of the last 15 years. AIP funding for hub airports has ranged between 51 and 66 percent of the total. Meanwhile, the funding share for nonhubs has consistently grown, while general aviation airports’ share has generally stayed between 21 and 33 percent of the total. In 1990, the Congress gave commercial airports the option to impose a PFC as an additional means to raise funds for development. Beginning in 1992, authorized airports were able to collect up to $3 per enplaned passenger to use for projects that are eligible for AIP and for certain other types of costs that are not, such as debt financing costs. Airports must apply to FAA for the authority to collect the charges. Large hub airports accounted for two-thirds of all PFC collections in 1996, while medium hub airports accounted for another 24 percent of total collections. Figure I.3 shows PFC collections by category of airport. The single largest category of airport funding is bonds. Even before 1982, airports were fairly sophisticated in their use of debt to finance future development. From 1982 through 1996, airports issued $53.6 billion worth of bonds. Roughly $17.3 billion, or one-third of this total, was to refinance existing debt, while the other $36.3 billion, or two-thirds, was new financing for airports’ capital development. As figure I.4 shows, the total amount of bonding, as well as the split between “refinance” and “new finance,” varies from year to year. For example, when interest rates fell in 1992 and 1993, many airports refinanced their outstanding bonds. The figure also shows that the amount of new finance for airport development increased from less than $1.5 billion in 1982 to more than $3.7 billion in 1996. Since 1982, the vast majority of airport bonds have been issued by large and medium hub airports, nearly $33 billion of a total $36 billion. Figure I.5 shows the distribution of new finance from airport bonds by the category of issuing airport. Despite the $25 billion in new bonds issued by large hub airports from 1982 through 1996, their capacity to issue new debt has not been harmed. As we reported in 1996, large hub airports’ operating ratios did not decline between 1988 and 1994, indicating that revenue kept pace with increased debt service costs. More than 95 percent of all airport debt issued since 1982 has been in the form of general airport revenue bonds (GARB), which are secured by an airport’s future revenue. Thirty years ago, general obligation bonds, which are backed by the taxing power of a governmental unit, were far more common because of their stronger credit standing and therefore lower financing costs. The decline in general obligation bonds reflects the improved acceptance of GARBs by investors. Today, general aviation airports have been the most common issuer of general obligation bonds for airport development. A special category of airport bonds is special facility bonds. While still issued by the airports’ sponsors in order to obtain tax-exempt status, the special facility bonds are secured by the revenue from the indebted facility, such as a terminal, hangar, or maintenance facility, rather than the airports’ general revenue. As figure I.6 shows, the amount of special facility bonds is especially volatile from year to year; they have tended to be issued by large hub airports. 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 The annual amount of special facility bonds is more volatile than that for regular airport bonds because fewer special facility bonds are issued for larger amounts than regular airport bonds; since 1982, 158 special facility bonds averaging $64.7 million have been issued versus 1,181 airport bonds averaging $38.8 million per issue. Nearly all states provide financial assistance to airports, primarily in the form of grants as matching funds for AIP grants or as separate state grants. States fund their grant programs through a variety of sources, including aviation fuel and aircraft sales taxes, highway taxes, bonds, and general fund appropriations. State funding data have been aggregated periodically by the National Association of State Aviation Officials, which began its current annual reporting of state data in 1996. States provided about $285 million to national system airports in the states’ fiscal year 1996.Figure I.7 shows the distribution of those grants by airport category. About 20 percent of state grants, or $57.3 million, was used to supplement AIP grants; the other 80 percent of grants, or $227.4 million, was provided as separate grants. In addition, states provided some funds to airports that are not part of the national system. Total state funding levels vary by state and by year. In the states’ fiscal year 1996, two states—Florida and Maryland—accounted for 45 percent of state grants, while six states provided no grants to national system airports. Also, states offer a slightly greater share of their grants to smaller airports than does the federal government grant program. In 1996, about 56 percent of state grants went to nonhub, other commercial service, and general aviation airports, while only about 42 percent of federal grants went to these same airports. Airports generate revenue from landing fees and terminal leases (both paid by airlines), concessions (such as parking fees), and other income (such as advertising and fuel sales). Airports’ operating revenue supports airports’ operating expenses, debt service costs, and, to the extent available, other nonoperating expenditures, such as capital development. Using airports’ operating revenue to fund development is sometimes referred to as “pay-as-you-go” financing, as opposed to leveraging future revenue to obtain bonds. In addition, to satisfy bond covenants and rating agencies, airports must reserve some portion of their operating revenue to ensure their ability to meet future debt service costs. Beginning in 1996, FAA required commercial airports receiving grants to report financial statement information. Of the 355 commercial airports, 63 percent reported audited financial information, from which we calculated airports’ net operating revenue (operating revenue minus operating expenses) and, in consideration of airports’ debt service and coverage requirements, any portion of net operating revenue in excess of 125 percent of debt service costs (the minimum coverage required by most bond agreements). While not an exact figure for revenue used to support capital development, it is a reasonable estimate of the revenue available for that purpose, according to rating agency officials. Figure I.8 compares airports’ mean net operating revenue in excess of 1.25 times debt service. Figure I.8: Airports’ Available Net Operating Revenue and Operating Ratios, 1996 Available net operating revenue (dollars in millions) Operating ratio (percent) As figure I.8 shows, on average, large and medium hub airports generated modest revenue that can be used for capital development. In addition, large and medium hub airports produced a mean operating ratio, a measure of operational liquidity, twice that of nonhubs and other commercial service airports, which generally operate at or below the break-even point. While operating revenue data for general aviation airports were not available, a recent study for FAA by Gellman Research Associates found that most general aviation airports operate at less than the break-even point, often having to rely on the local municipality for operating subsidies. According to rating agency officials, federal and state grants, PFCs, bonds, and airport revenue make up the vast majority of capital funding sources for airports. Although some local municipalities and outside developers may help finance airport development, little information exists that documents the magnitude or prevalence of these sources. According to FAA officials, local municipalities occasionally provide funding to airports, primarily to smaller airports and primarily as operating rather than capital subsidies. We found only one example of local financing for development. Planned development 1997 through 2001 (annualized) Planned development 1997 through 2001 (annualized) Planned development 1997 through 2001 (annualized) Planned development 1997 through 2001 (annualized) Planned development 1997 through 2001 (annualized) Planned development 1997 through 2001 (annualized) Data on general aviation airports’ revenue are unavailable. Special facility bonding was $250 million for one airport. To determine how much airports of various sizes are spending on capital development and from which sources, we sought data on airports’ capital funding because comprehensive airport spending data are limited and because, over time, funding and spending should roughly equate. We obtained capital funding data from the FAA, the National Association of State Aviation Officials, the Securities Data Company, the Airports Council International-North America, and the American Association of Airport Executives. We screened each of these databases for their accuracy to ensure that airports were correctly classified and compared funding streams across databases where possible. We did not, however, audit how the databases were compiled or test their overall accuracy, except in the case of state grant data from the National Association of State Aviation Officials, which we independently confirmed. We subtotaled each funding stream by year and airport category and added to other funding streams to determine the total funding. With FAA, bond rating agencies, bond underwriters, airport financial consultants, and airport and airline industry associations, we then discussed the data and our conclusions to verify their reasonableness and accuracy. To determine whether current funding is sufficient to meet planned development for the 5-year period from 1997 through 2001 for each airport category and overall, we compared total funding to planned future development as determined in our prior report on airport development. We also compared funding from AIP to higher-priority projects to assess the relative balance between federal funding and their primary intent. Additionally, we correlated each funding stream to the airports’ size, as measured by activity, and among other funding streams to better understand airports’ varying reliance on them and the relationships among sources of finance. We then discussed our findings with FAA, bond rating agencies, bond underwriters, airport financial consultants, and airport and airline industry associations to determine how our findings compared with their knowledge and experiences. Finally, to evaluate how funding shortages, if they were found to exist, might be reduced, we examined several proposals and initiatives for their effect on funding and the relative distribution among different categories of airports. To evaluate the effects of increased AIP funding, we applied the current AIP formula and 1997 distribution percentages to larger funding levels. To evaluate the Air Transport Association’s proposal, we categorized airports and distributions on the basis of 1996 enplanements and operational data. To evaluate the effects of raising the PFC ceiling, we estimated potential PFC collections under $4, $5, and $6 PFCs on the basis of 1996 enplanements and collection rates under two scenarios: (1) with only those airports currently with a PFC imposing one and (2) all airports imposing a PFC. To evaluate various innovative financing proposals, we first identified various proposals and initiatives from discussions with aviation industry associations and FAA and a review of prior studies and legislation. Specifically, the results of FAA’s March 1996 report to the Congress on innovative approaches to the use of federal funds formed the basis for the primary alternatives we considered. We discussed the potential for each of these alternatives with FAA, bond rating agencies, bond underwriters, airport financial consultants, and airport and airline industry associations. We also determined the status of FAA’s pilot programs for innovative financing and privatization. R. Jerry Aiken Paul Aussendorf Beverly Ann Bendekgey Jay R. Cherlow Lynne L. Goldfarb David K. Hooper Joseph D. Kile Kirsten Landeryou Sara Ann Moessbauer Stanley G. Stenersen Randall B. Williamson The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO assessed airports' capacity to finance their future development, focusing on: (1) how much airports of various sizes are spending on capital development and where the money is coming from; (2) whether current funding levels will be sufficient to meet capital development planned for the 5-year period from 1997 through 2001; and (3) the potential effect of various proposals to increase airport funding, if a difference exists between current funding and planned development. GAO noted that: (1) in 1996, the 3,304 airports that make up the national airport system obtained about $7 billion for capital development; (2) more than 90 percent of this funding came from three sources: airport and special facility bonds ($4.1 billion), funding made available from the Airport and Airway Trust Fund ($1.4 billion), and passenger facility charges paid on each airline ticket ($1.1 billion); (3) capital funding more than doubled from 1982 through 1992 and has since declined; (4) airports' 1996 capital funding of about $7 billion is less that the $10 billion per year that airports anticipate will be needed to fund the development planned for 1997 through 2001; (5) while this difference is not an absolute predictor of future funding shortfalls--both funding and planned development may change in the future--it does provide a useful indication of where funding differences may be the greatest; (6) the difference between past funding and planned development is especially acute for smaller commercial and general aviation airports, whose 1996 funding was a little over half of the estimated costs of their planned development; (7) the picture is somewhat brighter if the categories of planned development are narrowed to just those the Federal Aviation Administration (FAA) gives highest priority--that is, safety, security, and noise-mitigation projects and the maintenance of existing airfields; (8) with the exception of the small commercial airports, federal grants in 1996 matched or exceeded the planned development for such projects; (9) several proposals to increase funding for airports have emerged in recent years; (10) these include increasing the size of the federal grant program, raising the ceiling on passenger facility charges, and leveraging existing funding sources; (11) each proposal varies in its magnitude and in its effect on airports and their users; (12) increasing the size of the federal grant program would mostly help smaller airports, while raising passenger facility charges would mostly help larger airports; (13) GAO believes that the FAA's current pilot programs to use grants in more innovative ways and to privatize airports are likely to yield only marginal benefits; (14) however, another means to expand airport investment would be to use federal airport grants to capitalize state revolving funds; and (15) while not a currently permitted use for federal airport grants according to FAA officials, state revolving funds have proved successful in other infrastructure sectors. |
Servicemembers wounded in recent conflicts are surviving injuries that would have been fatal in past conflicts, in part because of advanced protective equipment and medical treatment. However, the severity of their injuries can result in a lengthy transition from patient status back to active duty or to veteran status. Most severely wounded servicemembers from the conflicts in Iraq and Afghanistan initially are evacuated to Landstuhl Regional Medical Center in Germany for treatment. From there, they are usually transported to military treatment facilities in the United States, with most of the severely wounded admitted to Walter Reed Army Medical Center, the National Naval Medical Center, or Brooke Army Medical Center. Acute medical treatment and stabilization is the first of three phases in the “continuum of care” experienced by severely wounded servicemembers. The second phase of the continuum is rehabilitation at a DOD, VA, or civilian facility. (The recovery needs of some servicemembers receiving rehabilitation may require their return to a medical center for acute medical care, such as surgical procedures.) The third phase of the continuum is reintegration—either return to active duty or to the civilian community as a veteran, where they may receive health care from DOD, VA, or civilian providers. From January 2008—when FRCP enrollment began—through September 2010, the FRCP provided services to a total of 1,268 servicemembers and veterans. As of September 2010, the program had 607 active enrollees, ranging in age from 19 to 61 years, with a median age of 27 years. About half of the enrollees were or had been married. Fifty-eight percent had designated another person as his or her primary caregiver, and 38 percent had delegated legal authority to another person. (See table 1 for additional demographic information about current FRCP enrollees.) FRCs are senior-level registered nurses and licensed social workers whose principal role is to coordinate services with case managers rather than provide services directly to enrollees. FRCs are expected to serve as the single point of contact for the enrollees and their families and to assist the enrollees in a number of ways. FRCP care coordination guidelines identify FRC activities, which are outlined in table 2. According to FRCP policy, the FRC’s primary responsibility is to develop and monitor progress of each enrollee as detailed in that person’s Federal Individual Recovery Plan, which is created and implemented by the FRC with input from the enrollee and his or her family and clinical team. This plan is to be a comprehensive, client-centered plan that sets individualized goals for recovery and is intended to guide and support the enrollee through the continuum of care. FRCs update Federal Individual Recovery Plans to reflect changing conditions or enrollee goals. Based on their diagnoses and other factors, enrollees are likely to require a complex array of clinical and nonclinical services from multiple providers and facilities. (See table 3.) In providing care coordination services, the FRC may engage with an enrollee’s health care providers, other care coordinators, and case managers, such as those with the military services’ wounded warrior programs. As care coordinators, FRCs are generally not expected to directly provide the services needed by enrollees. However, FRCs may provide services directly to enrollees in certain situations, such as when they cannot determine whether a case manager has taken care of an issue for an FRCP enrollee, when asked to resolve complex problems, or when making complicated arrangements, for example, identifying and arranging admission to a substance abuse treatment program for a veteran who was beginning to develop violent behaviors and had refused to complete a VA drug rehabilitation treatment program. It is unclear whether all of the eligible “severely wounded, ill, and injured” servicemembers and veterans who could benefit from the FRCP are being enrolled in the program. The FRCP cannot readily identify these individuals because the “severely wounded, ill, and injured” classification is not captured in existing data sources. Additionally, the program’s broad eligibility criteria cannot be used systematically to identify potentially eligible servicemembers and veterans. Instead, the FRCP must rely on referrals from others to identify these individuals, although the program has also taken steps to identify potential enrollees through the FRCs’ efforts at medical facilities and through a “look back” initiative to identify eligible veterans who were wounded prior to program implementation. In addition, the FRCs must exercise judgment in applying the program’s criteria for enrollment determinations, and FRCP leadership does not systematically review these decisions to ensure that these criteria are applied appropriately so that referred individuals who could benefit from the program are enrolled, and that individuals who could be served by less intensive services are referred to other programs. FRCP officials have experienced difficulties in identifying the potentially eligible population of “severely wounded, ill, or injured” servicemembers and veterans, and as a result, it is unclear whether all of these individuals who could benefit from care coordination services are enrolled in the program. The Senior Oversight Committee, which created the FRCP, developed a three-level care categorization system to differentiate the population of wounded servicemembers and veterans for different programs based on the severity of their conditions. In this system, Category 1 servicemembers are those with mild wounds, illnesses, or injuries who are expected to return to duty in less than 180 days; Category 2 servicemembers are those with serious wounds, illnesses, or injuries who are unlikely to return to duty in less than 180 days and possibly may be medically separated from the military; and Category 3 servicemembers are severely wounded, ill, or injured individuals whose medical conditions are highly likely to prevent their return to duty and also likely to result in medical separation from the military. Individuals who fall under category 3 may be considered for enrollment into the FRCP, while individuals falling under categories 1 or 2 may qualify for other types of programs. However, according to the FRCP Executive Director, these are administrative categories that are not captured in existing VA or DOD medical or benefits data systems or included in medical or benefits records. As a result, the FRCP cannot use this classification to systematically identify the population of potentially eligible severely wounded servicemembers and veterans using available data sources. In addition, the FRCP Executive Director and FRCs told us that the broad eligibility criteria developed for the FRCP must be used on a case-by-case basis to identify potentially eligible individuals for the program because these criteria require some judgment. Therefore, the criteria cannot be used systematically to identify the program’s potentially eligible population. These criteria include both specific medical diagnoses and requirements that are somewhat subjective, such as whether an individual may benefit from a recovery plan. To decide whether potential enrollees may benefit from a recovery plan, FRCs reported that they evaluate the complexity of a situation by examining issues such as future medical needs, family dynamics, and any financial or legal problems— information that is not readily available in any one data source. As a result, to identify potentially eligible individuals, the FRCP relies on referrals from others, including program officials and medical facility staff. Sources of referrals include, for example, wounded warrior program staff, Recovery Care Coordinators, and clinical treatment teams. Of the program officials and medical facility staff we spoke with who discussed referrals, more than half (25 of 47) had made a referral to the program. However, more than half (15 of 27) of the program officials and medical facility staff we interviewed who responded to questions on eligibility also felt that the FRCP eligibility criteria were unclear. In addition to relying on referrals, the FRCs also take steps to identify potential enrollees. Some FRCs stated that they review their facility’s list of incoming severely wounded servicemembers and attend weekly multidisciplinary team meetings where hospital officials and medical staff discuss severely wounded patients’ cases. In an attempt to ensure that eligible veterans who were wounded prior to the program’s inception are enrolled in the program, the FRCP conducted a “look back” initiative in May 2010. Because no single data source contains sufficient information, the FRCP Executive Director told us that she combined five DOD and VA data sets and used multiple “proxy” factors to narrow the data from 40,000 veterans’ records to the final list of potentially eligible veterans. For example, the Veterans Benefits Administration’s 100 percent disability compensation list and medical diagnostic codes were used to help identify this population. Based on this analysis, the FRCP Executive Director reported that the program contacted approximately 300 potential enrollees to determine whether they could benefit from an FRC’s assistance. As a result, 35 of those severely wounded veterans will be further evaluated for potential enrollment. According to the FRCP Executive Director, this analysis was prioritized to focus on severely wounded veterans who were most likely to need FRC assistance. The Executive Director told us that, as a result, the list was not comprehensive—for example, the program did not contact veterans who were already enrolled in VA’s OEF/OIF Care Management Program under the assumption that they were already receiving adequate case management. Additionally, the FRCP Executive Director told us that identifying 35 veterans indicated that the FRCP is not reaching all potentially eligible veterans through its normal referral process or that information about the program is not reaching severely wounded veterans. The FRCP Executive Director added that once it is complete, this effort will be assessed to determine whether another “look back” is needed, but as of February 2011, leadership officials had not yet determined whether they would conduct a subsequent “look back.” Following the identification of potentially eligible servicemembers and veterans, FRCs use a more thorough application of the program’s eligibility criteria to evaluate these individuals for enrollment. The eligibility criteria are broad and require FRCs to exercise judgment with their enrollment decisions. However, FRCP leadership does not systematically review these decisions to ensure that referred individuals who could benefit from the program are enrolled while those requiring less intensive services are referred to other programs. Eligibility criteria for the program—developed by the Senior Oversight Committee—specify that enrollees be receiving acute care in a military treatment facility; be diagnosed or referred for one or more of the following: spinal cord injury, burns, amputation, visual impairment, traumatic brain injury, or post-traumatic stress disorder; be considered at risk for psychosocial complication; or may benefit from a recovery plan. Because some of these criteria are subjective, particularly whether an individual is at risk for psychosocial complications or would benefit from a recovery plan, the FRCs must use their judgment when deciding whether an individual should be enrolled in the program. According to the FRCP Executive Director, the program’s criteria are intended to provide guidance for the FRCs, giving them the flexibility to enroll severely wounded servicemembers and veterans, rather than being restrictive. The Executive Director added that FRCs strive to enroll severely wounded servicemembers and veterans in cases where having an FRC can add value to existing case management efforts. To evaluate servicemembers and veterans for program eligibility, FRCs must make subjective assessments of the impact their care coordination efforts could have on potential enrollees. This involves FRCs making assessments of the severity of potential enrollees’ medical conditions to determine future medical needs—such as rehabilitation—and nonmedical issues—such as caregiver status. FRCs obtain information from a number of sources, including DOD and VA medical records, as well as records from private sector providers. They may also discuss potential enrollees’ situations with members of multidisciplinary teams providing medical treatment, family members, and the potential enrollees. At the end of the evaluation period, the FRC will consider a potential enrollee’s need for care coordination based on the collected information and determine whether the individual should be enrolled in the program, provided temporary assistance, or referred to another program. While it is necessary for FRCs to use their judgment in making enrollment decisions, the FRCP does not systematically review the factors and reasons for enrolling, providing temporary assistance, or referring potentially eligible servicemembers and veterans to other programs. Systematic review could involve the use of a defined protocol for the review of eligibility decisions made by FRCs. According to federal internal control standards, agencies should establish ongoing internal control activities to provide reasonable assurance that decisions are consistent with applicable criteria—in this case, criteria designed to ensure that those in need of care coordination services are enrolled in the program. While the FRCs indicate in their data management system—the Veterans Tracking Application—whether they decided to enroll an individual, FRCP leadership told us they do not require that the FRCs record the factors they considered to support this decision. Additionally, FRCP leadership told us that while they closely review all enrollment decisions made by new FRCs, they do not perform similar reviews of decisions made by more experienced FRCs. Instead, FRCP leadership and experienced FRCs discuss the FRCs’ recommended actions on newly referred individuals as part of weekly telephone conversations. However, FRCP officials acknowledged that these discussions with the FRCs may not be comprehensive and that there is no section in the Veterans Tracking Application dedicated to recording these discussions. Without specific documentation of the factors the FRCs considered when making their enrollment decisions and absent internal controls and systematic oversight of much of the enrollment process, it is difficult to determine whether severely wounded servicemembers and veterans who are referred and could benefit from the program are actually enrolled and severely wounded servicemembers and veterans who could be served by less intensive services are referred to other programs. Additionally, this issue could become even more problematic as the program’s enrollment continues to increase and FRCP leadership has to review more enrollment decisions. Several challenges confront the FRCP in determining staffing needs for the program, including how to manage FRCs’ caseloads, deciding when VA should hire FRCs, and determining where to place them in the field to best serve current and potential enrollees. The FRCP has not established a formal caseload size for FRCs because there are no comparable criteria upon which to determine caseload size because of the program’s unique care coordination activities. Also, while establishing an appropriate caseload size for FRCs may help FRCP leadership determine how many FRCs VA should hire, it remains difficult for FRCP leadership to determine when VA should hire FRCs. Finally, the FRCP lacks a clear and consistent rationale for making decisions about where to place FRCs in the field. The FRCs we spoke with expressed concerns about the high number of enrollees assigned to them and cited the need for improved caseload management. Specifically, 11 of the 15 FRCs we interviewed identified inadequate caseload management as a concern. Eight of these FRCs expressed concerns about the large number of cases assigned to them. As of September 30, 2010, FRCs’ caseloads ranged from 25 to 48, with two- thirds of the FRCs (10 of 15) having caseloads that exceeded the informal target ratio of 1 FRC for every 30 enrollees established by the FRCP Executive Director to manage FRC caseloads. Some FRCs told us that the large number of cases required them to work long hours and sometimes forced them to limit the amount of time that they could devote to an enrollee. In addition, more than half of the FRCs (8 of 15) expressed concerns that FRCP leadership does not adequately account for the services required by existing enrollees in their caseloads when assigning new cases. For example, one FRC told us that the types of cases assigned to her were stressful. She indicated that she had been assigned two enrollees with terminal conditions because she was skilled at managing the issues related to these types of cases, but she is now reluctant to take another terminally ill enrollee because it is emotionally draining to deal with end-of-life issues. However, an FRCP leadership official told us that FRCs have the flexibility to forward a referral to the FRCP central office for assignment to another FRC as a means of managing their existing caseloads. According to the FRCP Executive Director, an appropriate caseload is difficult to determine because care coordination is a new type of function, and there are no comparable criteria against which to measure and base caseload size for this program because of its unique activities. Additionally, the FRCs’ caseloads are dynamic in that the needs of each enrollee differ and may change over time. For example, out of a caseload of 30 clients, 5 may need intensive crisis management, while the remaining 25 enrollees may only need periodic contact or limited services. However, as noted by FRCP leadership and some FRCs, the needs of these enrollees, and consequently, the time required of an FRC, may change as enrollees move through different stages of the continuum of care. As a means of managing FRCs’ caseloads, the FRCP Executive Director cited two actions in particular that FRCP leadership uses to assess and manage FRC caseloads. FRCP leadership uses an informal FRC-to-enrollee target ratio of 1 to 30 (with a targeted range of 25 to 35 enrollees per FRC), which is based on the FRCP Executive Director’s experience in managing the program over time. Weekly telephone calls with each FRC are used by FRCP leadership to discuss issues related to their assigned cases and to gauge workload burden. The FRCP Executive Director told us that the program is developing a customized workload assessment tool to help balance FRCs’ caseloads— in other words, to ensure that an FRC’s caseload mix is manageable. The objective of the workload assessment tool is to identify specific enrollee characteristics, such as medical diagnosis, and to correlate each characteristic with the amount of time an FRC would be required to spend on addressing issues related to it. One method being considered is the assignment of a point value to each identified enrollee characteristic. Adding up the number of points for the characteristics of all enrollees in an FRC’s caseload would provide an estimate of that FRC’s workload burden. However, according to the FRCP Executive Director, the development of such a tool has been difficult, primarily because the enrollee characteristics that existing workload assessment tools use to determine how much time it takes to address an issue are not relevant to the care coordination activities that FRCs perform. As a result, program leadership continues to consider different methods of assessing FRCs’ workloads, including measurement tools that have already been validated for other purposes, to identify a method that could potentially be relevant for the program. The FRCP Executive Director is uncertain how long it will take to develop a workload assessment tool and has not established timelines to complete this effort. Without a workload assessment tool, the program does not have the data it needs to develop a more comprehensive caseload management strategy and to better determine appropriate caseload size for FRCs. While establishing appropriate FRC caseloads should help FRCP leadership better determine how many FRCs VA should hire, determining when VA should hire FRCs has been another staffing challenge. Currently, the FRCP Executive Director’s decisions about when VA should hire FRCs are based on various ongoing monitoring efforts. The FRCP Executive Director told us that staffing decisions regarding FRCs are difficult to make because the FRCP cannot predict the number of potentially eligible servicemembers and veterans, which is affected by the OEF/OIF conflicts. In the absence of being able to project the number of potentially eligible servicemembers and veterans, the FRCP Executive Director said she uses other methods to predict future trends and guide the staffing process. One method involves monitoring FRCs’ workloads as an indicator that workload levels are increasing and new FRCs are needed. In this regard, the FRCP Executive Director told us that FRCP leadership conducts weekly telephone calls with each FRC to discuss issues related to their caseloads. The FRCP Executive Director told us that another method she uses to predict staffing needs is through the analysis of the number of new referrals and enrollment rates in the program, which she uses to create a quarterly report that highlights the projected number of FRCs that the program may need. For example, the average number of new referrals grew from 25 a month in 2008 to 35 a month in 2009. VA hired five FRCs in January 2010 in part because of this increase in the number of referrals and the expected resulting increase in the number of enrolled servicemembers and veterans. The FRCP Executive Director told us that the referral data collected in 2010 show that the number of new referrals continued to increase and averaged 50 a month, which indicates a continuing need for more FRCs. According to the FRCP Executive Director, she routinely shares this information with the Secretary of Veterans Affairs as advance notice that a request for additional FRCs may be forthcoming because it takes about 6 months for VA to hire a new FRC. The FRCP Executive Director told us that the program’s ongoing monitoring efforts are the most logical approach for determining when and how many FRCs VA should hire in the absence of knowing the number of potentially eligible servicemembers and veterans. While these methods appear to be reasonable given the lack of overall data on the numbers of severely wounded servicemembers and veterans, the staffing process is not well documented. Internal control standards applicable to all federal agencies state that an agency should effectively communicate its policies and procedures by providing clear documentation that is readily available for examination. Consistent with this internal control standard, we would expect the FRCP to have documented procedures outlining its process for making staffing decisions. FRCP leadership documented staffing projections for fiscal year 2010 in the program’s annual operating plan, citing that ongoing analysis of referrals and enrollment rates was important in making those projections. However, the process used by program leadership—specifically how the referral and enrollment data are used in making staffing decisions—has not been clearly defined or documented in the operating plan or any of the other program policies or procedures. By documenting this information, the FRCP would have greater assurance that the process developed by the current leadership will be maintained during management changes. Deciding where to place FRCs to best serve current and potential enrollees’ needs is another key staffing issue, despite the fact that FRCs often coordinate services for enrollees who are located throughout the country and may not be receiving care at the facility where their assigned FRC is located. The FRCP’s basis for making decisions about where to place FRCs has varied over time, and the program currently lacks a clear and consistent rationale for making FRC placement decisions. As of September 2010, 20 FRCs were located at 10 facilities. (See fig. 1.) When the FRCP began operating in 2008, eight FRCs were placed at the three military treatment facilities where the majority of severely wounded servicemembers were receiving treatment. According to the FRCP Executive Director, the placement of FRCs at military treatment facilities helped with the identification of servicemembers who could benefit from FRCP services. In addition, some FRCs told us that being located at the military treatment facilities allowed them to develop relationships with the enrollees, their families, and the case managers who would be providing direct services to the enrollees. However, as the program expanded, placement of some FRCs was not based on a rationale or an analysis of where FRCs could provide the maximum benefit to severely wounded servicemembers and veterans. For example, some DOD and VA officials we spoke with expressed concerns about the FRCP’s placement decisions, particularly the placement of FRCs at facilities that do not treat a large population of severely wounded servicemembers or veterans. DOD officials told us that it was not clear why there were FRCs assigned to a military treatment facility that typically does not treat severely wounded servicemembers. Similarly, a VA medical center official stated that it was unclear why FRCs were initially placed at two VA medical centers that had few FRCP enrollees being treated there, rather than at VA medical centers where a significant number of severely wounded veterans may be receiving treatment. There was no official FRCP documentation that explained the basis for these decisions, which were made by FRCP officials who are no longer with the program. After the FRCP leadership changed in July 2008, decisions to place FRCs have been based on several factors. According to the FRCP Executive Director, some placement decisions focused on ensuring that enough FRCs were in place to meet the demands of the FRCP workload by replacing FRCs who had left the program and by adding FRCs at facilities where only one FRC was located. She explained that where possible, it is helpful to have at least two FRCs at each facility so that there can be backup support, particularly for administrative purposes such as coverage, when an FRC is on leave. However, the FRCP Executive Director told us that more recently—from March 2010 through September 2010—FRC placement decisions have primarily been based on requests or recommendations from DOD and VA officials. For example, in June 2010, the FRCP relocated an FRC to a military wounded warrior program headquarters facility in response to a request from the program’s director. FRCP officials have also decided to place some new FRCs at two VA medical centers where servicemembers and veterans with polytrauma injuries receive care, based on recommendations from DOD and VA officials. The FRCP Executive Director explained that the FRCP had not established a systematic rationale for FRC placement because the program initially lacked the data upon which to base these determinations. Additionally, she told us that every placement of an FRC at a VA or DOD facility is a negotiation and depends on the facility’s ability to accommodate an FRC, including the provision of work space and equipment. However, she told us that she and other FRCP leadership officials have begun to think about how to improve the FRCP’s process for deciding where to place FRCs. In August 2010, the FRCP Executive Director explained that a planned update of the Veterans Tracking Application would collect additional information that would allow FRCP officials to identify the location of individuals who refer potential enrollees. She anticipates being able to use these data to identify the locations and facilities where the most referrals are being made. According to the FRCP Executive Director, this information along with other factors, such as placement recommendations from DOD and VA officials, could be used in making future placement decisions. However, as of December 2010, she had not established a specific time frame for this effort. Developing a clear and consistent rationale for placing FRCs, which includes a systematic analysis of program data, should help ensure that FRCs are located where they could provide the maximum benefit to current and potential enrollees. FRCs and others identified challenges that can limit the FRCP’s efforts to coordinate the services needed by severely wounded servicemembers and veterans. One challenge involves limitations on the FRCP’s ability to share information with the large number of programs that provide care coordination and case management services to wounded servicemembers and veterans. These limitations—which are the result of restrictions on the disclosure of enrollee information and data systems’ incompatibility— have sometimes resulted in confusion and the duplication of services for enrollees. Efforts by the FRCP to improve information sharing are ongoing. Another challenge is that FRCs often have difficulty obtaining resources from the facilities at which they are located—such as telephones, computers, and private office space—that they need to perform their care coordination activities, including communicating with enrollees across the country. This can affect the quality of services to enrollees, and the FRCP is working to resolve these logistical issues. Coordination among DOD and VA programs that provide care coordination and case management is difficult because of the large number of such programs that exist to address the needs of wounded servicemembers and veterans and the limitations in the ability of these programs to share information. Although these programs vary in terms of the severity of the injuries among the servicemembers or veterans they serve and the specific types of services they coordinate, many programs have similar functions. (See table 4.) For the purposes of this table, we have categorized the severity of enrollees’ injuries according to the injury categories established by the DOD and VA Wounded, Ill, and Injured Senior Oversight Committee. Servicemembers with mild wounds, illness, or injury are expected to return to duty in less than 180 days; those with serious wounds, illness, or injury are unlikely to return to duty in less than 180 days and possibly may be medically separated from the military; and those who are severely wounded, ill, or injured are highly unlikely to return to duty and also likely to medically separate from the military. These categories are not necessarily used by the programs themselves. The military wounded warrior programs are the Army Wounded Warrior Program, Marine Wounded Warrior Regiment, Navy Safe Harbor, Air Force Warrior and Survivor Care Program, and Special Operations Command’s Care Coalition. An FRC placed at Special Operations Command’s Care Coalition headquarters coordinates clinical and nonclinical care for Care Coalition and other FRCP enrollees. OEF/OIF refers to Operation Enduring Freedom and Operation Iraqi Freedom. An OEF/OIF care manager supervises the case managers and transition patient advocates and may also maintain a caseload of wounded veterans. According to VA, in some instances, patients are transferred to VA medical facilities while still in the acute phase of the care continuum and may receive services from VA care management or polytrauma program staff. Many recovering servicemembers and veterans are enrolled in more than one program. For example, in September 2010, approximately 84 percent of FRCP enrollees were also enrolled in a military service wounded warrior program. According to one FRC, his enrollees have, on average, eight case managers who are affiliated with different programs. Individuals enrolled in multiple programs may have recovery plans or goals that have been developed by different programs. Moreover, some case managers of other programs consider themselves to be the single point of contact for their enrollees, even those enrolled in the FRCP. Because the majority of FRCP enrollees are enrolled in more than one program, there is a high likelihood that without adequate information exchange and coordination, FRCs and case managers could duplicate one another’s efforts, confuse enrollees and families, waste resources, or mistakenly believe that someone else has taken care of a task for an enrollee. The extent of overlap and the lack of information sharing by the FRCP have prompted some programs to limit FRCs’ involvement with servicemembers when they are receiving initial medical treatment at a military treatment facility. At two of the military treatment facilities we visited, for example, a military program serving wounded servicemembers delays referrals to the FRCP until a servicemember approaches the point when he or she is preparing to transition to another facility or VA. Prior to January 2011, VA had not completed public disclosure actions necessary to enable the sharing of information from the Veterans Tracking Application, the information system used by the FRCP that contains each enrollee’s personal information and Federal Individual Recovery Plan. As a result, VA management had advised the FRCP that the program could not provide staff of non-VA programs (such as those affiliated with DOD) with its enrollees’ personally identifiable information, such as names, addresses, Social Security numbers, and details of Federal Individual Recovery Plans. Specifically, VA had not completed the System of Records Notification process for the Veterans Tracking Application, a process required by the Privacy Act of 1974 that requires federal agencies to publish in the Federal Register a notice of the existence, purpose, and routine uses of every “system of records” that contains information that may be linked to individuals. Although this limitation did not prevent FRCs from performing their care coordination responsibilities, it has been a source of frustration for others. Specifically, officials of several of DOD’s wounded warrior programs contend that the inability to receive enrollment information from the FRCP has caused difficulties. The director of one program, for example, told us that not having the names of servicemembers enrolled in the FRCP resulted in a situation in which an FRC and a wounded warrior program Recovery Care Coordinator were not aware that the other was involved in coordinating care for the same servicemember and had unknowingly established conflicting recovery goals for this individual. In this case, a servicemember with multiple amputations was advised by his FRC to separate from the military in order to receive needed services from VA, whereas his Recovery Care Coordinator set a goal of remaining on active duty. These conflicting goals caused considerable confusion for this servicemember and his family. Furthermore, leadership officials of two of the military services’ wounded warrior programs told us that they have instructed their staff not to make referrals to the FRCP to avoid confusion and potential duplication of activities, citing issues associated with information sharing. In August 2010, prompted by the FRCP, VA initiated the public-disclosure process to facilitate information sharing. In December 2010, VA published a notice in the Federal Register that describes the compilation of information in the Veterans Tracking Application and routine uses of that information. VA received no comments on the notice during the public comment period, which ended on January 10, 2011. The new system of records became effective on that date and the FRCP was able to share certain enrollee information, such as the names of enrollees, with DOD programs. Another factor that limits information sharing is the inability of the information systems used by the FRCP, the DOD Recovery Coordination Program, and the five military services’ wounded warrior programs to exchange information directly with one another. As a result, FRCs cannot readily access information from data systems used by case management programs about their enrollees and information about an individual cannot be easily transferred among systems. To help address this issue, the FRCP has spearheaded an effort, known as the Information Sharing Initiative, to identify an approach for the direct exchange of information between DOD and VA care coordination and case management information systems in the future. The FRCP Executive Director explained that this initiative primarily includes identifying the data that need to be exchanged as well as identifying the data systems where these data originate and subsequently developing a technical solution to electronically exchange this information. Further, she noted that the Information Sharing Initiative is a grassroots effort and that work on the initiative has been performed by DOD and VA employees in addition to their normal duties, making a completion date difficult to estimate. An official from the Interagency Program Office, which oversees major information technology initiatives jointly undertaken by DOD and VA, said that the Information Sharing Initiative was a well-considered initial step but notes that the ultimate goal of direct information exchange among programs’ information systems faces daunting challenges, such as resolving conflicting DOD and VA policies pertaining to information exchange. We have previously reported on DOD’s and VA’s efforts to electronically exchange health care information, including the departments’ progress toward increasing their capabilities to share medical and nonmedical history and physical exam data. We have found that despite the departments’ progress, their efforts to meet clinicians’ evolving needs to exchange health information and to create a single lifetime electronic record for each servicemember, which is intended to streamline the transition of electronic records between the two departments, are ongoing. Recognizing that these limitations on information sharing exist, the FRCP is also taking steps to emphasize FRCs’ principal role of coordinating with case managers rather than providing services to enrollees themselves, which should help prevent unintentional duplication of effort. Because FRCs may provide a direct service in some instances, proper information sharing is necessary so that staff from multiple programs may not unknowingly perform the same task for an enrollee. For example, an FRC told us that in one instance there were five case managers working on the same life insurance issue for an individual. According to the FRCP Executive Director, the Federal Individual Recovery Plan process has been improved to encourage coordination by FRCs and also to reinforce their primary role as care coordinators. To accomplish these objectives, in January 2011 the FRCP upgraded the Veterans Tracking Application, in which Federal Individual Recovery Plans are maintained, by adding a record of the names of the case managers who are responsible for completing activities linked to enrollees’ planning goals. In addition, the Veterans Tracking Application began displaying indicators to inform each FRC about the completion status of every goal-related activity planned for each enrollee, based on the completion dates that the FRCs put into the system. The FRCP Executive Director believes that such an indicator system, when linked to the names of the case managers who are responsible for completing the activities, will reinforce the FRCs’ care coordination role by encouraging them to actively follow up with others on the status of individual tasks rather than taking on these tasks themselves. FRCs and others identified several types of logistical problems that have affected the FRCs’ ability to carry out their responsibilities in dealing with FRCP enrollees and coordinating with wounded warrior programs. These issues center around three specific areas: provision of equipment (such as computers, printers, landline telephones, and BlackBerrys), technology support (such as equipment maintenance, software upgrades, and systems security), and private work space at the medical facilities. Provision of equipment. Most of the FRCs’ work is done using computers, accessing data management systems, and communicating with enrollees and DOD and VA facility staff by e-mail and phone. However, about half of the FRCs told us that they have been hindered in their ability to perform their care coordination responsibilities by the lack of appropriate technology resources at the facilities at which they work. Some FRCs expressed frustration with delays in obtaining appropriate computer or communications equipment when they first reported to their facilities, and this experience was echoed by nearly all of the FRCs hired in January 2010. For example, one FRC said she waited more than 6 weeks at the facility to receive a DOD computer and landline telephone. Another FRC reported that he has found that e-mail is an effective mode of communication with enrollees with traumatic brain injuries because he can provide detailed instructions to them, but when he was hired he did not receive a DOD computer and a landline telephone with long-distance calling capability for 8 months. Consequently, he had to resort to mailing letters and brochures to current and potential enrollees. Technology support. In addition to the lack of equipment, some FRCs cited the lack of technology support as a factor that hindered their care coordination activities. Technological support includes functions such as connectivity to information systems, installing security systems, and equipment upgrading and repair. An FRCP deputy director told us that the lack of such support is often experienced by new FRCs, but it is also an ongoing issue for many, especially after a facility computer system is upgraded and the FRCs’ equipment becomes incompatible. Additionally, several FRCs have had difficulty with their BlackBerrys, either because the facility was unable to install a security patch needed to access e-mails or because poor reception made the device unusable. Some FRCs also reported their inability to access DOD medical records (although this issue is beyond the scope of a single program to address)—for example, FRCs located at VA medical centers must ask FRCs at military treatment facilities to access enrollees’ DOD records and then fax them to the FRCs at the VA medical centers. Finally, FRCP officials noted that equipment repair has been a problem for some FRCs—one FRC told us that she had to use a malfunctioning laptop computer issued to her by the local VA medical center for 8 months. Work space. Some FRCs noted that they had been assigned work space at the facility that was unsuitable for conducting sensitive conversations with enrollees, family members, and coworkers. At a major medical center, we observed that FRCs were located in tightly spaced cubicles that allowed nearby staff to easily overhear their conversations. A recently transferred FRC told us that when she arrived at her new medical center, she found that she had no office and had been located in an open room that serves as the call center for triage nurses. Lacking the privacy needed to make confidential calls to her enrollees, this FRC resorted to making sensitive phone calls from her car in the parking lot. At another treatment facility, an FRC who shared an office with staff from another program had to take phone calls with enrollees in the stairwell in order to have privacy. Finally, two recently hired FRCs were not only placed in the same office but also had to share the same desk. The provision of equipment, technology support, and work space is covered by memoranda of agreement between the FRCP and the DOD and VA facilities where FRCs are located. However, an FRCP deputy director told us that obtaining compliance with the memoranda of agreement at some facilities is an ongoing challenge and that equipment maintenance and systems upgrades are persistent issues for all FRCs. In some instances, after FRCs had made repeated requests for needed resources without result, the FRCP Executive Director intervened with medical center officials or through the Senior Oversight Committee to obtain a resolution. A leadership official for a wounded warrior program told us that some military medical centers have difficulty satisfying requests for equipment and space from programs such as the FRCP because these facilities house and support various DOD and VA support programs and all make requests for resources. This official pointed out that at one military treatment facility, a military case manager was relocated in order to make an office available to an FRC. An FRCP deputy director added that given the frequent turnover of military staff, medical center officials are sometimes unaware that their facility is responsible for providing resources and services to FRCs. FRCP officials reviewed existing memoranda of agreement between the FRCP and DOD and VA medical facilities to determine where improvements could be made to ensure that the FRCs have the tools and privacy required to do their work. The program has developed three new templates for memoranda of agreement that will be used when FRCs are located in new settings: one each for military treatment facilities, VA medical centers where servicemembers and veterans with polytrauma injuries receive care, and military wounded warrior programs. These new memoranda are more detailed than the previous versions, and they identify who is responsible for providing specific resources and services. The FRCP is using the revised agreements in its negotiations for logistical support for newly placed FRCs at two VA medical centers and with the Special Operations Command wounded warrior program. Following implementation of the new memoranda of agreement, the FRCP plans to revise existing agreements to make them consistent with the newer versions, but no specific timetable has been established to complete these revisions. Since its inception, the FRCP has increased the number of enrollees, enlarged its staff considerably, and expanded the number of locations where FRCs are assigned. However, the program faces significant challenges as it matures. As the first joint care coordination program for DOD and VA, the FRCP represents a new paradigm in patient support for the departments. Because of its unprecedented nature, the program cannot refer to preexisting data or policies and procedures to manage the program, and as a result, FRCP leadership had to develop management processes as the program was being implemented and has largely relied on informal processes to oversee and manage key aspects of the program. However, now that the program has been operating for several years and continues to grow, it has become apparent that the program would benefit from more definitive management processes to strengthen program oversight and decision making. While the program has overcome some early setbacks and has established processes related to enrollment and staffing, these processes are not clearly documented or systematic. Because enrollment decisions are not well documented or systematically reviewed by FRCP leadership, it is unclear whether referred servicemembers and veterans who need FRC services are being enrolled in the program. Additionally, as the number of individuals enrolled in the program steadily increases, it will be important for the FRCP to appropriately balance FRCs’ workload to ensure that enrollees receive the services they need and to prevent FRC burnout. While program leadership recognizes this issue and is developing a customized workload tool, there is no firm timeline for the completion of this effort. The FRCP also needs clearly documented processes and criteria for guiding staffing and placement decisions. Without this, it will be difficult to provide continuity to subsequent program leadership and to place FRCs where they would best serve the needs of current and future enrollees. Some of the daunting challenges facing FRCs and the program are beyond the capability of the program’s leadership to resolve. The exchange of information among DOD and VA data systems, in particular, has been a long-standing issue and will require interdepartmental action. Similarly, the duplication of effort resulting from the proliferation and overlap of DOD and VA programs that support recovering servicemembers and veterans can best be resolved through interdepartmental coordination and action. We recommend that the Secretary of Veterans Affairs direct the Executive Director of the FRCP to take four actions: 1. Ensure that referred servicemembers and veterans who need FRC services are enrolled in the program by establishing adequate internal controls regarding the FRCs’ enrollment decisions. To accomplish this, the FRCP leadership should require FRCs to record in the Veterans Tracking Application the factors they consider in making an enrollment decision, develop and implement a methodology and protocol for assessing the appropriateness of enrollment decisions, and refine the methodology as needed. 2. Complete development of the FRCP’s workload assessment tool that will enable the program to assess the complexity of services needed by enrollees and the amount of time required to provide services to improve the management of FRCs’ caseloads. 3. Clearly define and document the FRCP’s decision-making process for determining when and how many FRCs VA should hire to ensure that subsequent FRCP leadership can understand the methods currently used to make staffing decisions. 4. Develop and document a clear rationale for the placement of FRCs, which should include a systematic analysis of data, such as referral locations, to ensure that future FRC placement decisions are strategic in providing maximum benefit for the program’s population. DOD and VA each provided comments on a draft of this report. In its comments, DOD stated that it continues to work with VA to fully integrate their efforts and to increase collaboration between the two departments. (DOD’s comments are reprinted in app. II.) In its comments, VA stated that it generally agrees with GAO’s conclusions and concurs with our recommendations to the Secretary. (VA’s comments are reprinted in app. III.) VA’s responses to each of our recommendations are as follows: To ensure that referred servicemembers and veterans who need FRC services are enrolled in the program, VA indicated that the FRCP will document decisions and factors used to assess a potential enrollee’s eligibility for the program. In addition, the program will establish clear documentation requirements according to a defined protocol within the program’s data management system. To complete the development of the FRCP’s workload assessment tool, VA indicated that the FRCP will continue field-testing a new assessment tool, which will require at least a year to complete. To document the decision-making process for determining when and how many FRCs VA should hire, VA stated that the FRCP will clearly document the current process used for making staffing decisions. In addition, the staffing processes and plans will be updated annually in the FRCP business operation planning document. To develop and document a clear rationale for the placement of FRCs, VA indicated that the FRCP will develop an FRC placement strategy based upon a systematic analysis of data over the next 6 months. This process will be documented and updated annually in the FRCP business operation planning document. VA provided an additional comment regarding the progress made toward the exchange of data between VA and DOD’s wounded warrior information systems. VA stated that it anticipates that an initial set of data will be available for exchange between VA and DOD by the end of fiscal year 2011. The departments plan to expand the exchange of data to support improved collaboration on care plans in fiscal year 2012. We are sending copies of this report to the Secretary of Defense and the Secretary of Veterans Affairs and other interested parties. The report also is available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff members have any questions about this report, please contact me at (202) 512-7114 or williamsonr@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix IV. To conduct a content analysis of our interviews with program officials and medical facility staff, we used a qualitative data analysis software package. The software facilitated our analysis of over 150 of the 170 interviews we conducted and helped us to identify and quantify interviewees’ responses on various topics. The program’s coding capabilities allowed us to group our interviewees’ responses into categories. It also provided a centralized location where all of our documents could be reviewed and analyzed. We took a number of steps to ensure that our analysis was methodologically sound. First, we defined categories to organize the views of the Department of Defense and the Department of Veterans Affairs program officials and medical facility staff by specific topics, including the Federal Recovery Coordination Program’s (FRCP) eligibility criteria, the interviewees’ interactions with the Federal Recovery Coordinators (FRC), overlap and duplication of activities among the FRCP and the case management programs with which the FRCs interacted, knowledge of the FRC role, and challenges faced by the FRCs. These categories were chosen based on themes we heard during our interviews with the program officials and medical facility staff. We conducted an intercoder reliability check to ensure the accuracy of the category definitions. To do this, two analysts coded a sample of 15 interviews into the categories. A methodologist compared the analyses to determine where inconsistencies occurred and, as a result, what categories needed more specific definitions. Once the category definitions were finalized, the same two analysts divided the categories among them and coded their categories for all of the interview documents. When the coding was completed, both analysts reviewed every code made by the other analyst and indicated whether they agreed or disagreed with the code. Changes were then made accordingly. We subsequently analyzed the interviewees’ responses based on the defined categories. This analysis allowed us to quantify interviewees’ responses within each category. In addition to the contact named above, Bonnie Anderson, Assistant Director; Susannah Bloch; Frederick Caison; Elizabeth Conklin; Cynthia Gilbert; Deitra Lee; Lisa Motley; Kristina Martin; Steven Putansu; and Suzanne Worth made key contributions to this report. | In 2007, following reports of poor case management for outpatients at Walter Reed Army Medical Center, the Departments of Defense (DOD) and Veterans Affairs (VA) jointly developed the Federal Recovery Coordination Program (FRCP) to coordinate the clinical and nonclinical services needed by severely wounded, ill, and injured servicemembers and veterans. The FRCP, which continues to expand, is administered by VA, and the care coordinators, called Federal Recovery Coordinators (FRC), are VA employees. This report examines (1) whether servicemembers and veterans who need FRCP services are being identified and enrolled in the program, (2) staffing challenges confronting the FRCP, and (3) challenges facing the FRCP in its efforts to coordinate care for enrollees. GAO reviewed FRCP policies and procedures and conducted over 170 interviews of FRCP officials, FRCs, headquarters officials and staff of DOD and VA case management programs, and staff at medical facilities where FRCs are located. It is unclear whether all individuals who could benefit from the FRCP's care coordination services are being identified and enrolled in the program. Because neither DOD nor VA medical and benefits information systems classify servicemembers and veterans as "severely wounded, ill, and injured," FRCs cannot readily identify potential enrollees using existing data sources. Instead, the program must rely on referrals to identify eligible individuals. Once these individuals are identified, FRCs must evaluate them and make their enrollment determinations--a process that involves considerable judgment by FRCs because of broad criteria. However, FRCP leadership does not systematically review FRCs' enrollment decisions, and as a result, program officials cannot ensure that referred individuals who could benefit from the program are enrolled and, conversely, that the individuals who are not enrolled are referred to other programs. The FRCP faces challenges in determining staffing needs, including managing FRCs' caseloads and deciding when VA should hire additional FRCs and where to place them. According to the FRCP Executive Director, appropriately balanced caseloads (size and mix) are difficult to determine because there are no comparable criteria against which to base caseloads for this program because of its unique care coordination activities. The program has taken other steps to manage FRCs' caseloads, including the use of an informal FRC-to-enrollee ratio. Because these methods have some limitations, the FRCP is developing a customized workload assessment tool to help balance the size and mix of FRCs' caseloads but has not determined when this tool will be completed. In addition, the FRCP has not clearly defined or documented the processes for making staffing decisions in FRCP policies or procedures. As a result, it is difficult to determine how staffing decisions are made, or how this process could be sustained during a change in leadership. Finally, the FRCP's basis for placing FRCs at DOD and VA facilities has changed over time, and the program lacks a clear and consistent rationale for making these decisions, which would help ensure that FRCs are located where they could provide maximum benefit to current and potential enrollees. A key challenge facing the FRCP concerns limitations on sharing information needed to coordinate services for enrollees, who may be enrolled in multiple DOD and VA case management programs. These limitations are often blamed for duplication of services and enrollee confusion, prompting two military wounded warrior programs to cease making referrals to the FRCP. One such limitation existed because VA had not completed public disclosure actions necessary to enable the sharing of information from the FRCP's information system. In January 2011, VA completed the process needed to resolve this issue. In addition, incompatibility among information systems used by different case management programs limits data sharing. Although the ultimate solution to information system incompatibility is beyond the capacity of the FRCP to resolve, the program has initiated an effort to improve information exchange. GAO recommends that VA direct the FRCP Executive Director to establish systematic oversight of enrollment decisions, complete development of a workload assessment tool, document staffing decisions, and develop and document a rationale for FRC placement. GAO received comments from DOD and VA; VA concurred with GAO's recommendations. |
Under the Medicare inpatient PPS, hospitals receive a fixed, predetermined payment for each hospital stay. The payment is based on standardized amounts that are calculated separately for hospitals in large metropolitan areas (with populations of 1 million or more) and for hospitals in smaller metropolitan and nonmetropolitan areas. The standardized amounts are the average cost of hospital stays for Medicare beneficiaries based on historical data and are updated annually for inflation. For 2001, the standardized amount for hospitals in large metropolitan areas was $4,028 and for hospitals in other areas it was $3,965. To determine a hospital’s payment for a Medicare beneficiary’s stay, the standardized amount is adjusted to account for variation in the cost of providing care to specific patients in specific locations. The labor cost adjustment accounts for geographic variation in hospitals’ labor costs, because the wages hospitals must pay employees vary significantly by area. The portion of the standardized amount (71 percent) that reflects labor-related expenses is multiplied by the area wage index. The remaining portion of the standardized amount (29 percent) is not adjusted. This part of the payment—which covers drugs, medical supplies, utilities, and other nonlabor-related expenses—is uniform nationwide because prices for these items are not perceived as varying significantly from area to area. The case-mix adjustment accounts for differences in resource requirements across types of patients. It is based on the expected care needs of the patient as measured by the diagnosis-related group (DRG) patient classification system. Additional payments are made under PPS to compensate hospitals for costs they incur in performing certain missions beyond caring for individual patients. Teaching hospitals receive additional payments from Medicare to account for costs associated with training medical residents. Hospitals that serve a disproportionate share of low-income Medicare and Medicaid patients also receive additional Medicare payments. The combination of all these adjustments and additional payments may result in widely varying per-stay payments across different types of hospitals or geographic areas. The Medicare labor cost adjustment is based on a wage index that is computed for each of 324 metropolitan and 49 statewide nonmetropolitan areas using data that hospitals submit to Medicare. The wage index for an area is the ratio of the average hourly hospital wage in the area compared to the national average hourly hospital wage. The average hourly wage is calculated for each area by aggregating Medicare-allowable wages for all the hospitals in the area and then dividing that sum by the corresponding staff hours. The area’s average hourly wage is then divided by the national average hourly wage to produce the area’s wage index. For example, if the average hourly wage for all hospitals in a large metropolitan area was $22.59, the wage index for that large metropolitan area would be $22.59 divided by the national average hourly wage of $21.77, for a wage index of 1.04. The wage indexes ranged from roughly 0.74 to 1.5 in 2001. As currently calculated, the wage indexes vary because of geographic differences in wages paid and also because of variation in the mix of higher- and lower- skilled workers employed in an area, termed occupational mix. An area’s average hourly wage can be higher than the national average if hospitals in an area employ more highly skilled (and thus more highly paid) workers and lower if an area’s hospitals employ more lower- skilled workers than the national average. When one area’s hospitals have a larger proportion of more skilled, higher wage staff than another area, the former’s wage index will be higher, even if wage rates in both areas for staff with the same skills, such as registered nurses, are identical. While geographic differences in wages paid affect a hospital’s labor costs but are largely beyond an individual hospital’s ability to control, the mix of occupations employed in a hospital reflects managerial decisions. The Congress, in the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000 (BIPA), required the Secretary of Health and Human Services to collect data on hospitals’ mix of employees and their corresponding wages and calculate wage indexes beginning October 1, 2004, that are adjusted for occupational mix. (For a more detailed discussion of the impact of occupational mix variation on the wage index, see app. II.) The Medicare program uses OMB’s “metropolitan/nonmetropolitan” classification system to define its geographic areas for the labor cost adjustment. Each metropolitan statistical area (MSA) is defined as a metropolitan labor market and the residual area in each state is defined as a single, nonmetropolitan labor market. The current geographic areas will most likely change when MSA boundaries are updated in 2003 with population data from the most recent decennial census and revised standards for selecting counties for inclusion in an MSA. The Omnibus Budget Reconciliation Act of 1989 established an administrative process for geographic reclassification, in which hospitals meeting certain criteria can apply to be paid for Medicare inpatient hospital services as if they were located in another geographic area. Once reclassified, hospitals receive the higher labor cost adjustment and, where applicable, the large urban standardized amount. To reclassify, a hospital must submit an application to the MGCRB, which determines if the hospital meets the reclassification criteria (see fig. 1). The two standard criteria that individual hospitals must meet to reclassify for a higher wage index are intended to identify hospitals that have higher average wages than other hospitals in their area because they are competing for labor with hospitals in a different nearby area. The first criterion concerns the hospital’s proximity to the higher wage “target” area. The proximity requirement is satisfied if the hospital is within a specified number of miles of the target area or if at least half of the hospital’s employees reside in the target area. The second criterion pertains to the hospital’s wages relative to the average wages in the target area. The wage criterion is satisfied if the hospital’s wages are a specified amount higher than the average in its assigned area and its wages are comparable to the average wages in the target area. Wage index reclassifications are effective for 3 years. All hospitals in an urban county can reclassify as a group if together the hospitals meet certain criteria, as described in figure 2. Rural referral centers (RRC) and sole community hospitals (SCH) can reclassify by meeting less stringent criteria. These hospitals receive special treatment from Medicare because of their role in preserving access to care for beneficiaries in specified areas. RRCs are relatively large rural hospitals providing an array of services and treating patients from a wide geographic area. SCHs are small hospitals isolated from other hospitals by location, weather, or travel conditions. RRCs and SCHs do not have to meet the proximity criteria to reclassify. RRCs are also exempt from the requirement that their wages be higher than the average wages in their original area. Hospitals that have lost their RRC designation can continue to reclassify under these less stringent criteria. In 1992, the first year of reclassifications, 930 hospitals were reclassified under less restrictive criteria than those currently used. More than 75 percent of these hospitals were in nonmetropolitan areas. In the following year, almost 1,200 hospitals were reclassified (of which 69 percent were in nonmetropolitan areas). For 1994, HCFA established more restrictive criteria and the number of reclassified hospitals subsequently dropped by approximately 44 percent, to 667 (see fig. 3). From 1995 to 2002, wage index reclassifications became more predominant, increasing by an average of 6 percent annually, while standardized amount reclassifications fell by almost one-quarter. For 2002, 511 nonmetropolitan hospitals and 117 metropolitan hospitals were reclassified for Medicare payment purposes. Individual hospitals have also been reclassified through legislation. Recently, the BBRA reclassified all hospitals in 7 counties (this totaled 26 hospitals) for purposes of the wage index and the standardized amount. The geographic areas that Medicare uses for the labor cost adjustment include hospitals that pay wages that may be quite different from the average wage in the entire geographic area. Hospital wages within some Medicare geographic areas—either MSAs or states’ nonmetropolitan areas—vary systematically across certain parts of the area or across types of communities. While wages paid by individual hospitals within a labor market may vary, the observed systematic variation suggests that some Medicare geographic areas include multiple labor markets. For example, the average wages of the hospitals in outlying counties of metropolitan areas usually are lower than the average wages for the entire metropolitan area’s hospitals. As a result, the labor cost adjustment for hospitals in outlying counties of metropolitan areas is based on an average wage that is often higher than the wages paid by these hospitals. In contrast, the average wages paid by hospitals in large towns (nonmetropolitan communities with between 10,000 and 49,999 people) tend to be significantly higher than the average wage of all hospitals in nonmetropolitan areas in the state. Some MSAs are very large, encompassing a diverse mix of counties. Given the broad expanse of many large MSAs, the hospitals in the different parts of an MSA may not be directly competing with each other for the same pool of employees, and the wages they pay can vary greatly. The most populous MSAs typically cover a region of several thousand square miles (see table 1). Distances between points within an MSA can exceed 100 miles. For example, the Chicago MSA includes 8 counties and 5,065 square miles, and the distance from its northernmost to southernmost point is roughly 110 miles. Hospitals in central counties of an MSA typically paid higher wages than hospitals in outlying counties. In the most populous MSAs, average central county hospital wages ranged from 7 percent higher than outlying county wages in Houston to 38 percent higher in New York in 1997. In most of these MSAs, the average wage difference between central and outlying counties ranged from 11 to 18 percent. The Washington, D.C. MSA illustrates how hospital wages in a large MSA can vary across different counties (see fig. 4). It includes hospitals located in the central city of the District of Columbia, as well as 18 counties in Maryland, Virginia, and West Virginia. Hospital wages averaged more than $23 per hour in 1997 in the District of Columbia and in most of the adjacent suburban Maryland and Virginia counties, but averaged below $20 per hour in several outlying counties. One reason MSAs are so large is because they are composed of counties, which can also be quite expansive. As with MSAs, an individual county may subsume multiple labor markets within its boundaries. As an example, San Bernardino County, California extends over 150 miles—from the city limits of San Bernardino through the Mojave Desert to the Nevada border. While most of the population is concentrated in the southwest corner of the county, which includes the city of San Bernardino, even the sparsely populated desert and mountainous portions of the county are part of the MSA. As a result, a hospital in the desert community of Joshua Tree, California, receives the same labor cost adjustment as hospitals in the city of San Bernardino 70 miles away, even though hospital wages averaged $20.84 per hour in 1997 in Joshua Tree, 13 percent less than average wages paid in San Bernardino. The Medicare program groups hospitals in nonmetropolitan areas of each state into a single geographic area for the purposes of the labor cost adjustment. Given their vast size, each statewide nonmetropolitan area is not perceived to be a single labor market, but the same labor cost adjustment is applied to hospitals in these areas. However, there are significant differences in average wages across parts of these areas. For example, for all hospitals in the nonmetropolitan area of Washington state, Medicare payments for 2001 were adjusted based on an average wage of $22.71 per hour. Yet, nonmetropolitan hospitals in the western part of the state had average wages of $24.23 per hour. Wages for nonmetropolitan hospitals in the central and eastern parts of the state, however, averaged $21.15 per hour, or 13 percent lower than hospitals in the western part of the state. Other variation in average wages across the statewide nonmetropolitan areas is associated with the type of community. In three-quarters of all states, the average wages paid by hospitals in large towns are higher than those paid by hospitals in small towns or rural areas. As a result, the Medicare labor cost adjustment may be based on average wages that are below those paid by large town hospitals and above those paid by hospitals in small towns and rural areas. For example, the 2001 labor cost adjustment for hospitals in nonmetropolitan Nebraska was based on an average hourly wage of $17.65; yet, Nebraska hospitals in large towns paid an average wage of $19.54. At the same time, small town Nebraska hospitals paid an average of $16.83 and hospitals in rural areas paid an average of $14.87, or 5 and 16 percent lower, respectively, than the area average (see table 2). In 2001, 38 percent of hospitals in large towns paid wages that were at least 5 percent higher than the average wage in their area; 16 percent paid wages that were at least 10 percent higher than the area average. While reclassification results in more appropriate labor cost adjustments for some higher wage hospitals, the reclassification criteria prevent some of them from reclassifying and exceptions to the criteria allow some lower wage hospitals to do so. In 2001, 419 hospitals, less than 10 percent of all hospitals, reclassified to receive a larger labor cost adjustment. Most of these hospitals had average wages that were above their area’s average by enough to meet the standard reclassification wage criterion. Higher wage hospitals in large towns are likelier to reclassify than higher wage hospitals in other community types because many of them are RRCs, which are exempt from the reclassification proximity criterion. Other higher wage hospitals in large towns and many higher wage hospitals in metropolitan areas, small towns, and rural areas cannot reclassify. About one-quarter of hospitals that reclassified had wages that were not high enough to satisfy the standard reclassification wage criterion. These were primarily RRCs. Generally, hospitals that reclassify but do not satisfy the standard wage criterion receive a post-reclassification labor cost adjustment that reflects average wage levels much higher than the wages they actually pay. For hospitals that meet the standard wage criterion, however, reclassification results in an adjustment that better matches their actual labor costs than did their original one. Of the 756 hospitals that paid wages sufficiently higher than their area average wage to meet the reclassification wage criteria, 310 (41 percent) were reclassified in 2001 (see table 3). Hospitals that met the wage criteria, but did not satisfy the proximity criterion, did not reclassify. Just over one- quarter of the higher wage hospitals were in large towns, yet large town hospitals made up almost half of the higher wage hospitals that reclassified. Metropolitan hospitals made up 42 percent of the higher wage hospitals, but comprised only 12 percent of the higher wage reclassified hospitals. Higher wage hospitals in large towns are likelier to reclassify than other higher wage hospitals because many are RRCs, and so are exempt from the proximity criterion. Close to half of the higher wage hospitals in small towns and rural areas reclassify. Almost 39 percent of the reclassified higher wage small town and rural hospitals were exempt from the proximity criterion because they were RRCs or SCHs. Some nonreclassified, higher wage small town or rural hospitals that were SCHs may have opted out of the PPS to receive cost-based payments from Medicare, making reclassification irrelevant. In 2001, only 38 of the 317 metropolitan hospitals with wages that were at least 8 percent higher than the average for their area, thus satisfying the standard wage criteria, reclassified to receive a higher labor cost adjustment. Nearly two-thirds of all reclassified metropolitan hospitals were in two areas—California and the northeast. Metropolitan areas in these two regions are contiguous, so higher wage hospitals may be more likely than hospitals in other areas to satisfy the proximity criterion. In 2001, 109 (about 25 percent) of all hospitals that reclassified for the Medicare labor cost adjustment paid wages that were too low to meet the standard wage criterion for reclassification. Of these, 89 were RRCs. Roughly 42 percent of these RRCs that reclassified had wage costs below the average in their area. Some of the hospitals that were reclassified in 2001 but that did not satisfy the standard wage criterion were part of countywide reclassifications. Others had been reclassified via legislation. The relationship between a hospital’s wages and the average in its geographic area, before and after reclassification, depends on whether it was in a metropolitan or nonmetropolitan area and whether it satisfied the standard reclassification wage criterion (see table 4). Reclassification resulted in higher wage hospitals receiving a labor cost adjustment that more closely reflects the wages they actually paid. For example, prior to their reclassification, the higher wage metropolitan hospitals received a labor cost adjustment based on wages in their original area that averaged 10 percent lower than their own wages. After reclassification, the average wages paid by these hospitals did not differ from the average wages paid by the other hospitals in their area. Higher wage nonmetropolitan hospitals that reclassified joined areas with average wages about 4 percent higher than their own average wages. Before reclassification, the higher wage nonmetropolitan hospitals would have received a labor cost adjustment based on average wages that were much lower than what they actually paid. In contrast, reclassification resulted in hospitals that did not satisfy the standard wage criterion joining areas that, on average, had much higher average wages. Prior to reclassification, nonmetropolitan hospitals that did not satisfy the standard wage criterion paid wages near the average of their area. After reclassification, they received a labor cost adjustment based on wages that averaged 8 percent above their own average wages. While geographic reclassification increases the labor cost adjustment, and thus Medicare payments, to hospitals that reclassify, it does not raise total Medicare outlays because any payment increases must be offset by an across-the-board reduction to Medicare payments for all hospitals. In 2002, this budget neutrality adjustment reduced Medicare payments to nonreclassified metropolitan hospitals by about 1 percent and to nonreclassified nonmetropolitan hospitals by about 0.6 percent. If the budget neutrality adjustment were calculated and applied on a state- specific basis, the payment reductions would be different in each state. A state-specific budget neutrality adjustment would reduce payments more in some states and less in other states than the national adjustment. In states in which overall Medicare hospital payments increase more than the national average increase due to reclassification, a state-specific option would result in a bigger payment reduction. A state-specific adjustment would reduce payments less in states in which hospitals do not benefit as much from geographic reclassification as the average. Hospital payments would not be reduced in states that have no reclassified hospitals under a state-specific budget neutrality option. To meet the budget neutrality requirement, CMS annually calculates the increase in Medicare payments to reclassified hospitals. This increase is due to the use of a higher wage index or standardized amount, or both. CMS then calculates how much the standardized amount—the fixed, predetermined hospital payment—needs to be reduced so that total Medicare outlays for hospital services do not change because of reclassification. In 2002, Medicare payments to nonreclassified metropolitan hospitals were about 1 percent lower due to the budget neutrality provision than they would have been in the absence of any geographic reclassifications (see table 5). Payments to nonreclassified nonmetropolitan hospitals were about 0.6 percent lower. The effect of the budget neutrality adjustment on hospital payments varies annually depending on how much Medicare payments are increased due to hospitals being reclassified, compared to total Medicare payments to all hospitals. The budget neutrality adjustment will be higher in those years where reclassified hospitals account for a greater share of Medicare payments. A state-specific adjustment would reduce payments less than a national adjustment in states where reclassified hospitals account for a smaller share of the state’s Medicare inpatient hospital spending than the national average. For example, in Colorado, where 3 of 64 hospitals were reclassified in 2000, a state-specific budget neutrality adjustment would have reduced hospital payments by only 0.07 percent, compared to a 0.6 percent reduction under the national budget neutrality calculation. For the states that have no hospitals reclassifying, such as Nevada, there would be no budget neutrality adjustment under a state-specific approach. Conversely, a state-specific adjustment would reduce Medicare payments more than a national one in states where reclassified hospitals account for a larger share of Medicare inpatient hospital spending than the national average. In New Hampshire, for example, where a large share of the state’s hospitals was reclassified (4 of 26 hospitals) a state-specific adjustment would have reduced payments to nonreclassified hospitals by nearly 3 percent, compared to a 0.6 percent reduction under the national adjustment. Medicare’s PPS for inpatient services provides incentives to hospitals to deliver care efficiently by allowing them to keep any difference between their Medicare payments and their costs, and by making them responsible for their costs that exceed Medicare payments. To ensure that the PPS rewards efficiency rather than hospitals’ circumstances, payment adjustments are intended to account for cost differences across hospitals that are beyond the control of individual facilities. If these cost differences are not adequately accounted for by the payment adjustments, hospitals are inappropriately rewarded or put under fiscal pressure. The adjustment used to account for geographic differences in wages—the labor cost adjustment—does not adequately account for these cost differences because the geographic areas used to define labor markets are too large in many instances. As a result, refinements are needed to address systematic problems in defining hospital labor markets. Such changes could improve payment accuracy and reduce the need for geographic reclassification by grouping hospitals into areas with average wages that better match their own wages. RRCs and certain other specially designated hospitals have easier access to a higher labor cost adjustment because they are allowed to reclassify under less stringent criteria than other hospitals. These hospitals may face higher costs than other hospitals, but they do not necessarily have labor costs that are higher than the average in their geographic area. Reclassification potentially offers some financial relief to a share of these facilities, but it does not address the problem underlying their financial circumstances or assist all such facilities. Identifying the underlying cause of their higher costs is important to develop mechanisms to address their financial circumstances. To improve the adequacy of Medicare’s labor cost adjustments, we recommend that the Administrator of CMS refine the geographic areas used to more accurately reflect the labor markets in which hospitals compete for employees and the geographic variation in hospitals’ labor costs. This could include separating large towns in a state into their own labor market area and removing certain outlying counties in MSAs from the metropolitan geographic area if they exhibit wage costs that are significantly different from the rest of the metropolitan area. In its written comments on a draft of this report (see app. VI), CMS stated that it agreed with the problems we identified with the current labor market areas. CMS stated that it had conducted its own analyses of alternative approaches to defining geographic areas and consulted with hospital representatives and concluded that there is no consensus on an alternative to Medicare’s current geographic areas. CMS stated that it will consider whether changes in MSA definitions based on new census figures should be used for refining the geographic areas. CMS noted that a state- specific budget neutrality approach, which we were required to assess, would require statutory change and could make reclassifications within states highly contentious. We believe that Medicare’s current geographic areas could be refined to better reflect variation in area labor costs. While forthcoming changes to MSA definitions are important to consider in refining Medicare’s geographic areas, these changes are unlikely to improve the labor cost adjustment in most large towns. We recognize that consensus on any changes to the geographic areas would be difficult to achieve because any change would redistribute Medicare payments across hospitals so that hospital payments would increase in some areas and decrease in others. Yet, because the refinements would result in Medicare payments that better match the costs that hospitals face, they would strengthen the incentives of the PPS that encourage hospital efficiency and improve Medicare’s payment method. CMS also provided technical comments, which we incorporated as appropriate. We are sending copies of this report to the Administrator of CMS and interested congressional committees. We will also make copies available to others upon request. In addition, this report will be available at no charge on the GAO Web site at http://www.gao.gov. If you have any other questions about this report, please call me at (202) 512-7119. Jean Chung, James Mathews, Michael Rose, and Kara Sokol made key contributions to this report. To conduct this work, we recreated the 2001 labor cost adjustment for each hospital in the country prior to any reclassifications, using aggregated wage and hour data reported on 1997 Medicare hospital cost reports. We used data on reclassifications and hospital characteristics from the PPS Payment Impact Files created each year by CMS. Information on metropolitan areas, such as central and outlying counties and the criteria by which counties are included in an MSA, was obtained from the U.S. Census Bureau Web site as well as interviews with Census Bureau staff. We used RUCA codes, developed at the Washington, Wyoming, Alaska, Montana, & Idaho (WWAMI) Rural Health Research Center at the University of Washington, to examine segments of nonmetropolitan areas. We assigned 1 of 30 possible RUCA codes to each hospital based on its census tract. These 30 codes were then collapsed into 4 categories: urban, large town, small town, and rural. We calculated dollar-weighted average hourly hospital wages for each of the nonmetropolitan categories, nationally and by state, by dividing aggregate wages for all hospitals within a category by aggregate hours. We then compared the average hourly hospital wage for each nonmetropolitan subgroup within a state to the statewide nonmetropolitan average hourly wage. To evaluate the potential payment impact of applying a geographic reclassification budget-neutrality factor on a state-specific basis, we used the 2000 PPS Payment Impact File to calculate the Medicare payments to all hospitals within each state, before and after any geographic reclassifications. We then used the difference between pre- and post- reclassification payments to calculate a budget neutrality factor for each state. These budget neutrality factors were then used to estimate how payments to reclassified and nonreclassified hospitals in each state would differ under a state-specific budget neutrality adjustment, compared to the current national adjustment. In BIPA, the Congress required the Secretary of Health and Human Services to collect data on hospitals’ mix of occupations and their corresponding wages by September 30, 2003, and calculate wage indexes beginning October 1, 2004, that are adjusted to remove the effects of occupational mix on average wages. Occupational mix data for each acute care hospital will be collected and updated every 3 years. The methodology for adjusting the wage index for occupational mix will be determined after the data have been collected. Average hospital wages vary because of differences in wages paid across hospitals, but also because hospitals employ different mixes of occupations. As a result, average hospital wages are higher than the national average if the hospitals in an area employ more workers in highly skilled occupations and lower if the hospitals employ fewer workers in more highly skilled occupations. The current calculation of the Medicare wage index does not distinguish between wage differences due to geographic labor cost variation and wage differences due to geographic variation in the mix of more highly and less highly skilled occupations. Thus, Medicare’s wage indexes are too high in areas with a more highly skilled mix of hospital workers and too low in areas with a less skilled mix of hospital workers. While geographic differences in wages paid affect hospitals’ labor costs, but are beyond an individual hospital’s ability to control, occupational mix generally is within the control of a hospital. Changing the calculation of the wage index to eliminate the effect of occupational mix differences will raise the wage index for some types of hospitals and lower it for others. Wage indexes will be reduced for hospitals, such as metropolitan or teaching hospitals, that tend to hire more employees in highly skilled occupations with higher wages. Wage indexes for rural hospitals, which tend to employ a less skilled mix of employees, are likely to go up. While national data on the occupational mix of hospital employees are not available, data from California demonstrate the potential effects of changing the wage index calculation to eliminate the effects of occupational mix differences. Without adjusting for differences in occupational mix, the average hourly wage for hospitals in the Oakland MSA is 57 percent higher than the average hourly wage for nonmetropolitan California hospitals. Hospitals in the Oakland area generally employ a greater proportion of more skilled, and therefore more expensive, staff (see table 6). For example, in Oakland area hospitals, RNs account for approximately 25 percent more of the total hours worked by hospital employees than they do in nonmetropolitan California. Recalculating the wage indexes so that they reflect the same mix of workers in all areas reduces the difference between the Oakland area wages and those paid in nonmetropolitan areas to 50 percent. An occupational mix-adjusted wage index in nonmetropolitan California would be almost 4 percent higher than the current wage index calculation (see table 7). Across metropolitan areas, the change to the wage index would vary. Change in Medicare hospital payments (current law) Change in Medicare hospital payments (current law) | The Medicare program's prospective payment system (PPS) for inpatient hospital services provides incentives for hospitals to operate efficiently by paying them a predetermined, fixed amount for each inpatient hospital stay regardless of the actual costs incurred in providing the care. Although the fixed amount is based on national average costs, actual per stay payments vary widely across hospitals, primarily because of two payment adjustments in the PPS. One adjustment accounts for cost differences across patients due to their care needs and the other accounts for the substantial variation in labor costs across the country. The Medicare program's labor cost adjustment may not adequately account for geographic differences in hospital wages because of problems with the definition of labor markets. The geographic areas used by Medicare to approximate hospital labor markets often encompass large areas in which hospitals in different parts of an area or different types of communities pay widely varying wages. Geographic reclassification does not systematically address inadequacies in the way the Medicare program defines geographic areas, although it allows some, but not all, hospitals that may be in distinct labor market and pay wages above the average in their area to receive a higher labor cost adjustment. Geographic reclassification reduces payments to hospitals that do not reclassify because of the budget neutrality requirement, and the amount of this reduction would vary across hospitals under a state-specific budget neutrality approach depending on their location. In 2002, payments to metropolitan hospitals that were not reclassified were 1 percent lower and payments to nonmetropolitan hospitals that were not reclassified were 0.6 percent lower because of geographic reclassification. If the budget neutrality provision were calculated and applied within individual states instead of nationally, the adjustment would be smaller in those states in which hospitals did not benefit much from reclassification and higher in states where a higher proportion of hospitals reclassified. |
The R&D Centers, Regional Labs, and Comprehensive Centers share responsibility with other programs created by the Congress for education research, research-based activities, and technical assistance. Many of these programs are located in six different offices throughout Education. (See fig. 4.) For example, the Individuals with Disabilities Education Act established a special education research and innovation program as well as technical assistance centers to improve services and results for children with disabilities. The Rehabilitation Act established a National Institute for Disability and Rehabilitation Research. The Office of Special Education and Rehabilitative Services administers these programs. The amount of funding received by these programs in fiscal year 2000 ranged from $70,000 for 11 American Overseas Research Centers to $86.5 million for the National Institute on Disability and Rehabilitation Research. OERI is Education’s lead office for educational research and development. Its goals are to promote quality and equity in education by funding research; developing new learning materials, teaching techniques, and methods of organizing schools; demonstrating and evaluating promising educational practices; disseminating research-based information; and collecting data related to schools in the United States and other nations. Unlike other Education offices, OERI’s activities span all grade levels, from preschool through adult education, and all major content areas of instruction. The R&D Centers, established in the 1960s to increase fundamental knowledge in education, are administered by OERI. Over the years, legislative changes have repositioned their placement in Education. Most recently, the Educational Research, Development, Dissemination and Improvement Act of 1994 reorganized OERI, implementing measures that changed the way the R&D Centers related to Education. First, the act created the National Educational Research Policy and Priorities Board to work with the Assistant Secretary of OERI to establish a long-term national agenda for research, development, and dissemination activities. Unlike previous boards, this board was charged with improving research priorities and developing standards for evaluating OERI research, including that done by the R&D Centers. Second, the act established five national research institutes within OERI, each with its own research focus, and placed the R&D Centers, as well as field-initiated studies and other research-related programs, under the appropriate institute. The Congress is expected to start the reauthorization of OERI in 2002; the last reauthorization was in 1994 and was intended to be on a 5-year cycle. Figure 5 shows the locations of the R&D Centers. Legislation has changed the focus of the Regional Labs over time. Originally established to resemble the labs funded by the Atomic Energy Commission, such as the Los Alamos National Laboratory, the Congress created the Regional Labs in 1965 to conduct long-term activities to address national educational problems. However, funding to support these activities was never made available, limiting the scope of the Regional Labs to smaller-scale projects. Further, in the late 1970s, the Congress’s negative reactions to federally supported curriculum projects prompted Education and the Regional Labs to discontinue all large-scale nationally oriented curriculum projects. As a result, Regional Labs developed an increasingly regional agenda. In 1994, the Congress gave the governing board of each Regional Lab sole responsibility for determining if the Regional Lab fulfilled the terms of its contract with Education and its regional agenda. The Congress mandated that each governing board reflects a balanced representation of states in the region, as well as interests and concerns of regional constituencies, including teachers and education researchers. Figure 6 shows the states included in the regions of the 10 Regional Labs. Created in 1994, the Comprehensive Centers were established more recently than the R&D Centers and the Regional Labs. The Improving America’s Schools Act of 1994 consolidated the functions of 48 categorical technical assistance centers that supported programs authorized under the ESEA, including Title I, Migrant Education, and Indian Education, into 15 Comprehensive Centers. The Congress created the Comprehensive Centers to support comprehensive, cross-program assistance as envisioned in the law, and placed them under the Office of Elementary and Secondary Education (OESE) and the Office of Bilingual Education and Minority Languages Affairs (OBEMLA). Figure 7 show the states included in the regions served by the Comprehensive Centers. Region XV includes: American Samoa, Commonwealth of the Northern Mariana Islands, Federated States of Micronesia (Chuuk, Kosrae, Pohnpei, and Yap), Guam, Hawaii, Republic of Marshall Islands, and the Republic of Palau. The amount of resources allocated to the R&D Centers, Regional Labs, and Comprehensive Centers differs, but the overall investment in these programs is modest. For example, to operate in fiscal year 2000, each R&D Center received from $1.5 million to $6.6 million, each Regional Lab received from $3.8 million to $8.6 million, while each Comprehensive Center received from $0.9 million to $2.7 million. The core budget for all these programs totaled about $130 million. Table 1 shows the key features of the R&D Centers, Regional Labs, and Comprehensive Centers. Because of the importance of education research, research-based activities and technical assistance in improving schools, many studies have focused on the R&D Centers, Regional Labs, and Comprehensive Centers. Education has funded various assessments of the R&D Centers and the Regional Labs and has recently conducted an evaluation of the Comprehensive Centers. In addition, the R&D Centers and the Regional Labs have been studied and discussed by numerous independent organizations, including the National Academy of Sciences, the Brookings Institution, the RAND Corporation, and the National Educational Research Policy and Priorities Board. Several studies, including the 1992 National Academy of Sciences report and a 2000 Brookings Institute report, have concluded that the funds available to OERI to support research have been, and continue to be, insufficient to support long-term, large-scale efforts. Laws define different missions and roles for the R&D Centers, Regional Labs, and Comprehensive Centers, and these differences are reflected in how these programs spend their money. R&D Centers focus on national research priorities, such as student assessment. Although both Regional Labs and Comprehensive Centers have a regional orientation, Regional Labs focus on meeting the needs of the regions. Comprehensive Centers focus on assisting customers in their regions implement federal education agendas, such as helping to close the achievement gaps for federally targeted groups like disadvantaged students. Education uses funding documents and program oversight to direct and prioritize the activities of the R&D Centers, Regional Labs, and Comprehensive Centers and shape the agendas of the R&D Centers and Comprehensive Centers. Unlike R&D Centers and Comprehensive Centers, the Regional Labs have governing boards. Because these governing boards determine regional agendas and oversee Regional Lab activities, Education has limited ability to shape the agendas of Regional Labs or ensure accountability for their products and services. The Congress created a separate primary focus for the R&D Centers, Regional Labs, and Comprehensive Centers and gave them the responsibility of performing specific activities. Because of differences in their mandates, the R&D Centers, Regional Labs, and Comprehensive Centers have different roles in supporting Education’s research agenda. The Educational Research, Development, Dissemination, and Improvement Act of 1994 places the R&D Centers under education research institutes in OERI, each of which addresses a specific content area, and requires them to carry out the purposes for which the institutes were created by conducting research and development. In contrast, rather than requiring the Regional Labs to address a particular content area, the act requires them to use research-based knowledge to address the issues in the regions they serve and assigns them an expansive array of activities to perform. Like the Regional Labs, the Comprehensive Centers have extensive mandates that require the Comprehensive Centers to focus on certain customers by giving priority to schools with schoolwide programs and the highest number of children in poverty. The law, however, allows Education to guide the general direction of the mandated activities. Table 2 provides the primary focus of the R&D Centers, Regional Labs, and Comprehensive Centers and summarizes their mandated activities. The spending patterns of the R&D Centers, Comprehensive Centers, and Regional Labs reflect their mandates and missions. For example, the Congress authorized the R&D Centers to conduct research and development in order to increase fundamental knowledge of central issues in education. To support this mission, they reported spending 73 percent of the $35.5 million in core funding they received from OERI in fiscal year 2000 on research and an additional 14 percent on development and dissemination. The Congress created the Comprehensive Centers to help state educational agencies, school districts, and schools within an assigned region implement federal elementary and secondary school programs by providing technical assistance and training. Accordingly, the Comprehensive Centers reported spending most of their fiscal year 2000 $28.6 million core funds from OESE 83 percent on technical assistance. Regional Labs were authorized to conduct a wide range of research-based activities, including applied research, development, dissemination, and technical assistance. Their spending reflected these purposes: the Regional Labs reported spending 25 percent of their fiscal year 2000 $65.2 million core funds from OERI on research, 17 percent on development, 16 percent on dissemination, and 21 percent on technical assistance. Most R&D Centers, three Regional Labs, and two Comprehensive Centers used funds from other organizations, including other federal agencies, state and local educational agencies, and foundations, to expand work they were performing for Education. For example, one R&D Center, the National Center for Postsecondary Improvement, used funding from the Pew Charitable Trust to extend the scope of a study examining the impact of state and university admission standards on secondary schools and students. Similarly, 10 R&D Centers reported that they leveraged additional money from other federal and state agencies and not-for-profit organizations to promote Education’s research agenda. Two of these R&D Centers reported receiving more funding from these other sources of funding than from OERI. Three Regional Labs also reported receiving funds from non-OERI sources, with these funds composing between 1 to 27 percent of their budgets. Two Comprehensive Centers reported receiving external funding and these funds accounted for 0.3 to 2 percent of those Centers’ funding. Education shapes the activities of the R&D Centers, Regional Labs, and Comprehensive Centers through its funding documents and program monitoring. Education uses funding documents, such as grant announcements and statements of work, to guide and direct activities included in the mandates and to obtain help in implementing department activities. In addition, Education assigns program officers to oversee the activities of the R&D Centers, Regional Labs, and Comprehensive Centers. Although priorities in R&D Center cooperative agreements are broad and do not impose particular methods for researching a topic, they are specific enough to shape the direction and breadth of the R&D Centers’ research agenda. For example, in spelling out the priorities for the R&D Center for enhancing young children’s development and learning, Education identified topics, theories, and research areas that the Center should address. Similarly in spelling out the priorities for the R&D Centers for meeting the needs of diverse student populations, Education identified topics, theories, and student populations to be studied. In addition, the R&D Centers’ cooperative agreements require that a portion of R&D Centers’ funds be spent on tasks defined by OERI. R&D Centers’ cooperative agreements require them to reserve 5 percent of their core funds to carry out OERI initiated activities that assist OERI in carrying out its responsibilities. For example, OERI may require an R&D Center to write a briefing paper or conduct a research project. Contracts between Education and the Regional Labs give the Regional Labs control over most of their activities, but also give Education the opportunity to guide some of their work. Regional Labs identify the critical issues in their region and develop plans to address these issues. However, these contracts also allow Education to assign each Regional Lab a broad specialty area for example, early childhood education or educational technology—that reflects the national education agenda and is aligned with a dominant theme of an OERI research institute with which they are associated. In making Regional Labs responsible for a particular specialty area, the contracts required that they (1) conduct development, applied research, and dissemination in that area; (2) keep abreast of developments in their designated field; (3) provide subject area expertise to other labs; and (4) work cooperatively with OERI institutes as appropriate. Although Education defines the general type of activities for Regional Labs in specialty areas, the governing boards determine the focus of the activities and the extent to which they met the requirements of the contract. Additionally, the contracts require Regional Labs to work together on areas of concern to all Regional Labs, such as how to effectively disseminate their products and develop a telecommunications network. Regional Labs reported spending about 25 percent of their funds on these required national activities and on other activities with a national purpose. Unlike the cooperative agreements for the R&D Centers, Regional Lab contracts do not require them to do work to support OERI activities. Prior to 1995, the contracts required that the Regional Labs spend 1 percent of their core funds from OERI to support OERI. However, this requirement was dropped because of objections from the Regional Labs. Currently, Regional Labs may agree to perform work for OERI in return for additional funding. Although the Comprehensive Centers are not research entities, Education shapes their activities by setting priorities for them in funding documents. Such priorities include, for example, meeting with school district officials to review and provide advice on district procedures for meeting federal requirements and assisting school districts in the development of student assessments. In addition, Comprehensive Centers are required to engage in common activities identified by Education such as conducting annual conferences on school improvement activities that promote Education’s agenda. In addition to funding documents, program officers who administer grants and contracts may play an important role in influencing activities of the R&D Centers and Comprehensive Centers by ensuring that work performed is consistent with work proposed and that funds are being used as effectively and efficiently as possible. Program officers are supposed to perform the following types of activities: (1) help to develop funding documents and ensure that R&D Centers and Comprehensive Centers are in compliance with these documents; (2) review progress reports, financial reports, and products; and (3) approve dissemination plans, staffing changes, and activities funded by other sources. Program officers reported using these oversight functions to ensure that the activities of R&D Centers and Comprehensive Centers are consistent with their proposals. For example, Education program officials reported that they identified activities that were inconsistent with those in the R&D Centers’ proposals and subsequently negotiated alternative activities, indicated where collaboration between some R&D Centers would be beneficial, and encouraged R&D Centers to drop nonpromising lines of research. Program officers also may play a role in determining supplemental and future funding decisions. Even though regional governing boards are responsible for the oversight of Regional Labs, Education’s program officers, nonetheless, may still have the potential to influence the activities of these labs. A program officer assigned to a Regional Lab described their role as that of “critical friends” who use their professional expertise and interpersonal relationships with Regional Labs’ staff to influence the activities of the Regional Labs. In addition, Regional Lab program officers, like R&D Center program officers, may play a role in determining supplemental and future funding decisions. Time and authority may limit program officers’ ability to exercise their influence. In OERI, one program officer is generally assigned to each R&D Center and Regional Lab. The program officers we interviewed reported spending about 50 percent of their time on monitoring activities related to the R&D Centers and Regional Labs because of other assigned responsibilities. Because the R&D Centers and Regional Labs are complex organizations and prolific producers of products and services, officers have to be very selective in targeting their own time. Only one program officer is assigned to monitor all 15 Comprehensive Centers. The Congress has consistently given education program oversight to either a federal agency—usually Education or to the states, since the states are generally responsible for the education of their students. The Regional Labs are unlike other federal education programs because neither the federal government nor state governments have oversight responsibility for programs. Specifically, Education has little control over Regional Labs because regional boards govern them. This occurs in spite of the fact that the Regional Labs get the largest share of the federal dollars devoted to these three programs that conduct or support research. The law requires that the Regional Labs establish governing boards with regional representatives. The regional boards have sole responsibility for determining the regional agenda and for determining whether the Labs are fulfilling the responsibility of their contracts, even though these contracts are funded by Education. Even for the national specialty areas, Education sets only wide parameters while the governing boards determine specific activities. Education limits its communications with Regional Labs mainly to administrative issues, according to a director we interviewed at a Regional Lab. In addition, Education’s Regional Lab program officers told us their comments on Regional Lab work and products were only advisory. The law requires each Regional Lab to establish a governing board that reflects both a balanced representation of the states in the region and the concerns of regional constituencies and includes teachers and education researchers. In addition, Education requires that every chief state school officer in the area served by the Regional Labs be offered an opportunity to serve on the board or to designate a representative. These safeguards, however, do not assure that priorities of each state in the region will be met. A variety of factors potentially make achieving balanced regional and state representation challenging. First, Regional Labs are not given guidance on how to obtain balance between states and regional interests, and regions, unlike states, have no formal governing body to establish educational priorities. Second, the governing boards, not the states, have the authority to determine how members are nominated and selected. Finally, states within regions vary substantially in size, population density, poverty levels, and ethnic composition. These factors may determine state educational priorities in a way that would make them vary widely. For example, California—a diverse and large state in size and population—is included with Utah, Nevada and Arizona in one region. New York also a diverse and large state—is included with Vermont and other New England states in another region. R&D Centers, Regional Labs, and Comprehensive Centers provided many examples of collaborative and coordinated activities. R&D Centers, Regional Labs, and Comprehensive Centers are required to collaborate and coordinate with each other and Education. The programs reported they are most likely to engage in collaborative and coordinated activities when they share a common interest in a specific student population, such as English language learners, or a specific topic, such as assessment. Partnerships between programs, common memberships in consortia, and staff members who are employed by more than one of these organizations facilitate collaboration and coordination and help leverage resources. Education plays a proactive role in promoting collaboration and coordination by including requirements for certain activities as part of its funding documents and in its ongoing negotiations with these organizations. However, certain factors differences in student populations, funding uncertainty, and competition reduce opportunities for collaboration and coordination. R&D Centers, Regional Labs, and Comprehensive Centers collaborate and coordinate with each other and Education as required by law and the documents that control their funding. Certain types of collaboration and coordination flow naturally from overlapping needs, interests, and resources. For example, Regional Labs and Comprehensive Centers with populations of students with limited English proficiency would naturally make use of the R&D Centers, Regional Labs, and Comprehensive Centers that have expertise in that area. R&D Centers, Regional Labs, and Comprehensive Centers reported a variety of collaborative efforts for fiscal year 2000, including joint projects and training. Joint projects included: Two Comprehensive Centers (the Northern California Comprehensive Center and the Southern California Comprehensive Center) worked together to produce teleconferences to help low performing schools. Ten Regional Labs and an R&D Center (the National Center for Early Development and Learning) produced a training guide entitled Continuity in Early Childhood: A Framework for Home, School, and Community Linkages. An R&D Center (the Center for Research on the Education of Students Placed At-Risk) and a Regional Lab (the Northwest Regional Lab) produced a joint publication about parent involvement in schools. A Regional Lab (the Northwest Regional Lab) and an R&D Center (the Center for Research on Evaluation, Standards, and Student Testing) created the Classroom Assessment Tool Kit. Examples of training included: A Comprehensive Center (the Southern California Comprehensive Center) trained other Comprehensive Centers to teach instructors how to coach children learning to read. An R&D Center (the Center for Research on Evaluation, Standards, and Student Testing) trained a Regional Lab (WestEd) to use a data collection tool. An R&D Center (the National Center for Improving Student Learning and Achievement in Mathematics and Science) worked with a Comprehensive Center (the Region VI Comprehensive Center) to conduct a professional development project by teaching people to train math and science teachers. A Regional Lab (the Southwest Educational Development Laboratory) provided training to a staff member of a Comprehensive Center (the Southeast Comprehensive Center) in the use of Flashlight and Compass: A Collection of Tools to Promote Instructional Coherence a tool for establishing teacher study groups. R&D Center, Regional Lab, and Comprehensive Center staff also provided many examples of coordination efforts meant to ensure that each was aware of the others’ projects. Some examples include the following: As a follow-up to Education’s National Awards for Model Professional Development, three Regional Labs, WestEd, the North Central Regional Lab, and the Mid-Continent Regional Lab, studied how the awarded districts supported districtwide teacher and student learning. An R&D Center, the Center for the Study of Teaching and Policy, shared its data on resource allocation among some of the same districts and contributed to the research questions and design. The Regional Labs shared the findings with Education. An R&D Center, Center for Research on Education, Diversity, and Excellence, reviewed work on Spanish Writing Assessment done by a Regional Lab, Northwest Regional Lab. Legislation requires that the Regional Labs and Comprehensive Centers collaborate and coordinate with each other and with the R&D Centers, but their mandates differ in the amount of collaboration and coordination they require. The Regional Labs are required to collaborate and coordinate with each other, Education-funded technical assistance providers, and OERI institutes, and to share and plan joint activities with other Education- funded state, and federal programs. They are also required to establish a network for sharing information, planning activities involving multiple regions, and working on national projects. The Comprehensive Centers are required to share information, coordinate services, and work cooperatively with the Regional Labs, R&D Centers, Education’s regional offices, state and local educational agencies, and all other Education- funded research, development, dissemination, and technical assistance programs. The R&D Centers do not have specific legislative requirements to collaborate and coordinate with other Education-funded programs; however, they are required to do so in their cooperative agreements. R&D Centers, Regional Labs, and Comprehensive Centers collaborate and coordinate when they have an interest in the same student population. The following examples illustrate how student populations provide a focal point for collaboration and coordination. A member of the Mid-Continent Regional Lab, which includes states with large numbers of Native Americans, sat on the steering committee of the National Research Center on the Gifted and Talented, an R&D Center. This committee oversaw the production of a publication on talented American Indian and Alaskan Native students. Two R&D Centers, the Center for Research on Education, Diversity, and Excellence and the National Center for Improving Student Learning and Achievement in Mathematics and Science, published a newsletter on issues related to how diverse students learn math and science. All 15 of the Comprehensive Centers created an Internet mailing list about English-language learners to share information from their regions, identify staff with proficiency in meeting the needs of English- language learners, and disseminate information. The structures of the R&D Centers, Regional Labs, and Comprehensive Centers foster collaboration and coordination among them and other entities. In some cases, partnerships may encourage collaboration. In other cases, collaboration occurs because one single entity operates both an R&D Center, Regional Lab, and/or Comprehensive Center. Partnerships may encourage collaboration by establishing bridges between programs. R&D Centers, Regional Labs, and Comprehensive Centers are programs that operate through parent organizations universities, not-for-profit organizations, and educational agencies. These parent organizations create consortia—formal partnerships with other universities, not-for-profit organizations, for-profit corporations, and educational agencies—to run R&D Centers and Comprehensive Centers. Participants in partnerships that run R&D Centers range from 2 universities to 29 universities and not-for-profit organizations. Figure 8 shows one R&D Center that is a consortium of 5 universities and the affiliations they have with other R&D Centers. (Parent Organization) (NCPI) Center for Research on Education, Diversity and Excellence (CREDE) Unlike R&D Centers and Comprehensive Centers, the parent organizations that run Regional Labs do not form partnerships with other universities, not-for-profits, or educational agencies to run the Regional Labs. They may, however, be asked by a Comprehensive Center or R&D Center to enter into a partnership in order to provide specialized services. We found four parent organizations that operated Regional Labs that had formed partnerships with Comprehensive Centers. Figure 9 shows an example of such a partnership. Participants in partnerships that run Comprehensive Centers range from three to eight universities, not-for-profit organizations, and educational agencies. Most parent organizations that run Comprehensive Centers partner with other organizations to obtain expert services in specialized areas, such as migrant education or Indian education. Because each of these specialized partners may work with as many as four Comprehensive Centers, these partnerships may establish bridges that forge coordination in particular topical areas. For example, ESCORT, a former categorical technical assistance center, specializes in migrant education and partners with four Comprehensive Centers to provide services in that area. Collaboration also results when a single organization operates both an R&D Center, Regional Lab, or a Comprehensive Center. For example, staff from the Western Regional Lab and the Northern California Comprehensive Center participated in a workgroup that developed a guide on how schools could better obtain student perspectives and suggestions to improve school planning. As shown in figure 9, both the Western Regional Lab and the Northern California Comprehensive Center are run by the same parent organization. Five of the 10 parent organizations that run a Regional Lab also run a Comprehensive Center. One of the 12 parent organizations that operates an R&D Center also operates a Comprehensive Center. These interlacing organizational relationships allow R&D Centers, Regional Labs, and Comprehensive Centers to leverage resources. For example, an Education official working with the Comprehensive Centers told us that through their parent organizations and partnerships Comprehensive Centers are able to leverage the personnel and expertise needed to perform their work. In those cases where parent organizations run two programs, we were told that staff divide their time between programs to leverage expertise. Similarly, a director of a parent organization that runs a Regional Lab and a Comprehensive Center stated that the Comprehensive Center draws upon experts assigned primarily to other projects to obtain skills needed to implement particular activities. Individual relationships, such as staff members holding multiple appointments within R&D Centers, Regional Labs, or Comprehensive Centers also facilitate collaboration and coordination among R&D Centers, Regional Labs, and Comprehensive Centers. For example, a principal investigator for the Consortium for Policy Research in Education, an R&D Center, also works on projects for the National Center for Improving Student Learning and Achievement in Mathematics and Science, another R&D Center. The funding agreements between Education and the R&D Centers, Regional Labs, and Comprehensive Centers reflect Education’s interpretation and implementation of legislative requirements for collaboration and coordination. The cooperative agreements for the R&D Centers require them to collaborate and coordinate with Regional Labs, Comprehensive Centers, other federal programs, policy institutions, and advocacy groups. R&D Centers also agree, in their funding documents, to conduct an annual meeting to share research with Education and other research and development programs, and collaborate with OERI. Regional Lab contracts with Education require that they participate in at least two meetings a year convened to discuss issues related to Education-funded programs. In addition, Education’s contracts for Regional Labs require Regional Lab representatives to meet with OERI annually and chairpersons of the governing boards to meet with OERI when their contract begins. Regional Labs also have an option in their contracts with Education that allows them to earn supplemental funds by agreeing to perform work in collaboration and coordination with OERI, including sponsoring meetings and panels and writing briefs. Cooperative agreements for the Comprehensive Centers require them to meet with seven different Education-funded programs, including the Regional Labs, to discuss collaboration and coordination; plan a national conference; engage in a common project to improve teaching; and collaborate with each other and local and state educational agencies. Cooperative agreements for the R&D Centers and Comprehensive Centers also outline Education’s responsibilities to facilitate collaboration and coordination. For example, the funding agreement for the Comprehensive Centers specifies that Education officials will work with Comprehensive Centers in planning conferences and identifying areas for collaboration and coordination. The funding agreements that Education has with Regional Labs do not outline Education’s responsibilities for facilitating collaboration and coordination. Education officials may also identify appropriate areas for collaboration and coordination. For example, an institute director told us that she contacted an R&D Center in another institute to discuss possible areas for collaboration and coordination with the R&D Center in her institute. Similarly, some Comprehensive Center officials stated that the program officer assigned to them identified areas for cross-program collaboration and coordination and communicated with them frequently. For example, the program officer suggested collaborating and coordinating on the creation of a common framework for working with low performing schools in Comprehensive Center regions. Despite the efforts made by the R&D Centers, Regional Labs, Comprehensive Centers, and Education in fostering collaboration and coordination, barriers exist. R&D Centers, Regional Labs, and Comprehensive Centers cited differences in student populations and topics, uncertain funding, and competition as hindrances to collaboration and coordination. R&D Centers have different research focuses and conduct research on different topics. Regional Labs and Comprehensive Centers serve diverse geographical areas with different interests. These differences potentially reduce opportunities for collaboration and coordination. For example, an R&D Center with a focus on postsecondary education would have little or no reason to collaborate or coordinate on substantive issues with other R&D Centers, Regional Labs, and Comprehensive Centers that focus on research related to elementary and secondary education. Funding uncertainties also make collaboration and coordination difficult. Directors said they were often reluctant to write collaborative activities into their proposals because they did not know which organizations would win future funding competitions. Competition also limits collaboration and coordination. The education research and technical assistance business is a competitive industry. Like others in competitive industries, R&D Center, Regional Lab, and Comprehensive Center staff are protective of ideas and practices that give them advantages over other organizations that they perceive as competitors for future sources of funding. The recently funded evaluations of the R&D Centers, Regional Labs, and Comprehensive Centers provided limited information on outcomes of the activities conducted by these programs. OERI is required to use peer review to evaluate the R&D Centers and Regional Labs. Peer review is well accepted and widely used for assessing the merit of research proposals and the scientific soundness of research. Unlike the R&D Centers, research is only a relatively small part of what Regional Labs do. Their other activities dissemination and technical assistance would have been more appropriately evaluated using methods other than peer review. The peer review processes that Education used to evaluate the R&D Centers and Regional Labs have shortcomings that limited the usefulness of the findings. First, the peer review process used has the potential for bias because the R&D Centers and the Regional Labs selected most of the products that were reviewed. Second, the processes were cumbersome. For example, Education required each member of a review team to write an individual assessment report and to review all contracts, contract modifications, progress reports, and annual updates for a three-year period. In addition, with regard to the Comprehensive Centers, the customer satisfaction survey evaluations of the Comprehensive Centers did not provide information on individual centers. Traditionally, Education has used peer review to evaluate activities carried out by OERI, including those conducted by the R&D Centers and the Regional Labs. The Educational Research, Development, Dissemination, and Improvement Act of 1994 codified this practice by requiring OERI to develop peer review standards for evaluating and assessing the performance of recipients of grants, cooperative agreements, and contracts that exceed $100,000, as well as for selecting proposals for funding and identifying exemplary and promising educational programs. Historically, peer review has been used extensively in the selection of proposed research projects and, to a lesser extent, to evaluate research and development programs. Peer review entails an independent assessment of the technical or scientific merit of research by peers who are scientists with knowledge and expertise equal to that of the researchers whose work they review. It is sometimes used to evaluate research when the ultimate outcomes of the research are unpredictable. Peer review may be appropriate for evaluating some activities conducted by the R&D Centers but other evaluation techniques are better suited for evaluating the many activities of the R&D Centers and Regional Labs. According to the National Academy of Sciences, although peer review is well suited for assessing theory-driven research with potential long-term effects, it is less appropriate for assessing applied research, development, technical assistance, and dissemination efforts in which outcomes are anticipated and can be measured over a relatively short period of time.The R&D Centers and Regional Labs engage in many research activities that are designed to achieve practical outcomes. Evaluation methods that measure outcomes customer surveys, comparisons with similar programs, and controlled case studies may be better suited for evaluating these activities. Education’s peer review processes have the potential for bias and were cumbersome. Both of these conditions limited the usefulness of findings in addressing key issues. The self-selection of materials for review by R&D Centers and Regional Labs did not provide a representative cross section of their products and services. R&D Center staff were involved in deciding which products were to be reviewed and each Regional Lab nominated five or six major products or services, two of which were selected for review. While this approach allowed reviewers an in-depth look at major program initiatives, it did not provide reviewers with a cross section of the R&D Centers or Regional Labs’ work, nor did it allow them to generalize from these works to other activities. Selecting as reviewers for the R&D Centers and Regional Labs some individuals who have previously evaluated the merits of the grant applications or proposals raised questions about objectivity. It is likely that the individuals who selected these organizations as grant or contract recipients might want their original choices validated. Requiring peers to have a broad mix of skills made selection and scheduling of reviewers difficult. Unlike other agencies that select peers solely on the basis of their expertise in the area of work, OERI requires that review panels include individuals with a broad range of knowledge and experience. For example, OERI standards require peer review panels to include individuals with in-depth knowledge of education policy or practice and in-depth knowledge of theories and methods of study related to the subject area. These requirements complicated the identification of peer review panels. The amount of material to be reviewed was extensive in terms of the short time frames of the assessments and the complexity of the organizations. Over a short period of time—for example, 5 days on site for the review of a Regional Lab, with half that time devoted to data gathering and the other half to writing the reports—all reviewers were required to read immense amounts of material, including funding documents, statements of work, proposals, and progress reports, to learn, in detail, how the programs performed their work and to write individual reports. Some reviewers complained that they only had time to “scratch the surface,” and that much of the material they had to review was repetitive. If responsibilities could have been shared, peer reviewers would have been able to gather more in-depth knowledge. In some cases, materials chosen for review by the R&D Centers were incomplete. Reviewers of some programs noted that materials addressing the rationale, hypotheses, controls, and usefulness of studies were often insufficient for them to make informed judgements. Assessments took place midway in the funding cycles. The reviews generally took place during the third year of a 5-year contract. As a result, reviewers were hesitant to be critical in reports because many studies were on going and therefore could improve by the end of the contract. The dual purposes of the peer reviews inhibited candor. On one hand, Education designed the reviews to be formative evaluations— evaluations that were to focus on the performance of the programs in terms of their missions and the technical quality of their products. In this regard, reviews were to examine the overall quality of the work of the R&D Centers and Regional Labs, the extent to which R&D Centers and Regional Labs performed their work on time and met professional standards. However, the reviews were also designed to assess the usefulness, outcomes, and effects of their work to help OERI determine if the R&D Centers and Regional Labs merited continued funding. The peer review process depended exclusively upon expert opinion rather than directly measuring how useful the research was or its effects. Reviewers discussed “the potential” value of activities and were not able to predict the ability of the entities to contribute substantially to the field. Moreover, they did not believe their findings should have influence over funding decisions, which affected what they wrote in their reports. Education’s evaluation of the Comprehensive Centers met the requirements of the law but provided little information that would help Education determine if each Comprehensive Center was meeting the needs of its customers. The Elementary and Secondary Education Act requires the Secretary of Education to collect information about the availability and quality of services provided by the Comprehensive Centers and to conduct surveys to determine if populations served by the Comprehensive Centers are satisfied with their access to services and the quality of those services. As part of the year 2000 evaluation, a contractor surveyed the satisfaction of customers who had participated in either of two activities offered by a Comprehensive Center. These activities were selected from among all the activities offered by the Comprehensive Center because they were long-term or intensive. The contractors surveyed customers by randomly selecting them from a list prepared by each Comprehensive Center. The contractor also surveyed a nationally representative sample of state and district officials that they identified as a likely pool of customers for the centers and a sample of gatekeepers— individuals that had requested or negotiated for services on behalf of their school or school district. In randomly selecting customers to survey, the contractor did not choose a number large enough from each Comprehensive Center’s list to allow any reliable generalizations to be made about a particular Comprehensive Center. Likewise, the sample of potential customers and gatekeepers was not suitable to draw conclusions about an individual Comprehensive Center. The inclusion of representative activities and customers for each Comprehensive Center would have increased the cost of the mail survey. However, not drawing a sample that was representative at the center level reduced the usefulness of the evaluation because Education could not identify variation across Comprehensive Centers or obtain information to improve practices at individual Comprehensive Centers. The Regional Labs are unlike most federal education programs because neither the federal government nor state governments have oversight responsibility for their programs. Under the current structure, Education is accountable for the activities it funds through the Regional Labs, but current laws limit its ability to oversee those activities. Not only is federal oversight limited, states also have limited control over the regional agenda. Although the requirement that governing boards have a balanced representation of states in the region may ensure state input for the agenda of the Regional Lab, the regional priorities that the Regional Labs serve may not correspond to the educational priorities of all states in the regions. Thus, neither the states nor Education can ensure that the Regional Labs are meeting the needs of the states. Congressionally mandated peer reviews of the R&D Centers and Regional Labs have produced limited information about the overall performance of each organization, the services they provide, or the extent to which teaching and learning are improved by the products R&D Centers and Regional Labs produce. In part, this was because reviewers could not divide the tasks among themselves, as each reviewer was required to do a full, independent assessment. In addition, given the present practice allowing the Regional Lab directors to choose products and services for review, the potential exists for bias and therefore calls into question the quality of the assessments. As a result, Education lacks information that would be useful in making funding decisions or improving the performance of each organization. Unless standard program evaluation techniques, such as customer surveys or controlled case studies, are introduced into the evaluation process of these entities, these problems are likely to continue. The value of the mandated year 2000 evaluations of the Comprehensive Centers was limited. We recognize that addressing this problem would involve expanding the sample sizes and, for the consumer survey, the number of activities selected. However, currently, neither the Comprehensive Centers themselves nor Education can determine if the customers of a particular Comprehensive Center are satisfied with some or all of its products and services. As a result, problems at any given center could go unchecked. Moreover, Education cannot assess the relative strengths and weakness of individual Comprehensive Centers so it can improve the services in all centers and make better funding decisions. If the Congress wishes to ensure greater accountability to a governmental entity for the Regional Labs, it could consider either giving Education responsibility for determining the regional agenda and overseeing the products and services of the Regional Labs or Education could provide these funds to states, possibly as part of a larger formula grant, for subsequent distribution by each state. This would give states greater control in purchasing research-based educational products and services. If the Congress wants to increase the usefulness of the assessments of the R&D Centers and Regional Labs, the Congress should consider allowing Education to use methods other than peer review when such methods are more appropriate than peer review for evaluating the activities of R&D Centers and Regional Labs. To improve the assessments of the R&D Centers and Regional Labs, we recommend that the Secretary of Education direct the Assistant Secretary of OERI to use random selection of projects, services and products to be reviewed when conducting future evaluations of R&D Centers and Regional Labs, and revise the peer review standards to allow for division of labor and greater concentration on assessing the quality of projects, services and products. To improve the performance and usefulness of the Comprehensive Centers, the Secretary of Education should direct the Assistant Secretary of Elementary and Secondary Education and the Assistant Secretary of Bilingual Education and Minority Languages Affairs to design future evaluations of the Comprehensive Centers to produce findings pertaining to individual Comprehensive Centers. We provided a draft of this report to the Department of Education for comment. Education’s Executive Secretariat confirmed by e-mail that Department officials had reviewed the draft and had no comments except for a few technical clarifications regarding funding. We are sending copies of this report to the Secretary of Education, the appropriate congressional offices, and other interested parties. If you have any questions, please contact me at (202) 512-7015 or Eleanor L. Johnson (202) 512-7209. Other contributors can be found in appendix I. In addition to those named above, Malcolm Drewery, Jr., Tahra N. Edwards, Richard B. Kelley, and Sarah Moorhead made key contributions to this report. | Research and Development (R&D) Centers, Regional Labs, and Comprehensive Centers support the Department of Education's research agenda to various degrees. Because statutes define different missions and activities for these programs, the amount and focus of the research and other research-based activities they support varies. Education shapes the priorities that guide the research done by the R&D Centers and targets the technical assistance provided by the Comprehensive Centers through requirements in agreements with these entities. However, Education has little control over the activities of the Regional Labs because, unlike most federal education programs, neither federal nor state governments have oversight responsibility for their programs. The R&D Centers, Regional Labs, and Comprehensive Centers reported collaborating and coordinating with each other and Education and cited various factors that have either facilitated or hindered such activities. They said that they were most likely to engage in these activities when they shared a common interest in a specific student population, such as English language learners, or in a specific topic, such as assessment. Current evaluation practices for assessing the R&D Centers, Regional Labs, and Comprehensive Centers have provided only limited information about the performance of these organizations and have not been useful for making future funding decisions. |
Foreign schools can offer unique educational opportunities for U.S. residents, such as improved language proficiency and knowledge of other cultures, and help ensure that U.S. residents have a wide range of options in pursuing postsecondary education. The number of loans certified for U.S. residents attending foreign schools has risen from just under 4,600 in the 1993-94 academic year to over 13,000 in the 2000-01 academic year. Over 500 schools in 44 foreign countries are currently eligible to participate. About 9,000 of these students attend foreign medical schools and account for about three-quarters of the total loan volume. By country, the highest volume of FFELP loans—over $35 million—are for students attending school in Dominica; its sole eligible institution is a private, for- profit medical school. England, ranked fourth in loan volume, and Canada, seventh, have the largest number of institutions eligible to participate in the FFELP—182 and 108, respectively. Those countries participating in the FFELP and the top 10 foreign schools in loan volume for the 2000-01 academic year are indicated in figure 1. While a few foreign schools enroll large numbers of U.S. residents who receive FFELP funds, the majority of foreign schools enroll only a small number. For example, Queen’s College at the University of Oxford had just 3 students receiving FFELP funds in 2001. For more information on the ranges for numbers of U.S. residents receiving FFELP funds for attendance at schools in different countries, see appendix I. In order to participate in FFELP, foreign schools must submit a variety of documents, such as an application and a copy of the most recent course catalog. Once Education initially certifies the school for participation, the school enters into a Program Participation Agreement (PPA) with Education that requires it to comply with the laws, regulations, and policies governing FFELP. PPAs vary, but some may be valid for up to 6 years. To maintain its ability to participate, the foreign school must demonstrate that it is administratively capable of providing the education promised and of properly managing the program, and that it is financially responsible. Schools must submit program compliance audits and audited financial statement reports to Education on an annual basis. Program compliance audit reports are intended to demonstrate schools’ ability to administer FFELP in compliance with HEA and related Education regulations, while audited financial statements serve as evidence of schools’ financial responsibility. Schools must submit recertification materials to Education for continued participation in FFELP before the expiration of their current PPA. Education evaluates the application and accompanying documentation to determine whether a school is eligible to participate. Education’s Foreign Schools Team, consisting of eight staff members and one director, is responsible for assisting and overseeing foreign schools. Some of the ways in which the team oversees foreign schools, which are similar to the way Education oversees domestic schools, are presented in table 1. Education also has responsibility for maintaining information systems involved in the loan process, which is discussed more fully below. While Education and the foreign schools each have specific responsibilities, other parties are involved in the student loan process, including students, lenders, and guaranty agencies. Students are responsible for filing certain loan application materials, while lenders make loans, and guaranty agencies repay lenders the loan funds if the borrower defaults. Regardless of whether a student plans to attend a foreign or domestic school, a student applying for a FFELP loan is required to first submit a Free Application for Federal Student Aid (FAFSA). The student must also sign the Master Promissory Note (MPN), which outlines the students’ responsibilities for repaying the loan. The information provided by the student on the FAFSA is checked by Education against various information systems, including Social Security Administration databases and the National Student Loan Database System (NSLDS), to test the accuracy of information and to assess the student’s loan history. The administrators of the school the student plans to attend must certify the student’s eligibility for loans and the loan amount based on the output from the FAFSA. This output will indicate whether there are any issues with the student’s eligibility based on the information provided by the student and the edit checks against the various databases. For example, the output would indicate if the check against SSA’s databases revealed that the social security number provided did not match the name provided by the student, or if the check against Education’s NSLDS revealed that the student was in default on previous loans. In addition, the output includes the Expected Family Contribution, which is the amount the student and his family are expected to contribute to educational expenses. The administrators determine the student’s financial need based on this information, the cost of attendance, and the amount of financial aid other than FFELP funds that the student is expected to receive. Once the school has certified the student’s eligibility and loan amount and the student has signed the MPN, the lender can disburse the loan. Although lenders disburse loans for students attending domestic schools to the school, a chief difference for students attending foreign schools is that lenders may disburse loans either directly to students or to the foreign school the student is attending. The guaranty agency then sends to the student’s school a student status confirmation report (SSCR), which lists all students for whom loans were guaranteed for attendance at the school. School officials must indicate the enrollment status of these students and return the form to the guaranty agency. FFELP is vulnerable to fraud, waste, and abuse in several ways. Some school officials who do not have electronic access to Education’s information systems are improperly documenting and determining student eligibility for loans and are unaware of the proper procedures to do so, which could result in ineligible students receiving federal funds. In addition, Education has not conducted any on-site reviews to assess schools’ ability to administer FFELP since November 2000. Moreover, exposure to fraud, waste, and abuse is increased because students attending foreign schools, unlike students attending domestic schools, can choose to receive loans directly and in one lump sum for the entire academic year. Further, foreign schools do not submit required audited financial statements and program compliance audit reports, which compromises Education’s ability to monitor for and detect significant fraud or other illegal acts. Also, an investigation by our Office of Special Investigations revealed vulnerability in Education’s process for determining the eligibility of foreign schools to participate in FFELP. Interviews with foreign school officials and our review of school files revealed that some officials are not properly determining and documenting students’ eligibility for FFELP loans. As a result, FFELP funds may be provided to students who should not be receiving them. In particular, we found that several schools were using incorrect versions of documents Education generated to alert school officials to information that might indicate a student is ineligible for FFELP loans. We identified this problem among those schools that did not have electronic access to Education’s information systems that contain data needed to determine students’ eligibility for loans. Of the over 500 foreign schools participating in FFELP, only 32 can electronically access these information systems. However, these 32 schools certified about 70 percent of the total foreign school FFELP loan volume for fiscal year 2001. Electronic access to Education’s information systems can help ensure that schools use the correct information to determine whether students should be receiving FFELP loans. In accessing Education’s information systems, schools can obtain Institutional Student Information Reports (ISIR), which Education generates to help schools determine whether students are eligible for loans. ISIRs contain summary information provided on students’ FAFSAs as well as the results of various computer matches that Education conducts. ISIRs indicate, among other things, whether an applicant’s social security number reported on the FAFSA is valid, whether a loan applicant has ever defaulted on a student loan, and how much an applicant has previously borrowed. Electronic access to Education’s information systems under its existing procedures will not be granted unless a foreign school has among its staff a person who possesses a U.S. social security number. Few foreign schools meet this requirement. In the absence of obtaining ISIRs, foreign school officials must rely on and obtain from students a special eight-page version of the Student Aid Report (SAR), which is also generated by Education and contains information similar to that found in the ISIR. Education typically provides students with only an abbreviated two-page SAR, which summarizes information students submit on the FAFSA, but does not contain all of the information foreign school officials need to determine whether a student is eligible for a loan. Students must specifically request the special eight-page version from Education. Rather than documenting and determining student eligibility based on the eight-page SAR, we found that certain foreign school officials were improperly basing their student eligibility determinations on the two-page SAR. In reviewing files to determine if schools were properly determining and documenting students’ eligibility for FFELP loans, we found that the 2 schools with electronic access to Education’s information systems had copies of ISIRs for every student file we reviewed. Each of these schools had certified in excess of $30 million in FFELP loans and together certified about 30 percent of the total FFELP loan volume for fiscal year 2001. However, 5 of the 6 schools without access to Education’s information systems, which collectively certified over $3 million in FFELP loans for fiscal year 2001, did not have copies of ISIRs or eight-page SARs on file indicating that schools may have approved loans without obtaining the information necessary for determining student eligibility. Some school officials, in fact, told us that they verified students’ eligibility for loans based of the two-page SAR and were unaware that without the eight-page SAR or ISIR, students’ eligibility for loans could not be properly verified and documented. The inability of foreign school officials to electronically access Education’s information systems also creates the potential for delays in schools confirming and reporting student enrollment. Schools must confirm the enrollment of students who have borrowed FFELP funds through the use of a SSCR. Without timely and accurate reporting of student enrollment, detecting an individual who receives an FFELP loan but never enrolls in a foreign school is made more difficult. Schools that have electronic access to Education’s information systems can enter these data directly into Education’s information systems. Guaranty agencies can then retrieve data through these information systems and monitor whether students whose loans the agency has guaranteed are in fact enrolled in the foreign school. Schools that do not have electronic access to Education’s information systems, however, rely on guaranty agencies to send them SSCRs, which they then must return to guaranty agencies via postal mail. Several school officials told us that the inefficiency and lack of dependability of postal mail interfered with their timely submission of SSCRs. Education has not conducted an on-site program review—which is intended to assess, promote, and improve schools’ compliance with laws and regulations and help ensure program integrity—at a foreign school since November 2000. Program reviews can supplement the information provided to Education through the required annual audit reports and also help Education to monitor for fraud. Between March 1999 and November 2000, Education conducted six such program reviews at foreign schools (or the U.S. administrative office of the foreign school). As a result of these reviews, Education identified problems with how schools were administering FFELP. For example, the reviews revealed that some foreign school administrators had certified FFELP loans for students in excess of allowable loan limits and certified loans without verifying students’ eligibility for FFELP loans. However, a senior FSA official stated that because of budget constraints, on-site visits at foreign schools may not be a feasible use of Education’s funds at this time. Exposure to potential loss through instances of fraud, waste, and abuse is exacerbated by the fact that students attending foreign schools, unlike those attending domestic schools, may choose to receive loans directly from the lender rather than through their schools and may receive all loan proceeds in one lump sum for the entire academic year rather than receive the proceeds in multiple disbursements during the academic year. For example, Education’s OIG investigated a case in which a single individual submitted about 50 fraudulent loan applications for over $900,000 by falsely claiming enrollment at foreign medical schools. About 26 of the loans, totaling about $400,000 in FFELP funds, were disbursed to the individual before the fraud was detected. Such cases of fraud underscore the importance of foreign schools confirming and reporting student enrollment information to guaranty agencies. Over the past decade, Education’s OIG has investigated 90 cases of suspected FFELP fraud, many of which involved individuals requesting to receive loan proceeds directly and posing as foreign school students. During this same time period, according to an Inspector General official, the OIG recouped about $2.75 million in restitution from the successful prosecution of cases and prevented an additional $1.2 million from being disbursed. Many foreign schools have not submitted required annual audited financial statement and program compliance audit reports, which enable Education to monitor whether schools are using correct procedures to award, disburse, and account for the use of federal funds as well as help Education monitor for and detect significant fraud or other illegal acts. According to Education’s OIG Foreign School Audit Guide, the annual audit reports are the primary tools used by Education program managers to meet their stewardship responsibilities in overseeing FFELP. For fiscal year 2001, about 57 percent of foreign schools failed to submit audited financial statements. Collectively, these schools certified about $38 million in FFELP loans, about 17 percent of the total foreign school loan volume during the period. Further, Education regulations require foreign schools that certify $500,000 or more in FFELP loans during a fiscal year to have audited financial statements presented in U.S. Generally Accepted Accounting Principles (GAAP). For fiscal year 2001, nearly one- third of the foreign schools that certified $500,000 or more in FFELP loans failed to submit audited financial statements. Moreover, of those schools that certified $500,000 or more in FFELP loans and submitted audited financial statements for the period, over half did not submit statements presented into U.S. GAAP as required. (See table 2.) In addition to submitting audited financial statements, all foreign schools are required to submit program compliance audit reports on an annual basis. These reports address schools’ compliance with the laws and regulations that are applicable to FFELP. In fiscal year 2001, however, only 7 percent of all foreign schools submitted such reports. Of schools that certified $500,000 or more in FFELP loans, over 40 percent failed to submit program compliance audit reports. The vast majority of those schools that certified less than $500,000 in FFELP loans also failed to submit such reports. While those schools that submitted program compliance audit reports collectively certified about 75 percent of the total FFELP loan volume for fiscal year 2001, the remaining schools certified about $59 million in FFELP loans. (See table 3.) Interviews with foreign school officials and our review of school files revealed that some foreign schools do not provide loan counseling. Despite that default rates for foreign schools as a whole are relatively low, loan counseling is important because new students often have little or no experience with repaying and managing debt. Such counseling can help borrowers avoid defaulting on their loans, which can, in turn, help prevent waste from occurring in FFELP. Two of the schools we visited, which are also the schools with electronic access to Education’s information systems, had staff available to provide loan counseling and school officials reported doing so both prior to students’ arrival on campus and after students’ registration on campus. Other school officials, who had certified loan volumes ranging from $100,000 to about $1 million, stated that loan counseling was not provided as required by regulations. Education’s current eligibility certification process does not include conducting on-site visits to verify the existence of foreign schools. As we reported in November 2002, due in part to this weakness, Education granted approval to a fictitious foreign school that our undercover investigators created and which enabled our investigators to obtain approval for FFELP loans for fictitious students. To obtain approval to participate in FFELP, our investigators created various false documents required to be submitted with its PPA, including a course catalog, audited financial statements, and a letter purporting to be from United Kingdom government authorities acknowledging the school as a nonprofit, degree- granting institution. Education did not verify the existence of the school with foreign government officials or other parties or sources before certifying the school as eligible to participate in FFELP. After receiving approval of their fictitious school, our investigators also requested and obtained information necessary for the school to certify student eligibility for loans. Our investigators then sought FFELP loans by filing FAFSAs using three different fictitious student identities and applying for loans from three different lenders. Our investigators created false school certifications of these students’ eligibility for loans and also created false student enrollment reports. Two of the three lenders to whom our investigators submitted applications approved loans totaling, in the aggregate, $55,000, at which point we completed the investigation. Based on the results of our investigation, we recommended that Education implement a process, including conducting on-site visits, to ensure that a foreign school applying to participate in FFELP actually exists. Education has taken limited steps—since the beginning of 2002 and throughout the course of our audit work—to reduce the vulnerability of FFELP to fraud, waste, and abuse; however, its actions in some cases have been limited or have achieved limited results. In an effort to share more information with foreign school officials to help them comply with HEA and Education requirements, Education has increased the technical assistance it provides to foreign schools by publishing a reference guide and holding a series of training sessions. In addition, to assist foreign schools in complying with audit requirements, Education’s OIG issued a Foreign School Audit Guide in September 2002. However, interviews with foreign school officials and review of school files revealed that these efforts may not be sufficient to ensure that FFELP is being properly administered by the schools. Our review also found that the on-line training tutorial made available to foreign school officials on Education’s Web site does not contain information specific to foreign schools and even has information contrary to how foreign schools are to administer FFELP. Moreover, while Education requested that all foreign schools with overdue audited financial statements and certain schools with overdue program compliance audit reports submit them, it has not decided on the consequences for schools that do not comply with the request. Finally, in response to our fraud investigation, Education established new procedures for staff to use in certifying schools’ eligibility to participate in FFELP and provided its staff training on the new procedures yet no new foreign schools have been approved for participation in FFELP since the summer of 2002. Education has provided a reference handbook and training to foreign school officials; however, our interviews with several school officials and our review of schools’ files revealed that they remain unaware of how to properly administer FFELP, which may increase the risk of fraud, waste, and abuse occurring. In January 2002, Education issued the Student Financial Aid Handbook for Foreign Schools: 2001-2002. The Handbook was designed to help participating foreign schools achieve manageable, student-friendly administration of FFELP and to ensure that schools are aware of the legal requirements of participating in FFELP. According to FSA, the Handbook was mailed to all foreign schools participating in FFELP and it is currently posted to Education’s Web site. Education also held a series of training sessions for foreign school officials during 2002 in several locations, including Canada, Australia, England, Scotland, and Puerto Rico. Also, in September 2002, Education’s OIG issued a Foreign School Audit Guide, which assists foreign school officials in complying with the audited financial statement and program compliance audit requirements. To supplement this information, Education offers an on-line training tutorial, FSA COACH, for school officials’ use, although it was not specifically designed for foreign school officials. However, Education’s efforts to improve FFELP administration through training may have fallen short because knowledge of the training materials available was not widespread among the school officials we spoke to during our review. For instance, two foreign school administrators indicated that they had not received the Handbook from Education. In addition, as previously discussed, some foreign school officials were unaware of how to properly document and determine student eligibility for FFELP loans. Furthermore, although HEA regulations require training for officials at schools newly certified to participate in FFELP, Education officials did not provide information about training requirements or opportunities to our undercover investigators when we created the fictitious foreign school. An FSA official said that Education does not require foreign school officials to travel to the United States to attend available training before certifying a schools’ eligibility to participate in FFELP because of concerns about the financial burden on foreign schools. Instead, FSA provides training materials, along with information about how to use FSA COACH, to school officials. However, some administrators remain unaware of any on-line information, and when we interviewed foreign school officials at schools that have been participating in FFELP for a number of years, several indicated that they had not received training prior to administering FFELP. Even when training materials did reach FFELP administrators, these materials may have been insufficient to assist school officials. While some officials told us that they found the information and training useful, other officials told us that they did not. For example, several foreign school officials we spoke with indicated that the training sessions were very useful and indicated that holding such trainings more frequently would be valuable. One school official, however, commented that his peers found the regulatory and legislative information contained in the Handbook beyond their grasp, and that some of the information was confusing, especially for those school officials in countries where student financial aid is administered in an entirely different fashion than in the United States. Many other school officials commented on the need for better on- line information. Some found Education’s Web page difficult to navigate and some reported being unable to find needed information. Finally, while reviewing COACH, we found that much of the information contained within it was not applicable to foreign schools and, in some instances, it presents information that is contrary to how foreign schools operate. (See table 4.) While COACH was not designed specifically for foreign schools, Education directs foreign school officials to COACH for training materials upon certification, and the COACH tutorial states that it is a comprehensive introductory course on school requirements for administering FFELP and other student financial aid programs. In December 2002, during the course of our review, Education sent letters to all foreign schools requesting that they submit overdue audited financial statement reports. They also requested schools that certify $500,000 or more in FFELP loans to submit program compliance audit reports for the 4 most recent fiscal years. Education told the schools that failure to submit the requested documents within 45 days would result in consequences. Education is now considering revoking or denying schools’ certification to participate in FFELP if it did not receive overdue audited financial statement and program compliance audit reports. According to Education officials, FSA revised internal procedures for verifying schools’ legitimacy, and its foreign schools’ team was retrained. The retraining covered school eligibility requirements with an emphasis on the importance of validating with the appropriate foreign education office that a school is legitimate. To help staff verify that a school is legitimate, Education modified an internal checklist to include space for documenting the source and date of validation in the school’s file. Since learning of our investigation, Education verified the existence of all schools that are participating in FFELP, by either checking that the school is approved on an official Web site, or by corresponding or speaking with country education offices and ministries. Additionally, with respect to new applications from schools that have not previously participated in FFELP, Education no longer accepts a post office box address as the official location of a school or a third party servicer that administers FFELP for the school. Education has not yet implemented some planned changes in its procedures for determining FFELP eligibility of new foreign school applicants. Consequently, no new foreign schools have been certified to participate in FFELP since Education became aware of the school we created in May 2002, even though applications have been received from 19 schools. Education is currently considering implementing a process similar to that used when a domestic school applies for participation. This process would entail circulating the name of the school and its owners among a number of officials in FSA and other Education offices to determine whether staff have any information or knowledge that would affect a decision to certify the school’s eligibility to participate in FFELP. Education’s International Affairs staff, who coordinate the agency’s various international programs, would be among those to whom such information would circulate. If any staff were to raise concerns about the school or its owners, Education would consider conducting an on-site program review. Education could take additional action to address the goal of reducing the vulnerability of FFELP to fraud, waste, and abuse, such as more strictly enforcing school audit requirements or providing electronic access to information systems to help school officials more easily determine students’ eligibility for FFELP loans. However, any steps that Education takes will likely involve trade-offs that may affect access, accountability, and burden for various participants in FFELP. For example, Education could aggressively enforce foreign schools’ audit reporting requirements, but this may lead to unintended consequences, including limiting students’ access to such institutions if foreign schools withdraw from FFELP as a result. Other potential steps include changing disbursement procedures to help limit the federal government’s exposure to loss, but doing so may increase burdens for schools and students. In addition, providing foreign school officials with electronic access to information may help them properly determine student eligibility for FFELP loans, but may increase security risks. Additionally, we have developed tools that could help Education determine how to balance the objectives of providing U.S. residents with access to foreign schools while protecting the taxpayers’ investment that is intended to help provide that access. Education could more strictly enforce school audit report requirements, but doing so may limit U.S. residents’ access to foreign schools. FSA officials have stated that while Education is committed to maintaining the integrity of the FFELP program, it is also committed to providing access to international education opportunities for U.S. resident students and does not want to create barriers to those opportunities. As previously discussed, a large number of foreign schools have failed to submit required audited financial statement and compliance audit reports to Education in a timely manner. FSA officials told us that balancing enforcement of these statutory and regulatory provisions with providing students access to foreign schools is challenging. In their opinion, the current compliance audit requirements may place an undue burden and result in excessive costs for foreign schools that enroll few U.S. residents. Several foreign school officials we spoke to also told us that they found such audits to be costly, considering that students receiving FFELP loans constituted very small proportions of their student bodies. According to these officials, these audit requirements provide a disincentive to participate in FFELP in order to avoid what they perceive as an administrative and financial burden. Education officials are now considering whether to issue letters to foreign schools that certify less than $500,000 annually in FFELP loans requesting program compliance audit reports and whether an alternative approach to overseeing these schools should be taken. In addition to requiring foreign schools to submit audited financial statements and compliance reports, another potential step Education is considering relates to the requirement that certain schools submit audited financial statements under U.S. GAAP. Several school administrators and government officials in the United Kingdom told us that they found the requirement for schools to submit audited financial statements presented in U.S. GAAP to be burdensome, in light of the audit requirements of their home country. They stated that they believed that the United Kingdom’s accounting standards are sufficiently comparable to U.S. GAAP that Education should accept their statements for purposes of meeting FFELP statutory and regulatory requirements. Doing so, according to these officials, would reduce the administrative and financial burden associated with the requirement. Further, because Education’s regulation requiring that audited financial statements be presented under U.S. GAAP applies only to foreign schools that certify $500,000 or more in FFELP loans, these officials told us that foreign schools have an incentive to limit enrollment of students receiving FFELP loans so that they do not exceed this threshold. Education is currently considering whether to allow exemptions for foreign schools located in Canada and the United Kingdom—which collectively accounted for 314, or about 62 percent of the total foreign schools participating in the FFELP during academic year 2000-01—to its regulations requiring audited financial statements be presented into U.S. GAAP. According to an FSA official, the justification for such an exemption is based on the results of a comparison of several foreign countries’ auditing standards contained in Education’s policies and procedures manual, developed in consultation with a private accounting firm. While the purpose of the manual is to provide a methodology for FSA staff to use in assessing the financial health of foreign schools certifying less than $500,000 in FFELP loans, the manual does contain a limited analysis comparing the selected foreign countries’ accounting standards with U.S. GAAP and the potential effects of Education relying on foreign standards on the results of its analyses. Education could seek statutory, and consider regulatory, changes to loan disbursement procedures to address the potential for fraud, waste, and abuse; however, such changes could have a significant impact on schools and students. In our discussions with FFELP lenders and school officials, we found that disbursement methods and preferences vary among both lenders and schools. For example, representatives of one large FFELP lender told us that it is their standard operating procedure to disburse student loan proceeds directly to student borrowers by sending them checks. In contrast, a representative of another large FFELP lender, which specializes in making FFELP loans to students attending foreign medical schools, told us that it only (1) issues checks that are payable to both the student borrower and the foreign school and (2) sends these checks, or electronically transfers loan proceeds, to foreign schools, requiring student borrowers to obtain their funds through the schools. Some foreign school officials encourage students to receive their loan proceeds in this manner, as it helps the school maintain control of the funds. According to a guaranty agency official, a school official, and an FSA official some schools do not have financial aid offices or routinely carry out such functions at their institutions and therefore do not have the resources to be an intermediary between lenders and students. Other school officials told us that they are prohibited by local regulations from taking out student fees from loan checks and remitting the difference to students. In addition to receiving loan proceeds directly from lenders, students attending foreign schools may also receive loan proceeds in one lump sum rather than in multiple disbursements. According to many of the lenders and foreign school officials we spoke to, students frequently elect to receive their loan proceeds in this way, particularly students who are enrolled in 1-year graduate programs. Yet, several school officials told us that they prefer multiple disbursements for their students as the school is on a semester or trimester calendar and multiple disbursements provide them more assurance that expenses will be paid. One lending official, however, told us of an instance in which a student had trouble entering a country because she did not have sufficient proof that she had enough funds for the academic year. Thus, allowing students to receive loan proceeds in one lump sum might help students in such situations. Education is considering taking additional steps with respect to current disbursement procedures. As previously discussed and as documented by prior OIG investigations, the disbursement procedures used to provide loan proceeds to U.S. residents attending foreign schools exposes the federal government to increased risk for potential losses. Education is considering encouraging or requiring lenders to take steps prior to disbursing loan funds to students attending foreign schools. These steps could include (1) confirming that schools are eligible to participate in FFELP, (2) verifying that students are accepted for enrollment at foreign schools prior to disbursing funds, and (3) continuing to notify foreign schools when loan disbursements are made to student borrowers. Providing electronic access to Education’s information systems needed to determine student eligibility may help improve schools’ administrative capacity but may also increase information security risk. The lack of electronic access decreases schools’ administrative capacity, as foreign school officials have difficulty obtaining the documentation necessary to determine student eligibility and impedes the exchange of SSCRs with guaranty agencies. Education is currently working to address these issues and is considering providing foreign school officials with an alternative to requiring that someone on their staff possess a U.S. social security number to access its information systems. However, poor information security is a high-risk area across the federal government with potentially devastating consequences. Threats to the security of any data system may include attempts to access private information by unauthorized users, user error, as well as pranks and malicious acts. Potential damage arising from such threats could include, among other things, the disclosure of sensitive information, disruption of critical services, the interruption of services and benefits, and the corruption of federal data and reports. Therefore, Education needs to carefully weigh the benefits and risks of providing such access to foreign school administrators. We have found that conducting a risk assessment is one of several critical steps that agencies need to undertake to identify and address major performance challenges and areas that are at risk for fraud, waste, and abuse. We have also developed tools to assist agencies in undertaking such assessments. These tools provide a framework for identifying areas at greatest risk as well as various reports which can assist agencies in evaluating internal controls and addressing improper payments resulting from fraud, waste, and abuse. These tools could be useful to Education in weighing the advantages and disadvantages of various ways of overseeing and assisting foreign schools. Among other things, these tools highlight the importance of conducting risk assessments—comprehensive reviews and analyses of program operations to determine the nature and extent of program risks—and identifying cost-effective control activities to address identified risks. Foreign schools’ ability to participate in FFELP supports wide-ranging educational opportunities for U.S. residents and ensures that these students have a variety of options in pursuing postsecondary education. In light of recent events highlighting the vulnerability of FFELP with respect to U.S. residents attending foreign schools, Education has taken some important steps, and could take additional steps, both immediate and longer term, to decrease the vulnerability of the program. Ensuring that foreign school officials know how to properly administer the program, especially what steps they need to take to ensure that students are eligible to receive federal funds, is critical to reducing the program’s vulnerability to fraud, waste, and abuse. Education has taken steps to provide school officials with additional information concerning their responsibilities yet as we have shown, foreign school officials may need more information. Training that is convenient and specifically designed for foreign school officials could help bridge this information gap. Education is also considering what regulatory flexibilities it might extend to some foreign schools while also considering stricter enforcement of current statutory and regulatory provisions. The use of a risk assessment could help ensure that Education appropriately identifies the risks involved in the program and how best to balance the objectives of providing U.S. residents with access to foreign schools while protecting the taxpayers’ investment intended to help provide that access. In taking such actions, Education might identify alternative regulatory and oversight methods that would strike such a balance. To help ensure that foreign school officials have the knowledge necessary to properly administer FFELP, we recommend that the Secretary of Education develop on-line training resources specifically designed for foreign school officials. To better ensure that Education is adequately overseeing foreign schools participating in FFELP, we recommend that the Secretary of Education undertake a risk assessment to determine how best to ensure accountability while considering costs, burden to schools and students, and the desire to maintain student access to a variety of postsecondary educational opportunities. Further, after completing the risk assessment, if Education determines that legislative and/or regulatory changes are justified, we recommend that the Secretary seek any necessary legislative authority and/or implement any necessary regulatory changes. In written comments on our draft report, Education agreed with our reported findings and recommendations and, among other things, said that it has begun to reengineer its process for determining the eligibility of foreign schools to participate in FFELP. Education also provided technical clarification, which we incorporated where appropriate. Education’s written comments appear in appendix II. We are sending copies of this report to the Secretary of Education, appropriate congressional committees, and other interested parties. In addition, the report will also be available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staffs have any questions about his report, please contact me on (202) 512-8403 or Jeff Appel on (202) 512-9915. Gillian Martin and Cara Jackson made significant contributions to this report. | Recent events have increased concerns about the potential for fraud in Education's student loan programs related to loans for U.S. residents attending foreign schools. In 2002, GAO's Office of Special Investigations created a fictitious foreign school that Education subsequently certified as eligible to participate in the student loan program. GAO investigators subsequently successfully obtained approval for student loans totaling $55,000 on behalf of three fictitious students. Over the past decade, Education's Inspector General has investigated many instances of suspected student loan fraud involving individuals applying for loans for purported attendance at foreign schools. The conference report accompanying the 2001 Labor, Health and Human Services, and Education Appropriations Act mandated that GAO examine and report on fraud, waste, and abuse with respect to student loans for Americans attending foreign schools. Foreign schools offer unique educational opportunities for Americans and help ensure that U.S. students have a wide range of options in pursuing postsecondary education. Almost 70 percent of all U.S. residents receiving Federal Family Education Loan Program (FFELP) funds to attend foreign schools are in medical school and they account for three-quarters of the total loan volume. While some foreign schools participating in the FFELP enroll large numbers of U.S. residents, others enroll only a few, as seen in the table below, which also indicates the countries wherein FFELP loan volume is highest. We found that FFELP is vulnerable to fraud, waste, and abuse in several ways. For instance, many foreign schools do not submit required audited financial statements and program compliance audit reports, which would allow Education to monitor for and detect significant fraud or other illegal acts. For fiscal year 2001, about 57 percent of foreign schools failed to submit audited financial statements, while the vast majority of foreign schools failed to submit program compliance audit reports. Education has taken limited steps to address instances of vulnerabilities to fraud, waste, and abuse. For example, Education has issued a reference guide and conducted training for foreign school officials. However, a number of foreign school officials reported that they had not received training prior to administering FFELP funds. In addition, we found that some foreign school officials are not properly determining and documenting student eligibility for loans; as a result FFELP funds may be provided to students who should not be receiving them. We also found that the on-line training to which Education refers foreign school officials presents information in some cases that is contrary to how foreign schools are to administer FFELP. Education could take additional action to reduce the potential for fraud, waste, and abuse, but will have to address the trade-offs that arise from its actions that may affect student access and burden for various program participants. A comprehensive risk assessment is one method that Education could employ to determine how to balance an appropriate level of oversight with the desire to provide American students access to foreign educational opportunities. |
According to the Intergovernmental Panel on Climate Change—a United Nations organization that assesses scientific, technical, and economic information on the effects of climate change—global atmospheric concentrations of greenhouse gases have increased markedly as a result of human activities over the past 200 years. These gases trap heat that would otherwise escape the earth’s atmosphere, contributing to climate change. Climate change is a long-term and global issue because greenhouse gases disperse widely in the atmosphere once emitted and can remain there for an extended period of time. Among other potential impacts, climate change could threaten coastal areas with rising sea levels, alter agricultural productivity, and increase the intensity and frequency of floods and tropical storms. Carbon dioxide is emitted in by far the largest volume of any greenhouse gas, and most emissions are caused by fossil fuel combustion. According to the EPA, carbon dioxide emissions from fossil fuel combustion accounted for approximately 80 percent of all greenhouse gas emissions in 2007. Placing a price on emissions is likely to raise the cost of production of many goods and services. The size of the impact will depend on the price of allowances, as well as the ability of producers to substitute less emission-intensive processes and inputs. While some studies suggest that the overall impact would be modest, a cap-and-trade program could have a disproportionate effect on covered entities that rely heavily on fossil fuels, such as electricity generators. According to the Energy Information Administration (EIA), electricity generators derived about 49 percent of their electrical power from coal in 2007. The combustion of coal generates about twice as much carbon dioxide per unit of energy as the combustion of natural gas, the next most common fuel source for U.S. electricity generation, according to EIA. Due to changes in the regulation of electricity markets, certain companies may be limited in their ability to pass on emissions reduction costs to their customers. Historically, electricity was generated, transmitted, and distributed by local monopolies. These companies were overseen by regulators who restricted the entry of new companies, approved investments and retail prices, and determined profits. Since the 1970s, efforts have been made to “restructure” electricity markets by introducing more competitive conditions. At the wholesale level, federal regulators have introduced market-based pricing, although these markets can take a variety of forms. About half the states have made efforts since the 1990s to restructure how retail prices are set, generally seeking to increase competition in electricity sales. According to EIA, 14 of these states— located in New England and the upper Midwest, plus Texas—currently operate retail markets in which customers may choose among competing power suppliers. The other states where restructuring was introduced have either suspended or repealed these efforts. In the remaining states, regulators still approve utility costs and prices. In addition to covered utilities, which are mostly investor-owned, most states also have utilities that are owned either by the public (such as through a municipality) or cooperatively by customers themselves. Such utilities—which currently account for about one-quarter of electricity sales—generally set prices at cost instead of maximizing profits. In markets where regulation and international competition are not major factors, it is likely that consumers will ultimately bear most of the costs associated with pricing emissions. These costs are expected to disproportionately affect low-income consumers, who tend to spend a higher proportion of their incomes on energy products like electricity, heating, and gasoline. EPA has estimated that the cap-and-trade program in the American Clean Energy and Security Act would cost the average household $80 to $111 per year. A similar study by CBO estimated average household costs to be $175 per year, with some lower-income households receiving a net benefit. On the other hand, research suggests that policy makers could mitigate or eliminate these effects by selling allowances to covered entities through auctions and returning the revenue back to consumers in the form of lump-sum rebates or tax adjustments. If the government were to ‘recycle’ revenue through tax reductions, it could also realize benefits to the overall economy in the form of “economic efficiency.” Many economists view certain taxes as inefficient or “distortionary,” because they shift resources away from their most highly valued use. For example, efficiency costs may arise because taxes on labor income may affect job choices or hours worked. Most economists agree that minimizing the efficiency cost of the revenue raised to fund government services is an important objective of tax policy, among other objectives such as distributing the burden of taxation equitably. The effects of emissions pricing on consumers and industry will also vary by region. While some recent studies suggest that this variation would be minimal, it may be more substantial for low-income households. Areas that get most of their electricity from coal, the most emissions-intensive source, may see a greater electricity cost increase than areas that rely heavily on natural gas, nuclear energy, or hydropower. One study has estimated that the cost burden as a percentage of household income would range from about 1.9 percent in the East South Central region to about 1.5 percent in the West North Central region. A cap-and-trade program would also affect federal, state, and local governments, which purchase energy intensive goods and would be responsible for the program’s implementation. According to one study, governments produce approximately 13 percent of U.S. carbon dioxide emissions, and the allowance consumption associated with these emissions could cost governments an additional $16.6 billion. Furthermore, price increases could increase government payments—such as Social Security benefits and federal pensions, which are indexed to prices—and reduce personal income tax collections. Finally, depending on the details of the program, a cap-and-trade program could increase the administrative burden on the government relative to a business-as-usual situation. For example, markets for emissions allowances would require oversight, and the distribution of auction revenues could require additional personnel or a new entity to administer payments. The design of a cap-and-trade program’s allowance allocation plan—the ways in which tradable allowances are allotted to covered entities at the outset of the program—will help determine how costs and benefits are distributed across the economy, according to available literature. The method of allowance allocation will generally not affect the level of emissions reductions achieved by the program, because allocation is independent of the overall cap. Therefore, the principal consideration in designing an allowance allocation plan is how to distribute the allowances in a way that helps to achieve certain goals: for example, to offset the program’s economic impact on disproportionately affected industries or to generate revenue that could be redistributed to consumers or used for other purposes. To accomplish these goals, three basic design choices are available: allowances may be sold through an auction or other means, distributed for free, or dispensed using a combination of these methods. Selling allowances to regulated entities could provide several benefits. First, it would generate a source of revenue that the government could use to defray the economic costs associated with emissions reductions or direct toward other purposes. These revenues could be substantial: in June 2009, CBO reported that the American Clean Energy and Security Act would generate annual revenues of $45 billion by the year 2019 by auctioning of a percentage of the allowances. Earlier CBO estimates indicated that annual allowance revenues could range between $30 billion and $300 billion by roughly the same time period if all allowances were auctioned, although this proposal is not part of the bill. Some existing cap-and-trade programs have already sold allowances through auctions or commodity exchanges. For example, several member states participating in the European Union’s Emissions Trading Scheme (ETS)—including Ireland, Hungary, Lithuania, the United Kingdom, and Germany—have generated revenues from allowance sales; in Germany, these totaled approximately $1.2 billion in 2008. The level of auctioning is expected to increase as the program moves toward its third phase, which is to begin in 2013. In the United States, the Regional Greenhouse Gas Initiative—a regional cap-and-trade program involving 10 northeastern states—has conducted four auctions since it began auctioning allowances in 2008. These auctions, held quarterly, have each raised between $38 million and $118 million for programs to promote energy efficiency and assist low-income households with energy costs, among other things. Given the revenue generation potential of auctions, many experts we consulted as part of a prior study suggested that a cap-and-trade program should maximize the level of auctioning. Auctioning may confer other additional benefits, according to available literature and researchers we spoke with. For example, many economists favor auctioning because of its transparency and because it discourages behaviors motivated by a desire to gain free allowances, such as “baseline inflation.” This occurs when a firm attempts to boost the number of allowances it receives by increasing its emissions prior to the outset of a cap-and-trade program. Auctioning can also help ensure that new entrants to an industry face the same emissions reduction costs as existing firms. Finally, auctioning could decrease the possibility that covered entities earn windfall profits as a result of the cap-and-trade program, particularly in restructured regions where prices are determined largely by market factors. Covered entities could earn windfall profits if they pass along the “opportunity costs” of free allowances—that is, the revenue foregone by not selling them—in the form of increased electricity prices. For example, in the first phase of the European Union’s ETS, electric utilities that received free allowances reaped substantial profits by charging ratepayers for the opportunity cost of those allowances. On the other hand, auctioning does not offer compensation to covered entities, particularly those that face disproportionate costs due to a cap- and-trade program. The government will also incur certain administrative costs associated with designing and administering the auctions, although these activities could be funded using part of the auction revenues. Moreover, the effectiveness of allowance auctions will depend partly on their design. Free allocation could help establish political support at the outset of a cap- and-trade program and compensate covered entities for any decrease in profits they might experience as a result the program, but it could also have some disadvantages. Two principal options are available when allocating allowances for free: “grandfathering” or “output-based updating allocation.” Grandfathering involves allocating allowances based on historic (pre-regulation) emissions measures, while output-based updating allocation involves adjusting the number of allowances provided to an entity based on its recent production levels. Available literature indicates that since past emissions measures do not change, grandfathering may be less susceptible to manipulation than output-based updating allocation. However, research suggests that grandfathering is unlikely to prevent the “leakage” of economic activity—including production, jobs, and emissions—to countries where greenhouse gases are not regulated. As we have previously reported, leakage may be of particular concern to firms in certain energy-intensive industries that face international competition—such as primary metals, paper, and chemicals—as these firms could find it more difficult than other covered entities to pass on costs to consumers by raising prices. Grandfathering could also provide an advantage to existing facilities, which are more likely to have outdated, inefficient technologies in place. Output-based updating allocation could also present trade-offs. As we have previously reported, output-based updating allocation could provide incentives for covered entities to maintain or increase production, potentially reducing the likelihood that these entities would move production to countries that are not subject to emissions regulations. However, output-based updating allocation could also decrease incentives for covered entities to engage in conservation and reduce their energy intensity, depending on how the program is designed. Moreover, some research indicates that an output-based approach would subsidize entities in certain industries, forcing entities in other sectors to make deeper cuts in their emissions in order to meet the overall cap. Since these cuts may be more expensive than the reductions that would have otherwise taken place, the overall cost of the cap-and-trade program would increase. In addition, some research suggests that maintaining output may not always be a worthwhile goal: for example, the contraction of output from a high- emissions sector may be one of the most cost-effective means by which to reach the overall emissions target. Furthermore, attempts to keep energy prices low could increase the cost of the program to the economy. Rising prices for energy and energy- intensive goods are critical to the success of the program, because these “price signals” create incentives for both covered entities and consumers to conserve energy, and thereby reduce emissions of greenhouse gases. To the extent that price signals are not preserved, fewer households and businesses will change their behavior in response to these signals. This could reduce the economic efficiency of a cap-and-trade program, since some of the less costly emissions reduction opportunities would be forgone. The structure of the U.S. electricity generation sector—which represents roughly 40 percent of domestic carbon dioxide emissions—could affect whether price signals reach energy users. Since the price of electricity is regulated in certain regions, generators that receive free allowances in these regions may not be able to pass along the costs associated with an emissions price to residential and commercial electricity users. If costs are not passed through, incentives for conservation decrease. A diminished price signal could also have indirect effects—for example, if the price of energy intensive goods does not rise in relation to other goods, consumers have less of an incentive to purchase fewer of these goods. Considering the limitations of free allocation, some analyses have advocated limiting the use of free allowances to specific subsets of carbon intensive industries. Several studies suggest that freely allocating between 6 and 21 percent of all allowances would be enough to compensate these industries—which include coal-fired power plants, fossil fuel suppliers, and energy intensive manufacturers—for profit losses related to emissions regulation. In 2007, CBO reported that less than 15 percent of allowances would be sufficient to offset net losses in stock value as a result of the program. The establishment of a cap-and-trade program creates opportunities for the government to direct the value of allowances in a variety of ways. For the purposes of this testimony, we assessed five options that are frequently discussed in the economic literature, although numerous other options exist. First, the government could reduce the overall cost of the program by reducing taxes on capital or income that currently make the economy less efficient. Second, the government could distribute lump-sum rebates to consumers, who would likely pay the bulk of the economic costs associated with a cap-and-trade program. Third, revenues could be used to expand the Earned Income Tax Credit to assist low-income working families. Fourth, policymakers could compensate covered entities for their increased costs through free allocation—an approach equivalent to selling allowances on the market and transferring all the revenue to covered entities. Finally, revenues could help fund climate-related programs or activities, including research and development, energy- efficiency programs, or international aid to developing countries that face challenges in mitigating and adapting to climate change. Using program revenues to reduce marginal tax rates—whether from individual income or payroll or taxes, corporate income taxes, or taxes on capital gains or investments—can reduce economic distortions in the tax code and lower the overall cost of the program. The benefits of tax reduction depend on the extent to which these taxes currently distort economic activity, according to literature and economists we spoke with. For example, existing taxes on labor or capital can discourage individuals from participating in the labor force or investing money. The structure of the tax code can also create distortions by directing spending toward certain areas where the buyer has a tax advantage, such as homeownership or employer-provided medical insurance. A cap-and-trade program could further exacerbate these tax distortions, according to economic literature. This so-called “tax interaction effect” could occur because a cap-and-trade program may have some of the same effects as a tax. Specifically, covered entities that face additional costs due to an emissions price will generally pass on their increased costs to consumers in the form of higher prices, thereby reducing the amount of goods that consumers can purchase. Because a loss of purchasing power effectively represents a decrease in real wages, incentives to work may also decrease. These effects could ultimately raise the cost of the program to the economy, according to economic literature we reviewed. However, ‘recycling’ auction revenues through the tax code could partially or wholly offset costs that result from inefficiencies in the tax code, as well as potential costs imposed by the cap-and-trade program, according to a review of economic literature and interviews conducted with economists. For example, because an emissions cap could cause prices to rise—and real wages to fall—a reduction in labor, income, or capital taxes could provide efficiency gains and help reduce the overall cost of the program. These efficiency gains may present trade-offs. Economic analyses suggest that reducing tax rates would do little to compensate low-income individuals that may be disproportionately affected by the cap-and-trade program. According to these analyses, most benefits from reduced taxes would accrue to higher income households, regardless of the tax targeted for reduction. Moreover, in the absence of supplemental policies, the benefits of reducing labor taxes will not reach individuals who do not file tax returns. To close this gap in coverage, the government could supplement a tax reduction with payments issued through existing systems, such as the Electronic Benefit Transfer system or state-based food stamp programs. However, using a combination of systems could increase the administrative burden and complexity of the program, and may require additional governmental coordination. In addition, adjusting the payroll tax rate may be complicated since these taxes represent social security and Medicare financing contributions. Another way to distribute revenues to consumers would be to distribute lump-sum rebates to consumers. Such a program could take many forms, but the underlying goal would be to compensate consumers or households through rebates of a specific amount. The amount of the rebate could be based on a simple per-capita formula with checks of equal size—also known as “cap-and-dividend”—or could account for household size, region, or other factors. An important advantage of lump-sum rebates, according to many economists, is that they help offset the costs of a cap-and-trade program on consumers, particularly on low-income households. Depending on the design of the program, certain consumers may even experience a net benefit. However, research indicates that distributing lump-sum rebates would forgo the efficiency gains that could be achieved through tax reductions, making the program comparatively more expensive to the economy overall. The ultimate cost of lump-sum rebates and the resulting effects on consumers would depend in part on the program’s administration. The funds could be distributed, for example, using existing government programs, such as the income tax system or other benefit transfer programs. For example, one economist has proposed that the government could provide rebates for taxes paid on the first $3,660 of each worker’s earnings, leading to a maximum rebate of $560 per worker. Alternatively, the government could develop a new distribution mechanism, although this approach would carry additional administrative costs. While using a single existing mechanism for rebate delivery would be the simplest and most transparent option, it would exclude individuals that did not participate in that program—for example, rebates that use the tax system would exclude individuals that do not file tax returns. The government could encourage these individuals to file through outreach campaigns, a strategy used when stimulus checks were distributed under the Economic Stimulus Act of 2008. Evidence suggests that such efforts could encourage more individuals to file—for example, of the 150 million individual income tax returns processed for tax year 2008, approximately 9 million claimed only the economic stimulus payment. However, any outreach effort would entail additional costs and administrative requirements. Policymakers could also design a rebating system that uses a combination of mechanisms to maximize coverage, although this strategy would increase the program’s complexity, given the need for program coordination, as well as the risk of fraud or duplicate rebates. Several proposals for distributing revenue involve expanding the Earned Income Tax Credit (EITC) program. The EITC was enacted in 1975 and was originally intended to offset the burden of Social Security taxes and provide a work incentive for low-income taxpayers. It is a refundable federal income tax credit, meaning that qualifying working taxpayers may receive a refund greater than the amount of income tax they paid for the year. According to one study, approximately half of all households would benefit from this approach, with lowest-income households with children reaping the highest gains. However, this study suggests this option would affect low-income households differently depending on their location. Low-income households in the Northeast, for example, could see about a 2 percent gain in income, compared to a 7.4 percent gain in Texas. Some research also indicates that the EITC may encourage labor activity for low- income workers. Using the EITC to distribute revenue, however, may involve trade-offs. For example, as the Treasury Inspector General for Tax Administration has reported, the EITC has been vulnerable to taxpayer error in the past, due in part to changes in eligibility and the tax code. Prior reviews by the IRS and GAO also suggest that errors are common—for example, an IRS study has reported that the EITC program has an erroneous payment rate estimated to be between 23 and 28 percent. Allocating free allowances to covered entities can help establish political support at the outset of a cap-and-trade program and compensate covered entities for any increased costs they incur as a result. However, as noted earlier, free allocation can raise the cost of the program if such allocation decreases incentives to conserve energy and reduce emissions in one sector and forces other sectors to make less efficient reductions. In addition, economic literature suggests that a grandfathering approach to free allocation would do little to discourage the leakage of economic activity, jobs, and emissions, since covered entities’ variable costs of production would remain unchanged. An output-based approach to free allocation, on the other hand, could reduce the likelihood that covered entities would relocate or decrease production, although it could also reduce their incentives to decrease emissions. Most of the benefits of freely allocated allowances will accrue to the shareholders of entities that receive them by compensating shareholders for any declines in stock value they might experience as a result of the cap. However, consumers are unlikely to see these benefits in the form of lower prices, since most covered entities will pass on costs associated with a cap-and-trade program, even when they receive allowances for free. Free allocation is therefore likely to benefit those with higher incomes more than those with lower incomes. The administrative burden associated with free allocation of allowances depends primarily on how policymakers determine the relative allocations to each industry. A grandfathering approach, for example, would require the government to select a set of years with which to determine a baseline. An output-based approach would require the government to define a baseline, which could prove challenging. As one economist we interviewed pointed out, “output” could be subject to numerous interpretations, each with its own implications for equity. The government could also direct the recipients of free allowances to use these allowances for the benefit of consumers. For example, HR 2454, as passed by the House on June 26, allocates some allowances to electric and natural gas local distribution companies (LDC) for the benefit of retail ratepayers. Distributing free allowances through LDCs may go some way toward mitigating regional differences in cost impacts, according to some researchers. However, the overall effects of this approach would depend largely on the extent to which it creates incentives to reduce energy use, according to economists we spoke with. Importantly, if benefits to electricity customers were conferred in the form of decreased energy rates, the incentives for energy conservation may diminish and the overall cost of the program could increase. This may be particularly true for residential customers, according to economists we interviewed, since industrial customer may have other reasons to pursue efficient practices. To help preserve incentives, LDCs could allow electricity rates to rise and rebate consumers through the fixed portion of their utility bills—that is, the portion not based on energy use. However, this approach assumes that electricity customers will differentiate between the fixed and variable portions of their utility bill when assessing their costs, as opposed to simply looking at the bottom line amount, which could remain largely unchanged. Several economists and researchers we spoke with expressed skepticism that customers would react to the price signal if their total energy costs did not change, although some said that distributing rebate checks separately from the utility bill could address this concern. The effect of this approach on consumers will depend on other factors. If both residential and business customers receive benefits, for example, the benefits conveyed to businesses may not get passed along to their customers. According to a CBO analysis of H.R. 2454, most of the allowance value given to local distribution companies would benefit business customers. The analysis also estimates that 63 of percent allowance values conferred to businesses would ultimately benefit the highest earning 20 percent of households, since these households are more likely to be shareholders. In addition, the way in which benefits are conveyed to customers—for example, through lower prices, investments in energy efficiency, or other means—will depend on the state public utility commissions that regulate the LDCs. While some organizations have expressed concern that past regulation has been uneven, several economists and state officials we spoke with expressed confidence that the existing regulatory structure could effectively ensure that customers received the benefits. Revenues generated through allowance auctions could also be directed toward climate-related programs or activities, including the research and development of low-carbon technologies, programs to promote energy- efficiency, or mitigation and adaptation activities abroad. Beyond their environmental benefits, such programs could also convey efficiency gains, if they lowered the cost of emissions reductions. The development of renewable energy sources, for example, could ultimately lower covered entities’ total expenditures on emissions allowances. Funding for efficiency programs could also offset costs for households through reduced energy demand. Some research organizations have also suggested that funding in these areas could create job opportunities, and in the long run could help ensure greater economic stability due to energy security. Economic research suggests that an emissions price, on its own, will go some way toward promoting low-carbon technologies and the efficient use of energy. However, economists we spoke to said that there are certain instances—known as “market failures”—where opportunities for reduction may not be captured. For example, builders and owners of rental properties may not have incentives to consider energy efficiency in the construction and renovation of these properties, since they may not be responsible for paying electricity and heating costs. In these cases, subsidies for efficient construction or renovation may be appropriate. In addition, certain technologies—such as carbon capture and storage—may face cost barriers that could be mitigated through grants or subsidies. Other technologies may need nationwide infrastructure that could require additional funding at the federal level—for example, an enhanced transmission grid to transmit renewable energy. While many economists we spoke with said funding such activities could be beneficial, several also cautioned that selecting, implementing and evaluating these programs could pose challenges. from a power plant’s emissions, transporting it to an underground storage location, and then injecting it into a geologic formation for long-term storage. result, several economists we spoke with recommended allocating part of the allowance revenues for research and development to help overcome these cost barriers. However, several also noted that it is difficult to determine how to allocate such funds effectively. For example, selecting which technologies receive funding places the government in the position of attempting to choose the best technologies rather than allowing the market to make that determination. Overall, research suggests that funding technologies in the early stages of development may be more cost- effective than using revenues to commercialize existing technologies. Investments in energy efficiency have the potential to alleviate some of the effects of the cap-and-trade program on households. For example, using auction revenue to support weatherization improvements for homes or the purchase of energy-efficient appliances could lower these households’ energy consumption and expenditures. Some research suggests that tax credits, for example, can have a significant impact on efficiency investments by homeowners and businesses. However, several researchers have noted that the implementation of such programs has been unpredictable in the past, in part because it is difficult to determine whether these activities would have occurred anyway. Allowance revenues could also be used as aid to developing countries, either in the form of grants, loans, or other means of assistance. Such aid could target activities that reduce greenhouse gas emissions in these countries—for example, programs that aim to deploy low-carbon technologies in areas where they would not normally be financially feasible. Revenue could also support adaptation activities that could help these countries prepare for and adjust to the project effects of climate change. Several economists and researchers we spoke with supported directing some portion of auction revenue for international aid efforts. Some highlighted an obligation on the part of developed countries, which represent the bulk of greenhouse gas emissions to date, to help less developed nations deal with potential problems associated with climate change, such as food shortages, water quality problems, and the increased risk of malnutrition or disease. In addition, research indicates that the developing world presents low-cost opportunities for emissions reduction—for example, avoiding landfill waste through composting—as well as opportunities to prevent future emissions in those countries that are rapidly developing their energy, industrial, and transportation infrastructures. Furthermore, some researchers noted that the provision of mitigation or adaptation aid to developing countries may essentially be a prerequisite to these countries’ participation in an international agreement to limit emissions. Mr. Chairman, this concludes my prepared statement. I would be happy to respond to any questions that you or other Members of the Committee might have. For further information about this statement, please contact John Stephenson, Director, at (202) 512-3841 or stephensonj@gao.gov. Individuals who made key contributions to this testimony include Michael Hix, Assistant Director; Cindy Gilbert; Robert Grace; Richard Johnson; Jessica Lemke; Ben Shouse; Jeanette Soares; Ardith A. Spence; and Vasiliki Theodoropoulos. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Congress is considering proposals to establish a price on greenhouse gas emissions through a cap-and-trade program that would limit overall emissions and require covered entities to hold tradable emissions permits, or allowances, for their emissions. The purpose of such a program is to raise the cost of activities that produce emissions and thereby provide an economic incentive to decrease emissions. Carbon dioxide, which results from burning fossil fuels, is the primary greenhouse gas and accounts for about 80 percent of U.S. emissions. A cap-and-trade program would increase the cost of burning fossil fuels and other activities that generate emissions and potentially raise costs for consumers. A key decision is the extent to which the government offsets these costs. For example, the government could sell the allowances and then return the revenues to covered entities or households. The government could also give away some or all of the allowances. According to the Congressional Budget Office, the value of the allowances could total $300 billion annually by 2020. Today's testimony provides preliminary results of ongoing work assessing the potential effects of (1) allowance allocation methods, and (2) options for distributing program revenues or the economic value of allowances. GAO reviewed economic literature and interviewed experts in climate policy, including those involved in existing cap-and-trade programs. The method for allocating allowances in a cap-and-trade program can have significant economic implications for the government, regulated entities, and households. Most importantly, a cap-and-trade system would create a market for a valuable new commodity: emissions allowances. The government could allocate these allowances to regulated entities in three main ways. First, it could auction all of the allowances and collect a significant amount of revenue that it could use, for example, to compensate households affected by the cap-and-trade program. Second, it could give away the allowances to entities affected by the program and thereby transfer the value of the allowances to those entities. This could enhance the program's appeal to covered entities but could also increase the program's overall cost to the economy if it reduced incentives for those entities to decrease their emissions. Third, the government could give away some allowances and auction the rest. For example, studies have suggested that freely allocating 6 to 21 percent of the allowances created by a cap-and-trade program would be sufficient to compensate entities in energy-intensive industries for any profit losses incurred as a result of the cap-and-trade program. According to the economic literature and economists we interviewed, regardless of the mechanism for distributing allowances, consumers will bear most of the costs of a cap-and-trade system because most regulated entities will pass along their increased costs in the form of increased prices; however, these costs could be largely offset depending on how revenues are used. Available literature and economists we interviewed point to five main options for distributing a program's allowance revenues, although numerous other options exist. First, the government could lower the overall cost of the cap-and-trade program to the economy through accompanying reductions in taxes on income, labor, or investment. Second, auction revenues could be distributed to households through lump-sum payments, which could offset the higher consumer prices resulting from a cap-and-trade program and mitigate any disproportionate impacts on low-income households. Third, the government could expand the scope of the Earned Income Tax Credit to further benefit low-income working families. Fourth, the government could compensate regulated entities and their shareholders for lost profits by allocating them free allowances. Finally, revenues might be used to fund climate-related programs, such as research on low-carbon technologies, or used to support climate change mitigation activities in developing nations. Each potential use of revenues has trade-offs. For example, decreasing tax rates could lower the overall economic cost of the program; however, this approach may do little to compensate low-income consumers, who would receive greater benefit from a direct rebate. In addition, using revenues to dampen increases in energy prices may benefit ratepayers but reduce their incentives to conserve energy, potentially increasing the program's overall cost. |
As one of the largest and most complex organizations in the world, DOD spends billions of dollars each year to operate, maintain, and modernize IT systems that support core business operations. These business operations consist of various interrelated and interdependent business functions, including logistics management, procurement, health care management, and financial management. Further, execution of these business operations span a wide range of defense organizations, including the military services and their respective major commands and functional activities, defense agencies and field activities, and combatant and joint operational commands that are responsible for military operations in geographic regions or theaters. For fiscal year 2005, DOD requested approximately $13 billion to operate, maintain, and modernize about 4,150 business systems and related IT infrastructure. The Army’s share of this funding and systems is about $2.7 billion and 727 systems. In 1995, we designated DOD’s business systems modernization efforts as a high-risk program, and we continue to designate it as such today for several reasons, including the department’s challenges in implementing effective IT investment management structures and processes, developing and implementing an enterprise architecture, and implementing effective IT system acquisition and development processes. In the early 1990s, DOD recognized the need to integrate defense IT systems by ordering the consolidation of systems that serve multiservice purposes. Accordingly, it directed that approximately 150 operating transportation systems be evaluated for consolidation opportunities. In 1995, DOD initiated TC-AIMS II as one of a number of programs that were commenced following this evaluation to consolidate and replace existing systems, and to automate transportation management function areas for all services. In doing so, DOD designated the Army as the “executive agent” responsible for managing TC-AIMS II, including acquiring the system. Expected TC-AIMS II benefits included enhancing and improving the efficiency and effectiveness of the support planning needed to deploy and redeploy forces and equipment, both within the United States and abroad; improving the visibility of assets; and enhancing cargo and passenger receiving, controlling, and shipping. The system is also expected to enable decision makers at various command levels to access information about unit movement and installation transportation, and is to provide key shared functionality related, for example, to load planning. The Army defined the following five-block acquisition and implementation strategy for TC-AIMS II. According to the strategy, each block is to address a functional portion of the requirements. Block 1: This block is to import data about assets (people, equipment, and supplies) from the services’ existing systems and is to manage these data to (1) plan and execute organizational moves (i.e., provide documentation, packing and shipping labels, and orders) and (2) request and schedule strategic and local transportation vehicles (e.g., trucks, ships, airplanes, motor pools, and trucks). Block 2: This block is to add Web access to permit automated data exchange with related systems, such as airlift load planning systems and global transportation networks. This increment is to link and track cargo and personnel (via DOD’s standard smart cards), and their corresponding transports. It also is to generate and manage the documentation associated with these functions. Block 3: This block is to add interfaces to both supply and personnel systems to facilitate control over “in theater” operations, such as reception (unit arrival and off-loading in theater), staging (preparation for departure to assigned locations), onward movement (transit to assigned locations), and integration (coordinating and linking the movement of multiple related units). It also is to generate and manage documentation associated with these functions. Block 4: This block is to add the capability to manage maritime forces associated with prepositioned equipment and cargo (e.g., tank ammunition for an armored cavalry unit), including the ability to manage cargo logistics (loading and off-loading), as well as greater capability to manage theater operations, including the availability, distribution, and dispatch of vehicle fleets and drivers. It also is to generate and manage documentation associated with these functions. Block 5: This block is to automate the management of transportation assets and traffic operations at military installations both in the United States and worldwide. At this point, TC-AIMS II is to be able to generate and manage all data and documentation related to moving units within DOD’s transportation system. As part of this block, the interfaces to systems that are to be integrated with TC-AIMS II are to be upgraded or created, as required, in response to changes in technology. Figure 1 shows the Army’s timeline for acquiring the five TC-AIMS II blocks. In particular, the Army reports that it is currently acquiring Block 3 with the goal of making it operational by late 2006. The Army estimates the total life-cycle cost of TC-AIMS II to be $1.7 billion over 25 years, including $569 million for acquisition and $1,168 million for operations and maintenance. The Army reports that it has spent approximately $751 million on TC-AIMS II since its inception. A number of DOD and Army organizations are involved in managing the TC-AIMS II program and acquisition. Briefly, they are as follows: The Assistant Secretary of Defense for Network Information and Integration, who has overall responsibility for the program. The Army’s Program Executive Office for Enterprise Information Systems, Transportation Information Systems, oversees the management of TC-AIMS II, which is managed on a day-to-day basis by the Project Manager, Joint Program Management Office. The Joint Requirements Board represents TC-AIMS II users and includes stakeholders from the military services and other DOD organizations. The board is responsible for managing TC-AIMS II requirements development activities. These and other key organizations and their roles and responsibilities are described in more detail in table 1. The Army uses a system integration contractor to support the program office in managing the acquisition of TC-AIMS II. Most recently, it awarded a firm, fixed-price contract to a contractor in April 2004 to deliver the Block 3 solution. The contract has 5 option years, which are 1 year in duration. To help it oversee contractor activities, the program office has employed the help of the General Service Administration’s Federal Systems Integration and Management Center, whose mission is to provide acquisition management expertise and support to DOD and other federal agencies on large, complex IT projects. In July 2004, we reported that DOD’s revised systems acquisition policies and guidance incorporated many best practices for acquiring business systems, such as (1) justifying system investments economically, on the basis of costs, benefits, and risks, and (2) continually measuring an acquisition’s performance, cost, and schedule against approved baselines. However, the revised policies and guidance did not incorporate a number of other best practices, particularly those associated with acquiring commercial component-based business systems, and DOD did not have documented plans for incorporating these additional best practices into its policies. We also reported that the department’s revised acquisition policies did not include sufficient controls to ensure that military services and defense agencies would appropriately follow these practices. We concluded that until these additional best practices were incorporated into DOD’s acquisition policies and guidance, there was increased risk that department system acquisitions would not deliver planned capabilities and benefits on time and within budget, and increased risk that DOD organizations would not adopt and use best practices that were defined. Accordingly, we made 14 recommendations to the Secretary of Defense that were aimed at strengthening DOD’s acquisition policy and guidance by including additional IT systems acquisition best practices and controls for ensuring that these best practices were followed. DOD agreed with most of our recommendations and has since issued additional system acquisition guidance. The Army, as DOD’s acquisition agent for TC-AIMS II, has not managed this program in accordance with certain department acquisition policies, and related guidance, that are intended to reasonably ensure that a proposed system is the right solution to meet mission needs. In particular, the Army has not economically justified the program on the basis of reliable estimates of life-cycle costs and benefits, and it has not ensured that this program, which is intended to produce a departmentwide military deployment management system, is aligned with a departmentwide business enterprise architecture or an integrated blueprint for business systems modernization. As a result, the Army does not know that investment in TC-AIMS II as planned is warranted, represents a prudent use of limited DOD resources, and will be interoperable with and not be duplicative of related systems. Even if TC-AIMS II happens to be the right system, both economically and architecturally, the Army has not fully implemented risk management and has not fully employed performance-based contracting practices. However, it has largely managed system requirements, commercial components, and contractor tracking and oversight in accordance with DOD policies and related guidance. If the program can be economically justified and architecturally defined, then fully addressing these key acquisition management areas will be essential to the program’s success. DOD’s acquisition management policy and related guidance recognize the importance of developing reliable economic analyses to support investment decision making. Accordingly, they describe how these analyses should be developed and used. For TC-AIMS II, the Army has not adhered to important aspects of departmental policy and related guidance, such as updating analyses to reflect material changes in program scope and justifying investment in large, multiyear programs on an incremental basis. As a result, the Army currently lacks an adequate basis to justify its planned investment in TC-AIMS II. DOD policy and guidance state that projects should be economically justified on the basis of reliable analyses of expected life-cycle costs and benefits and that cost and benefit estimates should be based on realistic plans for the program’s scope. Federal guidance also advocates economically justifying proposed investments on the basis of a benefit-to- cost ratio that is greater than 1, basing that ratio on reliable analyses of expected, quantifiable life-cycle benefits, costs, and risks. The guidance also promotes the calculation and consideration of qualitative benefits in preparing an economic analysis. Further, the guidance calls for updating the analysis when significant program changes occur. These practices help to continually ensure a positive return on investment—even as program changes occur. The program office developed economic analyses to justify investment in the system in August 2002 and again in December 2003. The program office was supported in its efforts by the Office of the Deputy Assistant Secretary of the Army’s Office of Cost and Economics and the Office of the Secretary of Defense’s Office of Program Analysis and Evaluation, whose missions include verifying and validating the reliability of cost and benefit estimates found in Army and other services’ economic analyses. However, this most recent economic analysis is not based on reliable estimates of life-cycle costs and benefits. Specifically, the analysis was based on the assumption that TC-AIMS II would be used by all four military services, and, thus, it included life-cycle cost estimates showing all of the services’ data—as well as the benefits that would accrue through its use by all four. However, this assumption is not valid. Specifically, the Air Force’s Assistant Deputy Chief of Staff, Installations and Logistics, informed the program office in a September 2002 correspondence that it would not use the system, stating that the Air Force would rely instead on its existing system capabilities. Similarly, in March 2004, the Marine Corps’ Deputy Commandant for Installations and Logistics informed the program office that the Corps would not use the system because of concerns about whether the system would, among other things, deliver expected capabilities. Program officials told us that the latest economic analysis includes the costs and benefits for all services because, despite communications to the contrary from the Air Force and Marine Corps, the program office still considers TC-AIMS II to be a multiservice program and has not yet been directed to assume and plan otherwise. In addition to the unreliable cost and benefit estimates, the latest analysis does not show a benefit-to-cost ratio that is greater than 1. Specifically, the analysis identified present value, total expected benefits of $928 million versus present value, and total estimated costs of $929 million, for a benefit-to-cost ratio of slightly less than 1. This means that for each dollar spent on the system, slightly less than one dollar of benefit is returned. In addition, the analysis cited various qualitative benefits (e.g., increasing commander visibility of assets and enhancing worker productivity) that would be derived from the system. However, the analysis did not provide any underlying support or analysis for these qualitative benefits. Program officials said that they recognize that the system’s expected return on investment is marginal, but they nevertheless have been directed by the milestone decision authority—the Assistant Secretary for Defense, Networks and Information Integration—to deliver the system’s capabilities as planned. According to the officials in the Assistant Secretary’s office, the decision to continue investing in TC-AIMS II was based on quantitative and qualitative benefits, which they stated together exceeded costs, thus making the investment justified. However, these officials did not provide supporting justification for their qualitative benefit claims. In addition, their position does not recognize the intended change in the program’s scope that is not reflected in the latest analysis. Thus, the Army lacks the basis for knowing whether its planned investment in TC-AIMS II is justified. DOD policy and guidance, as well as other related federal and best practice guidance, state that large projects, such as TC-AIMS II, should be divided into a series of smaller, incremental subprojects or releases so that investment decisions can be made on each increment. By doing this, the tremendous risk associated with investing large sums of money over many years in anticipation of delivering system capabilities and associated expected business value far into the future can be spread across project increments that are smaller, of shorter duration, and capable of being more reliably justified and more effectively managed against cost, benefit, and risk expectations. The Army has neither analyzed TC-AIMS II costs and benefits, nor made associated investment decisions, on an incremental basis. Specifically, while the program office divided the system’s acquisition into five smaller increments (blocks), the economic analyses addressed life-cycle costs and benefits for the entire system. The analyses did not identify separately the costs and benefits for each of the five increments, and, thus, this information was not considered as program officials made decisions about TC-AIMS II at key milestones. Program officials stated that they have not analyzed costs and benefits on an incremental basis because once the program had been approved and initiated, they did not see the need thereafter to justify each increment. By not determining the costs and benefits of each of the system’s increments, the Army runs the risk of discovering too late (i.e., after it has invested hundreds of millions of dollars) that TC-AIMS II is not cost-beneficial. DOD’s acquisition policies and guidance, as well as federal and best practice guidance, recognize the importance of investing in IT business systems within the context of an enterprise architecture. Our research and experience in reviewing federal agencies show that not doing so often results in systems that are duplicative, are not well integrated, are unnecessarily costly to interface and maintain, and do not optimally support mission outcomes. TC-AIMS II has not been defined and developed in the context of a DOD enterprise architecture. Instead, the Army has pursued the system on the basis of an Army logistics-focused architecture. This means that TC-AIMS II as a DOD-wide program is based on a service-specific architecture and not a DOD-wide architecture, thus increasing the risk that TC-AIMS II, as defined, will not properly fit within the context of future DOD enterprisewide business operations and IT environments. A well-defined enterprise architecture provides a clear and comprehensive picture of an entity, whether it is an organization (e.g., a federal department) or a functional or mission area that cuts across more than one organization (e.g., personnel management). This picture consists of snapshots of both the enterprise’s current or “As Is” environment and its target or “To Be” environment, as well as a capital investment road map for transitioning from the current to the target environment. These snapshots consist of integrated “views,” which are one or more architecture products that describe, for example, the enterprise’s business processes and rules; information needs and flows among functions, supporting systems, services, and applications; and data and technical standards and structures. DOD has long operated without a well-defined enterprise architecture for its business environment. In 2001, we first reported that DOD did not have such an architecture, and we recommended that it develop one to guide and constrain IT business systems, such as TC-AIMS II. Over the next 4 years, we have reported that DOD’s architecture development efforts were not resulting in the kind of business enterprise architecture that could effectively guide and constrain business system investments, largely because the department did not have in place the architecture management structures and processes described in federal guidance. In particular, we most recently reported in July 2005 that despite spending about $318 million producing eight versions of its architecture, DOD’s latest version still did not have, for example, a clearly defined purpose that could be linked to the department’s goals and objectives, and a description of the “As Is” environment and a transition plan. Further, we reported that the description of the “To Be” environment was still missing important content (depth and scope of operational and technical requirements) relative to, for example, the actual systems to be developed or acquired to support future business operations and the physical infrastructure (e.g., hardware and software) that would be needed to support the business systems. Over the last several years, we have also reported that DOD’s efforts for determining whether ongoing investments were aligned to its evolving architecture were not documented and independently verifiable. On September 28, 2005, DOD’s Under Secretary of Defense for Acquisition, Technology, and Logistics and Under Secretary for Defense (Comptroller) issued the next version of its business enterprise architecture, which we are required to review, along with other things, such as the department’s efforts to review certain investments’ alignment with the architecture, pursuant to the Fiscal Year 2005 National Defense Authorization Act. According to program officials, the system has not been assessed against DOD’s business enterprise architecture because one has yet to be completed. Instead, as described in more detail below, program officials told us they have aligned TC-AIMS II with the Army’s enterprise architecture for logistics. However, without a well-defined architecture that sets the DOD-wide context within which a DOD-wide system is to fit, and a firm understanding of the extent to which TC-AIMS II, as defined, fits within that context, the program office risks acquiring a system that does not meet DOD business needs and introduces redundancies and incompatibilities that require costly and time-consuming rework to fix. In the absence of a DOD-wide architecture, program officials told us they have developed the system in the context of the Army’s enterprise architecture for logistics—the Single Army Logistical Enterprise, which is managed by the Army’s Chief Information Officer. For example, in 2004, program officials requested the Army’s Chief Information Officer to determine whether TC-AIMS II was aligned with the logistic architecture. The Office of the Chief Information Officer conducted an analysis during the fall of 2004 that focused on, among other things, tracing TC-AIMS II business processes and requirements to those specified in the logistics architecture. The Chief Information Officer reported in March 2005 that TC-AIMS II, as defined in Blocks 1 to 3, was aligned with the logistics architecture. We did not review, and thus do not question, this analysis because TC-AIMS II is intended to be a joint system, and thus aligning it solely with the Army’s logistics architecture will not provide the program with sufficient basis for managing the risk of not being able to adequately ensure that a DOD-wide system aligns with a DOD-wide architecture. Effective risk management is vital to the success of any system acquisition. Accordingly, DOD acquisition management policies and guidance, as well as other relevant best practice guidance, advocate proactively identifying facts and circumstances that can increase the probability of an acquisition’s failing to meet cost, schedule, and performance commitments and then taking steps to reduce the probability of their occurrence and impact. Effective risk management includes developing written policies and procedures; assigning roles and responsibilities for managing risk; developing a risk management plan that provides for (1) identifying and prioritizing risks, (2) developing and implementing the appropriate risk mitigation strategies, and (3) tracking and reporting on progress in implementing the strategies; and executing the plan. To its credit, the program office has satisfied the first three of these four key practices. First, it has adopted and uses the risk policies and procedures specified in DOD’s major IT systems acquisition guidance. Second, the program office has assigned risk management roles and responsibilities to the Government Risk Manager, who chairs the program’s Risk Management Board. The board, which consists of key program managers and stakeholder representatives, is responsible for ensuring that the risk management program is implemented according to the risk management plan. Among other things, this includes assigning individuals who are to develop and implement risk mitigation plans for all newly identified risks. Third, the program office has a risk management plan, dated October 2004, and this plan provides for (1) identifying and prioritizing risks, (2) developing and implementing the appropriate mitigation strategies, and (3) tracking and reporting on progress in implementing the strategies. For example, the plan defines procedures for program staff and the board to follow in prioritizing risks, including assessing its “risk exposure” on the basis of a probability of occurrence and potential impact. On the basis of the assessment, “risk exposure” is designated as either high, medium, or low. As of May 2005, the TC-AIMS II program office had identified 22 active risks in its risk inventory with the following exposures—1 high, 18 medium, and 3 low. Fourth, since the program office has not fully implemented its risk management plan, it has not satisfied the fourth risk management practice—execution. For example, of the 22 active risks, 6 did not have mitigation strategies. Of these 6, 1 is high exposure and 5 are medium exposure risks. As another example, the program office’s inventory does not include all key risks, such as the lack of reliable economic justification and alignment to a DOD-wide architecture as previously described in this report. Program office officials stated that these risk management practices have not been fully implemented because of other competing priorities. Further, officials attributed the lack of risk mitigation strategies to staff turnover, which resulted in relatively few staff receiving training on how to use the automated tool employed by the office to document and track risks. In particular, program officials stated that only about 20 percent of the program staff (40 individuals) has attended the program’s training on how to use the automated tool. When we concluded our audit work, program officials acknowledged the missing risk mitigation strategies and stated that they are in the process of addressing them. Until the program fully implements effective risk management practices, there is increased probability that program risks will become actual problems leading to program cost, schedule, and performance shortfalls. DOD policy and guidance, as well as other relevant best practices, recognize the importance of effectively developing and managing system requirements. According to the policy and guidance, well-defined requirements are important because they establish agreement among the various stakeholders on what the system is to do, how well it is to do it, and how it is to interact with other systems. Having this agreement is key to acquiring a system that meets end-user needs and performs as intended. Effective requirements development and management involve, among other things, (1) establishing a written policy on establishing and conducting requirements development and management activities; (2) eliciting desired and required system capabilities from users and translating them into system requirements; (3) documenting the requirements, including having users validate that they have been accurately captured; (4) ensuring that requirements changes are controlled as the system is developed and implemented; and (5) maintaining bidirectional traceability, meaning that a given requirement can be traced backward to its source and forward to both the component or increment that will satisfy the requirement and the test that will verify that it is satisfied. The program office is managing TC-AIMS II requirements in accordance with these five practices. First, the program office adopted DOD Instruction 5000.2 and related guidance, as the program’s policy for requirements development and management activities. The directive and guidance are also incorporated in program office plans that detail how the office develops and manages requirements during their life cycle—that is, from when requirements are first documented to when system responsibility is transferred to operations and maintenance support. According to program officials responsible for managing requirements, they use the guidance and plans to perform their work. Second, program officials elicited required system capabilities from end users through the program’s Joint Requirements Board, which consists of key system stakeholders and representatives and is chaired by the Joint Forces Command. The board is responsible for managing requirements development activities. Among other things, the board’s charter states that its duties include defining, reviewing, validating, prioritizing, and approving the system’s requirements. Board meeting minutes show evidence of it performing these duties, including, for example, reviewing the Block 3 baseline requirements document on September 8, 2004. Third, program officials showed us how they document the requirements, using an automated tool. Further, the Joint Requirements Board approved the requirements baseline for each of the three blocks. Fourth, to manage system requirement changes, the program office established a change control structure (configuration control board) and supporting processes, which are documented in its 2004 configuration management plan. Consistent with the plan, the board has six voting members and includes representatives from the four services. As shown in board meeting minutes and supporting documentation, the board meets monthly to consider and decide on change requests. Fifth, the program maintains bidirectional traceability for the requirements using an automated tool. Specifically, program officials demonstrated to us how each requirement can be traced back and to its source (e.g., TC-AIMS II’s operational requirements document) and forward to the block in which the requirement is to be satisfied. The quality of the processes and practices followed in acquiring software- intensive systems greatly influences the quality of the systems produced. Moreover, acquiring custom-developed system solutions is sufficiently different from acquiring commercial component-based systems that adherence to certain practices unique to the latter is key to their success. As we have previously reported, DOD’s revised systems acquisition policies and guidance require that commercial components be used to the maximum extent that is feasible. Collectively, they also provide for the following commercial component acquisition management best practices: modifying components only after a thorough analysis of life-cycle costs ensuring that integration contractors are explicitly evaluated on their ability to implement commercial components, and ensuring that project plans provide for the time and resources needed to integrate commercial components with existing (legacy) systems. The Army has largely complied with these practices for TC-AIMS II. First, program officials stated that it is their policy to discourage component modifications, which they have communicated in meetings and other program management directives. In addition, the program has established a joint configuration control board and supporting processes to limit and control changes to the commercial products being used. As stated in the program’s configuration management plan (dated June 30, 2005), the board, which consists of key program managers and stakeholder representatives, only approves proposed changes to those products if the change is justified by a thorough analysis of costs, benefits, and risks. The board meets monthly to discuss and decide upon such changes, and, according to program officials, this approach has resulted in no major modifications to date. Second, the program office evaluation of competing contractors included consideration of their ability to implement commercial components. For example, the TC-AIMS II request for proposal specified that the contractor’s ability to implement commercial products was to be a significant factor in evaluating overall performance. Third, the program’s work breakdown schedule (dated Nov. 9, 2004) specifies the time and resources needed for integrating commercial components with existing systems. By following these best practices associated with acquiring commercial- based systems, the program is increasing the likelihood that TC-AIMS II will be successfully implemented and effectively used. DOD policy and guidance and other relevant system acquisition management best practices recognize the importance of effectively managing contractor activities. To this end, policy and guidance call for, among other things, performing contractor tracking and oversight activities and using performance-based contracts to the maximum extent practicable. To its credit, the Army has performed key activities related to tracking and oversight. However, it has not fully implemented one key aspect of performance-based contracting. By not doing so, the program office has limited its ability to effectively influence contractor performance. Department policy and guidance and related best practices guidance identify effective contractor tracking and oversight as a key contract management activity and define a number of associated practices to follow, including establishing a written policy on contract tracking and oversight; designating responsibility for contract tracking and oversight activities and having contracting specialists perform these activities; and using approved contractor planning documents as part of periodic reviews and interchanges with the contractor, during which actual cost and schedule performance is compared with the contractor’s planned budgets and schedules to identify variances and issues. The program office’s tracking and oversight of the systems integration contractor from Block 3 largely satisfies these three practices. First, the program office has adopted the department’s acquisition management polices and guidance as its written program-specific policy on contractor tracking and oversight. Second, the program office assigned responsibility for contract tracking and oversight to its Budget Management Division, which has three staff devoted to these duties. The devoted staff are contracting specialists on detail from the General Services Administration’s Federal Systems Integration and Management Center. Third, the program office uses approved contractor planning documents as the basis for overseeing the contractor. These planning documents are specified in the contract and include a program management plan, a master project schedule, and program status reports. Collectively, these documents delineate when and how products and services are to be delivered. According to program officials, these planning documents are used during monthly status meetings with the contractor to track progress. In particular, they are used to compare actual cost and schedule performance to date with planned budgets and schedules to identify potential variances. Program officials attribute the state of their contract tracking and oversight capability to two factors. First, the program has been in existence for about 10 years, thus allowing time for the office to develop disciplined processes in this area. Second, in hiring the Federal Systems Integration and Management Center contract specialists, the program has subject matter experts who know the importance of, and who actively advocate, rigorously following the processes. By employing key contract tracking and oversight practices, the program office is increasing the likelihood that the contractor will perform as expected. Department policy, federal acquisition regulations, and other relevant best practices also advocate the use of performance-based contracting when acquiring services. Effective performance-based contracting includes defining clearly the work to be performed, specifying performance standards (quality and timeliness) that are tied establishing positive and negative incentives that are tied to the contractor’s performance, and having a quality assurance plan that describes how the contractor’s performance in meeting requirements will be measured against standards. The program office’s approach to managing the contract is largely consistent with these four practices. First, the contract for Block 3 (dated Apr. 30, 2004) includes a statement of objectives that define the work to be performed. For example, the contract identifies specific project tasks (e.g., establishing a Block 3 development help desk for system users) and associated project management products (e.g., program management plan, program schedule, and program status reports) that the contractor is to perform and deliver, respectively, in accordance with the contract. Second, the contract includes performance standards (measures) that are contractually required. For example, it states that the help desk is to be staffed 24 hours per day, 7 days a week, with help operators capable of resolving 80 percent of user calls within 1 hour, while limiting caller on- hold time to 3 minutes or less. Third, the contract provides positive incentives that are tied to contractor performance. That is, in addition to the base contract (cost plus fixed fee) amount, the contract includes an additional 8 percent fee that can be awarded to the contractor for satisfying specified performance criteria. Fourth, the latest quality assurance plan, dated August 2004, describes how the contractor’s performance will be measured and what remedial steps are to be taken in case the contractor does not meet contractual requirements. However, the contract does not specify effective disincentives or penalties for failing to meet the performance criteria. More precisely, the contract does not provide for forfeiting all or part of the fee if the contractor fails to perform as expected. Instead, program officials stated that their practice of withholding the additional award fee when expectations are not met, constitutes a disincentive. For example, they said that they withheld the award fee for the contractor’s first performance period (which totaled about $900,000) because the contractor failed to develop two deliverables on time (the program management plan and the performance metrics). We do not agree that this practice constitutes an effective negative incentive because it does not penalize performance that fails to meet expectations. Moreover, program officials told us that in the above example, withholding the award fee was not a forfeiture, because it was actually deferred to the second performance period when the contractor had another opportunity to earn it. According to program officials, during the second performance period, the contractor earned almost $780,000 of the withheld fee from the first period by completing, among other things, first period tasks. In the absence of specific negative incentives that are tied to intended performance, the program is at increased risk of contractor nonperformance. This practice puts the program at risk of taking more time and spending more money than necessary to acquire contract deliverables and services. It is unclear whether the Army’s planned investment in TC-AIMS II is warranted. Of critical concern is the absence of reliable analysis showing that further investment will produce future mission benefits commensurate with estimated costs. This is due in large part to the evident change in the scope of the program, which was initiated on the basis of the four military services using TC-AIMS II but has not been revised to reflect that two services have stated their intentions not to use the system. Compounding this uncertainty is the inherent risk of defining and developing this multiservice system outside the context of a well-defined DOD-wide enterprise architecture. Without this information, the Army cannot currently determine whether TC-AIMS II, as defined and as being developed, is the right solution to meet its strategic business and technological needs. If these uncertainties are addressed and the Army demonstrates that TC-AIMS II plans are the right course of action, the department largely has the capabilities that are essential to successful system acquisition and deployment, although there is room for strengthening its practices related to risk management and performance-based contracting. It is vital that Army and DOD authorities responsible and accountable for ensuring prudent use of limited resources first reassess whether allowing TC-AIMS II to continue as planned is warranted, and that the decision on how to proceed be based on reliable data about program costs, benefits, risk, and status. We recommend that the Secretary of Defense direct the Secretary of the Army to determine if continued investment in TC-AIMS II as planned represents a prudent use of the department’s limited resources. To accomplish this, the Secretary of the Army should take the following three actions: collaborate with the Office of the Assistant Secretary of Defense for Networks and Information Integration/Chief Information Officer, the Office of Program Analysis and Evaluation, and the Army Cost Analysis Division to prepare a reliable economic analysis; ensure that development of this economic analysis (1) complies with cost-estimating best practices and relevant Office of Management and Budget cost benefit guidance, (2) incorporates available data on whether deployed TC-AIMS II capabilities are actually producing benefits, and (3) addresses that the Air Force and Marines are not planning to use the system; and collaborate with the Undersecretary of Defense for Acquisition, Technology, and Logistics and the Under Secretary of Defense (Comptroller) to ensure that TC-AIMS II is adequately aligned with the evolving DOD business enterprise architecture. In addition, we recommend that the Secretary of Defense direct the Secretary of the Army to present the results of these analyses to the Deputy Secretary of Defense, or his designee, and seek a departmental decision on how best to proceed with the program. Until this is done, we recommend that the Secretary of Defense direct the Secretary of the Army to limit future investment in already deployed applications to essential operation and maintenance activities and only developmental activities deemed essential to national security needs. If—on the basis of reliable data—a decision is made to further develop TC-AIMS II, we recommend that the Secretary of Defense direct the Secretary of the Army to ensure that the program implements effective program management activities related to risk management and performance-based contracting. In its written comments on our draft report, signed by the Deputy to the Assistant Secretary of Defense for Networks and Information Integration (Command, Control, Communications, Intelligence, Surveillance, and Reconnaissance and Information Technology Acquisition) and reprinted in appendix III, DOD either agreed or partially agreed with our recommendations, and it stated that some of our findings have merit and that it has significant efforts ongoing to comply with applicable guidance. For example, while DOD agreed that TC-AIMS II was originally defined and implemented without a complete and formal enterprise architecture and that the program had not identified the lack of a mature architecture as a program risk, it stated that the program is taking steps to ensure alignment to relevant enterprise architectures and is thereby mitigating this risk. In addition, although the department agreed that the latest economic analysis shows a negative return on investment, it said that other intangible benefits not included in the analysis, such as improved planning and equipment tracking, bolster the value of TC-AIMS II. Further, it agreed that the system was originally envisioned as a single DOD-wide solution and that the Air Force and Marine Corps are no longer intending to use the system, but added that it was now evaluating whether these two services’ existing (legacy) systems should be enabled to share data with TC-AIMS II, rather than be replaced in their entirety by a joint system as was originally envisioned. According to DOD, the results of this evaluation are to be considered at the next milestone review, which is planned for early to mid- 2006. These comments are largely consistent with our report. Specifically, the report makes clear that DOD is managing TC-AIMS II in accordance with some key practices and has efforts under way, such as those previously cited, intended to bring the program into compliance with other key practices. However, the report also points out noncompliance in areas intended to provide sufficient information to determine how the TC-AIMS II program should proceed. Accordingly, it contains recommendations to ensure that the economic analysis is updated to (among other things) reflect current operating assumptions, and that the program is adequately aligned with the department’s evolving enterprise architectures. It is precisely because the program does not have this information that, despite having invested $751 million over 11 years, the department does not know whether the program as defined is the right solution to meet DOD-wide asset deployment needs. DOD offered five detailed comments on our four recommendations. The department’s comments and our responses are as follow: 1. DOD stated that the TC-AIMS II program has used best practices and followed all economic analysis requirements specified in relevant DOD and OMB guidance, and has successfully passed milestone reviews at which it was required to show costs and benefits for both the entire system and its increments. In addition, the department stated that even though its economic analysis showed a slight negative return on investment, the system’s intangible benefits bolster the program’s overall value. According to DOD, the program identified these intangible benefits to obtain department milestone approval, and this documentation was provided to us on October 31, 2005. This notwithstanding, the department stated the program will (1) conduct, in coordination with the Office of the Secretary of Defense’s Program Analysis and Evaluation, an analysis to determine the relationship of TC-AIMS II and other programs, such as the Global Combat Support System, that may provide similar functionality to ensure optimal return on investment and (2) brief the Networks and Information Integration Overarching Integrated Product Team in preparation for the next major milestone decision review. It also stated that these results will be used to justify further investment. We do not agree that the program has used all best practices and followed all requirements specified in relevant guidance. In particular, the Army did not economically justify its TC-AIMS II investment on the basis of reliable estimates of costs and benefits and did not invest in TC-AIMS II within the context of a well-defined architecture, both of which are recognized best practices as well as DOD and OMB requirements. In addition, while it is appropriate to consider intangible benefits in economically justifying an investment, these benefits are not in the economic analysis, and the separate document provided on October 31, 2005, did not contain sufficient rationale and justification for claimed qualitative benefits. Rather, the information provided was a restatement of the quantitative benefits in the economic analysis. Further, by stating that it is to conduct an analysis of the relationships of TC-AIMS II to other programs that provide similar functionality and use the results of this analysis to justify further system development, DOD is recognizing that it does not have the information it needs to make informed investment decisions. If done thoroughly, the analysis, which the department committed to conduct, should help to inform future program decisions. 2. Consistent with our report, DOD stated that TC-AIMS II was established as a joint military service program. However, in explaining why the Air Force and Marine Corps are not intending to use TC-AIMS II, it added that changes to operations throughout the world since then have necessitated changes to the services’ deployment processes. In light of this, the department stated it will evaluate other options for data exchange among Air Force and Marine Corps existing asset deployment systems and TC-AIMS II, instead of having all the services use a single, joint system. Further, the department said it will have the program, in coordination with Program Analysis and Evaluation, develop an economic analysis for TC-AIMS II that is based on the system being used solely by the Army and the Navy. DOD also commented that our report assumed a positive cost-benefit impact from including the Air Force and Marine Corps in the program. We support the department’s decision to develop an economic analysis that reflects this material change in the program’s scope, which is what we recommended. However, we would note that DOD’s comments do not explain how unspecified changes to operations throughout the world caused unspecified changes to each of the services’ deployment processes, which in turn required DOD to have to interface multiple service-unique systems, rather than adopting a joint system. We would expect such an explanation to be an integral part of updating the economic analysis, as will the costs of pursuing service-unique courses of action. With regard to the comment that we assumed the inclusion of the Air Force and Marine Corps would have a positive cost-benefit impact, we do not agree. We made no assumption about the impact of the program’s scope on the cost-to-benefit or return on investment ratios. Rather, our point is that DOD policy and related guidance state that programs should be economically justified by, among other things, cost and benefit estimates that are based on realistic plans for the program’s scope. Such realistic plans did not exist because DOD’s justification was based on the four services using the system, when in reality the Air Force and Marine Corps do not intend to do so. Relevant guidance calls for updating economic analyses when significant program changes such as these occur. Accordingly, we recommended that the department follow this guidance to ensure it had adequate information for informed decision making. 3. DOD commented that it has worked over the past several years to develop an enterprise architecture and is on a path to fully defining and effectively managing its enterprise architecture. It also said that it is working to align TC-AIMS II with relevant evolving architectures. In particular, DOD stated that Blocks 1 through 4 are aligned with the Joint Deployment Operational Architecture and the program office continues to work to align the program with the Joint Deployment and Distribution Architecture, which it said is currently aligning to the evolving Business Enterprise Architecture. We do not take issue with the comments that work has been and is under way in this area. Our point is that the Business Enterprise Architecture is still a work in progress and what has been available to guide and constrain business system investments, such as TC-AIMS II, has not been sufficient. Moreover, program officials told us that TC-AIMS II has not been assessed against the Business Enterprise Architecture because this architecture has not yet been completed. Without a well-defined DOD-wide architecture that sets the context within which such a joint military service system is to fit, and a firm understanding of the extent to which TC-AIMS II, as defined, fits within that context, the program office risks, among other things, acquiring a system that does not meet DOD’s strategic business needs. 4. DOD stated that while comprehensive mitigation and contingency strategies were not developed for all of the 22 risks cited in our report, the program office has subsequently developed strategies for them. DOD also commented that subsequent risk management data, along with evidence that risks are continuously reviewed and updated, was provided to us on October 28 and November 3, 2005. We acknowledge that program officials orally stated that this had occurred. However, we have yet to receive any documentation to substantiate these statements. 5. DOD stated that while the current TC-AIMS II contract does not include direct disincentives for contractor failure to meet performance criteria, its semiannual award fee mechanism serves a similar purpose. Nonetheless, DOD commented that it will address its award fee implementation on future TC-AIMS II contract reviews. We do not agree that this semiannual award fee approach is an effective disincentive because award fees are only being withheld when contractor performance does not exceed certain targets; conversely, direct disincentives penalize poor performance that falls short of meeting contractually required levels of performance. Without direct disincentives, the program is at increased risk of contractor nonperformance, and thus taking more time and spending more money than necessary to acquire contract deliverables and services. We support the department’s stated commitment to address this issue. We are sending copies of this report to interested congressional committees. We are also sending copies to the Director, Office of Management and Budget; the Secretary of Defense; the Deputy Secretary of Defense; the Undersecretary of Defense for Acquisition, Technology, and Logistics; the Assistant Secretary of Defense (Networks and Information Integration)/Chief Information Officer; the Secretaries of the Air Force, Army, and Navy; the Commandant of the Marine Corps; the Department of the Army’s Chief Information Officer; and the Program Executive Officer, Enterprise Information Systems, Transportation Information Systems, Department of the Army. We are also sending copies to other interested parties. This report will also be available at no charge on our Web site at http://www.gao.gov. Should you have any questions about matters discussed in this report, please contact me at (202) 512-3439 or hiter@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. A GAO contact and staff who made major contributions to this report are listed in appendix IV. Our objective was to determine whether the Transportation Coordinators’ Automated Information for Movements System II (TC-AIMS II) was being managed according to important aspects of the Department of Defense’s (DOD) acquisition policies and guidance, as well as other relevant acquisition management best practices. To accomplish this, we focused on the program’s (1) economic justification; (2) architectural alignment; (3) risk management; (4) requirements development and management; (5) commercial components; and (6) contract management, including contractor oversight and performance-based contracting. To determine whether the Department of the Army had economically justified its investment in TC-AIMS II, we reviewed the latest economic analysis to determine the basis for the cost and benefit estimates and net present value calculations. This included evaluating the analysis against DOD policies and guidance as well as other relevant best practices guidance. It also included interviewing responsible program officials, including the program and deputy program managers, regarding their respective roles, responsibilities, and actual efforts in developing and/or reviewing the economic analysis. In addition, we also interviewed the program and deputy program manager and the Office of the Assistant Secretary of Defense for Network Information and Integration about the purpose and use of the analysis for managing the Army’s investment in the TC-AIMS II program, including the extent to which measures and metrics showed that benefits projected in the economic analysis were actually being realized. To determine whether the Army had aligned TC-AIMS II to the DOD business enterprise architecture, we relied on our prior reports addressing DOD and Army architecture development and implementation efforts and documents on the Army’s efforts to align TC-AIMS II with its logistics architecture (the Single Army Logistics Enterprise), as well as documents on the Army’s efforts to align the logistics architecture with DOD’s business enterprise architecture. We also interviewed officials from the TC-AIMS II joint program management office, the Office of the Assistant Secretary of Defense for Networks and Information Integration/Chief Information Officer, and the Office of the Army’s Chief Information Officer about DOD and Army architecture efforts and TC-AIMS II’s alignment to them. To determine whether the Army was effectively managing key system acquisition activities, namely risk management, requirements development and management, commercial components, and contract management, we did the following: To assess risk management, we reviewed the program’s risk policies, processes, management plan, and related documentation. We also observed how the program used an automated tool (database) to document and track risks, including developing risk mitigation strategies. In addition, we analyzed a version of the database to determine, among other things, the state of risk mitigation strategies. We then compared these activities with DOD policy, guidance, and related best practices to determine whether variances existed, and if so, why. Moreover, we interviewed Army officials involved with risk management, including the program manager and risk manager, to discuss their roles and responsibilities for managing risk associated with acquiring and implementing TC-AIMS II. To assess requirements development and management capabilities, we reviewed program documentation, such as the official list of requirements, and system specifications and evaluated them against these relevant best practices for several characteristics, including traceability and prioritization. We observed program office staff trace requirements to both higher level documents and lower level specifications using an automated tool. We interviewed Army officials involved in the requirements development process, including the program and deputy program managers, to discuss their roles and responsibilities for developing and managing requirements. To evaluate the management of commercial components, we reviewed a variety of documents, such as program and contractor studies of commercial components’ products, as well as the program office’s project plans, and contract task order documentation. Additionally, we conducted interviews with the program officials, including the program and deputy program managers, responsible for selecting the commercial components for planned TC-AIMS II blocks, to determine whether the commercial component products had been acquired in accordance with DOD policy, guidance, and related best practices. To assess contract management, we focused on the program office’s contract tracking and oversight activities as well as its use of performance-based contracting. For tracking and oversight, we reviewed program management plans, contractor monthly status reports, and related documentation of the program office’s policies, processes, and activities for overseeing its system integration contractor for Block 3. We then compared these with DOD policies, guidance, and related best practices. We also interviewed program offices responsible for managing these activities. For performance-based contracting, we reviewed the Block 3 contract (issued in April 2004) and related documentation, such as contract modifications, quality assurance plan, and reports on determining contractor award fee, and then compared them with department policy and guidance, federal acquisition regulations, and related best practices. We also interviewed program management officials, including the contract specialist from the General Service Administration’s Federal Systems Integration and Management Center, who are responsible for performing these activities. We did not independently validate information on the program’s cost and budget as well as the number of problem reports. We conducted our work at DOD headquarters in the Washington, D.C., metropolitan area and at the Army’s Joint Program Management Office in Northern Virginia. We performed our work from October 2004 through October 2005, in accordance with generally accepted government auditing standards. One indicator of system quality is defect density, which can be measured in a number of ways, including the trend in the number of system defects being reported and resolved. The program office prepares problem reports to document and prioritize system defects, and it established a database to manage and track these reports. Further, the program office assigns priority (i.e., criticality) levels to problem reports, with category 1 being the most critical, and category 5 being the least critical. The program office’s definition of each category is as follows: Category 1: The system or function does not work and the problem must be addressed immediately, before any further work can be completed. Category 2: The system or function may work, but the problem is severe and must be addressed before the end of the current block. Category 3: Essentially, this is a category 2 problem with a known solution that would provide a temporary fix. Category 4: A routine problem that does not pose a significant threat and thus no immediate action is required. Category 5: A minor problem that does not pose a significant threat to system functionality. As depicted in figure 2, the data show that the number of open (yet to be resolved) problem reports over the first two development blocks has decreased. In particular, the total number of open problem reports has decreased by 321 (or 75 percent), from 426 in fiscal year 1999 to 105 at the end of fiscal year 2004. During the same period, categories 1 and 2 problem reports decreased by 216 (or about 100 percent), from 217 to 1. This downward trend reversed for the less critical defects in fiscal year 2005. Specifically, as of September 30, 2005, there were 534 total defects, 25 of which were categories 1, 2, and 3. Program officials attributed this reversal of categories 4 and 5 defects to current Block 3 testing activities and stated that they anticipated the increase and are actively managing it. In addition, they stated that while overall defects have recently increased, the number of critical defects has increased only slightly, and program staff are working to resolve these increases, with the goal of having permanent solutions implemented before Block 3 is deployed, which is scheduled to begin in October 2006. In addition to the contact named above, Gary Mountjoy, Assistant Director; Justin Booth; Hal Brumm Jr.; Joanne Fiorino; Anh Le; Kush K. Malhotra; Teresa M. Neven; Dr. Rona Stillman; Amos Tevelow; and Bill Wadsworth made key contributions to this report. | Because of the importance of the Department of Defense's (DOD) adherence to disciplined information technology (IT) acquisition processes in successfully modernizing its business systems, GAO was asked to determine whether the Transportation Coordinators' Automated Information for Movements System II (TC-AIMS II) program is being managed according to important aspects of DOD's acquisition policies and guidance, as well as other relevant acquisition management best practices. TC-AIMS II was initiated in 1995 as a joint services system to help manage force and equipment movements within the United States and abroad. The U.S. Department of the Army has the lead responsibility for managing the system's acquisition and estimates its life-cycle cost to be $1.7 billion over 25 years. The Army has managed the TC-AIMS II program in accordance with some, but not all, key aspects of DOD's system acquisition management policies and related guidance. These policies and guidance are intended to reasonably ensure that investment in a given IT system represents the right solution to fill a mission need--and, if it does, that acquisition and deployment of the system are handled in a manner that maximizes the chances of delivering defined system capabilities on time and within budget. The Army has not managed the program in accordance with those DOD policies and related guidance, including related federal and other best practice guidance, that are intended to reasonably ensure that a proposed system is the right solution to meet mission needs. Specifically, the Army has not economically justified its investment in TC-AIMS II on the basis of reliable estimates of costs and benefits. For example, the most recent economic justification included cost and benefit estimates predicated on all four military services using the system. However, two services (U.S. Department of the Air Force and U.S. Marine Corps) have stated that they do not intend to use it. The Army has not invested in TC-AIMS II within the context of a well-defined enterprise architecture, which is an institutional blueprint to control program investment decisions in a way that promotes interoperability and reduces redundancy among systems. The Army has instead focused on aligning TC-AIMS II with its logistics architecture; this means that even though TC-AIMS II is intended to be a DOD-wide program, it has been based on a service-specific architecture rather than a DOD-wide architecture. As a result, it may not properly fit within departmentwide plans. To its credit, the Army has largely managed the program in accordance with key policies and related guidance that are intended to reasonably ensure that the acquisition and deployment of a given system are handled in a manner that maximizes the chances of delivering defined capabilities on time and within budget. However, some aspects of this policy and guidance have not been followed. For example, the Army has not fully implemented risk management and has not adhered to a key feature of performance-based contracting. Reasons the Army cited for not following policies and guidance ranged from management inattention to lack of training. As a result, the Army, among other things, does not know whether the system is the right solution. Until this uncertainty and the previously discussed problems are addressed, it will remain unclear whether further planned investment in TC-AIMS II is warranted, and certain aspects of the program's management will be limited. |
Several key actions have been taken to implement the law since it was enacted over 14 months ago. Some of the actions taken to implement the act had a direct impact on the Service’s 2007 financial condition, while others facilitated the transition to a new financial, operating, and regulatory environment. Specific actions in the act that have affected the Service’s 2007 financial condition include: prefunding the Service’s significant unfunded retiree health obligations. While this change results in significant retiree health benefit cost increases for a decade, over the long term this action improves the fairness and balance of the cost burdens for current and future ratepayers. The Service’s 2007 payment of $5.4 billion was the first of 10 annual payments required under this change. expensing almost $3 billion in funds previously set aside in escrow (transferring them to the Treasury) and eliminating future escrow payments, including an estimated $3.3 billion payment that had been scheduled for 2007. transferring the estimated $27 billion funding obligation for selected military service benefits back to the Treasury. eliminating certain annual CSRS pension funding requirements, thereby saving the Service approximately $1.6 billion in 2007. The effects of these changes, however, must be put into context with other actions and events during this time to gain a comprehensive understanding of the Service’s financial and operating condition. For example, Mail volumes and revenues: Total revenues of nearly $75 billion dollars in 2007 represented an increase of 3 percent from 2006. This revenue increase, however, was largely attributable to the January 2006 and May 2007 rate increases—not mail volume increases. In particular, the Service experienced an overall decline in mail volume from 2006 of over 900 million pieces (a 0.4 percent decline), largely due to a decrease of 1.7 billion pieces of First-Class Mail and the smallest increase in Standard Mail volumes since 2001. Operating costs: Total operating expenses of over $80 billion in 2007 represented an increase of nearly 12 percent from 2006. This increase was largely due to a net increase of $6.8 billion in expenses that resulted from requirements of the act described earlier. The Service was also affected by increases in postal wage rates; rising fuel costs (its transportation costs grew by almost 8 percent); and the extension of mail service to an additional 1.8 million delivery points. Productivity and cost control: The Service was able to partially mitigate these cost pressures by improving productivity for an eighth consecutive year. The Service reported a 1.7 percent increase in productivity, which is equivalent to $1.2 billion in cost savings. The Service reduced over 36 million workhours, partly by downsizing its career workforce by over 11,000 employees. Labor agreements: The Service negotiated agreements with 3 of its 4 major unions on wages, many benefits, and conditions of employment in 2007. Debt: The Service’s outstanding debt increased $2.1 billion in 2007, doubling its 2006 debt balance to $4.2 billion. These increases were primarily used to finance year-end worker compensation and retiree health payments. Capital: The Service reported a slight increase in capital cash outlays of $2.6 billion in 2006 to $2.7 billion in 2007. These funds were used for such projects as new facilities, automation equipment, and carrier vehicles. Service performance: According to the Service, it reported record annual on-time performance for First-Class Mail measured by the EXFC system. The Service reported on-time performance for 2007 of 96 percent for its 1- day mail, 93 percent for its 2-day mail, and 90 percent for its 3-day mail. The net income reported for 2007 was a $5.1 billion loss. Removing the financial impact of the new law, the Postal Service’s net income would have been $1.6 billion (which was $100 million less than the $1.7 billion originally budgeted for the year). Because of the new law, however, the Service required an additional $500 million in cash to cover the differences between the net increase in retirement-related expenses of $3.8 billion and the expected $3.3 billion escrow payment that was avoided. Aside from its direct financial impact, the act required other actions to facilitate the transition to a new financial, operating, and regulatory environment. Table 1 summarizes key actions. As indicated in table 1, multiple stakeholders have taken actions to implement the requirements of the act. While each of these actions is important, I would like to highlight the efforts of the Postal Service and other stakeholders in modernizing service standards and of the PRC in transitioning to its new regulatory responsibilities. Service standards: The Service has made important progress in implementing the act’s requirements to establish modern service standards for market-dominant products. The Service’s approach to developing these standards incorporated a high level of collaboration with mailers, consultations with the PRC, and comprehensive review of its network capabilities. A workgroup involving nearly 200 representatives from the Service, mailing organizations and mailers, and other members of the mailing industry was particularly noteworthy for its efforts to identify issues and build consensus in this area. The result was the most sweeping update in delivery performance standards in many years. In particular, standards for Periodicals, Package Services, and Standard Mail that dated back many years were realigned with current postal operations. Regulation: A key tenet of the act was to provide the Postal Service with more flexibility to set prices and introduce new products. The act, however, balanced this flexibility by granting the PRC enhanced regulatory authority to regulate these activities, and to, among other things, regulate rates for market-dominant products and services; monitor financial and service performance; ensure financial transparency and data quality; and act on complaints against the Postal Service. The PRC has made good progress in its transformation, particularly in establishing a new ratemaking system. The PRC issued its first set of rate regulations almost 8 months ahead of the statutory deadline, and the Postal Service recognized this achievement and chose not to file another rate case under the old system. The PRC set up its new organizational structure, including the new PRC Inspector General, as well as revising key rules and regulations to reflect implementation of the act. The PRC has also taken an active role in consulting with the Service on its service standards and performance measures, solicited public comments, and held hearings and meetings to stimulate constructive dialogue among the parties. The financial, operational, human capital, and regulatory challenges facing the Service and other stakeholders as they take actions to continue implementing the act are exacerbated by the current uncertain economic environment. The Service noted in its 2008 budget that it expected a net increase in costs of $1 billion net for 2008 from changes in the law. In addition, a slowing economy has negatively affected the Service’s financial performance in the first quarter of 2008—its mail volumes and revenues were both below planned amounts. The Service was able to mitigate these challenges by managing cost growth while achieving record service delivery performance for the segment of First-Class Mail that is currently measured. To address concerns about its challenged financial outlook, the Service filed a rate increase February 11, 2008, averaging about 2.9 percent for the majority of its products. This increase is scheduled to take effect on May 12, 2008, and is the first of its type under the new law. According to Service officials, this increase is expected to contribute an additional $700 million in revenues for 2008, but will need to be supplemented by accelerated cost reduction strategies to achieve its year-end target of a $600 million net loss. The Service has updated its strategies for addressing challenges under the new law related to generating sufficient revenues, achieving efficiencies through automation, and improving service. In particular, to help address its revenue challenges, the Service has indicated that it plans to fully use its pricing authority under the rate- setting cap to implement smaller, more frequent, predictable rate increases, as well as work with its customers to develop new products and services. It also plans for $2 billion in cost reduction efforts in 2008. Some of the key areas for continued oversight include changes in mail volumes in response to more frequent, predictable rate increases; efforts to control costs by modernizing and optimizing the Postal Service’s infrastructure and workforce; the transition to new automation and mail-tracking systems; the level of transparency in measuring and reporting delivery performance; and the implementation of the new rate-setting processes and regulations. The Service continues to face challenges in generating sufficient revenue as mail volumes are declining and the mail mix is changing. This challenge became more evident after the Service’s revenue and volume results for the first quarter of 2008 were released. Volumes were down 1.7 billion pieces (3 percent) compared with quarter 1 in 2007 (see table 2), with notable declines in the two major mail classes: First-Class Mail and Standard Mail. These results are of particular concern because they occurred during a typically strong volume quarter that includes the holiday mailing season. Key declines during this time took place within the two largest categories, First-Class Mail and Standard Mail. In particular, volumes declined for flat- sized Standard Mail (e.g., catalogs) by 13 percent; for flat-sized First-Class Mail (e.g., large envelopes) by 15 percent; and single-piece First-Class Mail by nearly 7 percent. As a result of the overall declines in mail volume, revenues were $500 million less than planned. The Service attributed these volume declines and revenue shortfalls to multiple factors, including the effects of the May 2007 rate increase; a slowing economy with declines in the financial and housing industries, business and consumer confidence, and rising fuel and paper prices; increasing competition from other advertising media; and the continued diversion of single-piece First-Class Mail to electronic alternatives such as Internet bill payment and direct deposit. The declines in First-Class Mail volume in the first quarter of 2008 parallel the ongoing trends of First-Class Mail in general. This class of mail, once with the largest volumes and revenues, saw volumes decline by more than 7 percent between 2001 and 2007. The Service’s First-Class Mail volume estimate of 95.4 billion pieces built into its 2008 budget would be a slight decline from 2007 levels and would be the lowest volume level since 1994. These declines in First-Class Mail were mitigated in past years by growth in Standard Mail volumes and revenues. Standard Mail volumes exceeded those for First-Class Mail for the first time in 2005. This change was significant, in part because Standard Mail is more sensitive to prices and economic conditions and it takes about two pieces of Standard Mail to make the same contribution to the Service’s overhead costs as one piece of First-Class Mail. The Service’s 2008 budget planned a modest 1.1 percent growth in Standard Mail volumes because of such factors as the effects from the May 2007 rate increase and a projected slowness in the economy. The Service stated in its first quarter report for 2008 that these factors, among others, had an adverse impact on volumes as Standard Mail volumes declining by 2.6 percent compared to the first quarter of 2007. For the remainder of 2008, mail volumes and revenues will continue to face many of the same challenges that affected its first quarter results, particularly economic uncertainty and the impacts of rate increases. The Service recognizes that the law provides opportunities to address the revenue challenges it faces and that “cost cutting alone cannot sustain the business.” The act specifically provides tools and mechanisms to help promote revenue generation and retention of revenues. The act established more timely, flexible pricing mechanisms for the Service’s competitive and market-dominant products. For example, it allows the Service to use a streamlined process for raising the rates for its market- dominant classes, such as First-Class Mail, Standard Mail, and Periodicals, up to a defined price cap; to exceed the price cap should extraordinary or exceptional circumstances arise; and to use any unused rate authority within 5 years. For its competitive products, such as Priority Mail or Expedited Mail, the Service may raise rates as it sees fit, as long as each competitive product covers its costs and competitive products as a whole cover their attributable costs and make a PRC-specified contribution to overhead. The act also allows for new, customized products and services, as well as for the Service to retain any earnings, which may help finance capital investment and increase financial stability. In its updated Strategic Transformation Plan, the Service states that it plans on taking advantage of these new flexibilities through such actions as improving the value of its market-dominant products through such tools as Intelligent Mail, tailoring competitive products to market requirements, enhancing online postal services, and streamlining acceptance of mail at postal facilities for commercial mailers. The Postal Service applied its new rate-setting flexibilities when, on February 11, 2008, it announced its first rate increases under the act for its market-dominant products, including an increase in the cost of a First- Class stamp from 41 to 42 cents. The Service intends to raise the rates for each class, on average, close to the maximum allowed by the price cap (2.9 percent). Within each class, scheduled rate increases will vary for specific mailing services. For example, the rates for Standard Mail Flats are scheduled to increase 0.9 percent, compared with a 3.4 percent increase for Standard Mail Letters. These variable increases reflect the Service’s decision to moderate the increases for catalogs and other flats because of the large rate increases they experienced in May 2007. Furthermore, the Postal Service has recently notified the PRC of rate- setting initiatives for two of its competitive products. One notice pertained to establishing a premium for guaranteed delivery of Express Mail on Sunday and holidays, while the other notice pertained to establishing prices for a Priority Mail large-sized Flat Rate Box. Continued congressional oversight will be needed of the Service’s actions under the act to address its volume and revenue challenges. Particular attention should be paid to monitoring how the Service and mailers respond to a slowing economy and the implementation of the new rate process. Questions to address include the following: How will mailers and volume respond to changes in rates in the short term, as well as the Service’s intent to fully use its pricing authority under the rate-setting cap? To what extent will these changes affect the mail mix, including the type, size, and weight of mail? What types of innovative pricing methods will the Service offer? To what extent will customers’ desire for mail be affected by privacy concerns, environmental concerns, preference for electronic alternatives, or efforts at the state level to establish Do Not Mail lists? How will the Service be able to enhance the value of the mail (e.g., by providing more predictable and consistent service, tracking and tracing capabilities)? What will the Service do with any retained earnings (e.g., expand its capital program, save to weather downturns in the economy)? The Service faces multiple pressures in the short and long term associated with controlling costs and improving productivity while experiencing above-inflation cost growth in certain categories, revenue challenges, and an inflation-based price cap. In the first quarter of 2008, the Service reported responding to revenue shortfalls by cutting more than $300 million in costs compared to plan, including reducing over 10.5 million workhours compared to the first quarter of 2007. The Service indicated that continued vigilance on cost will be needed for the rest of the year, and it will prove increasingly difficult to reduce workhours at the same pace if revenue challenges continue. The Postal Service budgeted for a $1 billion reduction in expenses for 2008, to be achieved in part by reducing 28 million workhours and increasing productivity by 1 percent. Based on the first quarter’s performance, the Postal Service recognizes that it needs to more aggressively reduce expenses to mitigate the financial impact of the economic slowdown, and it has identified an additional $1 billion in cost reduction efforts, many of which are tied to reduced volumes. While actions taken to implement the reform act put pressure on costs—the Service expects a net increase of $1 billion in costs in 2008—the act also eliminates other payments and provides opportunities to offset some of these cost pressures through efficiency gains that could restrain future rate increases. It will be crucial for the Service to take advantage of this opportunity and achieve sustainable, realizable cost reductions and productivity improvements throughout its networks. Personnel expenses (which include wages, employee and retiree benefits, and workers’ compensation) have consistently accounted for nearly 80 percent of annual operating expenses. Growth in such expenses has exceeded inflation in each of the last 4 years, and the expenses are budgeted to increase by almost $660 million in 2008. The major drivers of the personnel expense increase include cost of living adjustments (COLA), general wage increases, and health benefit expense increases. For example, retirement health benefit costs have tracked well above the rate of inflation, and will remain high because of the new multibillion dollar payments required by the law. Another cost pressure the Service faces is to modernize and maintain its vast infrastructure and transportation system that supports its expanding delivery network—projected to increase by 1.9 million delivery points in 2008. The Service’s transportation costs have grown faster than the rate of inflation for the past 3 years and were budgeted to increase by 5.4 percent ($350 million) in 2008. The Service attributes these increases in part to contractual rate increases and rising fuel costs. We noted the Service’s vulnerabilities to rising fuel prices in a report issued last year. We have also reported on the challenges facing the Service in managing its 34,000 facilities nationwide, including the need to capture and maintain accurate facility data, adequately maintain facilities, address deferred maintenance issues, and align retail access with customer needs. The act provides an opportunity for the Service to address its cost challenges by establishing an inflation-based price cap for market- dominant products, which provides an incentive for the Postal Service to operate more efficiently. The act also requires the Service to develop a plan by June 2008 that includes its strategy for rationalizing the postal facilities network and removing excess processing capacity from the network. As part of this plan, the Service is to identify cost savings and other benefits associated with network rationalization alternatives. This plan provides an opportunity for the Service to make its case that realignment is needed to address infrastructure issues (e.g., excess capacity, maintenance needs, and facility locations) and reduce costs. It can also address concerns raised by Congress and the public about how decisions related to planned network changes are made and communicated to affected parties. We have reported our concerns that the Service’s strategy for realigning its processing and distribution network and workforce was not clear, and that its strategy lacked sufficient transparency and accountability, adequate stakeholder input, and performance measures for results and we have recommendations outstanding related to these concerns. The Service recognizes these cost challenges and plans to build on its progress in this area. We have reported on the Service’s progress in containing cost growth by reducing workhours, downsizing its workforce and improving productivity, and the Service’s ability to control cost growth during the first quarter was encouraging. Furthermore, the Service should benefit from agreeing with its four major labor unions reducing its future share of the contributions to the cost of health benefit premiums for many of its employees. The Service is planning to continue its cost cutting efforts as part of its Strategic Transformation Plan and is seeking efficiency gains from a variety of sources including more fully automating the sorting of flat mail—in 2008, the Service will deploy 100 machines to automate flat sorting in 30 to 35 facilities as part of its Phase One of the Flats Sequencing System (FSS); outsourcing certain activities, such as expanding contract delivery consolidating mail processing operations; optimizing retail resources using two scheduling tools to help managers align staffing to changes in customer demand; and working with members of the mailing industry to optimize mailer preparation requirements, including the use of Intelligent Mail Barcodes on mailpieces, to facilitate achieving the lowest combined mailing cost for all parties. Making progress in addressing cost challenges will be important as the Service is required to operate under the new price cap, particularly if the economy continues to weaken. Progress will also be needed in areas where it has been difficult to achieve. For example, we reported last summer that progress in consolidating mail processing operations among facilities has been slow due to several factors. In some cases, the Service was not ready to proceed with the consolidation, and other external factors have slowed the process, including union and community resistance. In addition, language in recent Senate Appropriations Committee reports has directed the Service not to implement consolidation decisions in certain locations until specific requirements have been met. Furthermore, in its first quarter financial report, the Service stated that if proposed legislation limiting its ability to contract out mail delivery and other postal activities is enacted, it would place significant restraints on its ability to achieve cost reductions. As actions are carried out to control costs in the future, continued oversight will be needed to ensure that the Service’s cost reduction strategies achieve their goals, without negatively affecting service. Specific oversight questions include the following: If volume shortfalls persist, will the Service be able to implement corresponding cost controls? If the economy continues to worsen and/or certain key costs continue to increase at levels above inflation (e.g., health benefit costs), how can the Service still meet its service goals and manage its costs under the rate cap? How will the new rate structure lead to efficiency improvements throughout the mail system? Will the Service’s implementation of its network realignment result in greater cost savings and improved efficiency? How do external constraints limit the Service’s ability to achieve cost savings through network optimization and what can be done to alleviate these constraints? Would the Service achieve its expected return on investment and improvements in operational performance in a second phase of automated flat sorting equipment? The Service will be challenged to manage its workforce as it transitions to operating in a new postal environment. The Service is one of the nation’s largest employers, with almost 786,000 full- and part-time employees at the end of 2007. As the Service continues to improve its operational efficiencies (i.e., rationalize its facilities, expand service measurement, increase automation, improve retail access, and streamline its transportation network), it will be challenged to realign its workforce in accordance with these changes. These challenges may be compounded by such factors as (1) changes in mailers’ behavior in response to the new rate structure and economic uncertainty that may reduce the level of processing needed at Postal Service facilities and (2) the expected retirement of a significant portion of its workforce, particularly at the executive level, within the next 5 years. These actions will require a different mix in the number, skills, and deployment of its employees, and may involve repositioning, retraining, outsourcing, and further reducing its workforce. The Service must describe, as part of the Facilities Plan required by the act, its long-term vision for realigning its workforce and how it intends to implement that vision. This plan is to include a discussion of what impact any facility changes may have on the postal workforce and whether the Postal Service has sufficient flexibility to make needed workforce changes. The Service recognizes the challenges in aligning its workforce with changing customer needs, new technologies, and emerging markets. In its updated Strategic Transformation Plan, the Service includes specific actions aimed at improving workforce flexibility, succession planning, and staffing efficiency. As it takes actions in this area, oversight will be important in several areas including: How will the Service’s workforce be affected by the implementation of new automation equipment that supports such initiatives as FSS or Intelligent Mail? How will the Service balance the varying needs of diverse customers when realigning its delivery and processing networks? How will employees and employee organizations be affected and informed of network changes and how will the Service monitor the workplace environment? How will the Service take advantage of flexibilities to deal with peak operating periods? The Service faces continued challenges in further updating its delivery performance standards, implementing representative measures of delivery performance, setting appropriate goals for delivery speed and reliability, and reporting results in a transparent and accessible manner. This information is critical for stakeholders to understand how the Service is fulfilling its mission of providing affordable, high-quality universal service on a self-financing basis—it would assist the Service and its customers in identifying and addressing delivery problems, and help Congress, the PRC, and others to hold management accountable for results and conduct independent oversight. In July 2006, we reported that the Service’s delivery performance standards, measurement, and reporting needed improvement. Among other things, we found that delivery standards for major types of mail had not been updated in a number of years and did not reflect current operations, including how mail is prepared and delivered. We also found that the Service does not measure the delivery performance of most types of mail, which limits transparency. Based on these and related findings, we recommended the Service take actions to modernize its delivery service standards, develop a complete set of delivery service measures, more effectively collaborate with mailers, and improve transparency by publicly disclosing delivery performance information. The act provided an opportunity to address these issues by requiring the establishment of modern delivery standards, the setting of goals for these standards, and annual progress reports. The act also established other requirements: The Service must issue modern service standards by December 2007 (these standards were issued); Within 6 months of issuing service standards the Service must, in consultation with the PRC, develop and submit a plan, with performance goals, to Congress for meeting those standards. Within 90 days after the end of each fiscal year, the Service must report to PRC on the quality of service for each market-dominant product in terms of speed of delivery and reliability, as well as the degree of customer satisfaction with the service provided. The act also identified four objectives for modern service standards: Enhance the value of postal services to both senders and recipients. Preserve regular and effective access to postal services in all communities, including those in rural areas or where post offices are not self-sustaining. Reasonably assure Postal Service customers delivery reliability, speed, and frequency consistent with reasonable rates and best business practices. Provide a system of objective external performance measurements for each market-dominant product as a basis for measurement of Postal Service performance. The Postal Service has taken an active role to address this challenge, including collaborating with mailers and the PRC on issuing the new service standards. The Service submitted to the PRC a proposal on service measurement using Intelligent Mail and is planning to expand the geographic coverage of its External First-Class Measurement System, and the PRC has put this proposal out for comment. The Service is also consulting with the PRC about other reporting issues. We are encouraged by the Service’s progress to date as well as its performance during the first quarter of 2008 for the segment of First-Class Mail that it currently measures. The delivery performance for mail measured by the Service’s EXFC system reported on-time deliveries for 96 percent of 1-day mail, 93 percent of 2-day mail, and 88 percent of 3-day mail, all of which were improvements over the first quarter of 2007. We continue to believe that the key principles of completeness, availability, and usefulness should guide future actions related to updating service standards and implementing performance measurement and reporting systems. Continued collaboration and oversight will be critical to making further progress as the system becomes more and more developed. In particular, questions will need to be asked, including the following: How should the standards and goals reflect different operational capabilities that affect the speed and reliability of delivery, such as presorting and separate processing streams? Given the different information needs of the various stakeholder groups— e.g., the Service, PRC, Congress, mailers, the American public—what are appropriate levels of transparency for each of the key groups? What level of detail should be available to each group? For example, some mailers have said they need detailed, real-time information to help identify and address delivery problems. In what format should information be available, and how should privacy be protected? How frequently should information be reported and/or accessible (e.g., quarterly, annually, or in real-time)? Should mailers pay for some of the information? How should mailer issues regarding the implementation of Intelligent Mail be addressed? What exclusions, if any, should be allowed under the Service’s reporting of annual results (e.g., exclusions for the holiday mailing period and incorrectly addressed and/or prepared mail)? The Postal Service and PRC will continue to be challenged to successfully implement the extensive regulatory changes required by the act. Currently, the PRC is reviewing the May 2008 rate increases filed by the Service and has asked for public comment on this filing. In addition to the PRC’s regulatory responsibilities for rate setting and monitoring service performance discussed earlier, these parties will be challenged to implement other requirements related to postal costing, accounting, and financial reporting. We have reported on specific challenges the Postal Service has faced in these areas. With respect to its financial reporting, the Service has made significant improvements in the frequency, content, and availability to address our earlier recommendations. Furthermore, in 2005 we reported on the long-standing issues of ratemaking data quality, many of which persist today. The act establishes new reporting and accounting requirements that should help to address these challenges. The major change is the establishment of, and authority provided to the new PRC to help enhance the collection and reporting of information on the Service’s postal rates and financial performance. The PRC has oversight responsibilities in such areas as: Market-dominant products: The PRC must prescribe by regulation the form and content of annual Service reports that analyze costs, revenues, and rates, using methods that PRC must also prescribe; specify which reported information shall be made public; initiate proceedings as necessary to improve the quality, completeness, or accuracy of this information; and assess compliance and complaints. Competitive products: The PRC must establish regulations that ensure that each competitive product covers its attributable costs, prohibit the cross- subsidization of competitive products by market-dominant products, and ensure that competitive products collectively cover what PRC determines to be an appropriate share of the Service’s institutional costs (overhead costs), as well as to assess complaints. Financial reporting: The PRC must (1) review annual, quarterly, and other periodic reports from the Service that contains information required by the Securities and Exchange Commission (SEC) for registrants, (2) review reports, due in 2010, on the Service’s compliance with rules prescribed by the SEC for registrants in implementing section 404 of the Sarbanes-Oxley Act of 2002, and (3) by December 2008, establish the accounting principles and practices that the Service must follow related to its competitive products, and in doing so, consider Treasury recommendations. The Service recognizes these challenges and the potential costs associated with meeting the new requirements. In its updated Strategic Transformation Plan, it laid out a timeline for implementing the Sarbanes- Oxley section 404 requirements and noted that it must manage the uncertainties related to the implementation of the new ratemaking process, the extent to which the PRC incorporates recommendations from the Treasury report, and any developments from the FTC report. The Service has not yet estimated the additional costs associated with these new regulatory requirements. We have reported that other federal agencies and smaller public companies have incurred significant costs associated with complying with SEC’s implementing regulations for section 404 of the Sarbanes-Oxley Act, but have also reported that costs are expected to decline in subsequent years given the first-year investment in documenting internal controls. In sum, these changes can help provide accurate and timely data on the Service’s costs, revenues, and mail volumes. This information can be used to enhance transparency and accountability for all postal stakeholders so that they have a comprehensive understanding of the Service’s financial condition and outlook and of how postal rates are aligned with costs. As the new regulatory framework is implemented, continued oversight may be required in several areas: How will the PRC use its discretion to continue defining and implementing the new regulatory structure? How effectively is the PRC carrying out its regulatory responsibilities regarding rate setting and monitoring service performance? Given the complexity of regulatory changes, how can the PRC balance the interests of all stakeholders, particularly those with less expertise and resources? What criteria will the PRC use for evaluating the quality, completeness, and accuracy of ratemaking data, including the underlying accounting data and additional data used to attribute costs and revenues to specific types of mail? Looking forward, how will the PRC, the Service, and other stakeholders consider and implement improvements to data quality over time? How will the PRC balance the need for high-quality ratemaking data with the time and expense involved in obtaining the data? How will PRC structure any proceedings to improve the quality of ratemaking data and enable the Service and others to participate in such proceedings? What proceedings might PRC initiate to address data quality deficiencies and issues that PRC has raised in its recent decision on the rate case? How will the Service be affected by the costs associated with complying with the SEC rules for implementing section 404 of the Sarbanes-Oxley Act, as well as the need for separate information on competitive and market-dominant products? Information required under the act can be used to facilitate constructive dialogue and debate about postal reform issues related to universal service, the postal monopoly, fair competition, consumer protection, and transparency and accountability. Specifically, the act included provisions for reports required over the next 5 to 10 years related to key postal reform issues aimed at continually examining and reporting on the Postal Service’s mission, role, and oversight structure in an increasingly competitive environment. The act required multiple stakeholders, including the Postal Service, PRC, Postal Service OIG, OPM, Treasury, and GAO, to issue these reports, and established a wide range of deadlines for this work. The information can be useful to Congress when it is considering key postal reform issues including: What universal postal service will be needed in the future and how should it be defined, given past changes and future challenges? To what extent should certain monopoly provisions be maintained or narrowed? What role should the Service play in providing universal postal services vis-à-vis its competitors? What are appropriate legal standards for fair competition in areas where the Service competes with private-sector providers? What transparency, oversight, and accountability are needed for the Service, particularly as long as it remains a federal entity with a monopoly to deliver letter mail? How appropriate are the new regulatory structure and rate-setting system? What barriers, if any, have prevented progress under the act (e.g., in optimizing the Service’s infrastructure network), and how can they be addressed? As outlined earlier, information related to some of these issues has already been published from the Treasury and FTC. Treasury issued a report on the accounting principles and practices that should be followed by the Service, and the FTC issued a report that analyzed laws that apply differently to the Service’s competitive products and similar products provided by private competitors, and estimated the economic burdens on, and advantages to, the Service due to these legal differences. This information provides a good starting point for discussions on broader reform topics such as the following: Universal postal service and the postal monopoly: The mission of the Postal Service revolves around providing affordable, high-quality universal postal services on a self-financing basis. While the act requires the PRC to provide annual reviews of service quality and the estimated costs of providing universal service, the act requires a more comprehensive study from the PRC on the scope and standards of universal postal service and the postal monopoly. This report, due by December 2008, is to describe any deficiencies in universal service and can include recommendations on future changes. The PRC is required to obtain public comments and consult with the Service in preparing this report. Accounting, financial transparency, and oversight: The PRC solicited public comments on Treasury’s report. In addition, the PRC raised questions about what financial transparency and oversight are appropriate for the Service’s competitive products fund and whether a public or private entity should conduct such oversight. These comments will assist the PRC in fulfilling the act’s requirement to establish accounting practices and principles for the Service to follow, and issue regulations for the Service’s reporting of its costs, revenue, rates, and volumes. Regulation of postal rates: The act requires the PRC to annually report to the President and Congress on the extent to which postal regulations, including those related to postal rates, achieve statutory objectives. Looking forward, the PRC is required to assess ratemaking and other provisions of the act every 5 years (with the first report due by December 2011), and review the system for regulating the rates and classes for market-dominant products by December 2016. At that point, the act empowers PRC to make changes to the system for regulating market- dominant rates. Future business model: GAO is required to issue a report by December 2011 that evaluates various options and strategies for the long-term structural and operational reforms of the Service. The requirement states that we may include, among other things, recommendations on how the Service’s business model can be maintained or transformed to assure continued availability of affordable universal postal service. We are encouraged by the early implementation steps that the Service, the PRC, the Department of the Treasury, FTC, and other stakeholders have taken. The Service, the PRC, mailers, and other stakeholders have found new ways to engage in constructive dialogue and debate and in some cases, reach consensus on how best to proceed. These actions—which contrast sharply with the adversarial ratemaking process abolished by the act—hold promise for future progress across a broad range of postal reform issues. Such progress will remain necessary as the Service, the mailing industry, and competitors transform themselves in response to the rapidly changing communications and delivery marketplace. Mr. Chairman, this concludes my prepared statement. I would be pleased to respond to any questions that you or the Members of the Subcommittee may have. For further information regarding this statement, please contact Katherine Siggerud, Director, Physical Infrastructure Issues, at (202) 512-2834 or at siggerudk@gao.gov. Individuals making key contributions to this statement included Teresa Anderson, Joshua Bartzen, Kenneth John, Summer Lingard, Jeanette Franzel, Shirley Abel, Scott McNulty, Brandon Haller, David Hooper and Kathy Gilhooly. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | In December 2006, Congress passed the first comprehensive postal reform legislation in over 30 years. The Postal Accountability and Enhancement Act (the act) provided opportunities to address many of the financial, operational, and human capital challenges facing the Postal Service (the Service), which contributed to GAO's decision to remove the Service's transformation efforts from its High-Risk List last year. Specifically, the act provides tools and mechanisms that can be used to establish an efficient, flexible, fair, transparent, and financially sound Postal Service--one that can more effectively operate in an increasingly competitive environment not anticipated under the Postal Reorganization Act of 1970. This testimony focuses on (1) the actions to date resulting from implementing the act and how it affected the Service's 2007 financial condition, (2) the implementation challenges and areas for continued oversight, and (3) how information required under the law can contribute to future postal reform decisions. The testimony is based on GAO's past work; a review of the implementation of the postal reform law, including actions already taken; and updated information on the Service's financial and operational condition. The Postal Service had no comments on this testimony. Over the last 14 months, key actions have been taken to implement the act. For example, a new rate-setting system and regulatory agency were established, the Service began prefunding its retiree health benefit obligations, service standards were updated, and key reports were issued. These actions have required the collective efforts of many postal stakeholders including the Service and the Postal Regulatory Commission. The Service reported a $5.1 billion net loss for fiscal year 2007. Some of the actions taken to implement the act, such as funding changes to its retiree health benefit obligations and pension requirements, directly impacted these results, as did other events such as the January 2006 and May 2007 rate increases. The uncertain economic environment serves to exacerbate the challenges facing the Service and contributed to lower than expected mail volumes and revenues in the first quarter of fiscal year 2008. The Service projects a $600 million net loss for 2008 as it faces challenges such as generating volumes as rates increase again in May; managing its costs and improving operational efficiencies through accelerated cost reduction strategies; maintaining, measuring, and reporting service; and managing its workforce. Some key areas for continued oversight include changes to mail volumes and revenues, efforts to control costs by optimizing the Service's infrastructure and workforce, transition to new automation and technology to enhance mail sorting and tracking, transparency in measuring and reporting delivery performance, and implementation of the new rate-setting regulations. Information required under the act can be used to facilitate constructive dialogue about complex postal reform issues that may eventually need to be revisited by Congress. The act requires multiple reports and studies over the next 5 to 10 years that can be used to continually examine and assess the Postal Service's position in an environment of increasing competition and technological advances. Specifically, these reports and studies will provide key information on the Service's mission and role, monopoly protections, universal service requirements, rate-setting and other regulatory issues, oversight structure, competition issues, and consumer protection. |
Created by the National Flood Insurance Act of 1968, the NFIP is designed to protect homeowners from flood losses while also minimizing the exposure of property to flood damage. To participate in the program, communities must adopt and enforce floodplain management ordinances to mitigate the effects of flooding on new or existing homes in special flood hazard areas (SFHA). Flood insurance is available in communities participating in the NFIP and is offered to eligible homeowners for homes and their contents. FEMA, through its Federal Insurance and Mitigation Administration, manages the federal flood insurance program and floodplain mitigation programs. When the program was created, the purchase of flood insurance was voluntary. To increase the impact of the program, however, the Congress amended it in 1973 and in 1994 to require the purchase of flood insurance by many homeowners and to place the onus for ensuring compliance upon lending institutions. Currently, homeowners in SFHAs in participating communities must purchase flood insurance as a condition of obtaining mortgages on their homes if the loans are made, increased, extended, or renewed by federally regulated lending institutions; sold to Fannie Mae or Freddie Mac; or made, insured, or guaranteed by a federal agency, such as the Small Business Administration, Federal Housing Administration, or the Department of Veterans Affairs. No definitive data on the number of mortgages meeting these criteria exist; however, on the basis of 1999 data reported by lenders, most mortgaged properties meet the above criteria and, if in a SFHA, would be subject to the requirements of the National Flood Insurance Act. Federally regulated lending institutions—which make most of the mortgages in the United States—and loan servicers must ensure that, where required, flood insurance is purchased at the time that the mortgage is obtained and maintained throughout the life of the loan, or added if the residence involved is reclassified as being located in a SFHA. They may not make, increase, extend, or renew a loan secured by a structure located in a SFHA in a participating community unless the structure is covered by flood insurance. Lenders generally purchase flood zone determinations from flood zone determination companies that use FEMA flood maps and other data to ascertain if properties are situated in flood zones. The companies record the results of their determinations on a standard flood hazard determination form (SFHDF) and provide this form to lenders, who are required to maintain it. Figure 1 shows the process that lenders generally follow for obtaining and recording flood zone determinations as part of their mortgage approval process. If a lender or servicer determines at any time during the life of the mortgage that a property is located in a SFHA—even if a flood zone remapping places it in a SFHA after the mortgage was first originated—the lender or servicer must ensure the purchase of the appropriate flood insurance. The federal agencies that regulate lending institutions and the GSEs were also given certain compliance responsibilities. As required by the 1994 amendments to the National Flood Insurance Act, the regulatory agencies established rules directing lending institutions not to make loans secured by improved real estate located in SFHAs unless flood insurance had been purchased. The regulatory agencies are to examine loans for compliance with these regulations during their periodic examinations of member financial institutions, using uniform policies and procedures for assessing lender compliance with flood insurance requirements. The GSEs were required to implement procedures reasonably designed to ensure that flood insurance coverage exists for any purchased loan that is secured by improved real estate located in a SFHA. The GSEs have established flood insurance purchase requirements to be followed by institutions that sell mortgages to them or service mortgages for them and have procedures in place to assess loans for compliance. Appendix II contains additional information on bank examination procedures and processes, and appendix III contains additional information on the review and audit procedures followed by GSEs. Officials involved with the flood insurance program developed contrasting viewpoints about whether lenders are complying with flood insurance purchase requirements primarily because the officials use differing types of data to reach their conclusions. Federal bank regulators and officials from Fannie Mae and Freddie Mac base their belief that lenders are generally complying with the NFIP’s purchase requirements on regulators’ examinations and GSEs’ reviews conducted to monitor and verify lender compliance. In contrast, FEMA officials believe that many lenders frequently are not complying with the requirements, which is an opinion based largely on noncompliance estimates computed from data on mortgages, flood zones, and insurance policies; limited studies on compliance; and anecdotal evidence indicating that insurance is not in place where required. Neither side, however, is able to substantiate its differing claims with statistically sound data that provide a nationwide perspective on lender noncompliance. On the basis of their bank exams and compliance reviews, bank regulators and GSE officials believe that the rates of noncompliance with flood insurance purchase requirements are very low. According to representatives of the regulatory agencies, very few violations of flood insurance requirements have occurred. Moreover, according to the bank regulators’ 1994-2000 annual reports to the Congress, most of the violations found during examinations have been of a technical nature, such as improperly completing necessary forms or not giving borrowers timely notification that the property is in a flood zone before the loan closing date. More serious violations, such as failure to confirm that insurance is in place when required, or failure to obtain a flood zone determination for a property, have been infrequently detected during examinations. For example, since 1996, the bank regulators have levied 51 civil monetary penalties on lending institutions that committed serious violations of flood insurance requirements. Fannie Mae and Freddie Mac may force lenders who sell loans to repurchase a loan if their requirements for flood insurance are not met. Both enterprises told us that this action has rarely occurred because instances of noncompliance are almost always corrected by the lenders. Similarly, the bank regulators’ examiners and managers responsible for conducting bank examinations that we talked to said lending institutions are doing a good job of complying with flood insurance requirements. In general, the examination process involves field examiners assessing a lending institution’s procedures for ensuring compliance with the flood insurance requirements and checking a sample of loan files to verify that the procedures are routinely followed. According to the examiners, lending institutions are familiar with the stipulations of various consumer compliance laws and fulfilling the requirements for flood insurance has become a standard procedure in mortgage lending. Bank regulatory and industry officials have stated that completing standard flood hazard determination forms and ensuring that borrowers obtain flood insurance for properties where it is required are standard business practices for lenders. They added that at larger lending institutions, automated systems typically track borrowers’ flood insurance policies throughout the life of a loan. Nevertheless, the regulatory agencies and GSEs acknowledge that complete data on compliance do not exist and the data they obtain from bank examinations are not statistically representative of compliance either for the bank or the country. Regulators and GSEs do not use a statistical sample of loans when examining for flood insurance compliance. Their findings at a lending institution, or even from all of the lending institutions that they regulate, therefore, cannot be generalized to the industry. Examining for compliance with flood insurance purchase requirements is but one of approximately 20 different laws and regulations—such as those relating to equal opportunity lending, truth-in-lending, and debt collection practices—included in compliance examinations. Staff of one regulatory agency told us that obtaining a statistically valid sample for any of these issues would require examination of many more loans than currently inspected and would seriously threaten their ability to examine bank compliance with other requirements. They do not believe it would be prudent or cost-effective to review a statistically valid sample of loans for compliance with NFIP regulations without stronger evidence of widespread noncompliance. FEMA officials disagree with bank regulators and GSEs about the level of overall lender compliance with flood insurance requirements. Although FEMA officials believe that lenders have been doing a better job of ensuring the purchase of flood insurance in recent years, they also believe that noncompliance rates are still significant. FEMA officials base their opinion on three factors: (1) their estimate of the aggregate number of homeowners who live in flood zones, have federally backed mortgages and, therefore, are required to have flood insurance compared with the number of flood insurance policies in force; (2) a small number of relevant, but limited, studies; and (3) anecdotal information obtained from conversations with local government officials and others knowledgeable about the flood insurance program. FEMA does not have information on the individual properties that should be covered by flood insurance, but it has estimated the number of properties that should be insured. Using that estimate, it developed an overall estimate of noncompliance indicating that many properties do not have the required insurance. According to FEMA’s estimate, in fiscal year 2000 nearly one out of three homes required to have flood insurance did not have it. On the basis of Mortgage Bankers Association data on homes and mortgages, U.S. Corps of Engineers data on the percentage of structures located in SFHAs, and its own insurance policy data, FEMA estimates that less than 2.9 million flood insurance policies have been issued for over 4.3 million mortgaged properties in SFHAs. According to this estimate, nearly 1.4 million—32 percent—may not have flood insurance. FEMA officials acknowledge that an accurate rate of noncompliance will not be known until mortgage data are linked to flood insurance policy data. They nevertheless believe that the estimate indicates a potentially significant noncompliance problem. FEMA officials point out that two studies the agency conducted also indicate noncompliance with the mandatory purchase requirement. A 1999 study conducted by a FEMA regional office and a 2000 study by the FEMA IG assessed specific areas to determine the number of homes that had flood insurance. The two studies examined different localities, but each found a portion of sampled properties—as high as 45 percent in the regional office study—that did not have the required insurance, as discussed below. A post-disaster compliance study issued in April 1999 by one of FEMA’s regional offices assessed rates of noncompliance after a 1998 flood in Vermont. The study examined 120 properties located in a SFHA and found that 54—or 45 percent—had mortgages from a federally regulated institution and should have had insurance but did not. Moreover, the study found that the federal government provided $500,000 in disaster assistance to these properties—assistance funds that would not have been paid had these properties been insured. An August 2000 IG study examined the rate of noncompliance for 4,195 residences located in SFHAs in 10 states. The study found that for these residences, about 416—or 10 percent—were required to have flood insurance but did not. For example, a North Carolina subdivision that had been built in a SFHA in 1996 contained 27 uninsured homes of which 20 had a mortgage from a federally regulated lending institution. The study also noted that statistics for that state showed that of about 150,000 structures located in SFHAs, only 33 percent were covered by flood insurance. In addition to these analyses, FEMA officials cite the results of studies conducted by private companies after presidentially declared disasters in North Dakota and Kentucky that found that from 28 to 38 percent of the properties sampled did not have flood insurance. While neither of these studies took into consideration whether the homeowners without flood insurance had mortgages from regulated lenders, FEMA officials believe that it is reasonable to assume that this indicates a potential problem with compliance since a large percentage of the homes in this country have mortgages from regulated lenders. Finally, FEMA officials stated that their concerns about lender compliance are bolstered by anecdotal evidence they have obtained during FEMA- sponsored workshops and field visits and from insurance and flood zone determination industry officials. For example, during a NFIP lender- training workshop that we observed, bank employees discussed with FEMA officials examples of noncompliance that were known but not corrected. The bank employees said that they have observed instances in which other bank officials discovered that a home required flood insurance when its owner sought to refinance the mortgage it had with the bank. When the owner decided not to refinance, the bank officials did not require the purchase of the flood insurance to protect the existing loan, even though the law mandates such purchase. Similarly, when providing assistance in presidentially declared disaster areas, FEMA officials have heard accounts of flood victims who should have been required by their lender to have flood insurance but were not told that they needed it. Moreover, representatives of the insurance and determination industries told us they consistently informed FEMA that they believe the extent of noncompliance is a problem. FEMA officials acknowledge that their opinions on compliance are based on limited data. Nevertheless, they believe that the preponderance of information available to them indicates that lenders are not fully ensuring compliance with flood insurance purchase requirements. To be in compliance, the purchase of mandatory flood insurance must occur when a mortgage is originated, and this insurance must be retained and renewed over the life of the loan. No data were readily available to enable us to assess noncompliance at both loan origination and over the life of the loan and in all geographic areas. However, other readily available data we obtained suggest that in highly flood-prone areas, noncompliance could be low at loan origination. We compared the number of new mortgages made in 1999 with new flood insurance policies issued in certain of the nation’s most highly flood-prone census tract areas. For that period and for most of the geographic areas we examined, this comparison did not suggest a major noncompliance problem at loan origination, because only 44—9 percent of the 471 census tracts we analyzed—showed fewer insurance policies than mortgages. Moreover, buildings such as cooperatives and condominiums—which are under different purchase requirements—exist in some of these areas and could account for fewer policies than mortgages being issued. To obtain a perspective on the level of noncompliance with the mandatory flood insurance purchase requirements at loan origination, we first identified highly flood-prone census tracts where 90 to 100 percent of the properties are located in SFHAs. We then compared data reported by lenders on new mortgages made in 1999 by federally regulated lending institutions or guaranteed by a federal agency with FEMA data on new flood insurance policies issued in that year in each selected tract. These data, in the aggregate, do not suggest that noncompliance is widespread at the time of loan origination in highly flood-prone areas. For the 471 census tracts we selected, located in 64 different counties, over 88,000 insurance policies were issued in 1999, or about 88 percent more than the 47,000 mortgages originated in those census tracts during that period. A summary, by county and state, of the census tract data we examined is contained in appendix IV. Most of the census tracts had more new insurance policies purchased than mortgages originated. Of the 471 census tracts we analyzed, 413 had more new insurance policies issued than mortgages, and in many of the census tracts substantially more flood insurance policies were purchased than mortgages originated. For example, nearly 4 times more flood insurance policies were purchased than new mortgages originated in census tracts in 6 of the 64 counties. Two reasons for this could be that some homeowners in the census tracts who did not have new mortgages purchased insurance or that unregulated lenders originated a significant number of mortgages in those areas and also required the purchase of flood insurance. We contacted flood zone administrators for 2 of these counties, and they attributed the large number of insurance policies purchased to many factors, including an increasing awareness of flood dangers resulting from hurricanes affecting their counties during 1999 as well as recent public outreach and education efforts. Our analysis does suggest that a noncompliance problem at loan origination might exist in some geographic areas. For counties containing 44 of the census tracts—9 percent of the tracts we examined—we were able to determine that there were fewer insurance policies issued than mortgages. For example: In a Hawaii county that has 5 census tracts that are virtually entirely in a SFHA, 357 loans were made but only 88 flood insurance policies were issued. In a New Jersey county with 4 census tracts that are virtually entirely in a SFHA, 291 loans were made but only 81 flood insurance policies were issued. In a New York county with 1 census tract with 97 percent of the properties in a flood zone, 98 mortgage loans were made but only one new insurance policy was issued. However, floodplain managers in the states mentioned above point out that these areas are densely populated and contain many buildings that are condominiums or, in the case of New York and New Jersey, cooperatives. These structures have different purchase requirements than other residential properties. For example, condominium owners may not have to obtain a flood insurance policy if the building’s condominium association has purchased one that covers the entire structure. Consequently, there would be fewer flood insurance policies than new mortgages in those situations. For the remaining 14 census tracts we analyzed, we also found that there were fewer insurance policies than mortgages issued. However, these census tracts are in counties where a number of insurance polices were issued that could not be identified with any specific census tract and which could account for the shortage of policies as compared with mortgages. For example, in a North Carolina county with one census tract that had 37 mortgages and only 15 insurance policies issued, there were also 201 insurance policies that could not be identified with any of the six census tracts in the county. It is possible that a portion of these policies were actually for properties in the highly flood-prone census tracts in these counties. We discussed our analysis of mortgage and insurance data with officials of the bank regulatory agencies, GSEs, and FEMA. The regulatory and GSE officials stated that the analysis supports their position that few concerns exist regarding noncompliance with flood insurance purchase requirements. They stressed that they are confident that their examination procedures are effective and appropriate. Additionally, they said that our data were from 1999, before some of the regulators had completed examinations of all institutions they oversee to assess them for compliance with the current requirements. They believe that any noncompliance that may exist will be further reduced after all institutions have been examined and made fully aware of the flood insurance compliance requirements. FEMA officials agreed that the analysis indicates relatively low levels of noncompliance at the time loans are made and that on the basis of our analysis, noncompliance at the time of loan origination is not an area of major concern. Nevertheless, they still believe that significant noncompliance problems exist with insurance policy retention and renewal. They believe that lenders and homeowners fail to ensure that the insurance policies remain in force for the life of the loan and pointed out that our analysis was unable to examine existing mortgages to determine if insurance policies are being retained and renewed as required over the life of the loan. Property-specific data on mortgages, flood zone determinations, and flood insurance policies—obtained both at loan origination and at various points during the life of the loan to ensure the insurance remains in force—would be needed to fully assess compliance. Comparing these data would allow computation of compliance rates nationally, regionally, or locally and would—with an additional piece of data, the mortgage lender identification numbers—identify specific noncomplying lenders. However, there are a number of challenges to obtaining and assessing these data. These include establishing data reporting requirements for lenders to provide relevant mortgage data, designating an organization to receive and compare these data, and determining the costs and benefits of obtaining these data. The regulators and GSEs, on the one hand, and FEMA, on the other, have differing viewpoints of the viability of and need for obtaining these data. The three data elements that would have to be linked to fully measure compliance are property addresses, flood zone determinations, and proof of flood insurance for those properties. Property addresses would need to be obtained for those properties financed by mortgages covered by the NFIP legislation, namely, those made by federally regulated lending institutions or guaranteed by federal agencies, or those purchased by GSEs. At the time the loan is made, flood zone determinations for properties associated with those mortgages are required to be performed by the lender—making it possible to identify the pool of mortgages for which flood insurance is required. Once this pool of mortgages is identified, it would be necessary to match the individual mortgages to the flood insurance policies issued to determine whether insurance policies had been issued for those properties required to have them. The results could be used to (1) compute compliance rates nationally, regionally, and locally and (2) identify individual properties without flood insurance at origination. To monitor the status of compliance over the life of the mortgage, updated mortgage and insurance information would be needed. Mortgages are frequently sold and therefore held by a different entity. Similarly, the status of flood insurance policies may change. Currently, there are requirements for various notifications when these events occur, and these changes are required to be recorded in loan files. Collection of an additional data element—lender identification numbers— would permit measurement of noncompliance on lender specific levels. Lender identification numbers are necessary for identifying the specific lender associated with a mortgage and for determining the appropriate federal regulator. Obtaining lender identification numbers would reveal which lenders did not ensure that flood insurance was purchased and maintained when required. A number of challenges exist to collect and assess the data needed to determine compliance. First, reporting requirements would be needed to centrally collect data components to determine if flood insurance is being purchased as required when mortgages are originated. As previously explained, for each mortgage originated by a regulated lending institution or purchased by a GSE, key data identifying the specific mortgaged property (i.e., property address), the flood zone determination for the property, and proof of insurance for properties in SFHAs would need to be compiled. Lenders, FEMA, and others currently hold all or parts of these data. For example, lenders maintain all of these data in their loan files, and FEMA maintains a database that contains all insurance policy data. Consequently, no new data would have to be generated, but they would have to be centrally reported. Second, a single organization would need to be assigned the responsibility for measuring compliance. This organization also would need to have appropriate authority to collect the data needed to measure compliance. Although some organizations have various authorities to obtain data—for example, the bank regulators can collect data from lenders, and FEMA can obtain data from insurers—no organization currently has the authority to collect from lenders, insurance companies, or other organizations all the data needed to fully measure compliance. Specific legislative authority may be needed to enable a single organization to collect the data necessary to measure compliance. Third, costs and benefits would need to be fully explored to determine whether establishing a new system for measuring compliance is justified. The regulators and GSEs, on one hand, and FEMA, on the other, have differing viewpoints on the viability of and need for obtaining compliance measurement data. In this regard, regulatory and GSE officials said that this effort would result in significant costs for lending institutions if they were required to report all flood insurance-related mortgage data. An official from one regulatory agency pointed out that although there are no data on the costs of implementing new reporting systems, any new data requirement placed on lenders would result in changes to the lenders’ information systems. The official estimated that costs could be in the millions of dollars for even minimal changes, and for institutions that have older systems, or that are not highly automated, additional data reporting requirements could have a significant impact. The regulatory and GSE officials also said that in addition to concerns that obtaining this compliance measurement data would be costly, they also believe that little benefit would be obtained through such action. They stated that there continues to be no empirical evidence that there is any widespread noncompliance with flood insurance requirements, and that in fact, our analysis points to a high level of compliance. Officials from these organizations added that as they do their compliance examinations and reviews of lenders, they look at a sample of a lender’s entire portfolio of mortgages—both new and existing—and if these examinations and reviews are ensuring compliance at origination, they are also ensuring compliance over the life of the mortgage loan. Consequently, according to these officials, without further evidence of noncompliance problems, establishing a new process to require reporting and monitoring of flood insurance data is not justified. Officials from many of the regulatory agencies believe that instead of establishing a new compliance measurement program, it would be better to have FEMA use the data it currently has to measure compliance and to conduct additional post-disaster compliance studies. The officials stated that FEMA has significant data on mortgages requiring flood insurance, and that additional data to measure compliance could be obtained from the insurance agents that sell flood insurance policies. They further said that conducting compliance studies after disasters have occurred could determine if there were any significant amounts of noncompliance in the affected area. They added that more of these studies would better determine if there actually are noncompliance problems and help pinpoint geographic areas in which they may need to shift greater focus to flood insurance compliance in their examination activities. FEMA officials, however, believe that establishing a comprehensive noncompliance measurement program may be beneficial and appropriate. They said that they remain concerned that insurance policies are not being retained and renewed where required on existing mortgages, as their data show that the retention rate for flood insurance policies is about 90 percent, which is below the insurance industries’ homeowners policy retention rate of 95 percent. The FEMA officials said that if lenders were adequately ensuring the renewal of flood insurance policies, the retention rate would be similar. Therefore, they believe that only a comprehensive program that gathers and analyzes data on existing mortgages will resolve the debate over noncompliance and ensure that both property owners and the federal government are adequately protected. FEMA officials added that because FEMA is involved in the selling of insurance policies and in the financial condition of the insurance program, a comprehensive noncompliance measurement program might more appropriately rest outside of FEMA in an organization that would be more independent. Additionally, FEMA officials stated that with the provisions in the president’s proposed budget that will significantly increase their efforts to remap and update the nation’s flood zones, establishing a process to identify noncompliance is even more critical. They pointed out that the remapping efforts will likely place more properties in SFHAs, and more property owners—including those with existing mortgages—will be required to purchase flood insurance. They expect that this remapping will result in potentially more noncompliance, and that a comprehensive noncompliance monitoring effort will be needed to ensure that all owners of properties requiring flood insurance purchase such insurance. Finally, FEMA officials said that conducting post-disaster studies is not the solution to the noncompliance debate. They said that post-disaster studies can only offer a limited perspective on noncompliance and do not address noncompliance issues for the nation as a whole. Moreover, a major flaw with this approach would be that FEMA would be identifying noncompliance when it is too late—after the disaster has occurred and uninsured properties are flooded. They said that it is much more important to identify noncomplying properties before they are damaged in a disaster, thereby providing the opportunity to ensure that property owners have insurance protection and minimizing the need for federal disaster assistance for these properties. Lastly, FEMA officials said that conducting post-disaster studies is very resource intensive. They said that they do not have the resources and capabilities to conduct any significant number of compliance studies. We provided a draft of this report to FEMA, regulatory agencies—FDIC, FRB, OCC, and OTS—and GSEs—Fannie Mae and Freddie Mac—that are responsible for the issues discussed in this report. All of the agencies generally agreed that the report presents an accurate and objective presentation of the differing perspectives on noncompliance with mandatory flood insurance purchase requirements. FEMA, FDIC, FRB, Fannie Mae, and Freddie Mac provided letters commenting on the draft that appear in appendixes V, VI, VII, VIII, and IX. OCC and OTS provided clarifying language and technical comments that were incorporated into the report as appropriate. Three organizations provided additional perspectives and comments. FEMA said that it continues to believe that significant problems exist with insurance policy retention. It stated, as an example, that last year’s gains in new policies were offset by attrition from the previous years’ number of policies in force. FEMA also described a number of strategies it has initiated to improve policy retention. These strategies include working with the regulatory agencies and GSEs to identify actions FEMA could take to improve lender compliance; assessing state escrow laws and systems to determine whether obstacles to flood insurance escrow exist and, where necessary, work with states to resolve these obstacles; and improving its flood insurance public education and advertising campaign. FRB maintained that compliance is generally satisfactory with the institutions they supervise. It said that our analysis suggests low levels of noncompliance at loan origination and that the report helped narrow any future inquiry on lender noncompliance to areas of policy renewal and retention. FRB added that on the basis of years of experience in examining state member banks, it believes that those banks have a good record of compliance with flood insurance purchase requirements not only at loan origination but also during the time the banks own the loan. Freddie Mac commented that certain facts contradict FEMA’s assertions that noncompliance is substantial. Specifically, Freddie Mac noted that FEMA acknowledges that compliance at origination is high and 90 percent of policies are renewed. Therefore, it sees no basis for FEMA’s belief that noncompliance is substantial; rather, the evidence suggests that noncompliance is marginal. Additionally, Freddie Mac commented that FEMA should take a more proactive role in compliance monitoring by collecting and analyzing data currently available to it and conducting investigations to determine reasons for noncompliance. Freddie Mac said that it does not share FEMA’s belief that having responsibility for compliance creates a conflict of interest with its responsibility for managing the National Flood Insurance Program. We will send copies of this report to the Director, Federal Emergency Management Agency; Chairman, Federal Deposit Insurance Corporation; Chairman, Board of Governors of the Federal Reserve System; Comptroller of the Currency; Director, Office of Thrift Supervision; Chairman and CEO, Fannie Mae; and Chairman and CEO, Freddie Mac. We will also make copies available to others upon request. In addition, this report will be available at no charge on the GAO Web site at http://www.gao.gov. If you have any questions about this report, please call me or John Schulze at (202) 512-2834. Key contributors to this report are listed in appendix X. Federal Emergency Management Agency (FEMA) officials and bank regulators disagree about whether lenders are fully complying with the flood insurance purchase requirements of the National Flood Insurance Program (NFIP). Given this disagreement, and concerned that lender noncompliance could be high, the Subcommittee on VA, HUD, and Independent Agencies, Senate Committee on Appropriations, directed us to examine whether lenders are complying with the purchase requirements. In response to that mandate and to requests from the Senate Committee on Banking, Housing, and Urban Affairs and its Subcommittee on Economic Policy, we focused our work on the following questions: 1. What are the bases for the differing perspectives on lender noncompliance? 2. What does other readily available data indicate about the level of noncompliance? 3. What data would be needed to fully measure compliance? To obtain an overall understanding of the NFIP, we analyzed the program’s history, regulations, policies, and procedures. We interviewed and gathered studies from FEMA officials, federal regulatory agencies, government-sponsored enterprises (GSE), flood zone determination companies, mortgage companies, mortgage servicers, insurance companies, and industry associations. We examined reports issued by the FEMA Inspector General and documents from FEMA’s Federal Insurance and Mitigation Administration. In addition, we interviewed this organization’s former Administrator, Acting Administrator, Director of Marketing, Lender Compliance Officer, and other officials responsible for administering the NFIP. We also interviewed FEMA’s Assistant Inspector General who is responsible for that office’s flood insurance compliance review. To address the first objective, we determined how four regulatory agencies and two GSEs monitor lender compliance. Specifically, we focused on the Federal Deposit Insurance Corporation (FDIC), the Federal Reserve Board (FRB), the Office of the Comptroller of the Currency (OCC), the Office of Thrift Supervision (OTS), the Federal National Mortgage Association (Fannie Mae), and the Federal Home Loan Mortgage Corporation (Freddie Mac). We interviewed officials from each of these agencies, including 25 field managers and bank examiners from the FDIC, FRB, OCC, and OTS. In addition, we analyzed examination files from the FDIC, Federal Reserve Bank of New York, OCC, and OTS to identify banks and thrifts that had been subjected to civil monetary penalties and other enforcement actions; we further examined correspondence between the GSEs and servicers. We also observed FDIC and OTS bank examinations in the Baltimore, Md., area to better understand the policies and procedures of such examinations. Additionally, we reviewed FEMA’s data and efforts to measure and assess noncompliance. We interviewed officials from FEMA’s Federal Insurance and Mitigation Administration and obtained information and documentation of FEMA’s estimates on overall levels of noncompliance; processes it uses to estimate structures in special flood hazard areas; processes it uses to collect, report, and share flood insurance policy data; and actions it has taken to inform the public about floodplain mapping and compliance with mandatory purchase requirements. We interviewed FEMA officials and flood zone determination industry officials and obtained information and documentation on studies about levels of participation in the NFIP and lender compliance issues. We attended meetings and training sessions held by FEMA with insurance officials, local government officials, and lender representatives and observed discussions of noncompliance. In determining what other readily available data indicates about the level of noncompliance, we found that the data necessary to assess lender noncompliance are currently not reported in a way to permit full evaluation of this issue. However, we did determine a methodology that would enable us to obtain a perspective on noncompliance at loan origination. We compared the number of new mortgages made with the number of flood insurance policies issued in the same locations, in certain of the nation’s highly flood-prone areas. This analysis required that we (1) identify flood-prone areas; (2) determine the number of new mortgages made in such areas that were subject to NFIP regulations; (3) determine the number of flood insurance policies written; and (4) compare the number of mortgages with the number of flood insurance policies in certain areas to infer levels of lender noncompliance in those selected areas. Data on properties with existing mortgages in these flood-prone areas were not available; therefore, we did not perform any analysis on the level of noncompliance on existing mortgages. Further, we did not perform any analysis of the accuracy of the determinations made by flood zone determination companies and used by lenders as the basis for whether flood insurance is required. For our review, we defined as “flood-prone areas” all census tracts in which 90 percent or more of the tract is in a flood zone. To identify these census tracts, we used the percentage of properties determined to be flood-prone as a proxy for the percentage of each census tract area that may be flood-prone. We obtained data on flood zone determinations from Transamerica Flood Hazard Certification, Inc., which has been collecting information on properties in the United States since 1977. Its database consisted of about 62 million properties nationwide for which flood determinations have been made. Of those properties, 2.8 million, or 4.6 percent, have been certified as located in flood hazard areas. We obtained this information by state, county, and census tract. Transamerica’s data covered properties in 59,506 census tract areas. About 1 percent (742) of these census tracts had at least 90 percent of the properties determined to be in SFHAs. We focused on these 742 census tract areas. To identify the number of new mortgage loans that would be covered by NFIP regulations in the flood-prone census areas, we used Home Mortgage Disclosure Act (HMDA) data that regulators collected for the 1999 calendar year. The HMDA information shows the number of loans granted during 1999. We refined the number of mortgages to be included in our analysis by choosing only owner occupancy and single-family loans subject to government regulation. We excluded business loans, unregulated loans not purchased in the secondary market by either Fannie Mae or Freddie Mac, and loans that were designated for home improvements, multifamily dwellings, and refinances. This resulted in a total of 3,717,735 mortgage loans in 1999. We then aggregated all loan originations at the census-tract level. To identify the number of flood insurance policies written for the flood- prone census-tract areas, we used NFIP data that FEMA collects from insurance companies that issue policies. We obtained from FEMA a database that aggregated the number of flood insurance policies written in 1999 according to the state, county, and census tract of the property. To obtain policy data that would be comparable to mortgage data, we refined the policy data to include only new policy transactions for principal residences. We excluded policies that were not for a homeowner’s principal residence or were only to cover the contents of a property. This resulted in a total of 549,255 policies for 1999. These data were aggregated at the census-tract level. To determine whether the number of flood insurance policies in force approximated the number of regulated mortgage loans made in flood- prone areas, we merged the three types of data described above by census tract within each state and county for 1999. We assumed that the distribution of the properties, loans originated, and insurance policies written were the same within each census tract area. As previously indicated, we focused on those 742 census tract areas that had at least 90 percent of their properties in flood areas. To provide a greater degree of confidence that the data we obtained were representative of the entire census tract, we developed additional selection criteria whereby at least 20 loans had been made in that census tract and at least 100 properties within the census tract had flood-zone determinations. As a result, the number of census tracts we examined totaled 471 covering 17 states. For these areas, we found that 46,965 loans had been made and 88,300 flood insurance policies had been issued. We did not attempt to independently verify the accuracy of the data sets used for our analysis. We did, however, to the extent possible, assess the reliability by (1) performing electronic tests (as described below) and (2) discussing results of the testing and analysis with knowledgeable individuals. We cross-checked Transamerica’s determination data with data from another flood zone determination company (Geotrac of America, Inc.) to determine if both companies’ data identified the same flood-prone census tracts. We found a high degree of correlation between the data of the two flood zone determination companies. We also determined that the variables we used from the HMDA data was complete in its coding and did not have questionable outliers. Therefore, we determined that the data were reliable enough for the purposes of this report. To determine what data would be needed to fully measure compliance, we analyzed the processes established by the participants in the NFIP to collect, report, and share data on lender compliance with flood insurance purchase requirements. We interviewed officials and gathered documents from the Federal Insurance and Mitigation Administration, federal regulators, GSEs, loan servicers, insurance companies and their servicers, and related industry associations, including the National Floodzone Determination Association, the Independent Bankers Association of America, and the Mortgage Bankers Association. We made site visits to flood zone determination companies in Austin and Arlington, Tex.; Lakewood, Colo.; and Hasbrouck Heights, N.J.; and interviewed officials with a flood zone determination company in Norwalk, Ohio. These companies represent about 80 percent of the flood zone determination market. After determining how information pertaining to compliance is processed, we developed a process that would allow better measurement of noncompliance with the mandatory flood insurance purchase requirements. We then discussed this process with FEMA and bank regulatory officials to obtain their perspective on its viability, costs, and benefits. We conducted our review from April 2001 through April 2002 in accordance with generally accepted government auditing standards. Federal financial institution regulators have primary responsibility for ensuring that the institutions they supervise comply with the requirements of the National Flood Insurance Program. The regulators have issued uniform flood insurance regulations and examination procedures for enforcing and monitoring lender compliance. The policies and procedures are designed to ensure that flood zone determinations for mortgaged properties are performed, flood insurance is obtained when required, and flood insurance policies remain in force for the life of the loan. The National Flood Insurance Reform Act of 1994 directed that regulatory agencies promulgate rules to implement the act’s provisions and to coordinate their development through the Federal Financial Institutions Examination Council. The agencies’ regulations became effective on October 1, 1996, and established, among other provisions, new requirements for escrowing flood insurance premiums, documenting flood hazard determinations on the Standard Flood Hazard Determination Form, and “force-placing” flood insurance under certain circumstances. In November 1996, the regulators adopted uniform procedures for assessing lender compliance with the new flood insurance regulations. These procedures require that for the flood insurance component of the examinations the regulators assess whether an institution performs required flood determinations for home mortgage loans, including mobile homes affixed to a permanent foundation; if the institution requires flood insurance in the correct amount when it makes, increases, extends, or renews a covered loan; if the institution provides the required notices to the borrower whenever flood insurance is required as a condition of the loan; if the institution requires flood insurance premiums to be escrowed when other items, such as hazard insurance and taxes, are required to be escrowed; and if the institution complies with the forced placement provisions in cases where flood insurance on the loan is not sufficient to meet the requirements of the regulation. To fulfill these requirements, bank examiners review a sample of loan files to verify that flood insurance requirements are met. For smaller banks and thrifts, which make few mortgage loans, the sample may consist of all loans made since the last examination, and all loans in the portfolio known to be secured by properties in special flood hazard areas. For larger institutions, examiners review a nonprojectable sample of loans. Depending on the findings from those files, an examiner may analyze additional loans for further examination. If a lender appears to have failed to require adequate flood insurance coverage on selected loans, it may be required to conduct a review of its entire loan portfolio and report the results to the bank regulator. If violations of the flood insurance requirements are detected during an examination, corrective action may be required of the lender, and fines can be levied against the lender by the regulatory organization if it finds a pattern or practice of violations. Bank regulators perform compliance examinations on a periodic basis, generally every 12 to 60 months. In addition to compliance with flood insurance requirements, these examinations cover compliance with other consumer laws and regulations. The length of time between examinations is determined by several factors, including the bank’s rating at the time of its last examination and the size of the institution. In addition to regular examinations, examiners are to follow-up with institutions in which violations have been found to verify that any violations noted during the most recent examination have been resolved. The 1994 National Flood Insurance Reform Act directed Fannie Mae and Freddie Mac to implement procedures designed to ensure that loans that they purchase are covered by flood insurance for the term of the loans. While GSEs have no regulatory authority over their sellers and servicers, they require their sellers and servicers to comply with the flood insurance requirements through their contracts with them. These requirements are spelled out in the GSEs’ Seller/Servicers Guides. Fannie Mae and Freddie Mac also require that servicers have processes in place that allow the servicer to identify map changes, determine which mortgaged dwellings affected by map changes need flood insurance, and to ensure that the affected borrowers obtain such insurance within 120 days of the effective date of the map change. If Fannie Mae or Freddie Mac finds that a lender is not complying with their requirements for flood insurance, they may require that the lender repurchase the loans and correct deficiencies in their system for ensuring compliance. Officials from both enterprises told us that this occurs very rarely. The act also directed the Office of Federal Housing Enterprise Oversight (OFHEO), an independent agency within the Department of Housing and Urban Development responsible for regulation of Fannie Mae and Freddie Mac, to assess whether they have adopted and are adhering to flood insurance compliance procedures, and to report on this assessment in OFHEO’s annual reports to Congress for 1996, 1998, and 2000. OFHEO reported that the policies and procedures established by Fannie Mae and Freddie Mac with respect to the flood insurance requirements under the Flood Disaster Protection Act were adequate and were being used. The review procedures for flood insurance established by Fannie Mae and Freddie Mac are explained below. Post purchase review: Flood insurance compliance is incorporated as part of the monthly quality control reviews of a nonprojectable sample of recently purchased or securitized mortgages. Lists of property addresses are sent to two flood zone determination companies to review the flood zone determinations on file for properties both in and outside of flood zones. The mortgage file is checked to verify that a copy of the special flood hazard determination form is present; the loan is coded properly; and, if appropriate, evidence that flood insurance coverage was obtained. Portfolio review: On an annual basis, Fannie Mae performs a review on a sample of all loans it owns or has securitized to verify that sellers and servicers appropriately obtained and have maintained flood insurance, as applicable, throughout the term of the mortgage. The scope of the review emphasizes areas where flood zone remapping has occurred, or communities whose participation status in the National Flood Insurance Program has changed, to ensure that sellers and servicers are in compliance with the requirement to have procedures in place to monitor such changes. For a nonprojectable sample of mortgages in special flood hazard areas, sellers and servicers are required to provide documentation to confirm that flood insurance is in force for each selected mortgage, and that the mortgages were properly identified at delivery. Quality control operational review: Fannie Mae regional offices perform regular quality control reviews that examine sellers and servicers’ management, policies, and procedures, rather than examining individual mortgage files. These reviews include on-site examination of processes for ensuring the accuracy of the flood zone determinations that are obtained. Generally, the largest sellers and servicers are evaluated every year; others are reviewed every 2 to 3 years. In addition, Fannie Mae is looking at various options to improve its methodology for performing its flood insurance reviews. One such option is the use of Geographic Information Systems data and flood maps to target loans in its portfolio for flood reviews. This effort is currently in the testing stages. Quality control program: The data file for each mortgage purchased by Freddie Mac must contain a special characteristic code describing the mortgage’s status regarding flood insurance, as follows: in a flood zone with insurance coverage in place; in a flood zone with no flood insurance coverage; not in a flood zone with flood insurance coverage; or not in a flood zone and no flood insurance coverage. As part of Freddie Mac’s quality control program, a statistical sample of newly delivered mortgages is reviewed for the correct special characteristic codes regarding flood insurance; proper documentation of the flood zone determination; and, if applicable, the flood insurance policy. Flood audit program: Freddie Mac auditors provide a list of addresses for all of a servicer’s mortgages to a flood zone determination company. The flood zone determination company reviews the addresses in the portfolio and arrives at a list of 25 properties located entirely within special flood hazard areas. The list is then sent to several other participating flood zone determination companies to verify that the identified properties are within SFHAs. Each company performs independent flood zone determinations for the listed properties; Freddie Mac eliminates any properties for which the “in” determination is not unanimous. At the sellers and servicers’ facilities during the audit, Freddie Mac audits the mortgage files for the selected properties to verify that all of the flood insurance requirements are met. This includes ensuring that flood insurance is in effect and that the coverage meets Freddie Mac’s requirements. Underwriting reviews: Freddie Mac reviews sellers and servicers’ management controls for identifying properties in SFHAs, ensuring that flood insurance is maintained, and ensuring that flood insurance coverage is at least equivalent to that provided under the NFIP. Servicing review: Freddie Mac auditors review the management controls that sellers and servicers have in place to (1) become aware of changes in SFHAs, (2) ensure that the borrower obtains flood insurance coverage if the sellers and servicers become aware that existing coverage does not adequately protect the mortgaged premises, (3) ensure that the borrower obtains the required insurance, and (4) ensure that the sellers and servicers obtain the required coverage if the borrower does not obtain it. In addition to those named above, Martha Chow, Lawrence D. Cluff, Colin J. Fallon, Kerry D. Hawranek, DuEwa A. Kamara, Signora J. May, John T. McGrail, Lisa M. Moore, Patricia D. Moore, Bob Procaccini, and John J. Strauss made key contributions to this report. | The Federal Emergency Management Agency (FEMA) run National Flood Insurance Program (NFIP) has combined flood hazard mitigation efforts and insurance to protect homeowners against losses from floods. The program provides an incentive for communities to adopt floodplain management ordinances to mitigate the effect of flooding upon new or existing structures. Virtually all communities in the country with flood-prone areas now participate in the NFIP, and over four million U.S. households have flood insurance. Nevertheless, the President's proposed budget for 2003 characterizes the NFIP as "moderately effective," because many at-risk properties remain uninsured. The proposed budget establishes a goal to increase flood insurance policies in force by five percent in 2003 and would increase funding for flood zone mapping activities to better identify at-risk properties. Although the assessment and goal described in the proposed budget apply to the entire NFIP, the success of a particular component of the program--the mandatory purchase requirement--has been the subject of debate for many years. The federal bank regulators overseeing lending institutions that hold or service mortgages on properties that must have flood insurance believe that there is a high level of compliance. However, others have questioned whether the requirements are being met. The different types of evidence collected by bank regulators and government-sponsored enterprises on the one hand, and FEMA on the other, are the bases for their opposing perspectives on lender noncompliance with flood insurance purchase requirements. Federal organizations overseeing lenders use bank examinations and loan portfolio reviews to examine a nonstatistical sample of loans for compliance. These organizations uncovered few significant violations, leading them to believe that lenders are complying with flood insurance purchase requirements. In contrast, FEMA relies on its own noncompliance estimates from data it generates itself, from other entities, limited studies it conducted, and anecdotal evidence from public officials and others with knowledge of the program to gauge noncompliance. These data indicate that lenders are not adequately complying with the requirements. GAO's analysis of readily available data does not suggest a major noncompliance problem at loan origination in highly flood-prone areas. Property-specific data on mortgages, flood zone determinations, and flood insurance policies--compiled at loan origination and at various points during the life of the loan--would be needed to fully measure compliance. These data are needed to ensure that homeowners purchase, maintain, and do not terminate required flood insurance. |
Energy commodities are bought and sold in several different physical and financial markets. Physical markets include the spot, or cash, markets where products such as crude oil or gasoline are bought and sold for immediate or near-term delivery. The United States has several spot markets. Examples are the pipeline hub near Cushing, Oklahoma for West Texas Intermediate crude oil and the Henry Hub near Erath, Louisiana, for natural gas. The prices set in the specific spot markets provide a reference point that buyers and sellers use to set the price for other types of the commodity traded in other locations. The prices established for energy commodities in the physical markets generally are determined by supply and demand. For example, when the demand for the product rises relative to supply because economies are growing, prices are likely to rise. Conversely, when demand falls relative to supply, prices are likely to fall. For energy products, demand and supply, and therefore price, can fluctuate on a seasonal basis. For example, consumer demand for gasoline in the United States is generally higher from May through early September—the summer driving season— and tends to flatten after Labor Day. Similarly, demand for natural gas and heating oil is highest during the heating season between October and March. The relative inelasticity of energy commodities means that small shifts in demand and supply can result in relatively large price fluctuations. In general, when the price of an energy commodity rises, the demand for that product is likely to fall in the long term, and vice versa. However, demand for energy commodities is price inelastic in the short term—that is, the quantity demanded changes little in response to a change in price. On the supply side, rising energy commodities prices motivate producers to increase the amount of commodities they supply to increase profits. However, because producers hold relatively low inventories of energy commodities in reserve, and finding and producing additional energy commodities takes a long time and is expensive, supply also is relatively inelastic. For example, supplies of natural gas from new production wells cannot be increased quickly to meet higher demand because of the time required to get the newly produced gas into the marketplace. Energy commodities also are traded in the financial markets, especially in the form of derivatives. Derivatives include futures, options, and swaps, whose values are based on the performance of the underlying asset. Options give the purchaser the right, but not the obligation, to buy or sell a specific quantity of a commodity or financial asset at a designated price. Swaps traditionally are privately negotiated contracts that involve an ongoing exchange of one or more assets, liabilities, or payments for a specified period. Futures and options contracts are traded on exchanges designated by CFTC as contract markets (futures exchanges), where a wide range of energy, agricultural, financial, and other commodities are bought and sold for future delivery. Commodity futures and options can be traded on both OTC and exempt commercial markets if the transactions involve qualifying commodities and the participants satisfy statutory requirements. Energy futures include standardized contracts for future delivery of a specific crude oil, heating oil, natural gas, or gasoline product at a particular spot market location. The exchange standardizes the contracts, and participants cannot modify them to their particular needs. For example, a standard gasoline futures contract traded on NYMEX is for 1,000 barrels (42,000 gallons), quoted in dollars and cents per gallon, and for delivery of up to 36 months into the future at New York Harbor. The owner of an energy futures contract is obligated to buy or sell the commodity at a specified price and future date. However, the owner may eliminate the contractual obligation before the contract expires by selling or purchasing other contracts with terms that offset the original contract. In practice, relatively few futures contracts on NYMEX result in physical delivery of the underlying commodity, but instead are liquidated with offsets. Options on futures contracts also are traded on exchanges such as NYMEX and foreign boards of trade that U.S. traders access directly. In addition to exchange-traded futures and options, the financial markets for energy commodities include derivatives traded among multiple traders on exempt commercial markets and derivatives created bilaterally in OTC transactions. As with futures, exempt commercial markets and other OTC derivatives allow producers and users of energy commodities to manage the risk of future changes in the price of a particular commodity. These contracts include options and swaps at an agreed-upon price. Appendix II shows some of the different types of contracts and transactions for energy commodities in the physical and financial markets. Market participants use futures markets to offset the risk caused by changes in prices, discover commodity prices, and speculate on price changes. Some buyers and sellers of energy commodities in the physical markets trade in futures contracts to offset, or “hedge,” the risks of price changes in the physical markets. The futures markets help buyers and sellers determine, or “discover,” the price of commodities in the physical markets, thus linking the two markets. Other participants—generally, speculators—that do not have a commercial interest in the underlying commodities but are looking to make a profit take varying positions on the future value of commodities. In doing so, speculators provide liquidity and assume risks that other participants, such as hedgers, seek to avoid. Arbitrageurs are a third group of participants that aim to benefit by identifying discrepancies in price relationships, rather than by betting on future price movements. Arbitrage is a strategy that involves simultaneously entering into several transactions in multiple markets to benefit from price discrepancies across markets. For example, traders can trade simultaneously in exchanges and OTC. Price risk is an important concern for buyers and sellers of energy commodities because wide fluctuations in cash market prices introduce uncertainty for producers, distributors, and consumers of commodities and make investment planning, budgeting, and forecasting more difficult. A statistical measurement of the degree to which prices fluctuate over time is known as “volatility” and can be applied to prices in both the physical and financial markets. There are two basic types of volatility measurements. Historical volatility measures are calculated on the basis of price changes, using data from market transactions. Implied volatility reflects market participants’ expectations of future volatility as derived from the prices of traded options (see app. III). This report presents data on the relative historical volatility of energy futures contracts, which we calculated from relative changes in daily prices. Futures and off-exchange derivatives markets provide participants with a means to hedge or shift unwanted price risk to others more willing to assume the risk or those having different risk situations. For example, if a petroleum refiner wanted to shed its risk of losing money as a result of falling gasoline prices, it could lock in a price by selling futures contracts to deliver the gasoline in 6 months at a guaranteed price. Likewise, a transportation company that knows it must refill its gasoline tanks in 6 months might want to offset the price risk associated with purchasing fuel by buying futures contracts to take delivery of gasoline then at a set price. Without futures contracts that help them manage risk, producers, refiners, and others likely would face uncertainty related to investment planning, budgeting, and forecasting—and potentially higher costs. Futures markets also provide a means of price discovery for commodities such as energy products. For price discovery, markets need current information about supply and demand, a large number of participants, and transparency. Market participants monitor and analyze the factors that currently affect, and that they expect to affect, the future supply and demand for energy commodities. With that information, they buy or sell energy commodity contracts on the basis of the price for which they believe the commodity will sell at the delivery date. The futures markets, in effect, distill the diverse views of market participants into a single price. In turn, buyers and sellers of physical commodities consider those predictions about future prices with other factors when setting prices on the spot and retail markets. A wide variety of participants hedge and speculate in energy derivatives markets. For the exchange-traded futures markets, CFTC categorizes traders in general terms as either commercial or noncommercial participants. CFTC identifies several subcategories of participants within the commercial category: producers, manufacturers, dealers/merchants, and swaps/derivatives dealers. Dealers and merchants include, among others, wholesalers, exporters and importers, shippers, and crude oil marketers. Typical noncommercial traders are entities such as those that manage money (“managed money traders”). These noncommercial traders include, among others, commodity pool operators (CPO) and commodity trading advisors (CTA), many of which advise or operate hedge funds. Other noncommercial traders include floor brokers and unregistered traders. The prices for energy commodities in the futures and in the spot or physical markets are closely linked because they are influenced by the same market fundamentals in the long run. Prices in the physical spot and futures markets for the four energy commodities we reviewed are highly correlated and rose dramatically from 2002 to 2006. As shown in figure 1, from January 2002 to July 2006, monthly average spot prices for crude oil, gasoline, and heating oil increased by at least 220 percent. Natural gas spot prices increased by more than 140 percent. At the same time that spot prices increased, the futures prices for these commodities showed a similar pattern of a sharp and sustained increase from January 2002 into 2006. For example, the price of crude oil futures increased from an average of $22 per barrel in January 2002 to an average of $74 per barrel in July 2006. Natural gas futures prices spiked rapidly in the fall of 2005 after several strong hurricanes raised concerns about supply disruptions for the winter of 2005-2006, then prices fell sharply due in part to a mild winter. Prices in the spot and futures markets show similar patterns because traders in those markets tend to rely on the same types of information when entering into transactions. The differences between futures and spot market prices for energy commodities narrow and the prices converge when futures contracts near expiration and physical delivery is required. As the expiration date nears, the physical delivery provision of the contract and the ability of traders to arbitrage combine to bring the futures and physical market prices together. Arbitrage plays a crucial role in moderating or removing price differences between spot and futures markets and contributes to the convergence of futures and spot prices at expiration. For example, if the price for a crude oil futures contract that would expire in 2 weeks were $62 per barrel and the spot market price were $60 per barrel, a trader could choose to buy oil now at the spot price and enter into a futures contract to deliver oil in 2 weeks at the futures price, thereby making a $2 profit. This and similar transactions by other traders would put upward pressure on the spot price and downward pressure on the futures price and move them toward convergence. Figure 2 provides an example of how the price of the April 2006 crude oil futures contract and the spot price for that commodity converged as the contract approached expiration. Between the creation of CFTC in 1974 and the year 2000, the CEA generally restricted commodity derivatives trading to futures and options entered into on exchanges and made all transactions in futures contracts subject to CFTC’s exclusive jurisdiction. However, in the late 1980s and early 1990s, commercial entities began entering into nonstandardized, off- exchange derivative contracts that had pricing characteristics similar to futures (i.e., pricing of the transactions derived from the prices of various commodities), and the instruments were used for risk shifting. According to CFTC officials, under exemptive authority provided in 1992 reauthorization legislation, CFTC announced that it would not take enforcement action against qualified commercial entities engaged in certain types of energy derivatives transactions, but the legality of instruments not covered by the exemption (i.e., their status as futures contracts subject to the CEA) remained unresolved. In 2000, the CFMA amended the CEA to provide for both regulated markets and markets largely exempt from regulation and to permit off- exchange trading of energy derivatives by qualified parties. The regulated markets include futures exchanges that have self-regulatory surveillance and monitoring responsibilities as self-regulatory organizations (SRO) and by CFTC. CFTC’s primary mission includes preserving the integrity of these futures markets and protecting market users and the public from fraud, manipulation, and abusive practices related to the sale of commodity futures and options. This mission is achieved through a regulatory scheme that is based on federal oversight of industry self- regulation. The CEA also permits derivatives trading in markets that are largely exempt from CFTC’s regulatory authority, including both OTC and exempt commercial markets, subject to statutory requirements governing the types of commodity and trader and the facility used for conducting the trades. The President’s Working Group’s 1999 report on OTC derivatives focused on changes to the CEA that in their view would “promote innovation, competition, efficiency, and transparency in OTC derivatives markets, to reduce systemic risk, and to allow the United States to maintain leadership in these rapidly developing markets.” Derivatives on energy commodities, which are within the act’s definition of “exempt commodity,” may be traded in exempt commercial markets by eligible commercial entities, a category of traders broadly defined in the CEA to include firms with a commercial interest in the underlying commodity as well as other sophisticated investors, such as hedge funds. Violations of the CEA and CFTC regulations may be remedied by imposition of civil monetary penalties, trading bans, restitution, and other appropriate relief. In addition to CFTC oversight, futures exchanges accept self-regulatory obligations as a condition of designation. For example, NYMEX, as an SRO, is responsible for establishing and enforcing rules governing member conduct and trading; providing for the prevention of market manipulation, including monitoring trading activity; ensuring that futures industry professionals meet qualifications; and examining exchange members for financial soundness and other regulatory purposes. CFTC oversees SROs to ensure that each has an effective self-regulatory program. Within CFTC, three of the commission’s six major operating units actively oversee futures exchanges and their derivatives clearing organizations. The Division of Market Oversight approves and oversees the futures exchanges, conducts its own market surveillance, conducts trade practice reviews and investigations, and reviews exchange rules. The Division of Clearing and Intermediary Oversight oversees, among other things, derivatives clearing organizations and the registration of intermediaries, which are persons such as futures commission merchants, CPOs, or CTAs that act on the behalf of others in futures trading. The Division of Enforcement investigates and prosecutes alleged violations of the CEA and CFTC regulations. At the beginning of fiscal year 2006, 167 (34 percent) of CFTC’s 490 full- time-equivalent (FTE) positions were allocated to the first two CFTC divisions; at the beginning of fiscal year 2007, that allocation declined to 162 (35 percent) of CFTC’s 458 FTE positions. These staff monitor the markets and market participants from CFTC’s headquarters in Washington, D.C., as well as from field offices in New York; Chicago; Kansas City; and, until recently, Minneapolis. About one-third of CFTC’s staff are located in the field offices. At the beginning of fiscal year 2006, 132 (27 percent) of CFTC’s 490 FTE positions were allocated to the Division of Enforcement; at the beginning of fiscal year 2007, that number declined to 120 of CFTC’s 458 FTE positions. The 2007 data are estimated. While CFTC staffing levels have declined, according to CFTC, futures and options trading volume for all commodities has roughly doubled from fiscal years 2002 to 2006 and is expected to continue to rise, as indicated in figure 3. Both physical and futures markets experienced a substantial amount of change from 2002 through 2006. Reasonable arguments have been made that events in both markets have contributed to rising energy prices, at least in the short term, but opinions vary regarding the extent that recent changes in the financial markets have influenced the prices of energy products in the physical markets over the long term. Because of these concurrent changes, identifying the causes of the increases in energy prices in both the physical and futures markets for crude oil, unleaded gasoline, heating oil, and natural gas is difficult. First, during this period, the physical markets experienced tight supply and rising demand from increasing global demand, ongoing political instability in oil-producing regions, and other supply disruptions. Second, annual volatility of energy prices remained above historic averages during the beginning of the period (although during 2006, volatility generally declined to levels at or near the historical average). Third, the volume of trading in energy futures increased as growing numbers of managed money traders viewed energy futures as attractive investment alternatives. The energy physical markets have undergone substantial change and turmoil from 2002 through 2006, which affected prices in the spot and futures markets. First, like many market observers and participants, we found a number of fundamental supply and demand conditions that could influence prices. Moreover, these parties have observed that the lack of spare capacity in certain areas, such as production, transportation, and storage, can affect prices. Second, over the short term, weather events also were a significant cause of rising energy prices because of their effects on energy supply, according to several of the market observers we interviewed. Third, many market observers also identified geopolitical uncertainty arising from the instability and insecurity of the world’s major oil-producing regions as a major factor affecting energy prices. Concerns about political events may manifest in the form of higher futures prices if traders predict that an event—such as a strike within the industry or pipeline sabotage by terrorists—will have an effect on future supply. Finally, on the demand side, a significant factor noted by observers was the increase in global consumption of petroleum products, primarily among industrializing Asian nations such as China and India. Analysis of world oil prices by EIA and us indicates that increases in crude oil prices occur if political instability, terrorist acts, or natural disasters create uncertainties about, or actual disruptions in, supply from countries that produce or refine oil. For example, according to EIA, in the early 2000s, cutbacks in the Organization of the Petroleum Exporting Countries (OPEC) production and rising demand caused oil prices to increase to more than $30 per barrel, only to fall precipitously when the global economy weakened following the September 11, 2001, crisis. Moreover, as we reported in 2005, rapid growth in oil demand in Asia contributed to a rise in crude oil prices to more than $50 per barrel during 2004. According to EIA, world oil demand that was about 59 million barrels per day in 1983 grew to more than 85 million barrels per day in 2006. The United States consumes nearly one-quarter of this amount—or more than 20 million barrels per day in 2006—and its demand has grown about 1.5 percent per year since 1983. The rapid economic growth in Asia also has stimulated a strong demand for energy commodities. For example, China has overtaken Japan as the second-largest consumer of crude oil, after the United States. According to EIA data, from 1983 to 2004, Chinese demand grew from about 1.7 million barrels consumed per day to about 6.4 million barrels consumed per day. This increase in the global demand for crude oil is shown in figure 4. The growth in demand does not, by itself, lead to higher prices for crude oil or any other energy commodity. For example, if the growth in demand were exceeded by a growth in supply, prices would fall, with other things remaining constant. However, according to EIA, the growth in demand outpaced the growth in supply, even with spare production capacity included in supply. Spare production capacity is surplus oil that can be produced and brought to the market relatively quickly to rebalance the market if there were a supply disruption anywhere in the world oil market. EIA estimates that global spare production capacity in 2006 was about 1.3 million barrels per day (see fig. 5). Most of that capacity was concentrated in the 12 OPEC countries that supply about 40 percent of the world’s oil, primarily Saudi Arabia. This compared with spare capacity of about 10 million barrels per day in the mid-1980s, or of about 5.6 million barrels a day as recently as 2002. Analysis by EIA indicates that the growth of oil production in non-OPEC nations, which produce most of the world’s oil and include countries such as Canada, China, Mexico, Norway, Russia, the United Kingdom, and the United States, has slowed relative to the growth in demand, and these nations have virtually no spare production capacity. As a commodity that is produced and traded worldwide, crude oil prices could be affected by the value of the U.S. dollar on open currency markets. For example, because crude oil is typically denominated in U.S. dollars, the payments that oil-producing countries receive for their oil also are denominated in U.S. dollars. As a result, a weak U.S. dollar decreases the value of the oil sold at a given price, and oil-producing countries may wish to increase prices for their crude oil to maintain purchasing power in the face of a weakening U.S. dollar, to the extent they can. Major weather and political events also can lead to supply disruptions and higher prices. In its analysis, EIA has cited the following examples: Hurricanes Katrina and Rita removed about 450,000 barrels per day from the world oil market from June 2005 to June 2006. Instability in major OPEC oil-producing countries, such as Iran, Iraq, Nigeria, and Venezuela, has lowered production and increased the risk of future production shortfalls. Oil production in Russia, a major driver of non-OPEC supply growth during the early 2000s, was adversely affected by a worsened investment climate as the government raised export and extraction taxes. The supply of crude oil affects the supply of gasoline and heating oil, and, just as production capacity affects the supply of crude oil, refining capacity affects the supply of products distilled from crude oil. As we have reported, refining capacity in the United States has not expanded at the same pace as the demand for gasoline. Despite a growth in the capacity of existing gasoline refineries, the growth in demand has meant that refineries have been running at an average of more than 93 percent of production capacity since the mid-1990s, compared with about 78 percent in the 1980s. Higher utilization rates can increase operating costs and lead to prices being higher than otherwise would be expected, as occurred in the second half of the 1990s. Another factor affecting the supply, and therefore the price, of petroleum products is the amount held in inventory. Inventory is particularly crucial to the supply and demand balance because it can provide a cushion against price spikes if, for example, a refinery outage temporarily disrupts production. We have reported that, as in other industries, the petroleum products industry has adopted “just-in-time” delivery processes to reduce costs, leading to a downward trend in the level of gasoline inventories in the United States. For example, in the early 1980s, private companies held stocks of gasoline in excess of 35 days of average U.S. consumption; while in 2004, those stocks were equivalent to less than 25 days consumption. Lower costs of holding inventories may reduce gasoline prices, but lower levels of inventories also may cause prices to be more volatile because when a supply disruption occurs or there is an increase in demand, there are fewer stocks of readily available gasoline from which to draw, thereby putting upward pressure on prices. Others have noted that higher prices for future delivery of oil have induced oil companies to buy more oil and place it in storage. They concluded that this practice has created a situation where oil prices are high despite high levels of oil in inventory. In addition to the supply and demand factors that generally apply to all energy commodities, there are specific conditions that apply to particular commodities. For example, to meet national air quality standards under the Clean Air Act, as amended, many states have mandated the use of special gasoline blends—so-called “boutique fuels.” As we have recently reported, there is a general consensus that higher costs associated with supplying special gasoline blends contributed to higher gasoline prices, either because of more frequent or more severe supply disruptions or because higher costs are likely passed on, at least in part, to consumers. As another example, according to EIA, the recent phaseout of a chemical used to improve gasoline performance—methyl tertiary butyl ether— increased the price of U.S. gasoline, in part because the chemical was replaced by ethanol, a more costly additive. As in the futures markets, the physical markets have undergone substantial changes that can affect prices. These specific factors affecting particular commodities, when combined with the general supply and demand conditions, contribute to increased energy prices and price volatility. However, market participants and other observers disagree on whether high energy prices were solely due to supply and demand fundamentals or whether increased futures trading activity also was fueling higher prices. The changes occurring in the physical markets have not happened in isolation; they have been accompanied by advances in technology, relatively high but falling volatility in energy futures prices, and a growing volume of trading in the derivatives markets. The effects of these changes on energy prices are not clear. Although energy futures prices increased from 2002 to 2006 (see fig. 1), the relative volatility of those prices for three of the four commodities generally declined. As shown in figure 6, the annual historical volatilities— measured using the relative change in daily prices of energy futures—from 2000 through 2006 generally were above or near their long-term averages, although crude oil and heating oil declined below the average and gasoline declined slightly. As we have reported, futures prices typically reflect the effects of such world events on the price of crude oil. Political instability and terrorist acts in countries that supply oil create uncertainties about future supplies, which is reflected in futures prices in anticipation of an oil shortage and expected higher prices in the future. Conversely, news about a new oil discovery that would increase world oil supply could result in lower futures prices. In other words, futures traders’ expectations of what may happen to world oil supply and demand influence their price decisions. The annual volatility of natural gas fluctuated more widely than that of the other three commodities and increased in 2006, even though prices largely declined from the levels reached in 2005. EIA has stated that the volatility of natural gas prices is due to factors in the physical marketplace, such as changing weather, producers’ inability to move natural gas quickly to areas in response to quickly rising demand, and limited local storage. A research director for a consumer advocacy organization who studied natural gas prices concluded that increased trading by speculators had increased volatility and prices. CFTC also has studied this issue and found that natural gas prices from August 2003 through August 2004 did not appear to be determined by any single category of market participant, although joint demand and supply of contracts by all participants clearly affected the change in price. In other words, managed money traders’ activity (including hedge funds), by itself, did not have a significant effect on price changes. While some often equate higher prices with higher volatility, an increase in futures contract prices does not necessarily mean that volatility will increase in a similar manner, and an increase in volatility does not necessarily mean that prices will rise. Price volatility measures the variability rather than the direction of price changes and is based on the standard deviation of those changes. Therefore, if futures contract prices change at a steady rate, the prices may have lower volatility than if large swings in prices occurred. At the same time that prices were rising and volatility was generally above or near long-term averages, futures markets also experienced an increase in the number of large noncommercial participants, such as managed money traders. The trends in price and volatility made the energy derivatives markets attractive for an increasing number of traders looking to either hedge against those changes or profit from them. According to CFTC large trader data, from July 2003 to December 2006, crude oil futures and options contracts experienced the most dramatic increase as the average number of noncommercial traders grew from about 125 to about 286. As shown in figure 7, over a similar period, the average number of noncommercial traders also showed an upward but less dramatic trend for unleaded gasoline, heating oil, and natural gas. Some market participants and observers have concluded that large purchases of oil futures contracts by speculators in effect have created an additional demand for oil that has led to higher prices; others disagree. The Senate’s Permanent Subcommittee on Investigations, Committee on Homeland Security and Governmental Affairs, issued a staff report in June 2006 that concluded that the traditional forces of supply and demand could not fully account for increases in the prices of energy commodities. Also, according to an energy firm, an investment bank, an academic, and hedge fund officials, increasing numbers of speculative traders in the market and rising trading volume placed upward pressure on futures prices. However, others, including investment bank and CFTC officials, have argued that speculators did not increase prices, but they provided liquidity and dampened volatility. Moreover, other investment banks, energy firms, and FERC officials told us that speculative trading in the futures markets can contribute to short-term price movements in the physical markets. However, they did not believe it was possible to sustain a speculative “bubble” over time because the two markets are linked and both respond to information regarding changes in supply and demand caused by such factors as the weather or geopolitical events. Therefore, in their view, speculation could not lead to artificially high or low prices over a long period. Within the noncommercial trader category, the largest increases came from managed money traders—which generally trade for their own accounts rather than for others. Specifically, for crude oil, the average number of managed money traders that trade daily increased significantly from about 62 in July 2003 to about 128 in December 2006. At the same time, the number of smaller traders also grew significantly from an average of about 26 per day in July 2003 to an average of about 111 per day in December 2006. The number of managed money traders and smaller traders for unleaded gasoline, heating oil, and natural gas also increased similarly during that period. The number of commercial futures traders generally did not increase in a fashion similar to that of noncommercial traders. As the number of traders has increased, so has the trading volume on NYMEX for all energy futures contracts, particularly crude oil and natural gas, as shown in figure 8. From 2001 through 2006, the average daily contract volume for crude oil increased by 90 percent and for natural gas increased by 93 percent. However, unleaded gasoline and heating oil experienced less dramatic growth in their trading volumes during this period. Along with the strong growth of energy futures trading, the amount of energy derivatives traded outside of exchanges also appears to have increased significantly. However, comprehensive data on the trading volume of energy-related OTC derivatives are not available because OTC energy markets are not regulated. The Bank for International Settlements publishes data on worldwide OTC derivative trading volume for broader groupings of commodities that can be used as a rough proxy for trends in the trading volume of OTC energy derivatives. According to these data, the notional amounts outstanding of OTC commodity derivatives— excluding precious metals, such as gold—grew by 854 percent from December 2001 through December 2005. From December 2004 through December 2005, the notional amount outstanding increased by 214 percent to more than $3.2 trillion. Despite the lack of comprehensive energy- specific data on OTC derivatives, the recent experience of individual trading facilities revealed the growth of energy derivatives trading outside of futures exchanges. For example, according to an annual financial statement of the IntercontinentalExchange (ICE), the volume of contracts traded on ICE—including financially settled derivatives and physical contracts—increased by 438 percent, from more than 24 million contracts in 2003 to more than 130 million in 2006. While some market observers believed that managed money traders were exerting upward pressure on prices by predominantly buying futures contracts, CFTC data reveal that, from the middle of 2003 through the end of 2006, the trading activity of managed money participants became increasingly balanced between buying and selling. According to basic futures market theory, a trader speculating and holding an outstanding position to buy the commodity—a long open interest position—expects that the price of the commodity will rise, while a trader holding an outstanding position to sell the commodity—a short open interest position—expects that the price will decline. As shown in figure 9, according to CFTC data, from July 2003 through December 2003 managed money traders’ ratio of long open interest in crude oil to short open interest was about 2.5:1, suggesting a strong expectation that prices would rise, on average, throughout that period, which they did. By 2006, this ratio fell to 1.2:1, suggesting that managed money traders as a whole were more evenly divided in their expectations about future prices. Managed money trading in unleaded gasoline, heating oil, and natural gas showed similar trends. Although for natural gas, open interest was more often short than long, suggesting a general expectation that prices would decline, which largely did not occur until 2006. Also, the relatively high percentage of open interest for natural gas held by these traders in 2006—surging to just over 40 percent—was perhaps due to the increased volatility of natural gas futures prices from 2005 to 2006, which provided traders with more opportunities for profit (or loss). Energy products are traded on multiple markets, which are subject to varying levels of CFTC oversight and regulation. Under the CEA, CFTC regulatory oversight is focused on conducting the surveillance of futures exchanges, protecting the public, and ensuring market integrity. CFTC collects and analyzes trading position information on futures exchanges, which is central to this oversight. The information is subsequently published at highly aggregated levels in the commission’s COT reports, and it helps to provide transparency to the market. However, these public reports have been criticized because the informational categories for traders do not accurately reflect energy market activity. While CFTC’s oversight is focused on futures exchanges, the number of exempt commercial markets for trading energy commodities, which are not subject to general CFTC oversight, have grown. However, traders in these markets are subject to the CEA’s antimanipulation and, where applicable, antifraud provisions. Also, exempt commercial markets must provide CFTC with data for certain contracts and notify CFTC if cash markets use exempt market prices to price their transactions (although that has not occurred). Energy products also are traded off exchange (referred to as OTC) and are not subject to direct CFTC oversight and regulation. However, as we have previously noted, certain types of off-exchange transactions are subject to antifraud and the antimanipulation provisions of the CEA, which CFTC has the authority to enforce. In addition, contract participants may be subject to other regulatory authority on the basis of their role in the physical market. To enhance its ability to detect and deter price manipulations, CFTC has published for comment a proposal to amend part 18 of its regulations to obtain from traders that have large (reportable) positions in an exchange-traded commodity information about their off-exchange positions in the same commodity. CFTC also held a hearing in September 2007 to examine trading on regulated exchanges and exempt commercial markets, which included an assessment of price discovery and the implications for CFTC oversight. Under the CEA, CFTC has general oversight authority over futures exchanges such as NYMEX. These exchanges receive CFTC approval to list futures and options contracts for trading and are subject to direct CFTC regulation and oversight. To be a regulated futures exchange, an exchange must demonstrate to CFTC that the exchange complies with (1) the criteria for designation under section 5(b) of the CEA for, among other things, the prevention of market manipulation, fair and equitable trading, the conduct of trading facilities, and the financial integrity of transactions conducted on the board; (2) the set of core principles under section 5(d) of the act establishing their regulatory responsibilities; and (3) the provisions on application procedures of part 38 of the CFTC rules. According to CFTC officials, following procedures in the CEA, these exchanges may list new contracts, after certifying that they are in compliance with certain core principles, including ascertaining that the contracts are not readily susceptible to manipulation and monitoring trading to prevent price manipulation. CFTC’s oversight is focused on fulfilling three strategic goals relating to futures exchanges. First, to ensure the economic vitality of the commodity futures and options markets, CFTC conducts its own direct market surveillance and also reviews on an oversight basis the surveillance efforts of these exchanges. According to CFTC officials, the commission monitors trading activity in futures markets and uses these trading data to analyze large positions that might be used to manipulate futures markets. In its oversight role, CFTC reviews new futures contracts to assess susceptibility to manipulation. To list a new futures contract, an exchange must file a written self-certification with CFTC and, if requested, must provide additional evidence, information, or data to CFTC on whether the contact satisfies CEA requirements and the commission’s regulations or policies. Second, to protect market users and the public, CFTC has promoted sales practices and other customer protection rules applicable to futures commission merchants and other registered intermediaries. In this connection, CFTC closely monitors the enforcement of registration and other requirements by the National Futures Association, which is an SRO responsible for regulating all firms and individuals conducting futures business with public customers. Third, to ensure the market’s financial integrity, CFTC reviews the audit and financial surveillance activities of SROs. It also periodically reviews registered derivatives clearing organizations to ensure that they are effectively monitoring risks and protecting customer funds. CFTC provides the public information on open interest in exchange-traded futures and options by commercial and noncommercial traders for various commodities in its weekly COT reports, which are relied upon by the public. Changing market conditions caused CFTC in 2006 to reassess COT reporting and its value to the public. A trading entity generally gets classified as commercial by filing a statement with CFTC that it is commercially “engaged in business activities hedged by the use of the futures or option markets.” To ensure that traders are classified with accuracy and consistency, commission staff review this self-classification and may reclassify a trader if staff have additional information about the trader’s use of the markets. A trader may be classified as commercial in some commodities and as noncommercial in other commodities. A single trading entity cannot be classified as both commercial and noncommercial in the same commodity. Nonetheless, a multifunctional organization that has more than one trading entity may have each trading entity classified separately in a commodity. For example, a financial organization trading in financial futures may have a banking entity whose positions are classified as commercial and have a separate money-management entity whose positions are classified as noncommercial. “In addition, the Commission will begin publishing a supplemental COT report that includes, in a separate category, the positions of commodity index traders in certain physical commodity futures markets. These so-called ‘Index Traders’ will be drawn from both the current Noncommercial and the Commercial categories. Coming from the Noncommercial category will be managed funds, pension funds and other institutional investors that generally seek exposure to commodity prices as an asset class in an unleveraged and passively managed manner using a standardized commodity index. Coming from the Commercial category will be entities whose positions predominantly reflect hedging of OTC transactions involving commodity indices—for example, swap dealers holding long futures positions to hedge short OTC commodity index exposure opposite institutional traders such as pension funds. These latter position holders are those traders described in the request for comments as ‘non-traditional commercials.’” CFTC stated that the pilot program for reporting of commodity index trading did not include energy and metals markets because the large trader data currently available to the commission would not permit an accurate breakout of index trading in these markets. According to CFTC, swap dealers, who use futures markets to hedge commodity index transactions in the OTC market, conduct most trading of commodity index-related futures. However, these swap dealers also may engage in OTC derivative transactions on energy or metals prices directly and conduct cash transactions in the underlying energy or metals markets. As a result of these activities, the overall futures positions held by swap dealers in energy and metal futures markets may not necessarily correspond closely with the hedging of OTC commodity index transactions. The commission stated that including these traders in the new index trader category would not enhance market transparency. Furthermore, it did not want to delay publication of the new COT report while it continues to study whether it is feasible to publish meaningful reports for other markets. The objective of the pilot program is to improve the transparency of an evolving market by separately reporting the positions of index traders. Similarly, the increasing volume of off-exchange trading in energy derivatives and the recent volatility of energy commodity prices justify considering whether a COT category of futures positions held by participants in off-exchange energy markets also could enhance transparency. CFTC said it will assess the relevance and usefulness of the new reporting and study whether it is possible and appropriate to expand the supplemental report to include data for other physical commodity futures markets. Significant changes in the energy markets also may lead CFTC to further examine the usefulness, accuracy, and relevance of reported information to users. According to CFTC officials, energy trading has seen the entry of new market participants. For example, investment banks, hedge funds, and swaps dealers have become significant market participants. Moreover, according to industry analysts and representatives from investment banks and large oil companies, some commercial participants only hedge, some only speculate, and others both hedge and speculate in the energy markets. While some commercial participants may hedge and speculate in the same energy market, CFTC classifies these entities as commercial participants. CFTC has not been able to identify new categories for traders of energy commodities. Such reporting can distort the accuracy and relevance of reported information to users and the public, thereby limiting the usefulness of the information reported to the public as well as information used by traders. In contrast to the direct oversight provided to futures exchange, exempt commercial markets are not subject to CFTC’s general oversight authority. According to CFTC officials, as these markets have grown in prominence, some market observers have questioned their role in the energy markets. Trading energy derivatives on exempt commercial markets is permissible only for eligible commercial entities. While not subject to general CFTC oversight, these markets are subject to CFTC rule 36.3, which provides for the dissemination of exempt commercial market trading data should exempt commercial market prices be used to price cash markets and contains notification, recordkeeping, and reporting requirements. Also, exempt commercial market participants are subject to CFTC’s enforcement authority for the antimanipulation and antifraud provisions of the CEA. These markets are not required to register with CFTC, but must notify CFTC that they are operating as an exempt commercial market and comply with certain CFTC informational, recordkeeping, and other requirements. Specifically, CFTC promulgated rule 36.3 under two subsections of the CEA. One subsection authorizes CFTC to prescribe rules if necessary to ensure the timely dissemination of price, trading volume, and other trading data for a derivative traded on an exempt commercial market if the commission determines that the electronic trading facility used by the market performs a significant price discovery function for transactions in the cash market for the commodity underlying the derivative. The other subsection establishes notification, recordkeeping, and reporting requirements for exempt commercial markets. The rule requires, among other things, that the electronic trading facility in an exempt commercial market must notify CFTC of its reliance on the exemption and provide CFTC with price, quantity, and other data on contracts that average five or more trades a day over the most recent quarter for which they are relying on the CEA exemption. The facility also must maintain a record of allegations or complaints they receive concerning instances of suspected fraud or manipulation and provide CFTC with a copy of the record. CFTC officials said that the reports include transaction-level data, such as quantity and price, for all trades in products meeting the criteria, but not the identities of counterparties to the trades. These officials said that three exempt commercial markets—ICE, the Natural Gas Exchange, and ICAP— currently provide the rule 36.3 trade information reports; in the past, the Optionable and ChemConnect exempt commercial markets also provided these reports. For example, ICE officials told us that for their OTC activities they keep records for all of the products traded on their platform, and report to CFTC on liquid markets (those averaging five trades a day) and any complaints received from market participants. ICE officials said that CFTC often asks ICE for detailed information about participants that are putting up bids and offers and about all of the trades executed in a day. CFTC officials said that the other electronic exchanges have provided notice that they are operating in reliance on the CEA exemption, but they have not provided rule 36.3 trade information reports. CFTC officials explained that an electronic exchange only has to provide information reports if it meets the threshold for reporting, which includes averaging five trades per day in the relevant contract. These officials also said that they do not actively check to determine whether the thresholds are being met. To date, no exempt commercial market or CFTC has determined if cash markets for energy commodities routinely use exempt market prices to price their transactions. According to CFTC officials, an exempt commercial market or CFTC may determine, using certain criteria, if the market serves such a price discovery function. Exempt markets that serve such a function become subject to certain public reporting requirements. According to CFTC officials, the commission has not made such a determination for two reasons. First, they said that the only consequence of serving a price discovery function under current law is that the exempt commercial market must publish its prices. They noted that this is a circular argument because it is the public availability of pricing information that enables the exempt commercial market to serve a price discovery function. Second, they said that this is a low priority. In their view, the current fiscal situation does not allow CFTC to send its economists into the field on matters such as this that would not go before the commission. Also in their view, even if the markets served a price discovery function, no significant consequence would entail because of the circularity argument. However, in light of the growth of trading on ICE and the lessons learned from the Amaranth crisis, CFTC held a hearing in September 2007 to examine trading on regulated exchanges and exempt commercial markets. The hearings included an assessment of price discovery in these markets and the implications for CFTC oversight of these markets. Since 2001, 17 facilities have notified CFTC that they had begun operating as exempt commercial markets (see table 1). According to CFTC officials, 11 of these markets currently offer, or had offered, transactions in energy products, with 8 now operational. Some of these markets have become important players in the trading of energy products. ICE, in addition to the exempt swap contracts it trades in its capacity as an exempt commercial market, is the trading platform for physical commodities, including spot and forward contracts, which routinely involve delivery. According to CFTC officials, some in the industry assert that ICE is the trading platform for an estimated 70 percent of the spot trading for natural gas. Another exempt commercial market, ChemConnect, advertises that data and news providers, such as Bloomberg and Dow Jones Energy Services, rely on it to provide accurate, timely information on energy products. Furthermore, the Web site for the HoustonStreet Exchange indicates that it serves as an electronic trading facility for crude oil and refined products also traded on NYMEX. While there has been significant growth in the number of electronic exchanges, CFTC officials said that they receive trade information reports from only 2—ICE and the Natural Gas Exchange. According to CFTC officials, they have no evidence that the others meet the minimum threshold trading volume for reporting. Energy derivatives also may be traded OTC, under the conditions and restrictions in the CEA for exempt commodities. The act exempts from most of its provisions transactions in exempt commodities into which large market participants enter and that are not traded on a trading facility. In addition, the act exempts from most of its provisions transactions in exempt commodities traded on an electronic trading facility, as long as large commercial traders (defined in the act as “eligible commercial entities”) enter into them on a principal-to-principal basis. Bilateral OTC derivatives contacts are viewed as private transactions between sophisticated counterparties, and there is no requirement for parties involved in OTC transactions to disclose details of their transactions. Because OTC derivatives are contractual agreements, each party is subject to and assumes the risk of nonperformance by its counterparty. This is different from exchange-traded derivatives, where a central clearinghouse stands behind every trade. Thus, according to officials of the International Swaps and Derivatives Association, in the OTC context it is vitally important that one has confidence in the creditworthiness and trustworthiness of one’s counterparty. While these markets generally are not subject to direct CFTC oversight, CFTC has the authority to enforce antifraud and antimanipulation provisions of the CEA in connection with transactions in exempt commodities that take place through an electronic trading facility, and that are entered bilaterally without being subject to negotiation. Several of the enforcement actions filed by CFTC since 2001 addressed the use of false reporting in an attempt to manipulate energy prices on NYMEX. In addition to being subject to certain provisions of the CEA, the participants in these contracts may be subject to other regulatory authorities on the basis of their activities in the physical market. For example, certain actions—such as the buying and selling of a physical energy commodity by traders, such as hedge funds—may fall under the regulatory authority of FERC, which regulates the interstate transmission of physical commodities, such as natural gas, oil, and electricity, to protect energy consumers. Also, certain OTC derivative activities conducted by commercial banks are subject to oversight by the appropriate bank regulator. For example, commercial banks that engage in OTC derivatives are overseen by their relevant regulator, such as the Office of Comptroller of the Currency or the Board of Governors of the Federal Reserve System with respect to how their derivatives trading satisfies requirements of the banking laws. Likewise, SEC also has oversight authority over investment banks’ activities that fall under its regulatory purview. These regulators do not regulate the specific transactions or maintain oversight of OTC derivatives as a class of instruments or markets; they regulate the entities that enter into the contracts or that act as dealers, counterparties, or both. While some observers have called for more oversight of OTC derivatives, most notably for CFTC to be given greater oversight authority of this market, others oppose any such action as unnecessary. Supporters of more CFTC oversight authority believe that regulating OTC derivatives markets is needed to protect the regulated markets and protect consumers from potential abuse and possible manipulation. One of their concerns is that because there is little information available about the size of this market or the terms of the contracts, CFTC may not be assured that trading on the OTC market is not adversely affecting the regulated markets and, ultimately, consumers. Specifically, some have mentioned that, unlike trading on a regulated exchange, OTC derivatives are not subject to any routine reporting requirements. Some have suggested that a combination of quantitative and qualitative information (such as whether derivatives are used mainly for trading or hedging purposes, and notional amounts by derivatives category) be collected. However others, including the President’s Working Group, have concluded that OTC derivatives generally are not subject to manipulation because contracts are settled in cash on the basis of a rate or price determined in a separate, highly liquid market and these OTC transactions do not serve a significant price discovery function. The Working Group also noted that if electronic markets were to develop and serve a price discovery function, then consideration should be given to enacting a limited regulatory regime aimed at enhancing market transparency and efficiency through CFTC, as the regulator of exchange-traded derivatives. However, because of the lack of reported data about this market, addressing concerns about its function and effect on regulated markets and entities would be a challenge. CFTC officials have said that they have reason to believe these off-exchange activities affect prices determined on a regulated exchange. In a June 2007 Federal Register release clarifying its large trader reporting authority, CFTC noted that having data about the off-exchange positions of traders with large positions on regulated futures exchanges could enhance the commission’s ability to deter and prevent price manipulation or any other disruptions to the integrity of the regulated futures markets. According to CFTC officials, the commission also has proposed amendments to clarify its authority under the CEA to collect information and bring fraud actions in principal-to-principal transactions in these markets, thus enhancing CFTC’s ability to enforce antifraud provisions of the CEA. Also, in August 2007, CFTC announced plans to conduct a hearing to begin examining more closely the trading on regulated exchanges and exempt commercial markets. The September 2007 hearing focused on a number of issues, including the current tiered regulatory approach established by the CFMA and whether this model is beneficial; the similarities and differences between exempt commercial markets and regulated exchanges, and the associated regulatory risks of each market; and the types of regulatory or legislative changes that might be appropriate to address any identified risks. CFTC provides oversight for commodity futures markets through routine surveillance, analysis of market data, and inquiries of market participants and others. The commission uses information gathered from surveillance activities to identify unusual trading activity and possible market abuse. In particular, CFTC’s LTRS provides essential information for surveillance, and LTRS provides information on the majority of all trading activity on futures exchanges. CFTC staff also rely on data from other sources and on their experience to identify potential problems, reporting unresolved potential market problems to the commission. NYMEX also conducts its own surveillance activities. According to CFTC and industry officials, CFTC and NYMEX contact traders to collect additional information about questionable trading practices. CFTC staff also said that they routinely investigate traders with large open positions, but the staff added that they do not routinely maintain information about such inquiries, thereby making it difficult to determine the usefulness and extent of these activities. In addition, CFTC coordinates its surveillance activities with other federal, state, and foreign authorities. While CFTC’s surveillance authority is limited to futures exchanges, the commission’s enforcement authority for manipulation and fraud extends to both exchange-based trading and off-exchange trading in exempt commodities, such as energy products. According to data provided by CFTC, in recent years, it has used its enforcement authority to file enforcement actions for almost 300 cases, more than 30 of which involved energy-related commodities. However, as with programs operating in regulatory environments where performance is not easily measurable, evaluating the effectiveness of CFTC’s enforcement activities is challenging because of the lack of effective outcome-based performance measures. CFTC’s enforcement program received mixed ratings in a recent OMB review because CFTC could not fully demonstrate the effectiveness of its enforcement activities. CFTC conducts regular market surveillance and oversight of energy trading on NYMEX and other futures exchanges. These activities include focusing on detecting and preventing disruptive practices before they occur and keeping the CFTC commissioners informed of possible manipulation or abuse. In addition to conducting direct surveillance of trading in energy futures markets on NYMEX, CFTC focuses on NYMEX’s compliance with appropriate CEA core principles, including monitoring of trading to prevent price manipulation and enforcing position limits and position accountability rules. In conducting its own surveillance activities, NYMEX may bring enforcement actions when violations are found. CFTC staff also investigate traders with large open positions and document cases of improper trading. According to CFTC officials, CFTC staff at three regional offices provide much of the market oversight and monitor daily trading activity. For instance, CFTC’s New York Regional Office employs seven economists, who look for unusual trading and potential market manipulations in all futures contracts traded on New York futures exchanges. The New York regional staff obtain information from both market participants and NYMEX to monitor energy trading activity. New York CFTC staff stated that each morning, about 160 firms electronically submit large trader position data from the previous day to CFTC. CFTC headquarters receives these data and makes them available on a network to its field offices. Staff review these data for potential errors or omissions and then populate the LTRS, a database that staff use in conducting their surveillance activities. CFTC staff also said that they rely on the commission’s integrated surveillance system (ISS), which contains surveillance data that CFTC has collected from the futures exchanges, clearing members, foreign brokers, and large traders. According to CFTC’s 2005 performance and accountability report, ISS is a critical application to support futures and options data market surveillance. This system provides continuously updated trading data on holders of large futures and option positions that CFTC staff uses daily to monitor futures and option trading, detect potential problems, and identify trends in the marketplace. According to CFTC officials, ISS also is used to facilitate analysis of data received from exempt markets as a result of special calls for information. For example, pursuant to separate special calls issued in April, September, and December, 2006, ICE now continuously provides the commission with large trader position data. The commission also issued enforcement- related special calls seeking data for two individual ICE market participants in September 2006 and February 2007. The LTRS, which is part of ISS, is a comprehensive system for collecting information on market participants, a key information source for CFTC’s market surveillance program and essential for monitoring markets and identifying and resolving potential problems involving market congestion, manipulation, and speculative position limits. Congestion may occur when traders holding short positions are attempting to cover their positions but are unable to find an adequate supply of contracts provided by traders with long positions or by new sellers willing to enter the market, except at sharply higher prices. In conducting their daily surveillance activities, CFTC officials said they analyze the trading data for indications that individual traders may be attempting to manipulate the market. This activity involves (1) looking for traders having unusually large market positions relative to open interest—the total number of futures contracts that have been entered into and not yet liquidated by an offsetting transaction or fulfilled by delivery—and deliverable supply and (2) examining the potential for disruption at expiration and sharp moves in the market. If certain positions pose concerns to CFTC staff, they can request additional information from a reporting firm or trader about trading and delivery activity. CFTC staff also analyze trading using data from other sources. CFTC officials said that the staff look at price movements and price relationships—especially in the natural gas, crude oil, heating oil, and unleaded gasoline markets—using commercial information sources, such as Bloomberg, Gas Daily, Reuters, and other market sources. They also obtain information about traders by monitoring their Web sites and use NYMEX’s and EIA’s Web sites and Lexis-Nexis, as well as firms’ Web sites. CFTC staff said that they are in regular contact with exchange officials, who have data on clearing members and trading activity. They also obtain surveillance information from other units within the commission and from tips by the public. While CFTC data and other market collections are focused on identifying potential market disruptions and manipulations, staff also rely on their experience to identify potential problems. According to CFTC staff, the New York Regional Office staff assigned to surveillance of energy trading have many years of experience, either doing surveillance work for CFTC or in the futures industry, in general. Experienced staff are needed because, according to CFTC staff, analyzing market data is an art as well as a science. CFTC staff referred to the traditional test for manipulation set forth in the commission’s Indiana Farm decision as a commonly recognized statement of the elements that are necessary to prove manipulation. According to CFTC staff, when a potential market problem has been identified, surveillance staff generally contact the exchange or traders to gather additional information. They said that surveillance staff may ask exchange employees, brokers, or traders questions to confirm positions and determine the intent of traders. They added that staff may express concern about the size of positions or possible actions by traders and caution traders to act responsibly. According to the staff, CFTC’s Division of Market Oversight may issue a warning letter or make a referral to the Division of Enforcement to conduct a nonpublic investigation into the trading activity. Markets where surveillance problems have not been resolved may be included in reports presented to the commission at weekly surveillance meetings. These reports provide information on traders with the four largest long and short positions; other market information, including delivery information; and background on the contact. According to CFTC staff, CFTC commissioners review the reports; discuss the situations with surveillance staff; and, if appropriate, consider other possible remedial actions, such as suggesting that the exchange take emergency action. If necessary, the commission itself may take emergency action. If these actions do not resolve the issue or if an exchange fails to resolve a problem by taking actions that the commission deems appropriate, CFTC can order an exchange to take emergency actions. These actions include limiting trading, imposing or reducing limits on positions, requiring the liquidation of positions, extending a delivery period, or suspending trading. The commission has taken such emergency actions four times in its history, but never for energy markets. In addition to CFTC’s surveillance of NYMEX and trading on the exchange, NYMEX conducts its own surveillance activities and, if violations are found, brings its own enforcement actions. NYMEX is responsible for enforcing its own standards and CFTC’s standards embodied in its rules governing the exchange, and its surveillance program is designed to monitor for possible manipulation by market participants. If NYMEX staff find potential violations, they will gather information and, if needed, take enforcement actions. For example, according to officials at a large refiner, NYMEX staff call them nearly every month about a large trade to make sure that their physical (or wet) barrels have moved and that their trade is not a price-setting mechanism or market ploy. Refiner officials added that even though NYMEX staff know they are a big refiner, they will examine their trades to see the actual signed contract to make sure it is valid. In their view, NYMEX staff are vigilant, as they should be. Officials from a hedge fund also said that both NYMEX and CFTC staff monitor their positions carefully and, as a speculator, would be notified immediately by NYMEX and CFTC if they were over the trading limits on any day. When asked about what weaknesses in the structure, monitoring, or enforcement mechanisms of derivative markets might allow for market manipulation, one market observer responded that he was not aware of any such weaknesses. Appendix III contains detailed discussion of NYMEX surveillance activities and enforcement actions. CFTC staff routinely make inquiries about traders with large open positions approaching expiration, but formal records of their findings are only kept in cases where there is evidence of improper trading. If LTRS data reveal that a trader has a large open market position that could disrupt markets if it were not closed before expiration, CFTC staff would contact the trader to determine why the trader had the position and what plans the trader had to close the position before expiration or to ensure that the trader was able to take delivery. If the traders provided a reasonable explanation for the position and a reasonable delivery or liquidation strategy, staff said that no further action would be required. CFTC staff said they would document such contacts on the basis of their importance in either informal notes, e-mails to supervisors, or informal memorandums. No formal record would be made of the inquiry, according to one CFTC official, unless there was a signal indicating improper trading activity. Without such data, CFTC’s measures of the effectiveness of its actions to combat fraud and manipulation in the markets will not reflect this surveillance activity, and CFTC management might miss opportunities to both identify trends in activities or markets and better target its limited resources. CFTC staff added that all surveillance projects and activities that require a minimum number of hours of work are tracked by quarterly statistical reports, including those futures expirations with large trader or deliverable supply problems. They said that expirations are routinely monitored by economists and reviewed with their supervisors through weekly surveillance reports. Economists are responsible for the analytical review of cash and futures market developments, including the assessment of supply and demand factors, basis and spread relationships, the adequacy of deliverable supply, large trader positions and position changes, large trader histories, and the potential for group trader activity. CFTC staff said that their economists keep their supervisors and the commission informed of potential problems as they arise. In addition to keeping CFTC commissioners apprised of surveillance activities and specific cases that may require action, CFTC coordinates its surveillance and oversight activities with other federal agencies, states’ attorneys general, and foreign regulators. CFTC officials told us that through the Division of Enforcement’s Office of Cooperative Enforcement, which was created in 2002, they conduct outreach efforts to other financial regulators at the federal and state levels. Specifically, CFTC and FERC coordinate oversight and enforcement activities and have a memorandum of understanding that provides for the exchange of data. FERC regulates the interstate transmission of natural gas, oil, and electricity, and it audits natural gas sellers’ compliance with the protocols outlined by FERC for reporting sales to index publishers like Platts, a company that compiles information on oil, natural gas, and electricity and other energy commodities and provides industry reports on commodity prices. If futures transactions are thought to affect transactions within FERC’s jurisdiction, then FERC and CFTC may coordinate their oversight and enforcement work by sharing data as provided in the memorandum. In pursuing potential market abuse cases, such as individuals trying to manipulate energy spot prices to benefit their futures market positions, FERC officials said that FERC will tend to take the lead when abuses occur in the physical markets. FERC officials also said that CFTC will tend to take the lead when abuses occur in the futures markets. In July 2007, FERC filed two market manipulation cases that, according to a commission announcement, was the first time the agency used its enforcement authority under the Energy Policy Act of 2005 and its former market manipulation rule. According to CFTC officials, CFTC has filed 38 cases over the past 6 years that have focused on conduct in both the cash and futures markets (see app. IV). CFTC and FERC also may work with DOJ on certain cases. In addition, CFTC officials said that, on occasion and when warranted by the circumstances, CFTC has shared large trader information with certain agencies, such as the Department of the Treasury, the Board of Governors of the Federal Reserve System, and the Federal Reserve Bank of New York, to address issues of common concern to the agencies. For example, in the aftermath of the financial difficulties in 1998 of Long Term Capital Management, a large hedge fund, CFTC shared information on the hedge fund’s exchange trading activity with members of the President’s Working Group. Because coordinating requires judgments about what information would need to be and could be shared and about how best to share it, we concluded in a 1999 report that the regulators are in the best position to determine the most effective ways to enhance their coordination. CFTC also shares information with other members of the President’s Corporate Fraud Task Force at their quarterly meetings on antifraud cases. CFTC’s Division of Enforcement is charged with enforcing the antimanipulation sections of the CEA, including sections 6(c), 6(d), and 9(a)(2). In particular, section 9(a)(2) sets forth the commission’s antimanipulation and false reporting authority in cash and futures markets. In determining whether violative conduct has occurred, CFTC officials told us that the Division of Enforcement has broad investigatory authority to obtain records and testimony, including subpoena authority, under a commission order. They added that upon conclusion of an investigation, which is routinely nonpublic, the division may recommend enforcement action if warranted. The enforcement actions CFTC has taken in its energy-related cases generally have involved false public reporting as a method of attempting to manipulate prices on both the NYMEX futures market and the off- exchange markets. CFTC officials said that from October 2000 to September 2005, the commission initiated 287 enforcement cases and more than 30 of these cases involved energy trading, including actions against Enron and others. For example, according to CFTC data, from 2001 through 2005, CFTC levied fines totaling $305 million in actions alleging attempted manipulation of the price of natural gas (see app. IV for more detailed information). Most of these cases charged attempted manipulation by means of falsely reporting natural gas trading information to energy index firms, such as Platts, that calculate surveys or indexes of natural gas prices for various physical delivery points (hubs) throughout the United States. Generally, these cases involved allegations of various defendants knowingly disseminating false information in an effort to skew the indexes for their financial benefit or for other reasons. Participants in the natural gas markets use the indexes for price discovery and assessing price risk. Many of the actions were initiated on the basis of information that came from sources other than CFTC surveillance activities, or those of NYMEX, because they involved activities outside of NYMEX. As one major oil company official told us, in his view, CFTC and FERC vigorously pursued attempts by traders to manipulate the market. Most recently, on August 1, 2007, the commission entered an order imposing a $1 million penalty against Marathon Petroleum Company, LLC, for attempting to manipulate spot cash crude prices by attempting to influence the Platts market assessment. On July 25, 2007, the commission commenced an action against Amaranth and others for attempted manipulation of NYMEX natural gas futures prices. Also on July 26, 2007, the commission commenced an enforcement action on Energy Transfer Partners, L.P., and others for attempted manipulation of physical natural gas prices. Regarding energy futures, CFTC coordinates its enforcement activities with NYMEX officials and various other federal, state, and foreign authorities. CFTC staff stated that they meet periodically with NYMEX Compliance Department officials to discuss enforcement activities, as appropriate, and have formal quarterly meetings to discuss mutual involvement in specific cases, including energy products. In addition to coordinating energy enforcement matters with NYMEX, as a regulator of derivatives trading, CFTC often will work with the regulator of the underlying commodity or affected market, whether the Department of Agriculture, FERC, or Treasury. CFTC does not have criminal authority but often works with DOJ on those cases involving violations of the CEA that DOJ believes warrant criminal prosecution. DOJ officials stated that their focus has been on natural gas cases, which began with cases involving Enron. According to DOJ officials, their role complemented the regulatory roles of FERC and CFTC, and they have an effective working relationship with CFTC in terms of sharing case information. For example, pursuant to a memorandum of understanding with CFTC, in May 2006, FERC obtained information about trading in natural gas futures contracts that FERC used in support of an enforcement action against Amaranth that was initiated in July 2007. On July 25, 2007, CFTC filed an action in the United States District Court for the Southern District of New York against Amaranth Advisors, L.L.C., Amaranth Advisors (Calgary) ULC, and Brian Hunter alleging, among other things, that the defendants intentionally and unlawfully attempted to manipulate the price of natural gas futures contracts on NYMEX on February 24, 2006, and April 26, 2006. In another case, on June 28, 2006, CFTC brought an enforcement action against BP Products North America, Inc., alleging, among other things, that BP cornered the physical propane market and manipulated the price of propane in February 2004. Also on June 28, 2006, DOJ announced that a former BP trader had pled guilty to conspiracy to manipulate and corner the physical propane market. FTC also has exercised its authority in the energy arena. Since 1980, FTC’s focus in energy has been in reviewing mergers and acquisitions for anticompetitive behavior and investigating instances of possible collusion, price fixing, and other anticompetitive conduct. However, FTC staff told us that they generally did not coordinate their work with CFTC, but added that they would turn over any evidence of futures manipulation to CFTC. CFTC staff said that, as appropriate, CFTC also coordinates its antifraud enforcement activities with states’ attorneys general, who often will assist in a case by acting as a co-plaintiff with CFTC. In turn, CFTC may detail an attorney to a state. CFTC staff said that they also may work with international authorities, such as the United Kingdom’s Financial Services Authority, on cases involving activities in more than one nation. Although CFTC has undertaken enforcement actions and levied fines, OMB’s most recent 2004 Program Assessment Rating Tool (PART) assessment of the CFTC enforcement program was mixed. OMB designed PART to provide a consistent approach to assessing federal programs in the executive budget formulation process. PART is a standard series of questions meant to serve as a diagnostic performance tool, drawing on available program performance and evaluation information to form conclusions about program benefits and recommend adjustments that may improve results. In the assessment, OMB rated the enforcement program as “Results Not Demonstrated” and said that the enforcement program lacked performance measures that illustrate whether the program meets its overall objective. However, CFTC’s existing performance measures show that it brings substantive cases in a timely manner and “is well designed to meet its objectives [of protecting commodity futures and options market users and the public from fraud, manipulation, and abusive practices related to the sale of certain commodities through the enforcement of laws against such practices] and to maximize the use of its resources.” According to the PART assessment, the enforcement program has a clear purpose, addresses the public interest by ensuring adherence to the CEA and CFTC’s regulations, is not duplicative of other government programs, is free of major design flaws, and is effectively targeted so that the resources address the program’s purposes. OMB scored CFTC at 100 percent for the dimensions of both program purpose and design and program management, 71 percent for planning, and 67 percent for results and accountability. Compared with the other 96 programs that OMB identified as similar to CFTC’s program, the comparable programs have much lower average scores for the dimensions of purpose and design (82 percent), program management (84 percent), and results and accountability (50 percent) and have a similar score for planning (73 percent). CFTC’s score of 71 percent for the planning dimension reflected OMB’s assessment that CFTC included performance measures in its annual reports; used the actual results it achieved during the preceding fiscal year as a baseline for all of its performance measures and strove to set ambitious targets for its performance; was scrutinized on a regular basis by CFTC’s Office of the Inspector General; had budget requests that were explicitly tied to accomplishment of the annual and long-term performance goals, and resource needs that were presented completely and transparently in the program’s budget; and had taken meaningful steps to correct its strategic planning deficiencies. However, OMB also concluded that regarding the strategic planning dimension, the program had a limited number of long-term performance outcome measures that did not fully reflect the program’s goals, and that the long-term measures and targets did not fully reflect the program’s purposes. These measures included the percentage growth in market volume, the increase in the numbers of exchanges and clearinghouses, the percentages of SROs and clearing organizations that complied with the requirement to enforce their rules, and the percentage decreases in both the number of customers who lost funds because of alleged wrongdoing and the amount of funds that these customers lost. CFTC enforcement staff stated that they face challenges in establishing measures to determine whether the enforcement program achieves its goal of deterring people from engaging in market manipulation or other abusive behavior. According to OMB, CFTC’s score of 67 percent on the program results and accountability dimension reflected its assessment that CFTC’s enforcement program had demonstrated both (1) improved time efficiencies and cost-effectiveness in achieving its program goals and (2) our several evaluations of CFTC indicating that it was effective and achieving results. OMB also reported that for fiscal year 2004, the enforcement program met all of its outcome measures and came close to meeting all of its output measures, with one exception. OMB further stated that the outcome-related measures established for enforcement do not fully reflect progress on meeting the program’s overall goals. While CFTC satisfied most but not all of OMB’s PART criteria, it has fallen short in its ability to develop long-term performance outcome measures that are reflective of its program’s goals and purposes. As OMB identified, CFTC has substituted proxy measures for outcome measures: that is, using measures such as percentage growth in market volume and increase in the number of exchanges and clearinghouses as proxies for protecting market integrity, and percentage decreases in both the number of customers who lost funds because of alleged wrongdoing as proxies for both protecting market integrity and consumers. We have found that managers in a regulatory environment where programs and activities are not easily measurable, as is the case with CFTC enforcement, have reported that it is particularly challenging to measure outcome-oriented performance and collect useful data. However, there are a number of other ways to evaluate program effectiveness, such as using expert panel reviews, customer service surveys, and process and outcome evaluations. We have found with other programs that the form of the evaluations reflect differences in program structure and anticipated outcomes, and that the evaluations are designed around the programs and what they aim to achieve. Without utilizing these or other methods to evaluate program effectiveness, CFTC is unable to demonstrate whether its enforcement program is meeting its overall objectives. The rise in energy prices can be and has been attributed to a variety of factors. From January 2002 through June 2006, the physical and derivatives markets both underwent substantial change and evolution. The physical energy markets experienced tight supply and increasing global demand, ongoing political instability in oil-producing regions, and other supply disruptions, which affected the prices of energy products. At the same time, increasing numbers of and different types of market participants were trading futures in search of higher returns, thereby increasing contract volume. Substantial growth in the exempt commercial and OTC markets also occurred. Determining the impact of any one factor is complicated because price changes in the physical and futures markets are closely linked and in the long run are influenced by the same market fundamentals. Generally, futures prices reflect traders’ views of the impact of changes in the physical markets and spot prices are affected by these expressed views and vice versa. Given this interrelationship, it is not surprising that some market observers point to the changes in the energy futures and other derivatives markets as a possible explanation for price increases, while others, primarily the regulators, look to changes in the physical markets to explain the increases. However, given the changes in both markets, attributing causality to any one factor—much less a particular type of trading activity—is difficult. Regardless of the reason for the increases in prices, ongoing monitoring of both markets is warranted to ensure that the public interest is being protected as well as the integrity of the markets. Related to concerns about rising prices, some market observers and others have questioned whether CFTC’s authority is broad enough to protect investors from fraudulent, manipulative, and abusive practices. The scope of CFTC’s authority varies, depending on the market where the commodity is traded. Some markets are available for retail trading and receive direct CFTC oversight, while others are limited to professional traders (such as OTC energy derivatives markets) and receive less oversight. Other markets are largely unregulated. Given the changes in these markets in general and the growth in off-exchange trading in particular as well as ongoing questions about the relationship between exchange-traded and off- exchange markets, a reexamination of the scope of CFTC’s authority is warranted. The results of CFTC’s hearings on its existing regulatory structure and the similarities and differences between exchange-traded and exempt markets may be instructive for such a reexamination. While participants on all sides of this issue have perspectives that call for further consideration, these are public policy decisions that ultimately will be made by Congress. Unless resolved, questions will continue about the scope of CFTC’s authority. In the interim, we have identified a number of process issues that CFTC can address to strengthen its enforcement and surveillance programs. First, CFTC has attempted to provide the public with more meaningful information through the COT reports. While this effort has expanded the reporting for some agricultural commodities, it has remained virtually unchanged for energy commodities that have a high level of public and industry interest. Not having complete information on trading in energy commodities impairs the ability of traders to make fully informed decisions. Second, CFTC’s oversight of regulated exchanges involves a range of surveillance activities that have resulted in a number of commission- related enforcement actions. However, CFTC does not maintain complete records of its surveillance activities. Currently, the commission does not maintain written records on all surveillance follow-up activities, particularly in instances where no potential violation was found. Without such records, CFTC staff cannot fully demonstrate the actions they are taking to combat fraud and manipulation in the markets. Third, as is the case with most enforcement agencies, CFTC has had limited success in identifying meaningful outcome-based performance measures. However, agencies can use a variety of methods to evaluate program effectiveness, such as expert panel reviews, customer service surveys, and process and outcome evaluations. Without meaningful measures for program effectiveness, CFTC may be missing opportunities to identify significant trends in certain activities or markets and to better target its limited resources. In light of recent developments in derivatives markets and as part of CFTC’s reauthorization process, Congress should consider further exploring whether the current regulatory structure for energy derivatives, in particular for those traded in exempt commercial markets, provides adequately for fair trading and accurate pricing of energy commodities. To improve the oversight and available information on energy futures trading, we recommend that the Acting CFTC Chairman take the following three actions: reexamine the classifications in the COT reports to determine if the commercial and noncommercial trading categories should be refined to improve the accuracy and relevance of public information provided to the energy futures markets; explore ways to routinely maintain written records of inquiries into possible improper trading activity and the results of these inquiries to more fully determine the usefulness and extent of CFTC’s surveillance, antifraud, and antimanipulation authorities; and examine ways to more fully demonstrate the effectiveness of CFTC enforcement activities by developing additional outcome-related performance measures that more fully reflect progress in meeting the program’s overall goals. We provided a draft of this report to the Commodity Futures Trading Commission for comment. In its written comments, CFTC said that the commission will reexamine classifications in the COT reports. CFTC also said that the commission will explore additional recordkeeping procedures for its staff, but that it must balance the time required for such additional tasks against the need to undertake market surveillance by an already-stretched surveillance staff. CFTC added that it has included the development of measures to evaluate the effectiveness of its enforcement program in its most recent strategic plan. The commission’s comments are reprinted in appendix V. CFTC staff provided technical comments and corrections that we have incorporated in this report where appropriate. We will provide copies of this report to interested congressional committees. We are also sending a copy of this report to the Acting Chairman of the Commodity Futures Trading Commission, the Secretary of the Department of Energy, the Chairman of the Federal Energy Regulatory Commission, the Chairman of the Federal Trade Commission, the Acting U.S. Attorney General, and the Chairman of the Securities and Exchange Commission. We will make copies available to others upon request. In addition, this report will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staffs have any questions about this report, please contact me at (202) 512- 8678 or williamso@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix VI. To examine trends and patterns of trading activity in the energy derivatives markets and physical markets, we analyzed data on futures, spot, and over-the-counter (OTC) derivative markets. We gathered information on spot prices for crude oil, unleaded gasoline, heating oil, and natural gas from the U.S. Department of Energy’s Energy Information Administration (EIA). We obtained daily futures settlement prices and average daily volume data for the four commodities from the New York Mercantile Exchange, Inc. (NYMEX). We collected data on the size of the global OTC commodity derivatives market—including energy, but excluding precious metals—from the Bank for International Settlements. We also obtained information on the numbers of participants and outstanding positions in energy futures markets by different categories of traders from the Commodity Futures Trading Commission (CFTC). These CFTC data cover the period from July 2003 through December 2006. We determined that data from these sources were sufficiently reliable for the purposes of this report. We used monthly averages of the EIA spot prices and NYMEX futures prices to depict price trends over the past 20 years and illustrate the strong relationship between spot and futures prices. Also, we adjusted the prices to remove the effects of inflation so that prices would be comparable across years. We also adjusted the prices using monthly deflation factors that we derived from the seasonally adjusted implicit price deflator for gross domestic product from the Bureau of Economic Analysis, as of February 28, 2007. We used the futures price data obtained from NYMEX to calculate the volatility of energy futures prices. These data covered the period from January 1987 through December 2006 for crude oil, unleaded gasoline, and heating oil. The period for natural gas was from April 1990 through December 2006, when that contract began trading on NYMEX. We calculated the historical volatility of the futures prices as the standard deviation of the natural logarithm of relative changes in daily settlement prices. Monthly volatility figures were calculated from the trading days of each month and expressed on an annual basis. We annualized the monthly figures by multiplying daily volatility by the square root of 250, which represents an approximation of the number of trading days in a year. We also calculated annual volatility for each of the four commodities as the average of the monthly mean volatilities. We used the front month futures contract—that is, the nearest traded contract month—because it is the most frequently used maturity for measuring price and volatility and is the most heavily traded contract. To identify the opinions of market participants and analysts about the effect of energy derivatives trading on prices, we interviewed officials from CFTC, the Federal Energy Regulatory Commission, and EIA; managers from trading facilities, including NYMEX and the IntercontinentalExchange (ICE); academics knowledgeable about energy and finance; and market participants representing investment banks, hedge funds, and oil producers and refiners. We selected banks to interview on the basis of their perceived level of involvement in energy markets. The hedge funds we interviewed were identified through the assistance of the Managed Funds Association—a membership organization representing the hedge fund industry—which contacted its members involved in energy trading to identify hedge funds who were willing to be interviewed. We selected oil producers and refiners on the basis of their size and role in U.S. energy markets. We also gathered information from several trade associations, including those representing users of energy commodities, and interviewed former CFTC officials. Although we gathered the views of a wide range of market participants and observers, these sources do not necessarily represent the views of all market participants and observers. We also reviewed studies by governmental and nongovernmental observers, including CFTC; NYMEX; the Senate’s Permanent Subcommittee on Investigations, Committee on Homeland Security and Governmental Affairs; and a report prepared for the attorneys general of four midwestern states. In addition, we reviewed public statements from relevant government officials, such as the current Federal Reserve chairman and his predecessor. To understand the effects of supply and demand conditions in the physical energy markets, we examined data and analysis from EIA and prior GAO reports. To examine CFTC’s resources and authority for protecting market users from fraudulent, manipulative, and abusive practices in the trading of energy futures contracts, we describe CFTC’s current and past regulatory authority and approach by reviewing the Commodity Exchange Act (CEA), as amended; CFTC’s President’s Budget and Performance Plan for fiscal year 2007; CFTC’s 2004 Annual Report; and other information from CFTC. We obtained information on CFTC’s regulatory role and the exempt commercial and OTC markets from officials at CFTC, EIA, and NYMEX and from market participants. In addition, we reviewed information on CFTC’s regulatory role contained in the Federal Register and congressional hearing testimony. To describe the concerns regarding OTC derivative trading and the scope of CFTC’s regulatory authority, we obtained information from federal agency officials and an industry trade association. To describe the hedging and speculative trading of market participants, we reviewed various reports that addressed those concerns, and we interviewed several market participants. To examine how CFTC monitors and detects market abuses in the trading of energy futures, and enforcement actions taken in response to identified abuses, we gathered information from officials at CFTC headquarters and the New York Regional Office. We reviewed CFTC regulations and other documents on its surveillance and enforcement programs and observed a CFTC monthly surveillance meeting. We gathered information from market participants and experts regarding CFTC’s oversight activities. To examine CFTC’s enforcement program and how CFTC coordinates with other regulators and authorities, we gathered and analyzed data on CFTC’s enforcement cases, interviewed CFTC and other federal agency officials and staff on coordination activities and agreements, and reviewed CFTC Office of the Inspector General reports. We also reviewed the Office of Management and Budget’s PART assessment of CFTC’s enforcement program. Appendix II: Types of Contracts and Transactions for Energy Commodities in the Physical and Financial Markets Bilateral over-the-counter (OTC) transactions for immediate delivery or near-term delivery and payment representing a specific price and location. Industry analysts publish price data gathered from market participants. Bilateral OTC transactions in which the seller agrees to deliver to the buyer a specified quantity and quality of an asset or a commodity at a specified date at an agreed-upon price or pricing formula and where delivery is contemplated. Futures contracts are standardized contracts for a specific product at a specific location, where delivery is not usually made and contracts are offset prior to expiration. Transactions are executed on an exchange regulated by the Commodity Futures Trading Commission (CFTC), such as the New York Mercantile Exchange. The exchange publicly disseminates price and other data. Options on futures contracts are contracts that give a buyer the right, but not the obligation, to buy or sell a specific quantity of futures contracts within a designated period at a designated price. As in the United States, standardized contracts for a specific product at a specific location, where delivery is not usually made and contracts are usually offset prior to expiration. Sales are executed on an exchange, such as the IntercontinentalExchange (ICE) Futures in London that is subject to regulation by the U.K. Financial Services Authority. Foreign boards of trade are able to provide direct access to U.S. market participants by obtaining “no-action” relief from CFTC staff. Standardized contracts for a specific product at a specific location, where delivery is not usually contemplated because contracts are usually offset prior to expiration or are cash settled and are based on prices from a regulated futures exchange or another source. Transactions are executed on an electronic trading platform, such as ICE, involving “eligible commercial entities.” Exempt commercial markets may offer a clearing service for certain derivatives contracts. Exempt commercial markets may offer trading both in contracts that are subject to the Commodity Exchange Act and contracts that are not. Derivatives contracts that are privately negotiated, bilateral contracts between eligible counter parties, often involving a swap dealer. The contracts are financially settled and are based on prices from a regulated futures exchange or another source. OTC swaps are a promise between two parties to make a series of payments to each other, of which at least one series is based on a commodity price. OTC options: OTC markets also offer options to buy or sell other assets. Under CFTC regulations, NYMEX is responsible for establishing and enforcing rules governing its member conduct and trading, preventing market manipulation, ensuring that futures industry professionals meet qualifications, and examining members for financial strength. In carrying out these responsibilities, NYMEX officials told us that NYMEX’s surveillance program is designed to monitor market and trade practices. They said that NYMEX relies on automated detailed information for each transaction to identify the buyer, seller, and clearing members who maintain customer accounts, and to identify whether a person is a member of NYMEX. The market surveillance activities focus on monitoring for possible manipulation by market participants. Specifically, NYMEX officials monitor large trader data, the “street book” speculative position limits and accountability levels, exemptions to speculative position limits, position concentrations, and the relationship between cash and futures prices. A speculative position limit is the maximum net position that a market participant may hold in a specified contract month of a listed NYMEX contract, and is set by NYMEX. Market participants that are bona fide hedgers are eligible to apply for, and receive under certain conditions, limited exemptions from speculative position limits. NYMEX officials said that they monitor speculative position limits, and if a limit is about to be reached or has been hit, they record the overage and contact the customer to find out if there is a logical explanation for the overage that is linked to a bona fide commercial exposure. They added that if there is a logical explanation, the customer may be allowed to keep the position and file for a formal exemption; however, if there is no logical explanation, then NYMEX officials will direct the position to be reduced. The officials said they follow such directives with a warning letter to the customer. If the position is not reduced, the officials said that another warning letter is then to be issued. As a final step, NYMEX could hold a hearing before its business conduct committee to deny the customer access to the exchange. However, this has never happened, according to NYMEX officials. They also said that customers may obtain exemptions to speculative position limits on a case-by-case basis and for a period of 12 months. The officials said that exemptions are not given if they could disrupt the markets, and that the exemptions are monitored for possible changing circumstances, such as reorganizations or Moody’s downgrading a party’s equity rating. NYMEX also monitors trading by customers using multiple brokers. For example, NYMEX officials stated that large oil companies may use several brokers to trade—that is, each broker may trade using its own account to hide what it is doing from other companies. However, NYMEX officials said that all of the company’s accounts are aggregated, and that its staff will analyze the trading and large trader data and contact the customer if there are any surveillance issues. In addition, NYMEX has established the maximum daily trading limits for each commodity contract. These limits are the maximum price advance or decline from the prior day’s settlement price and, if exceeded, trading stops for a period. NYMEX officials said that the exchange has changed these limits over the years, and the officials could not recall the last time a trading limit was reached. Furthermore, the limits were completely eliminated for NYMEX’s New York Commodity Exchange (COMEX) division, which trades precious metals. According to NYMEX officials, suspending trading would give traders time to count and balance their positions before resuming trading. Unfortunately, the officials continued, if trading on NYMEX is suspended, the price discovery mechanism on which the OTC and cash markets depend is also suspended. They added that CFTC does not provide any requirements for price limits and, in fact, favors ongoing price discovery. In addition to market surveillance, NYMEX conducts trade practice surveillance. According to NYMEX officials, this surveillance focuses on persons who handle contract orders either on the trading floor or electronically, as well as on persons and firms engaging in proprietary trading for their own accounts. NYMEX seeks to identify trading practice abuses, such as prearranged trading, front running, providing tips on proprietary information, and accommodating trades. Prearranged trading is the noncompetitive trading between brokers in accordance with an expressed or implied agreement or understanding and is a violation of CEA and CFTC regulations. Front running is taking a futures or option position based on, for example, a customer order in the same or related future or option. This practice is also known as trading ahead. Accommodation trading is noncompetitive trading entered into by a trader, usually to assist another with illegal trades. NYMEX officials stated that trade practice surveillance information may be used as part of market surveillance, but this surveillance is not as focused on price movements and involves different types of monitoring, such as physically observing floor trading by people entering orders, and, in effect, is similar to “police on the beat.” The officials added that with their recent use of the Chicago Mercantile Exchange’s Globex electronic trading system to trade energy futures, their trade practice surveillance system is changing, with more emphasis on monitoring access to, and activity on, the Globex system in NYMEX contracts. As a self-regulatory organization (SRO), NYMEX has the authority to pursue instances of suspected manipulation or other attempts of fraudulent or abusive trading. NYMEX officials stated that the exchange conducts its own enforcement activities, and CFTC expects NYMEX as an SRO to handle issues relating to exchange members. However, NYMEX will request CFTC’s assistance if needed, especially for issues relating to nonexchange members. If NYMEX officials become aware of potentially abusive practices outside of their authority, they will notify the appropriate federal regulator, such as CFTC. Information from NYMEX surveillance activities and other sources is used to investigate potential abuses. Other sources of information include referrals from CFTC, traders, and customers. NYMEX officials investigate these referrals; if there is evidence of wrongdoing, they may open an investigation case. For each case that is pursued, they record information, including the source of the referral and investigation activities. Interviews conducted during an investigation may be taped and transcribed. From January 2000 through May 2006, NYMEX opened 706 investigations. However, if a referral did not result in a case that was pursued, NYMEX does not document how each referral was handled. For example, if a referral regarding a trade resulted in a NYMEX official making a telephone call and finding that there was no apparent violation, NYMEX officials said staff would not create a written record to log how the referral was handled or the result of the inquiry. NYMEX officials told us that when NYMEX pursues a case, such as prearranged trading, the case is brought before the exchange’s business conduct committee (BCC). According to NYMEX, the BCC (which includes three public members and other committee members), is structured in a manner analogous to a grand jury proceeding and determines, on the basis of the evidence in an investigative report and any written response from the person accused in the case, whether there is a reasonable basis to believe an exchange rule violation occurred. NYMEX officials told us that a BCC meeting is scheduled for every other month; however, in some months, there are no cases to discuss and no meetings are held. They added that the BCC hears about two or three dozen cases annually. Once a case is heard, the BCC may then direct the compliance department to issue a complaint, or, in the case of minor violations, a written warning. The person named in the complaint has 10 days to respond, and he or she may make a settlement offer at any time prior to the conclusion of a hearing. Settlement offers must be approved by NYMEX’s board of directors. From January 2000 through May 2006, NYMEX opened more than 700 investigations, most involving trading violations, and, of those, 125 were BCC-issued complaints (see table 2). Settlement of NYMEX complaints may result in settlement offers, fines, or disciplinary actions. Settlement offers exceeding $25,000, or cases contested by the respondent, are referred to NYMEX’s adjudication committee for settlement consideration or for a full disciplinary hearing. The adjudication committee is authorized to conduct hearings where the facts of the case are presented and argued by the respondent or their attorney and exchange compliance counsel. At the conclusion of the hearing, the committee issues a decision regarding any sanctions. Sanctions can include a cease-and-desist order, a fine of up to $250,000 for each rule violation, suspension, or expulsion from membership. For example, NYMEX settled a $100,000 case on Morgan Stanley in 2002, a $2.5 million case on BP Product North America, Inc., in 2003, and a $300,000 case on Shell in 2005. NYMEX officials stated that no one has been expelled from the exchange since 1998 for failure to pay a fine. According to the officials, the adjudication committee is scheduled to meet at least once a month. They added that about 40 percent of the cases are resolved at adjudication and cases rarely go to a full hearing. If a hearing does occur, the decisions can be appealed to NYMEX’s appeals committee, which makes a final determination within the exchange. Cases then can be appealed further to CFTC, but cases involving an appeal of an exchange appeals committee decision rarely are contested or proceed to a hearing. NYMEX officials said that the commission rarely, if ever, overturns a NYMEX ruling. NYMEX publishes its final disciplinary actions of the exchange. All settlements or adjudications are published in its monthly publication The Open Interest (formerly, Barrels, Bars and BTUs) and sent to the National Futures Association, where they are included in the publicly accessible disciplinary log called “BASIC,” which contains reports from all U.S. futures exchanges and from CFTC. Warning letters are not reported, but they are used internally in prosecuting disciplinary cases. NYMEX also reports its enforcement actions to CFTC. To maintain its designation as a contract market, NYMEX must demonstrate to CFTC its capacity to comply with the CEA’s core principles. In addition, CFTC conducts rule enforcement reviews and publicly reports on how NYMEX exercises its enforcement authority and other areas of operations. In 2004, CFTC reported that NYMEX’s disciplinary program provided reasonable sanctions for a majority of the cases where the exchange took disciplinary action, and that its dispute resolution program had fair and equitable procedures. The report was generally positive and reported that NYMEX’s procedures provided for the recording and safe storage of trade information. Furthermore, NYMEX’s surveillance of trade practices was deemed to be adequate, with thorough and well-documented investigations. NYMEX officials told us that a CFTC rule enforcement review was recently initiated at the exchange. Table 3 reflects information obtained from CFTC showing the energy- related enforcement actions it took by CFTC from August 2001 through September 2006. The enforcement cases were against individuals and companies, and the information used to initiate the investigation originated from both within CFTC and from outside of the commission. The actions mostly focused on attempts to manipulate energy commodity prices through alleged attempts of false reporting; some also involved alleged wash sales or trades—transactions intended to give the appearance that purchases and sales have been made, without incurring market risk or changing the trader’s market position, prearranged trading, and recordkeeping violations. CFTC’s information also shows a wide range of civil monetary penalties. In addition to the individual named above, Godwin Agbara, Kevin Averyt, Ross Campbell, Emily Chalmers, Jordan Corey, Patrick Dynes, Philip Farah, John Forrester, Austin Kelly, Marc Molino, Paul Thompson, Rich Tsuhara, and John Wanska (retired, Assistant Director) made key contributions to this report. Energy Markets: Factors Contributing to Higher Gasoline Prices. GAO-06-412T. Washington, D.C.: February 1, 2006. Natural Gas and Electricity Markets: Federal Government Actions to Improve Private Price Indices and Stakeholder Reaction. GAO-06-275. Washington, D.C.: December 15, 2005. SEC and CFTC Penalties: Continued Progress Made in Collection Efforts, but Greater SEC Management Attention Is Needed. GAO-05-670. Washington, D.C.: August 31, 2005. Mutual Fund Industry: SEC’s Revised Examination Approach Offers Potential Benefits, but Significant Oversight Challenges Remain. GAO-05-415. Washington, D.C.: August 17, 2005. National Energy Policy: Inventory of Major Federal Energy Programs and Status of Policy Recommendations. GAO-05-379. Washington, D.C.: June 10, 2005. Motor Fuels: Understanding the Factors That Influence the Retail Price of Gasoline. GAO-05-525SP. Washington, D.C.: May 2005. Mutual Fund Trading Abuses: SEC Consistently Applied Procedures in Setting Penalties, but Could Strengthen Certain Internal Controls. GAO-05-385. Washington, D.C.: May 16, 2005. Financial Regulation: Industry Changes Prompt Need to Reconsider U.S. Regulatory Structure. GAO-05-61. Washington, D.C.: October 6, 2004. Natural Gas: Domestic Nitrogen Fertilizer Production Depends on Natural Gas Availability and Prices. GAO-03-1148. Washington, D.C.: September 30, 2003. SEC and CFTC Fines Follow-up: Collection Programs Are Improving, but Further Steps Are Warranted. GAO-03-795. Washington, D.C.: July 15, 2003. Natural Gas: Analysis of Changes in Market Price. GAO-03-46. Washington, D.C.: December 18, 2002. SEC and CFTC: Most Fines Collected, but Improvements Needed in the Use of Treasury’s Collection Service. GAO-01-900. Washington, D.C.: July 16, 2001. Energy Markets: Results of Studies Assessing High Electricity Prices in California. GAO-01-857. Washington, D.C.: June 29, 2001. Commodity Exchange Act: Issues Related to the Regulation of Electronic Trading Systems. GAO/GGD-00-99. Washington, D.C.: May 5, 2000. CFTC and SEC: Issues Related to the Shad-Johnson Jurisdictional Accord. GAO/GGD-00-89. Washington, D.C.: April 6, 2000. Financial Regulatory Coordination: The Role and Functioning of the President’s Working Group. GAO/GGD-00-46. Washington, D.C.: January 21, 2000. The Commodity Exchange Act: Issues Related to the Commodity Futures Trading Commission’s Reauthorization. GAO/GGD-99-74. Washington, D.C.: May 5, 1999. CFTC Enforcement: Actions Taken to Strengthen the Division of Enforcement. GAO/GGD-98-193. Washington, D.C.: August 28, 1998. OTC Derivatives: Additional Oversight Could Reduce Costly Sales Practice Disputes. GAO/GGD-98-5. Washington, D.C.: October 2, 1997. CFTC/SEC Enforcement Programs: Status and Potential Impact of a Merger. GAO/T-GGD-96-36. Washington, D.C.: October 25, 1995. Financial Market Regulation: Benefits and Risks of Merging SEC and CFTC. GAO/T-GGD-95-153. Washington, D.C.: May 3, 1995. Energy Security and Policy: Analysis of the Pricing of Crude Oil and Petroleum Products. GAO/RCED-93-17. Washington, D.C.: March 19, 1993. Securities and Futures: How the Markets Developed and How They Are Regulated. GAO/GGD-86-26. Washington, D.C.: May 15, 1986. | Prices for four energy commodities--crude oil, heating oil, unleaded gasoline, and natural gas--have risen substantially since 2002. Some observers believe that higher energy prices are the result of changes in supply and demand. Others believe that increased futures trading activity has also contributed to higher prices. This report, conducted under the Comptroller General of the United States' authority, examines (1) trends and patterns in the physical and energy derivatives markets, (2) the scope of the Commodity Futures Trading Commission's (CFTC) regulatory authority over these markets, and (3) the effectiveness of CFTC's monitoring and detection of market abuses and enforcement. For this work, GAO analyzed futures and large trader data and interviewed market participants, experts, and officials at six federal agencies. Rising energy prices have been attributed to a variety of factors, among them recent trends (2002-2006) in the physical and futures markets. These trends include (1) factors in the physical markets, such as tight supply, rising demand, and a lack of spare production capacity; (2) higher than average, but declining, volatility (a measure of the degree to which prices fluctuate over time) in energy futures prices for crude oil, heating oil, and unleaded gasoline; and (3) growth in several key areas, including the number of noncommercial participants in the futures markets (including hedge funds), the volume of energy futures contracts traded, and the volume of energy derivatives traded outside of traditional futures exchanges. Because these changes took place concurrently, the effect of any individual trend or factor is unclear. On the basis of its authority under the Commodity Exchange Act (CEA), CFTC focuses its oversight primarily on the operations of traditional futures exchanges, such as the New York Mercantile Exchange, Inc. (NYMEX), where energy futures are traded. Energy derivatives are also traded on other markets, namely, exempt commercial and over-the-counter (OTC) markets, that are exempt from CFTC oversight. Both types of markets have seen their volumes climb in recent years. Exempt commercial markets are electronic trading facilities where certain commodities, such as energy, are traded between large, sophisticated participants. OTC markets allow eligible parties to enter into contracts directly, without using an exchange. While the exempt commercial and OTC markets are subject to the CEA's antimanipulation and antifraud provisions and CFTC enforcement of those provisions, some market observers question whether CFTC needs broader authority to oversee these markets. CFTC is currently examining the effects of trading in the regulated and exempt energy markets on price discovery and the scope of its authority over these markets--an issue that will warrant further examination as part of the CFTC reauthorization process. Moreover, because of changes and innovations in the market, the methods used to categorize these data can distort the information reported to the public, which may not be completely accurate or relevant. CFTC conducts daily surveillance of trading on NYMEX that is designed to detect and deter fraudulent or abusive trading practices involving energy futures contracts. To detect abusive practices, such as potential manipulation, CFTC uses various information sources and relies heavily on trading activity data for large market participants. Using this information, CFTC staff may pursue alleged abuse or manipulation. However, because the agency does not maintain complete records of all such allegations, this lack of information makes it difficult to determine the usefulness and extent of these activities. In addition, CFTC's performance measures for enforcement do not fully reflect the program's goals and purposes, which could be addressed by developing additional outcome-based performance measures that more fully reflect progress in meeting the program's overall goals. |
IRS expects individual taxpayers to pay the full amount of tax owed when they file their tax return. However, Section 6159 of the Internal Revenue Code provides for taxpayers to pay their taxes in installments when full payment is not possible. Taxpayers who cannot pay their taxes are subject to IRS’ collection process. IRS’ routine collection process can involve three stages, and taxpayers may enter into an installment agreement during any stage. In the first stage, IRS is to send a series of notices to taxpayers requesting payment. Cases unresolved by these notices may be referred to the second stage—ACS telephone call sites. IRS employees at these sites are to contact taxpayers to secure payments and contact banks and employers to identify levy sources. The employees are to also answer calls from taxpayers who have been subjected to collection actions, such as levies and liens. If ACS call site employees cannot resolve a case, it may be referred to an IRS district office—the third stage in the collection process—for further collection action. IRS made changes to its installment agreement program in April 1992 making it easier for taxpayers to obtain agreements. IRS made the following changes under the revised program. Staff in all offices with taxpayer contact were delegated authority to approve installment agreements of up to $100,000. The thresholds for which IRS can approve an installment agreement without filing a lien increased from $2,000 to $10,000 and without requiring taxpayers to provide financial information demonstrating the need for an agreement increased from $5,000 to $10,000. IRS refers to these agreements as streamlined agreements. IRS allowed taxpayers to request installment agreements when they file by attaching a Form 9465, Installment Agreement Request, or note to their tax returns (see app. I). These agreements, called pre-assessed agreements, do not go through IRS’ routine collection process. IRS made these changes to the installment agreement program to (1) improve voluntary compliance by emphasizing to taxpayers that installment payments are an option available to them to pay their tax debts if full payments cannot be made on time; (2) expedite the IRS collection process by bringing more delinquent accounts into a current status sooner, thereby increasing collections on such accounts in the form of installment payments; and (3) enhance its “one-stop service” concept by allowing staff in all functions with taxpayer contact to approve agreements. In addition, IRS recently developed procedures to implement another change in the program—charging user fees for installment agreements. The Treasury, Postal Service, and General Government Appropriations Act of 1995 (P.L. 103-329) authorized the Secretary of the Treasury to supplement IRS’ appropriations through the imposition of user fees for services performed by IRS. As a result, in March 1995, IRS began charging a $43 user fee for a new installment agreement and a $24 user fee for amending or reinstating an existing agreement. Our objectives were to review (1) the increase in installment agreements following changes IRS made in April 1992; (2) the effects these changes had on IRS’ collection activity; (3) concerns raised by IRS’ internal auditors regarding these changes; and (4) administrative aspects of the program, such as the information IRS provides taxpayers regarding installment agreements, and opportunities for improvement. We also collected descriptive information on taxpayers who elected to pay past due taxes through installment agreements. To address the four objectives, we did the following: We met with officials in IRS’ National Office of the Assistant Commissioner for Collections to discuss IRS’ objectives for the installment program and gather information about the program’s administration and progress. We met with IRS officials in the Cincinnati Service Center (Collections Branch) and the Cincinnati District Office (including the Taxpayer Service, Collection, and Examination divisions) to obtain information on the experiences of these IRS personnel with the revised procedures. We selected the Cincinnati Service Center because we had available experienced staff in that area. We discussed the installment agreement program changes with members of the American Institute of Certified Public Accountants, the Federation of State Tax Administrators, the National Consumer Law Center, the Consumer Federation of America, and a Cincinnati-based commercial bank. We reviewed IRS internal audit and management reports concerning the installment agreement program. To accomplish our objective of collecting descriptive information on taxpayers, we analyzed the 722 useable cases from our random sample of 900 installment agreements from IRS’ Service Centers in Atlanta, GA; Cincinnati, OH; and Fresno, CA. The sample was taken from the inventory of 628,285 agreements at the 3 service centers as of April 23, 1994. In reviewing the sample, we obtained information about the agreements, such as (1) the tax periods covered by the agreements; (2) the original amount and current balance of the agreements; (3) the agreement payment history; and (4) certain taxpayer financial characteristics, such as adjusted gross income and source of income. The sample results were indicative of the installment agreement inventory at the three IRS service centers as of April 23, 1994. However, the results are not projectable to IRS’ total universe of installment agreements and may not be indicative of the inventory at any other point in time. Appendix II describes our sampling methodology in more detail. We did our audit work from January 1994 through November 1994 in accordance with generally accepted government auditing standards. On March 27, 1995, we met with IRS National Office officials to obtain their comments on a draft of this report. IRS representatives at that meeting included the Acting Assistant Commissioner, Collection. Their comments are summarized on page 16 and incorporated in this report where appropriate. As stated earlier, the installment agreement program grew rapidly during fiscal years 1991 through 1994. As shown in table 1, significant growth took place between fiscal years 1992 and 1993. While part of this growth may be attributable to the April 1992 program changes, IRS officials said that changes to federal tax withholding tables in 1992 caused many taxpayers who would normally have received refunds to owe taxes when they filed their returns in 1993. IRS estimated that installment agreement requests would increase by about 39 percent because of the changes to the withholding tables. Since 1992, installment agreements have played an increasing role in IRS’ collection process. Table 2 shows that installment agreements have made up an increasing percentage of IRS’ accounts receivable inventory for individual taxpayers since 1992. Table 3 shows that installment agreements have accounted for an increasing percentage of IRS’ total collections. IRS National Office collection officials said that the increase in total collections between fiscal years 1993 and 1994 was a positive result of the changes made to the installment agreement program. In raising service center and taxpayer service authority to grant installment agreements from $5,000 to $10,000, IRS wanted to shift some of the collection work from ACS call sites to service center collection and district taxpayer service offices. Some shift in workload has occurred. Taxpayer service and service center staffs originated about 82 percent of all agreements in fiscal year 1994, up from 43 percent in fiscal year 1991. Most of the increase was due to these two staffs granting streamlined agreements. ACS call sites granted fewer agreements in fiscal year 1994 (about 381,000 or 15 percent of agreements) than in fiscal year 1991 (about 579,000 or 53 percent of agreements). In shifting the installment agreement workload to service center and taxpayer service staff, IRS hoped to increase ACS collections by freeing up ACS staff to work on higher dollar cases. However, the expectation of increased collections had not materialized since ACS collections per staff year dropped between fiscal years 1992 and 1994 from $1.71 million to $1.44 million in current year dollars. Overall, the amount collected at ACS sites decreased from $4.88 billion to $3.30 billion during the same period. IRS collection officials said that the lower ACS collections are due, in part, to their decision during this same period to increase the deferral level—the amount under which delinquent accounts are not pursued beyond the notice stage, except to offset refunds. According to IRS officials, this increase reduced ACS workloads more than expected and contributed to reduced ACS collections. IRS subsequently lowered its deferral level in April 1994, a move IRS officials expect will increase ACS workloads and the resulting collections. Another indication of the greater impact the installment agreement program is having on collections was the increased number of balance due accounts that are being resolved earlier in the collection process. IRS data show that more balance due accounts are being favorably resolved by taxpayers paying or arranging to pay the amounts owed without IRS resorting to collection notices and subsequent collection actions. Between fiscal years 1991 and 1994, accounts favorably resolved before a collection notice rose from 41 percent to 52 percent, with installment agreements accounting for a growing share of the favorable early resolutions. In fiscal year 1994, installment agreements accounted for 41 percent of the favorable early resolutions, up from 14 percent in fiscal year 1991. IRS’ internal audit raised several questions regarding the installment agreement program since IRS implemented the April 1992 changes. In September 1994, IRS’ internal auditors raised concerns about whether (1) the new procedures for granting installment agreements were encouraging taxpayers to choose installments when they could be paying their taxes in full and (2) the program was leading some taxpayers to accumulate debt they may not be able to pay. The installment agreement program was established to provide taxpayers who cannot pay their taxes in full, when they are due, an opportunity to pay in installments. Before the program was revised in April 1992, taxpayers whose agreements were for tax debts above $5,000 were required to provide financial data to substantiate their inability to pay in full and on time. Under its streamlined procedures, IRS relies on taxpayers to determine for themselves whether they can pay in full or want to pay in installments, as long as the amount owed does not exceed $10,000 and the payment period meets IRS requirements. IRS recognizes that this may result in some taxpayers opting for installment agreements even though they could have paid in full and on time. IRS collection officials told us they did not know the extent to which this occurs, but they believed the benefits from streamlining outweigh any unintended negative effects of making installment agreements available to taxpayers who can pay in full. This issue was raised by IRS’ internal auditors in their September 1994 report on using and processing installment agreement requests received with tax returns. The auditors reviewed a sample of 2,824 pre-assessed installment agreements—those agreements requested by taxpayers when they filed their tax returns—granted in 1993 and found that 22 percent of the taxpayers had paid in full with their return the previous year, while another 10 percent had fully paid later during the collection notice process. The auditors reasoned that on the basis of the taxpayer’s payment history, some of these taxpayers could have paid their taxes in full in the year they entered into an agreement. In addition, to obtain more conclusive information on whether some taxpayers who entered into installment agreements could have paid their taxes on time, the auditors contacted 87 taxpayers who received pre-assessed agreements. The auditors were told by 41 percent of these taxpayers that they could have fully paid their accounts by withdrawing savings, liquidating assets, or borrowing. In response to IRS’ internal audit findings, IRS has begun taking steps to discourage taxpayers who can pay in full from using installment agreements. For example, IRS revised Form 9465 to advise taxpayers to consider other alternatives, such as bank loans, before requesting an installment agreement when filing a tax return. IRS’ streamlined procedures allow for new tax debt to be added to an existing installment agreement without a review of a taxpayer’s financial condition or of possible remedies to incurring future debt, provided the new aggregate balance does not exceed $10,000 and the payment period meets IRS requirements. IRS data show that about 16 percent of the agreements granted in fiscal year 1994 were added to existing installment debt. In their September 1994 report, IRS internal auditors expressed concern that IRS procedures allowed taxpayers to incur tax debts beyond what they could reasonably be expected to pay. The auditors stated that allowing taxpayers to add tax debt beyond what could be paid within a reasonable time would increase IRS’ accounts receivable balance and its costs to collect taxes. According to an IRS official, one explanation for why the practice of accumulating debt was allowed is that IRS accommodated taxpayers who had balance due accounts because of 1992 changes in the withholding tables. During that period, IRS assumed that many taxpayers would owe taxes unexpectedly and allowed taxpayers to pay those taxes by adding debt to existing agreements under streamlined procedures. That policy continued through 1994 when, as a result of concerns from IRS’ internal auditors, IRS took several steps to address the issue for the 1995 tax filing season. For example, the monthly installment agreement reminder notices for December 1994 included a special reminder to taxpayers. The reminder stated that under the terms of their existing agreements the taxpayers were not eligible for another installment agreement while their existing agreements were in effect and that all federal tax liabilities must be paid in full or they would be in default. Also, Form 9465 was revised to inform taxpayers that they are not to use the form if they were currently making installment payments and that the agreement would be in default if all liabilities were not paid in full. In addition, in January 1995, IRS issued procedures to field personnel working installment agreements to prevent taxpayers from adding new tax debt to existing liabilities. IRS plans to revisit this issue before the 1996 tax filing season. IRS internal auditors also reviewed whether taxpayers with installment agreements acted on their own to avoid the need for future agreements. In doing so, IRS contacted 87 taxpayers who were granted pre-assessed installment agreements and found that 39 percent had not adjusted their withholding or estimated payments to avoid owing money on successive tax returns. Although acceptance letters to taxpayers contain instructions for making such adjustments, no such instructions appear on Form 9465, which is the initial form completed by taxpayers who request pre-assessed agreements. The auditors recommended that IRS include a section on Form 9465 for taxpayers to identify the cause of their tax debts and the steps taken to ensure that the condition does not recur the following year. Additionally, IRS’ Research Division is studying a group of taxpayers to determine the cause for their delinquencies and through that study intends to determine how best to solve the problem of underwithholding and underestimated payments. As we discussed earlier, the installment agreement program grew rapidly during fiscal years 1991 through 1994. Because of this growth, we believe the administrative operation and efficiency of the program has become increasingly important. In that regard, we reviewed certain aspects of the program and offer several suggestions that could lead to more efficient administration. Specifically, we are concerned about the (1) lack of information taxpayers receive about the length and costs of installment agreements, (2) extent IRS has taken advantage of opportunities to improve program efficiency, and (3) amount of installment agreement debt by taxpayers with agreements lasting more than 5 years. During the term of an agreement, a taxpayer continues to accrue interest and penalty charges on the unpaid balance of the debt. IRS advises taxpayers in its agreement application forms and acceptance letters that interest and penalties will continue to accrue while they are making installment payments. However, IRS does not tell taxpayers the total estimated cost of the agreement, including interest and penalty accruals, nor does IRS tell taxpayers how long it will take to pay off their debt. The lack of information provided by IRS contrasts with prevailing private sector practices, which are governed by truth-in-lending laws. Penalty and interest accruals add considerably to the cost and payoff period of an agreement. For example, assume a taxpayer agrees to make $100 monthly payments on a $2,800 tax debt (the median amounts from our sample). To pay off interest and penalty accruals, the taxpayer would need to make 6 additional monthly payments (34 payments versus 28 payments) and pay an additional $544 (assuming the 0.5 percent per month penalty rate and monthly compounding of interest at 9 percent). Under the Internal Revenue Code, the rate of interest is determined on a quarterly basis computed from the federal short-term rate based on daily compounding. Representatives of consumer groups told us that providing taxpayers with better cost information could lead some taxpayers to consider and use other options to pay their tax debts before choosing installment agreements or assuming long payoff periods. By providing more information to taxpayers on the cost of agreements, IRS may be able to limit agreement use more closely to taxpayers who are unable to pay their taxes on time in a lump sum. IRS officials agreed that more information would be a factor in deterring some taxpayers with other available resources from using IRS’ installment agreement program to pay their taxes. However, IRS officials are reluctant to provide specific cost and payoff period projections to installment agreement applicants. The officials are concerned that the estimates they provide at the beginning of an agreement could be misleading because these estimates could change over the life of the agreement. For example, the estimates could be affected by interest rates that are subject to quarterly adjustment and by missed taxpayer payments. A method to alleviate this concern would be to include revised cost and payoff period projections in the monthly statements that IRS sends taxpayers during the course of an agreement. These notices already include information specific to a taxpayer’s agreement, such as the amount of payment due and the current unpaid balance. IRS offers taxpayers the option of making installment payments automatically by directly withdrawing funds from the taxpayer’s bank account. According to IRS, the advantages to direct debit agreements are they (1) are cheaper to service than the standard agreement, (2) eliminate the chance that a taxpayer will forget to make a payment or send less than the agreed upon payment amount, (3) eliminate the float time associated with processing paper remittances, and (4) may reduce default rates. Although direct debit agreements offer these advantages to IRS, they made up less than 2 percent of the agreements in IRS’ inventory at the end of fiscal year 1994. IRS collection officials explained that one reason for the low rate of direct debits is that staff who set up installment agreements view direct debits as additional work because the processing involves performing more steps and gathering more information from taxpayers. According to IRS data, pre-assessed installment agreements accounted for almost 31 percent of the 2.6 million new agreements in fiscal year 1994. Form 9465, the form used by taxpayers to request pre-assessed agreements, does not mention the direct debit option (see app. I), leaving it up to IRS staff to pursue this option later. As noted, however, IRS collections officials indicated that this reliance on staff has limited the number of direct debits. Because of the advantages of the direct debit payment method, it may be worthwhile for IRS to test a revision of the form to include space for taxpayers to authorize direct debits and supply the information necessary to set them up. When taxpayers default on their installment agreements, service center staff must review the taxpayers’ accounts and send them a default notice. That notice is a statement informing the taxpayer that the account is not current and, if not corrected, the taxpayer is subject to levy action. Depending on how taxpayers respond to these notices, IRS staff may reinstate the agreement; take some other collection action, such as a levy or lien filing; or place the account in a deferred status where only passive collection actions, such as refund offsets are taken. One service center has suggested that some of the costs incurred on defaulted agreements may be reduced if IRS could send default notices by regular mail rather than certified mail. This change would replace a $1.29 letter with a 23-cent letter (presorted rate) and also reduce some handling costs. Since defaults are not uncommon, the savings in postage costs could be significant. The service center estimated its own annual savings would be nearly $150,000. Our analysis suggests that the service center suggestion has merit. We realize that some default notices may still need to be sent by certified mail because they are used to give taxpayers notice of impending levy actions against their assets. Section 6331(d) of the Internal Revenue Code requires that IRS inform taxpayers in person or by certified mail 30 days before taking levy actions. Other defaulted installment agreements, however, are placed in deferred status, and no levy action is pending because collection action is limited to periodic notices and offsets against future refunds. Our review of the October 1 through 7, 1994, listing of 1,933 defaulted agreements in the Central Region showed that 62 percent of these agreements were below the deferral level. IRS officials agreed that taxpayers in these instances would not need to be sent default notices by certified mail. The officials added that notices for cases subject to levies could continue to be coded to send by certified mail. According to IRS data, in 1994 the average installment agreement was paid off in about 9.5 months. However, a large portion of IRS’ installment debt is for agreements that will run for more than 5 years if payments are made at their current levels. Such agreements are costly to administer, and IRS considers them riskier than shorter term agreements. IRS’ manual states that installment agreements lasting more than 5 years are not likely to be paid off. Thus, these protracted agreements are candidates for other options, such as an offer in compromise, which is a program that allows taxpayers to liquidate their tax liability through a lump-sum settlement for less than the amount owed. About 18 percent of the agreements we reviewed from three service centers were protracted agreements. These agreements accounted for approximately 53 percent, or $846 million, of the installment debt owed at the three centers. For these agreements, the median balance owed was $5,780, the median monthly payment was $100, and the projected median payoff period was 7.6 years. Some of the protracted agreements in our sample are unlikely to ever be paid off because current payments were not keeping up with accruing interest. These agreements made up about 16 percent of the protracted agreements we reviewed, with a current unpaid balance totaling $292 million. The median monthly payment on these agreements was $50, and the median unpaid balance was $17,530. At IRS’ 9-percent annual interest rate as of December 1994, 1 month’s interest on the median unpaid balance amounted to $131. IRS officials acknowledged the existence of these types of agreements in their inventory, adding that sometimes accepting a taxpayer’s installment payment may be its best option. The officials explained that these payments often represented amounts that may otherwise not be collected. IRS’ procedures require staff to periodically review agreements with terms exceeding 3 years to reevaluate taxpayers’ financial conditions. Presumably, if the review indicated a better payment option, IRS would pursue it. Since 1991, taxpayer use of installment agreements has grown considerably, and installment agreements have accounted for a growing portion of IRS’ collection activity. The program grew more rapidly after IRS made changes in April 1992. IRS internal auditors, however, reported that some taxpayers are using installment agreements even though they were able to fully pay their taxes. This practice conflicts with IRS’ intent to reserve installment agreements for taxpayers who cannot otherwise pay their taxes in full when they are due. The auditors also raised concerns about the ease with which taxpayers could accumulate additional tax debt by adding new income tax liabilities to existing agreements. IRS has acted to address the internal auditors’ concerns. Our work surfaced several other concerns about certain administrative aspects of the program. For example, IRS does not tell taxpayers the extent that interest and penalty accruals will add to the costs of installment agreements and the payoff time. Providing this information may influence some installment payers to pay in full or make larger monthly payments. In addition, IRS could reduce some of the administrative costs of servicing agreements by (1) encouraging more taxpayers to make installment payments by direct debit and (2) mailing some default notices by regular mail instead of certified mail. Our sample of installment agreements contained a substantial amount of installment debt associated with agreements having payoff periods longer than 5 years. This length of time makes collection of the debt risky and more expensive to administer, according to IRS. IRS reviews agreements every 3 years to explore payment options. To improve the information provided to taxpayers and the administration of the installment agreement program, we recommend that the Commissioner of Internal Revenue notify taxpayers about projected total costs and payoff periods when setting up agreements with taxpayers and when mailing monthly reminder notices, experiment with Form 9465 to test whether having space for taxpayers to authorize direct debit installment payments increases the frequency with which this option is used, and send agreement default notices to taxpayers by regular mail instead of certified mail unless an account is being referred for levy action. Responsible IRS officials, including the Acting Assistant Commissioner, Collection, reviewed a draft of this report and provided oral comments in a meeting on March 27, 1995. The officials said that the report was a fair and accurate assessment of the installment agreement program and that they generally agreed with our recommendations. In response to the recommendations, the Acting Assistant Commissioner said that: IRS will study the feasibility of notifying taxpayers about the projected costs and payment periods of installment agreements. If notification is not feasible under its existing computer systems, IRS will pursue changes as part of its Tax Systems Modernization. In the interim, IRS plans to modify the monthly reminder notice in 1996 to provide taxpayers with a breakdown of the current balance due and penalty and interest charges. IRS will consider options, including modification of Form 9465, to encourage taxpayers to authorize direct debit payments on their installment agreements. IRS will develop methods to identify defaulted installment agreement accounts it does not intend to take levy action against and send default notices to these taxpayers by regular mail. We believe the actions that IRS proposes, if properly implemented, would be responsive to our recommendations. We are sending copies of this report to other congressional committees; the Secretary of the Treasury; the Commissioner of Internal Revenue; the Director, Office of Management and Budget; and other interested parties. We will make copies available to others upon request. The major contributors to this report are listed in appendix III. If you or your staff have any questions, you can reach me at (202) 512-5407. To profile characteristics of taxpayers with installment agreements, we used the active inventory of installment agreements as of April 23, 1994, at three IRS service centers—Atlanta, GA; Cincinnati, OH; and Fresno, CA. We selected these centers because of their larger inventories and the availability of our staff to review the sample. We identified active installment agreements by obtaining IRS’ Taxpayer Service and Returns Processing Categorization of Accounts Receivable (TRCAT) file for each of the three service centers. We isolated installment agreements from the individual master file and then took a stratified random sample of 900 installment agreements from the TRCAT file—300 from each service center. We were able to analyze 722 of these cases. Table II.1 gives information on the total number of cases we sampled from the three service centers. Since we used a probability sample of installment agreements to develop our estimates, each estimate has a measurable precision or sampling error that may be expressed as an upper and lower limit. A sampling error indicates how closely we can reproduce from a sample the results that we would obtain if we were to take a complete count of the study population using the same measurement methods. The difference between the upper and lower limits is called a confidence interval. Sampling errors and confidence intervals are stated at a certain confidence level—in this case 95 percent. For example, a confidence interval at the 95-percent confidence level means that in 95 out of 100 instances, the sampling procedures we used would produce a confidence interval containing the population value we are estimating. We attempted to gather profile information from IRS’ records on all 900 sample cases. Due to a 2-month time lag in processing sample information and obtaining the records from IRS’ Integrated Data Retrieval System, we were unable to obtain profile data for certain cases. Therefore, we excluded these cases from our sample analysis. We found the majority of the excluded cases involved installment agreements that had been paid off by the taxpayer during the period we were gathering our data. We also excluded installment agreements cases that had been paid off but a new agreement was created during the 2-month lag in obtaining IRS records because these new cases were technically not part of the April 23, 1994, inventory. We excluded other agreements granted to pay off trust fund recovery penalties because these types of debts are not related to payments of personal income taxes and would have biased our analysis. Finally, we excluded cases that IRS was unable to locate in its records. Table II.2 gives the reasons for excluding sample cases and the estimate for each category of excluded cases. Table II.3 contains profile information we gathered on the sample installment agreements taken from the TRCAT file. This file is a snapshot of the installment agreements IRS was managing on April 23, 1994. The sample results may not be indicative of all installment agreements managed by IRS at the three service centers at any other point in time. Median age of agreements (months) Median payoff period (months) Median number of tax periods in Percent of agreements with added tax Percent of agreements covering: Three or more tax years Percent of agreements that have been in default status at least once Reason agreement was added (percent): Collection status when entering agreement (percent): Agreements granted for tax year 1992 only Median adjusted gross income of Percent with adjusted gross income over (continued) Percent of total agreements made up by Percent of total unpaid balance made up agreements (millions) Median unpaid balance for protracted Median monthly payment for protracted Median payoff period for protracted agreements (years) Percent of protracted agreements made up by agreements with monthly payments insufficient to keep up with accruals Unpaid balance on agreements with insufficient payments (millions) Robert I. Lidman, Regional Assignment Manager Richard C. Edwards, Evaluator-in-Charge Donald L. Allgyer, Evaluator Jennifer C. Jones, Evaluator Mary Jo Lewnard, Technical Advisor Kenneth R. Libbey, Evaluator The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 6015 Gaithersburg, MD 20884-6015 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (301) 258-4066, or TDD (301) 413-0006. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (301) 258-4097 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO reviewed the Internal Revenue Service's (IRS) Installment Agreement Program, focusing on the: (1) increase in installment agreements since IRS streamlined the processing and approval of taxpayers' requests for installment agreements; (2) effects these changes had on IRS collection activities; (3) internal auditors' concerns regarding these changes; (4) information that IRS provides taxpayers on their liabilities under installment agreements; and (5) administrative practices that could improve the installment agreement program. GAO found that: (1) the 2.6 million new installment agreements approved in fiscal year (FY) 1994 represented a 136-percent increase over the number of FY 1991 installment agreements; (2) the amount of taxes paid through new installment agreements increased 135 percent between FY 1991 and 1994 and accounted for 33 percent of the delinquent taxes paid in FY 1994; (3) installment agreement program changes have affected IRS collection activities by reducing Automated Collection System collections and the routine collection process workload; (4) IRS internal auditors have raised concerns about the ease of entering into installment agreements and IRS failure to instruct taxpayers to amend their withholding or estimated taxes to prevent future delinquencies; (5) IRS may be permitting financially capable taxpayers to avoid paying their tax debts in one on-time payment and to accumulate tax debts by adding new tax balances to existing agreements; (6) IRS is taking steps to reduce the problems the auditors identified; and (7) administrative changes to improve the installment agreement program and reduce costs include informing taxpayers of the applicable penalties and interest that will be added to their installment agreements, allowing taxpayers to make electronic direct debit payments, and sending some default notices by regular mail instead of certified mail. |
BIA administers funding for the operation, maintenance, construction, and repair of school facilities at 171 elementary and secondary schools in 23 states. These schools are located primarily in rural areas and small towns and serve Indian students living on or near reservations. Many of these schools include not only educational buildings, but also dormitories and supporting infrastructure such as water and sewer systems. BIA operates 64 of the schools directly while the others are operated by tribes through separate grant or contract agreements. We previously reported on issues related to the condition of BIA school facilities in 1997 and 2001. BIA’s Office of Facilities Management and Construction (OFMC) is responsible for overseeing FMIS. At both BIA-operated and tribal-operated schools, it is the responsibility of the facility managers to enter data about the inventory and condition of their schools into the system. Prior to acceptance into FMIS, these draft data entries are reviewed and approved by facility managers at BIA agency and regional offices respectively, before final review and approval by a BIA contractor and BIA’s central office. For 22 years BIA relied on its Facility Construction Operations and Maintenance (FACCOM) system to maintain inventory data for its annual O&M program, as well as “backlog” data that reflect repairs and improvements needed outside of the annual maintenance program to improve the facilities’ condition now and in the future. These data assist BIA in monitoring the status of facilities repair and new construction projects and identifying funding needs for O&M and renovation. However, as BIA’s needs began to change, BIA managers realized that FACCOM had limitations and acknowledged that there were serious concerns with the accuracy and completeness of these data. As shown in figure 1, BIA’s efforts to replace FACCOM began in 1995 when one of its contractors issued a report about the FACCOM system’s shortcomings and recommended actions for improvement. In 1995, BIA entered into a contract with Anteon Corporation, a system developer, to design the new management information system. Relying on government standards, BIA worked with Anteon Corporation to design the new system and address FACCOM’s shortcomings. In 1999, BIA contracted with an engineering firm, Applied Management Engineering, Inc. (AME), to conduct a survey at all school sites in order to validate the schools’ condition data and to verify the presence of buildings and their use. According to BIA officials, after AME validated each school’s inventory and condition data, and BIA approved it, the data were accepted into FMIS. From fiscal years 1995 through 2002, BIA spent nearly $12.5 million to develop and begin implementing FMIS. These costs include about $8 million for contractor expenses and over $2.6 million for BIA in-house expenses, which covered the design of FMIS and ongoing technical support. During fiscal years 1999- 2003, BIA spent about $13 million for the AME contract covering the validation of inventory and condition data and other engineering support activities. To operate FMIS, BIA expects to spend about $1.7 million annually through fiscal year 2006 for contractor expenses and about $250,000 for in-house expenses. In fiscal year 2007, BIA hopes to move to an annual steady rate of about $750,000 for contractor costs and about $250,000 for in-house costs. To continue having AME reassess and validate the schools’ inventory and condition data, BIA projects to spend over $8.3 million from fiscal years 2004 through 2006 on contract expenses. Recognizing the FACCOM system’s shortcomings, BIA worked with its system developer to design a new management information system that would assist in resolving many of the weaknesses identified with the old system, including those related to difficulty of use and accuracy of data. FMIS is more user friendly and it is designed to meet facility managers’ needs at all levels within BIA by serving as both an information management system and as a project management tool. FMIS incorporates modules, including the inventory and backlog modules, which help facility managers make decisions regarding the condition of the school facilities to provide a safe environment for their students. The inventory module contains information such as the physical characteristics and use of buildings and is used to make funding decisions for annual operating expenses and routine maintenance. The backlog module contains data that tracks detailed information about the physical condition of a school’s facility and is used to prioritize and fund repair projects, capital improvements, and construction. FMIS is designed to better support the day-to-day activities of the facility management staff by being more user friendly. FMIS is a Windows-based system that provides a point-and-click feature, which makes it easy to navigate the system without having to remember codes. This is important for FMIS users, because some facility managers have little prior exposure to computers. Facility managers we interviewed at the school sites and agency and regional offices said that compared to FACCOM, FMIS is better and easier to use. One facility manager in Arizona said FMIS is easier to use because the system automatically sends messages to him when changes are made to the data, allowing him to instantly see when updates have been made to his school’s data. Another facility manager in Arizona said FMIS’s automated functions, such as drop-down menus, make it user friendly. Finally, according to BIA officials, FACCOM was only accessible by about 3-4 percent of facility managers, which did not include facility managers located at the schools. FMIS is designed to be accessible by all of BIA’s facility managers, including those at the school sites, via an Internet connection. Although most of the facility managers we visited at the schools said FMIS was better compared to FACCOM, many had been unable to access FMIS at school sites since December of 2001 due to a court order that shut down access to BIA’s Internet site. These facility managers had to travel to agency or regional offices to enter data or had to forward data to these offices for data entry. FMIS is also designed to help BIA employees improve the accuracy of the data, in part through automated mechanisms that help facility managers consistently describe the category and rank of backlog entries and the funding needed to address them. One difficulty under FACCOM was that entries listed in the backlog were often categorized incorrectly with inflated priorities, making it difficult to determine which projects needed immediate attention. To address this problem, FMIS is designed to restrict who can enter safety deficiencies, which are given first priority for funding. In addition, FMIS is designed to only accept entries that have been reviewed and approved by three different levels of BIA management and the contractor. Further, BIA refined the definitions of how backlog items should be categorized and ranked to help facility managers use the definitions consistently. The definitions now include nine categories and a ranking system for determining the priority of the items entered into the backlog (see table 1). These nine categories describe whether the deficiency at the school affects safety and endangers students’ lives; violates an environmental, disability, or energy standard; is a maintenance or capital improvement item; or requires an emergency repair. In addition, the system ranks items using a scale from one through three—with one describing the most severe deficiency. The backlog entries ranked as a “1” will most likely be funded first because they are the highest priority. FMIS requires facility managers to enter an associated category and rank for all items entered into the backlog. Such a process, with its greater specificity in how to categorize and rank deficiencies, can help facility managers improve consistency in the data entered for all 171 schools. U-1—an unforeseen event in which danger exists that could reasonably be expected to cause death, physical harm, or property damage. S-1—serious deficiency that poses a threat to safety and health, such as fire safety violations. S-2—moderate deficiency such as poor lighting or trip or fall hazards. (maintenance) M-1—deficiency related to the structural, mechanical, or electrical systems that render it inoperable, such as the deterioration of a roof that causes interior building damage. M-2—deficiency related to the facility, systems, or grounds, such as replacing worn door locks that are inoperable M-3—functional facility equipment exceeds its normal life expectancy. H-1—serious code deficiency, such as the lack of accessible door hardware. H-2—violation of codes and standards, such as the lack of code compliant accessible handrails. X-1—serious code deficiency that poses a threat to life or property, such as removing friable asbestos in occupied areas. X-2—code deficiency, such as removing asbestos floor tiles from a building. R-3—backlogs identified for future planning to determine the life cycle needs beyond the 5-year plan. C-1—to replace buildings with serious code/safety deficiencies or to abate numerous high cost code violations that meet or exceed the replacement cost rule. C-2—to accommodate functional or programmatic needs, such as replacing an undersized dining room to accommodate the student population. P-2—to change the functional space or to accommodate programmatic space needs, such as retrofitting an existing classroom into a computer laboratory. E-2—violation of energy codes and standards, such as upgrading or replacing inefficient heating systems. E-3—deficiencies, which when corrected will reduce energy use, such as replacing weather seals on exterior doors. To help facility managers develop accurate and consistent cost estimates to address backlog items, FMIS is designed to operate with a software program that helps facility managers accomplish industry standard cost estimates for replacement, renovation, or construction projects over $5,000. A facility manager at a school site we visited said that this software tool eliminates the need to make calculations by hand, and thus greatly assists him in estimating accurate costs for school projects. Another accuracy-related area that plagued FACCOM was that projects would continue on the active backlog list even after completion. The FMIS backlog module is designed with a “backlog completion screen” that stores completion dates, costs, and narrative comments. This function helps facility managers monitor the length of time that funded backlog items remained on the backlog without being completed. Unlike FACCOM, FMIS is designed with a tracking function that identifies the name of the person who entered or updated a particular backlog item. Managers can use this function to seek clarification or justification for items that have not been completed within a reasonable timeframe. The FMIS inventory module, one of six modules in the system, is designed to assist BIA in determining operations and maintenance funding for its school facilities. Specifically, the inventory module is designed to manage information about all of BIA’s school buildings, rooms, towers (such as a water tower), and grounds, along with their associated inventory items, such as stairs, sidewalks, or playgrounds. The inventory module also details if the property is owned, operated, or maintained by BIA directly or under contract or grant. The inventory module is designed with the capacity to integrate with other FMIS modules in order to generate reports and provide detailed documentation for federal funding purposes. At the schools we visited, several facility managers said the inventory function in FMIS helps them to better manage their school facilities. For example, one school facility manager in Arizona said FMIS helps him more accurately keep track of his school’s inventory and allowed him to enter the information that is necessary for BIA to make good funding decisions. Another facility manager in South Dakota said that because only one staff person can enter information into the inventory module, his school is able to maintain consistency in inventory changes and additions. The FMIS backlog module is designed to help BIA officials prioritize and make funding decisions for needed repair projects and capital improvements. The backlog collects and tracks condition data related to deficiencies, capital improvements, or construction for specific inventory items, such as classrooms, sidewalks, or utility systems. These data are entered into FMIS by a facility manager, by safety officers as a part of a safety inspection, or by BIA’s contractor. Although the backlog module can store information about any identified deficiency, the only items that are reported as part of BIA’s backlog are those with an estimated cost of more than $1,000 to fix. These deficiencies, which may be grouped together to form repair, replacement, or construction projects are maintained in the backlog until funded and complete. During our site visits, many of the facility managers said the backlog module helped them to better manage their facilities. For example, one school facility manager in Arizona said FMIS’s ability to store digital pictures was helpful because a picture of a deficiency could be sent to the regional facility manager and reviewed without the facility manager traveling nearly 284 miles to the school. Facility managers at schools use the information in the backlog module to justify funding needs for repair projects and capital improvements at their schools. BIA management officials allocating funding among the schools said the data in the backlog module allow them to determine which deficiencies are related to student safety and need to be addressed immediately, and which are related to capital improvements, such as roof replacement, that are planned for the future. BIA officials said they use the backlog data to help improve the physical condition of their schools in order to provide a safe and healthy learning environment for the students. BIA’s engineering contractor has corrected the inventory and backlog data that existed in the old data system, but BIA has not transferred all corrected information to the new FMIS. The contractor completed its validation of the inventory data in February 2003, and BIA plans to transfer the corrected data into the FMIS by August 2003. BIA officials expect that the corrected inventory data, in conjunction with improvement to the existing funding formula, will enhance their ability to better match funding with annual expenses for utilities and routine maintenance at each school site. For the backlog data, BIA’s contractor is in the third year of the second cycle review of the condition of BIA-funded schools as planned. In fiscal year 2002, the contractor visited 33 schools and identified corrections to add, delete, or adjust the FMIS data. BIA, however, had not entered the results of these condition assessments into FMIS for over 1 year. BIA officials attribute the long delays in correcting the FMIS condition data to a revised process for verifying contractor data and to software compatibility problems that they say are being addressed. The FMIS inventory module contains data on BIA facilities, including almost 2,200 separate buildings that are occupied by or used for BIA- funded schools. More than 50 percent (1,146) of the buildings are used directly by children, as shown in table 2. Accurate and up-to-date inventory data are crucial to the operation of the entire FMIS because other modules rely on inventory data for planning and prioritizing the work and for identifying and prioritizing funding needs. For example: School facility managers access inventory data when planning and scheduling routine maintenance of their facilities, grounds, and equipment. For example, one FMIS module acts as a scheduling tool to inform facilities managers about work, such as preventive maintenance, that needs to be done to buildings, equipment, and other physical assets listed in the inventory. BIA’s Office of Facilities Management and Construction uses inventory data in the formula that determines the amount of operations and maintenance funding allocated to each school location. Distribution of this funding is calculated using a formula that includes such inventory data as the square footage of rooms in each building and systems that support the facility such as heating and cooling systems. Funding distributions have been a particular source of contention for Office of Indian Education officials, who told us that inaccuracies in the inventory data have led to inequities in how the money is apportioned. Accurate data that are collected using a methodology that is consistent from site to site is a necessary component for demonstrating the fairness of the process. BIA’s engineering contractor, AME, has remained on schedule in its effort to improve the accuracy of the inventory information currently in FMIS. In conducting this effort, AME visited each school and collected inventory data using a standardized, industry-based approach to help ensure that information on all facilities is uniformly collected and recorded. AME completed the first phase of this effort in 2000, when it visited each school to verify and update the inventory data listed in the FACCOM before its transfer to FMIS in that year, according to a BIA official. This phase, which focused on the more general aspects of the inventory, was aimed at such matters as identifying which buildings were still in use, the use of the facility, and who owns it. AME completed the second phase of the improvement effort in February 2003. This second phase, which took longer than the first phase, involved a more extensive measurement of the buildings and updated the drawings of floor space, grounds, infrastructure, and utility lines. BIA’s current plans call for replacing the existing data in FMIS with this updated and corrected data by August 2003. Our preliminary review of the new data generated during the second phase of AME’s work indicates that the inventory figures may change considerably for some schools. At our request, AME officials provided revised square footage data for more than 90 buildings (such as classroom buildings, dormitories, multipurpose buildings, and offices) at 13 different schools. Overall, the revised measurements for these buildings decreased the total square footage by about 3 percent, but the range in increases and decreases at each school varied significantly. For 9 of the schools the decrease in square footage ranged from less than 1 percent to more than 13 percent; increases for the remaining 4 schools, ranged from less than 1 percent to almost 18 percent. We do not know if these results will be typical for all schools. One BIA official indicated, however, that some schools were likely to experience greater changes than others in the square footage that would qualify for O&M funding. BIA officials said that the improved data, along with improvements to the funding formula, will help ensure the various schools that their share of O&M funding was objectively and accurately determined. However, whether this corrected inventory data will be transferred to FMIS in time for making fiscal year 2004 funding decisions is uncertain. BIA may yet face some implementation problems as it moves into the final months of putting this information in place. For example, during 2002, BIA attempted to run the O&M funding formula using the existing FMIS inventory data for the first time and experienced problems with the data and software. While the agency has had a year to work out these problems, introduction of the updated inventory data may hold its own unforeseen problems. If such problems are encountered and remain unresolved, a BIA official told us that the agency would continue to use the data currently in the system to allocate the O&M funding for fiscal year 2004. The backlog data in FMIS reflects actions and funding needed to improve the condition of facilities and infrastructure at the various schools now and in the future. Most of the items listed in the backlog provide detail for repairs needed over the next 5 years to correct what is wrong with a facility such as a leaky roof, the presence of asbestos, or a violation of handicapped codes and standards. However, FMIS also includes entries for capital improvements that will need to be addressed beyond 5 years to upgrade specific building components such as replacing lighting and power systems, siding, and carpeting as well as future construction to replace, renovate, and add buildings to accommodate program needs. Accurate and up-to-date backlog data are important because FMIS contains formulas that use these data to allow BIA to make informed decisions not only about which projects are in greatest need of attention, but also how much money is needed to fund them each year. As of October 2002, for example, BIA schools had a backlog of unfunded repairs and improvements with an estimated cost over $640 million (see table 3); FMIS shows that almost two-thirds of this amount may be needed within the next 5 years. AME currently validates the backlog for each school in a 3-year cycle. During its first review, in 1999, AME conducted a 100-percent validation of the backlog data prior to transferring that data from the old FACCOM system to the new FMIS backlog module. In that review, AME updated the backlog data by confirming entries already in the system, updating the costs estimated to conduct the work, deleting entries for duplicate items or completed work, and identifying new entries. Results of the validation effort, as shown in table 4, increased the backlog by more than $265 million; almost 28 percent of the total backlog of $960 million that existed in 1999. AME has since started the second review of each school, which consists of updating this information for a certain percentage of the facilities each year. AME updated 20 percent of the facilities in each year 2001 and 2002. For 2003, AME is on track to increase the percentage of facilities reviewed each year to 34 percent to comply with recent changes in the law. These updates involve visually inspecting the architectural, structural, mechanical, and electrical components of each facility to determine if action is still needed and to update the estimated costs. In addition, AME identifies new deficiencies, including the extent to which handicapped accessibility requirements are met, and verifying estimated costs. Our review of some updates conducted during fiscal year 2002 indicates that AME’s reviews will continue to result in substantial changes in backlog data. We obtained data for 14 of the schools reviewed between February and April 2002. For the 14 schools, the AME update resulted in a net increase of almost $11 million (see table 5) in the unfunded FMIS backlog of more than $39 million, an increase of about 28 percent. Part of the change involved modifications to deficiencies already in the backlog inventory, such as revising cost estimates and deleting projects that had been completed or no longer needed but were still listed in the backlog as ongoing. However, a large part of the change involved adding new deficiencies to the backlog. In all, there were almost 650 new backlog entries, and more than 75 percent of these entries were for deficiencies identified at school facilities and dormitories, where children are the primary occupants. For example, there were more than 200 entries for dormitories with an estimated cost of almost $4 million. None of these entries were for urgent or safety deficiencies that needed immediate attention; most were maintenance deficiencies that will need attention over the next 2 to 5 years, such as repairing lighting and plumbing systems, carpeting, and ceilings. While the contractor’s field assessments are proceeding on schedule, there have been significant delays in incorporating the 2002 updated backlog data into FMIS. In fiscal year 2001, the first year of the updates, the contractor assessed the condition of 39 schools and transferred the data from its information system to the FMIS without problems, according to BIA and contractor officials. However, officials said that in fiscal year 2002, the untimely transfer of data from the field assessments to the contractor’s information system and software compatibility problems between the contractor’s information system and FMIS delayed the update of backlog data for 33 schools for over a year. These implementation problems occurred for such reasons as the following: BIA added a new function to the FMIS, which took 6 months to implement, rather than the 2 months that they had planned, according to a BIA official. This new function involved using the FMIS, for the first time, to generate the O&M funding amounts to be distributed to the various schools. A change to the backlog category and ranking system in FMIS created duplicate entries that took time to resolve. In the 2002 assessments, some deficiencies that were already in the system were recategorized, and when the update was transferred to FMIS a duplicate entry was created instead of overwriting the old entry. According to BIA and contractor officials, this software compatibility problem has largely been addressed. As of April 2003, data for 27 of the 33 schools had been entered into FMIS. A BIA official told us that delays in introducing these updates into the FMIS backlog could have some impact on their ability to prioritize or fund repair and capital improvement projects, but not a significant impact for two reasons. First, the deficiencies that receive the highest points in the project ranking system are safety deficiencies, which are identified and updated in an annual safety inspection by BIA’s safety officers and not AME. Second, the most critical deficiencies at the schools were identified in the first assessment in 1999, and during this second assessment, AME is finding very few deficiencies that would be funded within 1 to 2 years that were not already in the system. However, one official acknowledged that since the FMIS ranks the schools for major repair and capital improvement projects based on the points applied to each deficiency, if AME’s assessment indicated an increase in the severity of a deficiency and that change was not reflected in the system, that school could be ranked lower than if the data were up to date. Our site visits to 8 BIA-funded schools did not disclose any instances where serious problems were not being addressed. While facilities management staff and school principals pointed out problems with their facilities, we did not observe that the children were in an unsafe learning environment with obvious safety or repair issues. The problems we observed were either of a less serious nature, or if serious, were being addressed in some form. For example, at a boarding school in Arizona the principal said that the fire alarm system for the school building and dormitory had been improperly installed and had to be replaced. While waiting for the funding for a new system, which had been approved, they had to use funds from the school budget to hire extra people to stand fire watch 24 hours a day. The ability of FMIS to provide accurate and complete data depends on BIA employees entering correct and timely information, but review processes and training programs BIA has established for ensuring data quality have been largely ineffective. Although adherence to federal control standards are a major part of providing reasonable assurance that the objectives of the agency are being achieved, half of the entries proposed by BIA employees are incorrect or incomplete and are flagged by BIA’s contractor. Discussions with BIA employees indicate that some employees are unclear about their responsibilities in maintaining the accuracy of FMIS data, and the high error rates in data entry indicate that additional training is needed in some locations to improve performance. BIA has not analyzed the information it has available about the content and origination of the data errors to determine the type of additional training that might be needed or to target locations with the highest error rates for technical assistance. Further, BIA has not established criteria or performance goals that define its expectations for the accuracy and completeness of the data for employees that enter and review this information. BIA’s Office of Facilities Management and Construction, which manages FMIS, did not until recently have authority to establish criteria or performance goals for agency and regional office personnel that are responsible for reviewing and approving data entries by school facility managers. BIA’s OFMC still does not have similar authority over school facility managers who originate most data entries. Under BIA’s current organization, such action would have to be taken by the Office of Indian Education Programs. BIA established a multilevel review process as a control to prevent problems related to inaccurate and incomplete data entries to BIA’s information system. In this process, each entry that school facility managers propose for the backlog is first reviewed and approved by BIA facility management personnel at an agency and a regional office before it is sent to BIA’s contractor, AME, for review and approval, with final approval by BIA’s central office. After approval has been obtained from each level, the status of the entry is changed from “draft” to “accepted” in FMIS, according to BIA. Although BIA established this multilevel review process to improve the quality of the data entered into FMIS, AME continues to reject half of the proposed entries because they are inaccurate and incomplete. For example, between August 2001 and December 2002, out of more than 650 entries to the backlog made by facilities management staff from 102 schools, more than 300, or almost 50 percent, were rejected by the contractor. BIA documents show that the incidence of the errors among the 102 schools was widespread. In all, 73 of the 102 schools entering data had one or more entries disapproved, and 33 of them had all entries disapproved (see table 6). Facility management staff at the regional and agency offices appear to have the necessary background to fulfill their responsibilities for screening and correcting inaccurate and incomplete data entries; however, it may be unclear to some reviewers what their role is in this process. Federal control standards require that employees have the requisite knowledge, skills, and abilities to perform their job appropriately and have clear roles and responsibilities outlined in their job descriptions. Among the facilities management staffs in the agency and regional offices that we interviewed, most had an engineering background or had significant experience in facilities management and they were aware that maintaining accurate and complete data in the FMIS was important. However, facilities managers we interviewed at two regional offices each had a different view about their role in the review process. One said that it was important that the backlog entries are consistent and that two staff had been designated to review the FMIS entries from the schools and agencies in the region before being sent on to the central office for review by the contractor. The second manager said he reviewed the entries so that he knew what new additions were being proposed for the backlog; he did not consider it his role to critique the entries for accuracy and completeness. Although most BIA locations have staffs that have received training to use FMIS, the extent of the errors indicates that employees may not be receiving the kind of training needed to create accurate and complete data entries. Standards for internal control in the federal government include a commitment to competence, which includes the provision that employees receive the appropriate training necessary to improve their performance. BIA has developed a training program that is intended to provide FMIS users with sufficient information to operate the system. All FMIS users receive 40 hours of training before they are given a password that allows them to access the system, according to BIA; a review of BIA’s training log indicates that about 70 percent of the schools have at least one staff member onsite that has received this training. While training appears to be adequate in terms of providing staff with the basic skills needed to use the computer-based applications, what appears to be lacking in the training program is more specific instruction and guidance on the kinds of information that are needed to enter an accurate and complete deficiency to the backlog. BIA officials acknowledged that a user manual to provide this type of guidance was lacking and should be developed. BIA officials said developing the FMIS training has been a challenge because the facilities management staffs have different levels of knowledge about computers—for some facility managers, FMIS training was their first exposure to using a computer. The training programs were developed to meet the needs of this diverse group of users, according to the contractor that developed and provides the training. In our site visits we asked users about the training that they had received for using FMIS; almost all of the staff that we interviewed said that they were pleased with the training they received and believed that it had prepared them for using FMIS. In addition, they said that when they did have problems they contacted central office with questions and/or problems with using the system, the response was prompt and very helpful. However, the engineering contractor indicated that the varying levels of experience and expertise is a difficulty affecting staff’s ability to input data successfully. He said that there are more than 180 sites with education buildings for which data must be entered and the level of knowledge and expertise about facilities and the kinds of information needed for an FMIS entry varies widely—particularly in those sites where the turnover rate of facilities management staff is high. We were unable to obtain comprehensive data on the turnover rate for the facility management staff, but among the schools that we visited, the facilities managers’ length of employment at their current location ranged from 2 years to 17 years. BIA officials have information on the number and reasons that data entries are rejected at each location, but said that they had not used this information to provide performance counseling to employees or modify the training and guidance in this regard. Standards for internal control in the federal government include the provision that employees receive the feedback necessary to improve their performance. Analyzing the extent and content of data errors would be helpful to determine the type and amount of additional training and guidance needed to improve employee performance at schools, agencies, and regional offices and target them appropriately for technical assistance. BIA data we reviewed indicated that the reasons for disapproval generally fell into one of four groups and were consistent with historical data entry problems experienced under the old FACCOM system. For example, more detailed description was needed for the deficiency; a roof repair, for example, required specific information about the kind of roof and its size; questionable cost estimates involving labor rates or material costs; duplicate entry for a deficiency already in the backlog; and wrong category and/or rank for the entry such as categorizing the replacement of asbestos floor tiles as maintenance rather than environmental, which are funded from different sources. Of these problems, most of the rejected backlog entries generally related to insufficient detail to accurately estimate the cost to address the deficiency, according to an AME official who reviewed and rejected many of these entries. The kind of detail that is needed to successfully enter a deficiency can be seen in an example involving the repair of a leaking roof. To adequately estimate the cost for this repair, information is needed about the size of the area that needs repair or replacement, the composition of the roof (such as asphalt shingles or tile), and other associated components (such as whether skylights or gutters are present and whether they also need to be replaced). We reviewed one entry that had been rejected by the contractor because it came through the review process without information about the roof’s size or its composition. The ineffectiveness of BIA’s guidance, training, and review processes to minimize inaccurate or incomplete data entries by its employees suggests that accountability is another issue that deserves attention. Federal standards require that agencies clearly establish authority and responsibility for achieving agency goals and hold their employees accountable for performing their assigned responsibilities in a competent manner. During our review, the organization of BIA was such that the office that manages FMIS did not have line authority to establish performance criteria and standards for the BIA employees that entered and reviewed the FMIS data. In April 2003, BIA announced a new organization plan. This plan may offer OFMC greater opportunity to establish performance criteria and standards for facility managers at agency and regional offices. BIA officials told us the reorganization would not provide OFMC with line authority for facility managers at schools. However, the Director of the Office of Indian Education programs said that his office would work with OFMC to establish comparable performance criteria and standards for school facility managers. FMIS is designed to assist BIA employees improve the quality of information used to manage school facilities, but the quality of the decisions that BIA makes for managing the operations and maintenance, repair, and construction of its facilities is directly dependent upon BIA employees entering correct and timely information. Currently, the FMIS data that BIA uses for making its decisions are improving as the data are updated by its contractor and entered into FMIS—to date the inventory data have been updated and the contractor is in the third year of assessing the condition of the schools and updating the backlog for the second time. However, challenges remain in BIA’s efforts to improve the quality of data entered by its employees. Although BIA has implemented controls for ensuring the accuracy and completeness of the FMIS data entered and reviewed by BIA employees, they do not work effectively. Without the role of the contractor as a reviewer of new entries by field staff and in conducting site visits to verify and update the data, the quality of the FMIS data could quickly become inaccurate and out of date. BIA has not taken the necessary steps to hold its staff accountable for data accuracy or to use the available information on why the problems exist to develop training programs and target technical assistance where it is needed. Such actions are needed if BIA is to rely on its employees, rather than the contractor, to ensure that it provides a safe and quality learning environment for Indian children. To better enable BIA to rely on its employees for maintaining accurate and complete information in the FMIS, we recommend that the Office of the Assistant Secretary of Indian Affairs establish data standards for accuracy and completeness of FMIS data and related performance criteria for BIA employees who are responsible for entering and reviewing the data and analyze available error data and use this information to provide its employees with the necessary training, guidance, and technical assistance to improve performance. We provided a draft of this report to the Department of the Interior for its review and comment. Interior’s comments are provided in appendix II. In its written comments, Interior agreed with our findings and recommendations and said that BIA is establishing a special working group to develop a plan to address our recommendations. In addition, BIA will consider our comments and observations as it continues to develop and implement the FMIS. We will send copies of this report to the Secretary of the Department of Interior, relevant congressional committees, Indian education organizations, and other interested parties, and will make copies available to others upon request. In addition, the report will be available at no charge on GAO’s Web site at http://www.gao.gov. Please contact me at (202) 512-6778 if you or your staff have any questions about this report. Other major contributors to this report are listed in appendix III. The objectives of this study were to determine (1) whether the Bureau of Indian Affairs (BIA) new facilities management information system (FMIS) addresses the former system’s shortcomings and meets BIA’s needs for managing school facilities; (2) the status of BIA’s effort to validate the accuracy and completeness of the data being transferred from the old system into FMIS; and (3) how well BIA’s quality control measures are working to ensure that new data entered into FMIS are accurate and complete. To determine the extent to which FMIS was designed to address weaknesses of the previous data processing system and how the new system meets BIA’s facility management needs we reviewed contractor reports, BIA documentation on the FMIS, and interviewed contractors and BIA headquarters, regional, agency, and school facility management staff. First, we reviewed the needs assessment studies conducted by independent contractors to identify old system weaknesses and the recommendations made for addressing the system problems. We then reviewed the FMIS documentation to determine whether the system addressed weaknesses identified in the needs assessment. In addition, we conducted interviews with some of the contractors hired by BIA to build and implement the system. We also interviewed BIA officials at the Office of Facilities Management and Construction, the Division of Safety and Risk Management, and the Office of Indian Education Programs about improvements in the new system and how it meets their management needs. Finally, to understand how well school facility management staff received FMIS, we conducted site visits to 8 schools in Arizona, New Mexico, and South Dakota. We selected these schools to obtain a mix based on their differences in size, geographic location, type of school (i.e., grade level, day school, or on-reservation boarding school), and whether it was BIA-operated or tribal-operated. At these schools, we interviewed facility managers, education line officers, principals, and tribal officials. We also interviewed facility management staff at two regional and five agency offices that provide facility management services to the schools. To determine the extent to which the FMIS inventory and backlog data are accurate and complete we used three methods. First, we obtained data from two sources (1) a copy of the FMIS database and (2) a copy of the contractor file of backlog data from the fiscal year 2002 condition assessments of 14 schools. The data in these files were assessed for reliability, which included looking for missing data, the relationship of one data element to another, values beyond a given range, and dates outside of valid time frames. We determined that the data were sufficiently reliable for the purposes of this report. We also calculated summary statistics of the data from these files. Second, we evaluated BIA’s multilevel review process. For this analysis, we obtained data on the number of backlog entries made by facility managers and the number that had been accepted and rejected by the contractor. In addition, we reviewed the log of entries that had been rejected by the contractor to understand the reasons for these rejections. We also interviewed the contractor and BIA central office and regional and agency facility management staff about this review process. Finally, we accompanied BIA’s engineering contractor on site visits to 2 schools in Arizona where we evaluated their methodologies for data collection and validation of the inventory and backlog data. To determine how well BIA’s internal control measures are working for its FMIS we first reviewed our standards on internal controls to identify controls that apply to an organization’s management of an information system. We then compared the BIA controls with those identified in our reports to evaluate the effectiveness of the BIA controls. Additionally, we interviewed staff at the BIA central office about the internal controls they had in place for the FMIS. We also interviewed staff at the regional and agency offices and the schools about one of the controls—the effectiveness of the training they received. In addition to the individuals mentioned above, Jessica Botsford, Maya Chakko, Terry Dorn, David Gill, Barbara Johnson, Nathan Morris, Stan Stenerson, and Michelle Zapata made significant contributions to this report. | The Bureau of Indian Affairs (BIA) is responsible for providing over 48,000 children with a safe place to learn. In response to concerns that data in its old information system did not accurately reflect the condition of facilities, BIA acquired a new system, called the Facilities Management Information System (FMIS). GAO was asked to determine whether FMIS addresses the old system's weaknesses and meets BIA's management needs, whether BIA has finished validating the accuracy of data entered into FMIS from the old system, and how well the quality control measures are working for ensuring the accuracy of new data being entered into the system from individual schools. FMIS is designed to address the previous data system's shortcomings and appears to have the capability to meet BIA's management needs if the data that are entered into FMIS are correct and timely. The old system was hard to use and did not readily provide data for maintenance and repair efforts. FMIS's design appears to overcome these weaknesses. For example, FMIS has features that help facility managers make data more consistent, as well as tools for helping managers develop cost estimates for maintenance and construction projects. BIA's contractor has been correcting the data that were transferred to FMIS from the previous system, but issues such as software compatibility problems between the contractor's system and FMIS have delayed entry of some of the data for more than 1 year. BIA officials say that these problems are being addressed. They said the delay has not affected their ability to prioritize or fund repair and construction projects, and our review of the data indicated that most newly identified deficiencies will not need to be addressed for 2 to 5 years. Our review of data from 14 BIA schools and observations during site visits disclosed no instances in which these data problems resulted in an unsafe learning environment for children. Most measures for controlling the quality of new data BIA employees are entering into the system for individual schools are not working well. BIA has established a multilevel review process and training programs to help ensure that such data entries are complete and accurate, but BIA's contractor, in reviewing data at the end of this process, continues to find that nearly half of the proposed data entries coming through the system are inaccurate and incomplete. Data entries from one-third of 102 schools that entered data show a 100-percent error rate. As a result, BIA officials continue to rely on their contractor to ensure that FMIS reflects accurate and complete data on the condition of BIA's facilities. |
The procedures under UMRA for the identification and analysis of intergovernmental and private sector mandates are very complex. Moreover, some potential mandates are enacted through procedures that never require them to be reviewed under UMRA. For example, UMRA does not require the automatic review of potential mandates contained in appropriation bills, nor does the act cover rules that were issued as final without having been preceded by a notice of proposed rulemaking. Even if proposed legislation or regulations are reviewed under UMRA, those provisions are subject to various definitions, exclusions, and exceptions before being identified as containing mandates at or above UMRA’s cost thresholds. For example, UMRA does not apply to legislative provisions that cover constitutional rights, discrimination, emergency aid, accounting and auditing procedures for grants, national security, treaty ratification, and certain parts of Social Security. As figure 1 illustrates, a provision in legislation must pass through a multiple step process before the Congressional Budget Office prepares required statements identifying and estimating the costs of mandates in legislation that meet certain criteria and determines whether or not those estimated costs meet or exceed UMRA’s thresholds. Based on UMRA’s requirements, we found that few provisions in statutes or rules are considered mandates as defined by UMRA. As mentioned previously, in 2001 and 2002, the period of our review, only 5 of the 377 statutes enacted and 9 of the 122 major rules issued contained federal mandates at or above UMRA’s thresholds. All 5 statutes and 9 rules contained private sector mandates and only one final rule—an Environmental Protection Agency standard on arsenic in drinking water— contained an intergovernmental mandate. Despite the determinations made under UMRA, nonfederal parties affected by federal actions viewed many more federal actions in statute and regulation as containing unfunded or under funded mandates. When we explored this issue, we found that some of the statutes and rules that had not triggered UMRA’s requirements appeared to have potential financial impacts on affected parties similar to those of actions that had been flagged as containing mandates at or above UMRA’ s thresholds. Specifically, we identified at least 43 statutes and 65 rules issued in 2001 and 2002 that resulted in new costs or other negative financial impacts on nonfederal parties that the affected parties might perceive as unfunded or under funded mandates even though they did not meet UMRA’s definition of a mandate or did not meet or exceed UMRA’s thresholds. For these statutes and rules, CBO or federal agencies most often had determined that the estimated direct costs or expenditures, as defined by UMRA, would not meet or exceed the applicable thresholds or that one or more of the other definitions, exclusions, or exclusions applied. These findings raised the question of whether UMRA, given its procedures, definitions, and exclusions, adequately captures and subjects to scrutiny federal statutory and regulatory actions that might impose significant financial or other burdens on affected nonfederal parties. To begin to address this question, you asked us to obtain the views of a diverse group of parties knowledgeable about UMRA and federal mandates. Parties from the various sectors provided a variety of comments but they generally fell into several broad themes. UMRA’s coverage was the most frequently cited theme, with comments provided by all the sectors (academic/think tank, business, federal agencies, public interest advocacy groups, and state and local governments). Issues involving enforcement were the second most frequently cited but with far fewer parties providing comments. Other themes that emerged from the comments included the use and usefulness of the information generated under UMRA, UMRA’s analytic framework, and consultation under UMRA. Finally, issues involving the design and funding and evaluation of federal mandates also emerged as themes. Given the findings from our May 2004 report, it’s not surprising that UMRA’s coverage, including its numerous definitions, exclusions, and exceptions, was the most frequently cited issue by parties from all five sectors. Most parties from the state and local governments, federal agency, business, and academic/think tank sectors viewed UMRA’s narrow coverage as a major weakness that leaves out many federal actions with potentially significant financial impacts on nonfederal parties. However, a few parties, from public interest advocacy groups and academic/think tank sectors, considered some of the existing exclusions important or identified UMRA’s narrow scope as one of the Act’s strengths. The comments about weaknesses in UMRA’s coverage ranged from general to specific. For example, some parties commented, in general, about the number of exclusions and exemptions. Others provided more specific comments, including points regarding issues with the exclusion of indirect costs and UMRA’s cost thresholds for legislative and regulatory mandates, which result in excluding many federal actions that may significantly impact nonfederal entities. Others raised more fundamental concerns about the exclusions for appropriations and other legislation not covered by the Act and for rules issued by independent regulatory agencies, which are also not covered by UMRA but can result in provisions that contain mandates. CBO estimates that in 2004, 5 of the 8 laws containing federal mandates (as defined by UMRA) that it did not review before enactment, were appropriations acts. Finally, parties from the state and local government sector also identified concerns about gaps in UMRA’s coverage of federal preemption of state and local authority. Although some preemptions are covered by UMRA such as those that preempt state or local revenue raising authority, they are covered only for legislative actions and not for federal regulations. According to CBO’s 2005 report on unfunded mandates, “Over half of the intergovernmental mandates for which CBO provided estimates were preemptions of state and local authority.” Despite the widespread view in several sectors that UMRA’s narrow coverage leaves out federal actions with potentially significant impacts on nonfederal entities, there was less agreement by parties about how to address this issue. The options ranged from general to specific but those most frequently suggested were: Generally revisit, amend, or modify the definitions, exceptions, and exclusions under UMRA and expand UMRA’s coverage. Clarify UMRA’s definitions and ensure their consistent implementation across agencies to ensure that all covered provisions are being included. Change the cost thresholds and/or definitions that trigger UMRA by, for example, lowering the threshold for legislative or executive reviews and expanding cost definitions to include indirect costs. Eliminate or amend the definitional exceptions for conditions of federal financial assistance or that arise from participation in voluntary federal programs. Expand UMRA coverage to all preemptions of state and local laws and regulations, including those nonfiscal preemptions of state and local authority. As I mentioned previously, while most parties thought UMRA’s narrow coverage was a weakness, a few parties from academic/think tank and public interest advocacy groups sectors view UMRA coverage differently. They viewed UMRA’s narrow scope as one of its primary strengths. In fact, rather than expanding UMRA’s coverage, these parties said that it should be kept narrow and that the exceptions and exclusions are needed. Between 1996 and 2004, CBO reports that of 5,269 intergovernmental statements, 617 had mandates and of 5,151 private sector statements, 732 had mandates. Of the mandates identified by CBO, about 9 percent of the intergovernmental mandates and almost 24 percent of private sector mandates had costs that would exceed the thresholds. As discussed at our January 26, 2005, symposium on UMRA and federal mandates, some parties also identified a number of suggestions that they could not support, namely any attempt to expand UMRA to cover constitutional or civil rights or excluding private sector mandates. Issues involving UMRA enforcement were the second most frequently cited issue but with far fewer parties from each sector commenting. Parties across and within sectors had differing views on both the enforcement mechanisms provided in the law itself and the level of effort exercised by those responsible for implementing UMRA’s provisions. Many of the comments focused on the point of order—one of the primary tools used to enforce UMRA requirements in title I of UMRA. Although the point of order provides members of Congress the opportunity to raise challenges to hinder the passage of legislative provisions containing an unfunded intergovernmental mandate, views were mixed about its effectiveness. Those representing state and local government and federal agency sectors said that the point of order should be retained because it has been successful in reducing the number of unfunded mandates by acting as a deterrent to their enactment, without greatly impeding the process. Conversely, some parties primarily from academic/think tank, business, and federal agency sectors did not believe the point of order has been effective in preventing or deterring the enactment of mandates and suggested otions to strengthen it. Moreover, others commented about its infrequent use. Some parties said the point of order needs to be strengthened by making it more difficult to defeat. One suggested revision was to require a three- fifths vote in Congress, rather than a simple majority, to overturn a point of order. This change was believed to strengthen the “institutional salience of UMRA” and to ensure that no mandate under UMRA could be enacted if it was supported only by a simple majority. As you know, on March 17, 2005, the Senate approved the fiscal year 2006 budget, which included a provision that would increase to 60 the number of votes needed to overturn an UMRA point of order in the Senate. A few parties from the federal agency and academic/think tank sectors commented on another enforcement mechanism for regulatory mandates—UMRA’s judicial review provision, which subjects any agency compliance or noncompliance with certain provisions in the Act to judicial review. Most felt that this mechanism does not provide meaningful relief or remedies if federal agencies have not complied with the requirements of UMRA because of its limited focus. Specifically, judicial review is limited to requirements that pertain to preparing UMRA statements and developing federal plans for mandates that may significantly impact small government agencies. Furthermore, if a court finds that an agency has not prepared a written statement or developed a plan for one of its rules, the court can order the agency to do the analysis and include it in the regulatory docket for that rule; but the court may not block or invalidate the rule. The few parties commenting about judicial review suggested expanding it to provide more opportunities for judicial challenges and more effective remedies when noncompliance of the Act’s requirements occur. A few parties primarily from the academic/think tank and public interest advocacy groups sectors said that efforts to limit or stop implementation of mandates through legal action might be unwarranted, because UMRA was not intended to preclude the enactment of federal mandates. They were primarily concerned about litigation being used to slow down the regulatory process. Commenting parties from business, federal agency, and state and local governments sectors questioned some federal agencies’ compliance with UMRA requirements and the effectiveness of enforcement mechanisms to address this perceived noncompliance. They mentioned the failure of some agencies to consult with state, local and tribal governments when developing regulations that may have a significant impact on nonfederal entities. Likewise, at least one party of each of the three sectors expressed concerns about the lack of accurate and complete information provided by federal agencies, which are responsible for determining whether a rule includes a mandate and whether it exceeds UMRA’s thresholds. The perceived lack of compliance with certain UMRA requirements generated several suggested changes to UMRA. However, the only suggestion that had support across parties from multiple sectors was to replicate CBO’s role for legislative mandates by creating a new office within OMB that would be responsible for calculating the cost estimates for federal mandates in regulations. Parties from all sectors also raised a number of other issues about the use and usefulness of UMRA information (in decreasing the number of unfunded mandates), UMRA’s analytic framework, and federal agency consultations with state, local, and tribal governments, but there was no consensus in their views about how these issues should be addressed. The parties provided mixed but generally positive views about the use and usefulness of UMRA information. Some parties commented that the Act does increase awareness of unfunded mandates but thought more could be done to increase its usefulness. However, the only option that attracted multiple supporters was a suggestion for a more centralized approach for generating information within the executive branch similar to the suggestions mentioned about improving enforcement. Parties also provided a number of comments about the provisions of UMRA that establish the analytic framework for cost estimates, which generated a few suggested options aimed at improving the quality of information generated such as including indirect costs for threshold purposes and clarifying certain definitions (e.g. “federal mandates” and “enforceable duty”). UMRA’s consultation also emerged as a recurring theme within and across certain sectors. The comments generally were about a perceived lack of consistency across agencies when consulting with state and local governments. Parties from all sectors also raised a number of broader issues about federal mandates—namely, the design and funding and evaluation of federal mandates—and suggested a variety of options. While most of the comments were about the evaluation of federal mandates, some parties also raised concerns about the design and funding of mandates, which varied across sectors. Issues raised include: (1) costs for mandates may vary across different affected nonfederal entities, (2) mismatches between the funding needs of parties compared to federal formulas, and (3) effects of the timing of federal actions and program changes on nonfederal parties. Most often, the comments focused on a perceived mismatch between the costs of federal mandates and the amount of federal funding provided to help carry them out. Others raised issues about the need to address the incentives for the federal government to “over leverage” federal funds by attaching (and often revising) additional conditions for receiving the funding. Parties, primarily from the academic/think tank sector, suggested a wide variety of options to address their concerns, but there was no broad support for any option. For example, while some parties across four sectors suggested providing waivers or offsets to reduce the costs of the mandates on affected parties or “off ramps” to release them of some responsibilities to fulfill the mandates in a given year if the federal government does not provide sufficient funding. Others said that compliance with federal mandates should not be made contingent on full federal funding and off ramps and waivers can introduce other issues. The option of building into the design of federal mandates “look back” or sunset provisions that would require retrospective analyses of the mandates’ effectiveness and results was also suggested. About half the parties, representing most sectors commented on the evaluation of federal mandates and offered suggestions to improve mandates, whether covered by the Act or not. Not surprisingly parties in the academic/think tank sector, who felt that the evaluation of federal mandates was especially important because there is a lack of information about the effects of federal mandates on affected parties, provided most of the comments. The issues raised included concerns about the lack of focus on evaluating the effectiveness (results) of the mandates; the questionable accuracy and completeness of cost estimates, particularly ones prepared by federal agencies, and the lack of evaluation of the impact of mandates. All of these issues are related and the concerns expressed touched upon the need to adequately evaluate mandates in the context of costs, benefits, impacts, and effectiveness of the mandated actions to achieve desired goals. Parties across the sectors suggested that various forms of retrospective analysis are needed for evaluating federal mandates after they are implemented. Some suggestions for retrospective analysis focused on costs and effectiveness of mandates, including comparing them to the estimates and expected outcomes. Others from the state and local governments sector also suggested focusing on the cumulative costs and effects of mandates—the impact of various related federal actions, which when viewed collectively, may have a substantial impact although any one may not exceed UMRA’s thresholds. Finally, parties primarily from the academic/think tank sector suggested examining local and regional impacts of mandates and analyzing the benefits of federal mandates, when appropriate, not just costs. As Congress reevaluates UMRA on its 10-year anniversary, the information we provided over the past year provides some useful insights. First, although parties from various sectors generally focused on the areas of UMRA and federal mandates that they would like to see fixed, they also recognized positive aspects and benefits of UMRA. In particular, they commented about the attention UMRA brings to potential consequences of federal mandates and how it serves to keep the debate in the spotlight. I also found it notable that no one suggested repealing UMRA. Second, when considering changes to UMRA itself, UMRA’s narrow coverage stands out as the primary issue for most sectors because it excludes so many actions from coverage under UMRA and contributes to complaints about unfunded or under funded mandates as discussed in both of our reports. Even with an issue such as coverage on which there was some general agreement across most sectors, the variety of suggested options indicates that finding workable solutions will require including all affected parties in the debate. Third, one of the challenges for Congress and other federal policy makers is to determine which issues and concerns about federal mandates are best addressed in the context of UMRA and which are best considered as part of more expansive policy debates on federal mandates and federalism. On broader policy issues concerning federal mandates, various parties recognized that UMRA is only part of the solution and the issue raises broader public policy questions about structuring and funding mandates in general. These parties made it clear that retrospective analysis is needed to ensure that mandates are achieving their desired goals, which could help provide additional accountability for federal mandates and provide information that could lead to better decisions regarding the design and funding of mandate programs. Finally, as we move forward in an environment of constrained fiscal resources, the issue of unfunded mandates raises broader questions about the assignment of fiscal responsibilities within our federal system. Reconsideration of such responsibilities begins with the observation that most major domestic programs, costs, and administrative responsibilities are shared and widely distributed throughout our system. Part of this public policy debate includes a reexamination of the federal government’s role in our system and a need to sort out how responsibilities for these kinds of programs should be financed in the future. If left unchecked, unfunded mandates can weaken accountability and remove constraints on decisions by separating the enactment of benefit programs from the responsibility for paying for these programs. Likewise, 100 percent federal financing of intergovernmental programs can pervert fiscal incentives necessary to ensure proper stewardship at the state and local level for shared programs. Mr. Chairman, once again I appreciate the opportunity to testify on these important issues and I would be pleased to address any questions you or other members of the committee might have. If additional information is needed regarding this testimony, UMRA or federal mandates, please contact Orice M. Williams at (202) 512-5837 or williamso@gao.gov or Tim Bober at (202) 512-4432 or bobert@gao.gov. Other key contributors to the work which was associated with this testimony were Tom Beall, Kate Gonzalez, and Boris Kachura. | The Unfunded Mandates Reform Act of 1995 (UMRA) was enacted to address concerns about federal statutes and regulations that require nonfederal parties to expend resources to achieve legislative goals without being provided funding to cover the costs. UMRA generates information about the nature and size of potential federal mandates but does not preclude the implementation of such mandates. At various times in UMRA's 10-year history, Congress has considered legislation to amend aspects of the act to address ongoing questions about its effectiveness. This testimony is based on GAO's reports, Unfunded Mandates: Analysis of Reform Act Coverage ( GAO-04-637 , May 12, 2004) and Unfunded Mandates: Views Vary About Reform Act's Strengths, Weaknesses, and Options for Improvement ( GAO-05-454 , March 31, 2005). Specifically, this testimony addresses (1) UMRA's procedures for the identification of federal mandates and GAO's analysis of the implementation of those procedures for statutes enacted and major rules issued in 2001 and 2002, and (2) the views of a diverse group of parties familiar with UMRA on the significant strengths and weaknesses of the act as the framework for addressing mandate issues and potential options for reinforcing the strengths or addressing the weaknesses. The identification and analysis of intergovernmental and private sector mandates is a complex process under UMRA. Proposed legislation and regulations are subject to various definitions, exceptions, and exclusions before being identified as containing mandates at or above UMRA's cost thresholds. Also, some legislation and rules may be enacted or issued via procedures that do not trigger UMRA reviews. In 2001 and 2002, 5 of 377 statutes enacted and 9 of 122 major or economically significant final rules issued were identified as containing federal mandates at or above UMRA's thresholds. Despite the determinations under UMRA, at least 43 other statutes and 65 rules resulted in new costs or negative financial consequences that affected nonfederal parties might perceive as unfunded or underfunded federal mandates. GAO obtained information from 52 knowledgeable parties, who provided a significant number of comments about UMRA, specifically, and federal mandates, generally. Their views often varied across and within the five sectors we identified (academic/think tank, public interest advocacy groups, business, federal agencies, and state and local governments). Overall, the numerous strengths, weaknesses, and options for improvement identified during the review fell into several broad themes, including, among others, UMRA-specific issues such as the act's coverage and enforcement, and more general issues about the design, funding, and evaluation of federal mandates. UMRA's coverage was, by far, the most frequently cited issue by parties from the various sectors. Parties across most sectors said that UMRA's numerous definitions, exclusions, and exceptions leave out many federal actions that might significantly impact nonfederal entities and suggested that they should be revisited. However, a few parties, primarily from the public interest advocacy sector, viewed UMRA's narrow coverage as a strength that should be maintained. Another issue on which the parties had particularly strong views was the perceived need for better evaluation and research of federal mandates and more complete estimates of both the direct and indirect costs of mandates on nonfederal entities. The most frequently suggested option to address these evaluation issues was more post-implementation evaluation of existing mandates or "look backs" at their effectiveness. Going forward, the issue of unfunded mandates raises broader questions about assigning fiscal responsibilities within our federal system. The long-term fiscal challenges facing the federal and state and local governments and the continued relevance of existing programs and priorities warrant a national debate to review what the government does, how it does business, and how it finances its priorities. Such a reexamination includes considering how responsibilities for financing public services are allocated and shared across the many nonfederal entities in the U.S. system. |
GPRA was intended to address several broad purposes, including promoting a new focus on results, service quality, and customer satisfaction. GPRA requires executive agencies to develop strategic plans, prepare annual performance plans, measure performance toward the achievement of goals in the annual plans and report annually on their progress. Building on GPRA and in carrying out the principles of the National Performance Review (NPR), on September 11, 1993, President Clinton issued Executive Order 12862, which remains in effect. The order created several requirements for agencies related to customer service. Specifically, the order stated that all executive departments and agencies that provide significant services directly to the public shall provide customer service equal to the best in business and shall take the following actions: identify the customers who are, or should be, served by the agency survey customers to determine the kind and quality of services they want and their level of satisfaction with existing services post service standards and measure results against them benchmark customer service performance against the best in business survey frontline employees on barriers to, and ideas for, matching the provide customers with choices in both the sources of service and the make information, services, and complaint systems easily accessible provide means to address customer complaints On March 23, 1995, President Clinton issued a presidential memorandum on improving customer service, which stated that the standards agencies had begun issuing in response to Executive Order 12862 had told the federal government’s customers for the first time what they had a right to expect when they asked for services. The memorandum instructed agencies to treat the requirements of Executive Order 12862 as continuing requirements and stated that the actions the order prescribes, including surveying customers and employees and benchmarking, shall be continuing agency activities. The memorandum further provided that standards should be published in a form readily available to customers, and that services that are delivered in partnership with state and local governments, services delivered by small agencies and regulatory agencies, and customer services of enforcement agencies are covered by the requirement to set and publish customer service standards. It further stated that agencies shall, on an ongoing basis, measure results achieved against the customer service standards and report those results to customers at least annually in terms readily understood by individual customers, and that measurement systems should include objective measures wherever possible, but should also include customer satisfaction as a measure. Finally, the memorandum stated that agencies should publish replacement customer service standards if needed to reflect customer views on what matters most to them. According to the last NPR report on customer service in 1997, 570 federal departments, agencies, organizations, and programs issued a total of 4,000 customer services standards. The NPR report was accompanied by a database through which customers could search for standards of interest or browse standards by customer group, such as beneficiaries, veterans, and businesses. For example, the Immigration and Naturalization Service issued a standard that it would complete action on naturalization applications within 6 months, and the Department of Health and Human Services issued a standard for Medicare and Medicaid that 80 percent of callers would be on hold for less than 2 minutes. In addition, the report stated that for 2,800 of the 4,000 standards, results of performance against the standards had been identified and were the basis for changes and improvements in the delivery of service. All 13 services we surveyed had established customer service standards. Officials from the Indian Health Service (IHS), which provides medical care for American Indians and Alaska Natives, told us their customer service standards are established by their area offices and facilities. The other services had standards that applied service-wide. The services’ standards varied in their form. Some standards were general commitments to qualitative standards, often stating commitments to meet certain identified customer rights. Other standards were structured quantitatively and based on daily, monthly, or annual averages. The Bureau of Consular Affairs, for example, had standards that include general commitments to qualitative standards for Passport Services, such as “We will provide service in a courteous, professional manner,” as well as commitments to meet customer rights, such as “You have the right to speak with management if you are not satisfied with the service you have received.” Some of the standards from IHS area offices and facilities also stated general commitments to qualified standards such as, “have pride in our appearance, and provide a pleasant greeting,” while other IHS area offices and facilities had individual level quantitative standards, which included, “answering phone calls within three rings.” The Veterans Benefits Administration (VBA) has established several quantitative standards for customer service related to their Veterans’ Group Life Insurance (VGLI) program, including averages such as the speed of answer for their call center, which is set at 20 seconds, as well as call abandonment rates, which they set at 2 percent. Veterans Health Administration (VHA) has similar quantitative standards for their Medical Care program, such as a standard for the number of new patient appointments completed within 30 days of the appointment create date. All services reported that they had reviewed their standards within the last year. For example, VGLI told us that they review standards annually, and 2 years ago changed the standard for e-mail response time from 48 to 24 hours based on this review. Similarly, Internal Revenue Service (IRS) officials told us that standards for taxpayer assistance are reviewed to see if they need updating three times per year by executives, directors, and managers, and the Forest Service told us that they review standards for recreational facilities and services as part of their annual performance reporting required by GPRA. All 13 of the surveyed services reported having measures of customer service, and 11 of the services reported having measures of satisfaction. Customs and Border Protection (CBP) has methods of gathering input from customers, such as through their comment card program and the online question and comment section of their CBP Info Center Web page, and is planning to measure customer satisfaction through a customer satisfaction survey in 2011. CBP submitted the planned survey for review and approval by the Office of Management and Budget (OMB) on September 1, 2010. The Transportation Security Administration (TSA) has methods of gathering input from customers by phone or e-mail through their Customer Contact Center, through the “TSA Blog” and “Talk to TSA” feedback Web site, or through speaking with a TSA manager at the airport; however, they do not have measures of satisfaction. All the services reported they measure customer service at least annually, and all 11 services that measured customer satisfaction reported doing so at least annually as well. In addition, many services measured more frequently. Eleven of the surveyed services measure customer satisfaction daily, for example, and 8 reported that they measure customer service daily. The surveyed services reported measuring various aspects of their customer service. Customer service measures can include measures related to customer access to services, wait times, accuracy and other factors. Eleven surveyed services had measures related to quality or accuracy of service. For example, IRS measures the accuracy of responses provided by service representatives to questions on both tax law and taxpayers’ accounts by listening to and reviewing phone calls. Ten services had measures related to customer wait times, eight had measures of processing time, and eight had measures related to access times. For example, Passport Services measures wait times for customers applying in a passport agency office, application processing times, and the number of passport book re-writes due to errors made by Passport Services. Similarly, for VGLI, VBA measures the percentage of calls that receive busy signals and the average speed of answer, as well as processing time for disbursements, applications, and correspondence. All surveyed services but TSA and CBP reported gathering input from customers regarding their level of satisfaction through surveys. The National Park Service (NPS), for instance, reported using a visitor survey card program to survey visitors at over 320 of its points of service. These surveys are conducted annually at NPS units to measure performance related to visitor satisfaction and visitor understanding of park significance. Four surveyed services stated that they use the American Customer Satisfaction Index (ACSI) to gather and report results. ACSI is a standardized customer satisfaction survey that measures customer satisfaction for a particular service based on drivers of satisfaction, including customer expectations and perceived quality. Overall results of the ACSI survey are made available on the ACSI public Web site, http://www.theacsi.org, where results can be compared among federal agencies that use the survey. Participating agencies receive more detailed results that contain information, trends, and recommendations for areas to work on to improve results. Federal Student Aid (FSA), which provides student loans under the direct loan program, for instance, uses an independent third-party contractor to conduct the ACSI survey on FSA’s behalf. Officials stated that from the results of ACSI, they determined that some of the responses that they were providing customers on telephone calls were not achieving satisfactory marks, which led FSA to start a task force to address the issue. In addition to customer surveys, most services also employed other methods through which they gathered customer input. The most common of these were comments cards or suggestion boxes (seven services), telephone numbers (seven services), and e-mail or written correspondence (five services). For instance, Social Security Administration (SSA) officials stated that customers have the opportunity to either complete comment cards at field offices, or send letters to field offices regarding their satisfaction with the service received. Passport Services officials stated that they gather input from customers by having them complete comment cards while being served at a passport agency. Additional methods services reported using for gathering input include focus groups, e-mail lists, conferences, and outreach campaigns. IHS, for instance, uses patient satisfaction surveys, telephone surveys, a director’s blog, personal interviews, an incident reporting process, tribal negotiation meetings, suggestion boxes, and a patient needs task force to gather input from customers. All surveyed services reporting having methods to receive customer complaints. In addition to general methods of gathering input described above, many agencies reported using additional methods to receive complaints. These included complaint forms on the agency Web site, identifying service mailing addresses to receive complaints, providing customers with additional telephone numbers to provide feedback, and general feedback forms and survey cards about the service they received. For example, VBA reported that they use a public message management system called the Inquiry Routing and Information System (IRIS) for receiving inquiries, including complaints, for beneficiary medical care and disability service. All electronic messages received from the public through VA Web sites are directed through the IRIS system, and a set of policies and procedures is in place that guides the agency in sorting through and responding to submissions. Before requiring or requesting information from the public, such as through customer satisfaction surveys, federal agencies are required by the Paperwork Reduction Act (PRA) to seek public comment as well as approval from the Office of Management and Budget (OMB) on the proposed collection of information. The PRA requires federal agencies to minimize the burden on the public resulting from their information collections, and to maximize the practical utility of the information collected. To comply with the PRA process, agencies must develop and review proposed collections to ensure that they meet the goals of the act. Once approved internally, agencies generally must publish a 60-day notice in the Federal Register soliciting public comment on the agency’s proposed collection, consider the public comments, submit the proposed collection to OMB and publish a second Federal Register notice inviting public comment to the agency and OMB. OMB may act on the agency’s request only after the 30-day comment period has closed. Under the PRA, OMB determines whether a proposed collection is necessary for the proper performance of the functions of the agency, including whether the information will have practical utility. The PRA gives OMB 60 days to approve or disapprove a proposed collection, however, OMB can also instruct the agency to make a substantive or material change to the proposed collection. There are two primary ways that the normal clearance process can be altered. First, agencies may request and OMB may authorize emergency processing under certain circumstances. Second, agencies may submit generic information collection requests, which are requests for OMB approval of a plan for conducting more than one information collection using very similar methods, such as a plan to gather views from the public through a series of customer satisfaction surveys. The plan itself is subject to the standard 60- and 30-day public comment periods, but, if approved, each specific collection under the plan requires only OMB review and approval without additional public comment, subject to the terms of the original generic clearance. According to some agency officials, the Paperwork Reduction Act clearance process may, in certain instances, make obtaining customer input difficult. A March 2009 report from the Government Contact Center Council, an interagency group of contact center directors and managers sponsored by the General Services Administration’s (GSA) Office of Citizen Services, notes that although the PRA is important and affords many safeguards and benefits, the council commonly hears from agencies that the approval process takes 9 months or longer, in fact, so long that sometimes a survey is no longer relevant. The report recommended that OMB lead a task force to help to fully understand the implications of the Paperwork Reduction Act for agencies and to identify ways to minimize the length of time it takes for an agency to get approval for surveys intended for citizens. Two services we surveyed told us that the PRA clearance process made obtaining customer input difficult. NPS officials referred us to the letter of input they provided in response to OMB’s Federal Register notice seeking comment on, among other things, reducing paperwork burdens and avoiding adverse consequences of the PRA clearance process. In the letter, NPS stated that lengthy delays in obtaining approval of information collections such as visitor surveys under the PRA sometimes causes research to be postponed or even abandoned. In addition, Forest Service officials told us that the time needed to obtain clearance for surveys is a major barrier in gathering input from customers on their level of satisfaction. Forest Service officials told us that because approval for new collections of information often takes in excess of a year, which is in addition to the time for collection, data entry, and reporting, it is not possible to include customer input in many time-sensitive decisions. OMB representatives noted that the reported lengths of time to clear submissions include the time for internal agency development and processing and that some delays can be the result of inappropriate submissions. OMB representatives told us that once an Information Collection Request (ICR) is submitted to OMB, the average time for review at OMB is typically less than the required sixty days. According to OMB representatives, the time to obtain clearance could be reduced by using generic clearances, if the collections meet the criteria for generic clearances. Some other agencies we contacted also told us that the generic clearance process made PRA requirements manageable for surveying. For example, SSA officials told us that obtaining OMB clearance for its surveys had not presented a challenge, and that this was facilitated by a generic clearance for satisfaction surveys that SSA has had since the mid-1990s. Likewise, FSA officials told us that the expedited review process associated with a generic clearance allows it to get surveys approved within 10 to 15 days and that most problems with surveys that arise during the process are quickly resolved. As required by a presidential memorandum issued on January 21, 2009, OMB issued an Open Government Directive, which among other things, instructs the Administrator of the Office of Information and Regulatory Affairs (OIRA) to review existing OMB policies, such as PRA guidance, to identify impediments to open government and the use of new technologies, and where necessary, issue clarifying guidance and/or propose policy revisions. OMB recently released three memorandums containing clarifying guidance to improve the implementation of the PRA. The memorandums provide information for federal agencies to facilitate their understanding of PRA clearances and when and how they can be used. The most recent memorandum, issued on May 28, 2010, outlines the availability and uses of generic information collection requests. The other two memorandums, issued on April 7, 2010, relate to the central requirements of the PRA and the treatment of social media and Web-based interactive technologies under the PRA. All surveyed services reported that within the last 12 months they gathered ideas for improving customer service from frontline employees who are in contact with customers. The Executive Order is not prescriptive in the method agencies should use to survey employees, and the services reported employing a variety of methods to survey their employees for ideas for improving customer service. These included staff meetings, blogs, employee suggestion programs, and employee surveys. Passport Services officials, for example, stated that each regional Passport Agency location is staffed with at least one customer service manager charged with reviewing the ideas and suggestions from staff within the agency or center. The manager then communicates the information to colleagues at other locations and the customer service division at headquarters to initiate further discussion and action on suggestions. To encourage customer service managers to gather and share ideas, there is a bi-weekly teleconference, an e-mail list, a monthly customer service report, and an annual conference. SSA officials stated that some of its offices have established councils, which include frontline employees, to look for new ideas to improve customer service. At the Forest Service, input from employees is gathered primarily at the individual forest level with no formal mechanism in place to communicate employee input at a national level. The five additional services we interviewed that had less widespread contact with the public and different customer groups or missions than those we surveyed perform many of the same customer service management activities as those we surveyed. All five had customer service standards. For example, the Centers for Disease Control and Prevention (CDC) has a target that 75 percent of customers will report being very satisfied, and the Nuclear Regulatory Commission has a standard that 90 percent of power plant licensing actions will be completed within 1 year of receipt. All five services reported measuring service results. Four of these had measures related to customer satisfaction, and most used the ACSI as a measure. In addition, four had measures of customer access, wait times or other measures. For instance, CDC measures the percentage of phone calls answered in less than 30 seconds. All of the additional services reported having methods to receive and act on customer complaints, such as through distributing feedback forms in person or through the agency Web site, as well as through designated e-mail and phone numbers or call centers, which were also used by the surveyed services. All five services gather ideas from employees on improving customer service. For example, GSA’s Federal Acquisition Service is using the social network tool “Yammer,” as a mechanism to gather internal employee input. About half of surveyed services’ standards were not in a form readily available to the public, as the March 1995 presidential memorandum on improving customer service requires. We found that the standards were often either not made available to the public at all or were made available in a way that would not be easy for customers to find and access. Ten services reported making standards available to customers through both government Web sites and government publications; one service, TSA’s Passenger and Baggage Screening, reported making standards available through only a government Web site; and two services did not make standards available at all. Five of the 11 services that reported they make standards available do so in documents or Web sites that are not likely to be viewed by customers. These services made their standards available in long, detailed documents mostly focused on other topics, such as Annual Performance Plans, Performance and Accountability Reports, and budget justifications. The two services that did not make standards available to customers were Federal Student Aid (FSA) and Forest Service for recreational facilities and services. FSA officials reported that they do not make their standards for the Direct Loan Program, which are standards for contractor performance, available to customers because they were not intended to inform the public. Similarly, the Forest Service officials told us that they have standards that are part of a contract with an external service provider, but because they are not in an understandable format and presentation for customers they are not made available. Forest Service officials also reported that Forest Service has standards employees must meet in the operation of recreation program areas that are not consistently shared with the public, such as the frequency with which rest rooms must be cleaned. Forest Service officials feel the standards would not be helpful to the visitors who evaluate such things as “cleanliness” of rest rooms against their own standards rather than the frequency of cleaning. Some services made their standards more readily available to customers. For example, U.S. Customs and Border Protection (CBP) has a Web page called “Know Before You Go,” http://www.cbp.gov/xp/cgov/travel/ vacation/kbyg/, which communicates regulations for international travel by U.S. residents. This Web page, which is also available as a document, specifically identifies customer service standards, such as a pledge to explain the CBP process to customers, and provides information and instructions to customers on how they can express their service concerns to CBP representatives. The standards are also available through a link on CBP’s travel customer service page, http://www.cbp.gov/xp/cgov/travel/ customerservice/. Five services, including the Indian Health Service, the Passport Service, the Veterans Health Administration, CBP, and SSA, make standards available by posting them in government offices open to the public. In addition to standards, all services we surveyed stated that they report customer service results to customers through a government Web site, and 11 of these services told us that they also make results available through written publications. Further, the Indian Health Service and the Veterans Health Administration also post service results in government offices open to the public. However, most services did not post service results in government offices. Two services, IRS and Forest Service, told us their results are too lengthy to post in their public offices. Similar to the reporting of customer service standards, about half of services posted customer service results in documents that may not be easily accessible to customers. Six services we surveyed only report customer service results in their Performance and Accountability Reports, Congressional Budget Justifications, or other documents that are targeted to larger or different audiences than customers, or through Web sites such as the ACSI Web site that customer may not know to visit to find customer service results. However, some agencies made their customer service results more readily available to customers. For example, CBP has a web page with airport and border wait times that can be accessed directly from CBP’s main travel page, http://www.cbp.gov/xp/cgov/travel/. In another example, the Veterans Benefits Administration has a customer service page for Veterans Group Life Insurance with service results that can be accessed from the insurance home page, http://www.insurance.va.gov/. As required by the presidential memorandum, all services report results to customers at least annually. In addition, more than half of services post some customer service results as frequently as monthly, and in one case, results are posted as frequently as hourly. Customs and Border Protection collects hourly wait time estimates for land border ports of entry, from 70 of the largest land border crossings on the Northern and Southwestern borders. This information is posted directly from the ports of entry to CBP’s wait times Web site on an hourly basis so that the public can use the information in trip planning. For airline travel, CBP collects detailed daily flight information at the arrival terminals of 48 of the busiest air ports of t air ports of entry. This information is assembled into an historical database that entry. This information is assembled into an historical database that provides hourly flight processing time estimates for any tim provides hourly flight processing time estimates for any time of day, and is available to the public on the airport wait times Web site. e of day, and is available to the public on the airport wait times Web site. All services we surveyed reported that they compare their performance to their service standards, as Executive Order 12862 requires for agencies that provide significant services directly to the public. All services but two reported doing so at least monthly, and five services reported making daily comparisons. For example, SSA officials said their performance on customer service measures, such as the average speed of answer and busy rate for calls to their national phone number are tracked and compared to monthly standards, and the results are reported on an internal tracking report that is distributed to SSA’s Commissioner and executive staff. CBP produces daily wait times exception summary reports for internal Headquarters and regional management of the ports of entry. Officials said wait times in excess of 1 hour at either air or land border ports require explanation and are tracked to monitor ongoing problems and develop mitigation strategies. Further, all services reported that they use the results of their customer service measures to improve customer service. Several services reported using performance data to improve training, allocate staff, and improve phone systems and Web sites. For example, the Passport Services reported that they use measures of passport center workloads to ensure timely processing of applications by transferring application processing work between centers at various times, such as during inclement weather. In addition the Internal Revenue Service said that it analyzed wait times for taxpayer assistance to improve service by making staffing adjustments and routing certain types of activities to specific employees. CBP officials said land border, and airport wait time patterns have been studied, which led to facility enhancements and staff assignment changes. The officials told us that changes at one port of entry, the Detroit Ambassador Bridge reduced wait times and recurring traffic delays by more than half. Medicare officials said the 1-800-Medicare call center used customer service measure results to improve and clarify the script used by its customer service representatives to answer beneficiary questions. Medicare officials also stated that their review of the volume, nature, and turnaround time for complaints about private companies marketing Medicare products led them to apply a more rigorous turnaround time requirement for staff handling these complaints. Seven services reported that they had compared customer service performance against performance in the private sector in the prior 12 months, and these seven also reported that they had used this comparison to improve their customer service. For example, officials from the Direct Student Loan Program (DLP) said they obtained information from the private sector that borrowers using electronic services (such as electronic debit and services on Web sites) were among the most satisfied. Subsequently, DLP implemented initiatives to increase borrowers’ awareness of online services and provided information on how to enroll in these services. Similarly, officials from Veterans Benefits Administration’s Veterans Group Life Insurance (VGLI) said comparisons against performance in the private sector offered ongoing confirmation that their measures and practices remain competitive against industry standards. For example, officials said awareness of changing technology in the private sector has led to enhanced self-service Web-based features that include the ability to pay VGLI premiums, add or change beneficiaries, update customer demographic information, and print certificates of coverage online. Officials from SSA told us that their national call centers operate efficiently by using sophisticated call forecasts and changing agent shift assignments to better match projected call patterns based on methodology used in the private sector. Finally, Medicare officials said after comparing the program’s turnaround time for urgent complaints with that of private sector health plans, they made a decision to revise their internal standard to align with the private sector plans. However, four of the services reported they had not compared their customer service performance against performance in the private sector as Executive Order 12862 requires, and one other service had not done so recently. Officials of three of the four services that had not compared customer service performance against the private sector, CBP, TSA, and the Bureau of Consular Affairs, told us that they do not have a comparable private entity to compare with. CBP officials, however, also told us they have worked with a number of private companies to identify improvements that could be made to the federal inspection service areas that would reduce confusion, shorten delays, and improve customer service, such as better managing passenger lines and improving CBP signs at airports. In addition, CBP is working closely with several private sector regional organizations, such as the U.S.-Canada Border Trade Alliance, to develop technology solutions for improving vehicle processing and traffic management at the ports of entry. Forest Service officials told us that they had worked with a private company that runs campgrounds to develop standards in the late 1980s, but have not had funding for benchmarking against the private sector since then. Most surveyed services told us that they base performance appraisals for their employees, in part, on customer service measures; these include members of the Senior Executive Service (SES) or equivalent, managers, supervisors of frontline employees, and frontline employees themselves. Office of Personnel Management regulations require agencies to establish performance management systems that evaluate SES performance using measures that balance organizational results with customer, employee, and other perspectives. Eleven of the 13 services we surveyed reported that they based performance appraisals for all SES in part on customer service performance measures, and two services reported that they based some SES performance appraisals on customer service performance measures. In addition, all services reported that performance appraisals for managers and supervisors of employees in contact with customers were based in part on customer service performance measures. Finally, 11 surveyed services reported that performance appraisals for all employees in contact with customers were based in part on customer service performance measures. Among those services that based performance appraisals on customer service performance measures, the extent to which they did so varied by service and job type. For example, performance appraisals for all employees in Federal Student Aid’s Direct Loan Program have two sections: organizational priorities and customer service. Officials from FSA told us that some staff are assessed for service to internal or external customers depending on the assigned job and some lower level employees have quantitative measures in their appraisals, such as accuracy and the average time it takes to resolve issues. For other services, performance appraisals are based on qualitative measures of customer service. For example, officials from IRS told us that, for taxpayer assistance, customer satisfaction knowledge and application have been critical job elements in performance appraisals for all employees, including those in their bargaining unit, for at least 10 years. Figure 5 shows descriptions of IRS’s customer satisfaction knowledge and application critical job elements. In addition, all employees are rated on whether they have met the standard for the fair and equitable treatment of taxpayers developed in accordance with the Internal Revenue Service Restructuring and Reform Act of 1998. The standard states, “Administer the tax laws fairly and equitably, protect all taxpayers’ rights, and treat each taxpayer ethically with honesty, integrity, and respect.” The inclusion of customer satisfaction in employee performance appraisals is part of the agency’s collective bargaining aining agreement with the union. agreement with the union. This individual performance critical job element describes how the employee promotes the satisfaction of taxpayers and customers through professionally and courteously identifying customers’ needs and/or concerns and providing quality products and services. Communication to the customer is appropriate for the issue and encourages voluntary compliance. This individual performance critical job element describes how the employee promotes the satisfaction of taxpayers and customers by providing the technical expertise to serve the customers with professional and helpful service. Accurate identification and resolution of issues and the correct interpretation of laws, rules, regulations and other information sources are key components of this critical job element. Officials from the National Treasury Employees Union (NTEU) and the American Federation of Government Employees, the two unions that represent the largest number of federal employees, both expressed concerns about the use of customer service performance measures in performance appraisals. Officials from both unions told us that many employees do not have control over the customer service results achieved, and one said that customer service performance is best addressed at the agency level. This official also cautioned that using customer service measures, such as the time to handle a case, in performance appraisals could lead to employees overlooking details of the case as they attempt to save time. Nevertheless, agency officials did not report recent issues related to including customer service as part of performance appraisals. Similarly, IRS and NTEU officials told us there had not been any recent concerns or issues related to the way customer service measures are used in performance appraisals. Most of the services we interviewed that had less widespread contact with the public and often different customer groups or missions than those we surveyed reported making customer service standards and results available to customers and using customer service results for various purposes. Four of the five services reported making their customer service standards available to customers, but only three of the four services post the standards where they are likely to be viewed by customers. Similarly, three of the five services reported making customer service results available to customers, but two posted the results in documents that may not be readily available to customers. Three services reported comparing the results to customer service standards, and two reported comparing results to the private sector. All five services reported using results of its customer service measures to improve customer service. Finally, three services reported basing performance appraisals in part on customer service measures. For example, performance appraisals for employees staffing CDC’s information contact center contain elements based on ensuring customer satisfaction for internal and external customers. A number of approaches have been used by state, local, and other national governments to improve customer service. Several approaches, which are also employed by some federal agencies, were identified as good practices in our literature review of customer service and customer service management and by knowledgeable current and former researchers and practitioners in these fields. These practices include methods to better understand customers’ needs, facilitate improved customer decision making, and provide citizens with the information necessary to hold government accountable for customer service performance. OMB is planning several initiatives designed to facilitate the use of many of these approaches across federal agencies. Organizations may be able to increase customer satisfaction by better understanding customer needs and organizing services around those needs. Governments of both the United Kingdom (UK) and New Zealand reported that research into customer needs helped them establish drivers of customer satisfaction. Once they identified the drivers, they used them to set customer service standards to better meet customer needs. In the United Kingdom, the Cabinet Office commissioned a nationally representative survey that yielded a set of five drivers of customer satisfaction in the United Kingdom. These drivers, listed in order of impact on customer satisfaction, include: Staff deliver the outcome as promised and manage any problems Staff address customer requests quickly and directly Information given to customers is accurate and comprehensive Staff are competent and treat customers fairly Staff are friendly, polite and sympathetic to customers’ needs In 2008, the UK government created a standard called Customer Service Excellence (CSE) that includes five criteria for evaluating customer service quality based on the five drivers of customer satisfaction. Organizations can apply for formal CSE certification in which their performance is measured against 57 sub-elements of the five CSE criteria by licensed certification bodies, accredited by the United Kingdom Accreditation Service. For example, one sub-element asks organizations to demonstrate that they evaluate customer satisfaction by asking customers specific questions related to the drivers of satisfaction, such as timeliness, delivery and information. The New Zealand Government’s State Services Commission (SSC), New Zealand’s central public service management agency, works with government service providers to monitor and improve performance. SSC conducted its own citizen survey, Kiwis Count 2007, to determine citizens’ perspectives on drivers of customer satisfaction. Although similar to the United Kingdom’s drivers in many respects, the SSC reported that New Zealanders also considered the value of a service relative to taxpayer investment when evaluating the overall customer service experience. The six drivers are: Service experience met citizen expectations Staff were competent Staff kept their promises—they did what they said they would do Citizens were treated fairly Citizens felt their individual circumstances were taken into account Citizens felt the service was an example of good value for tax dollars The SSC reported that respondents rated the “service met your expectations” driver as the most important driver of customer satisfaction, followed by “staff competency.” Although the New Zealand government does not have national standards, SSC worked with agencies to help them better understand what the drivers mean and how to appropriately set agency level standards. Additionally, SSC, in collaboration with other central agencies, developed a Performance Improvement Framework (PIF) to assess performance and drive improvements. According to an SSC official, the PIF includes a component that examines how well agencies meet customer expectations. In the future, this component may be expanded to examine the impact of basing standards on the drivers of satisfaction. In addition to establishing drivers of satisfaction, segmenting the population into groups and providing differentiated service delivery can be an effective strategy to better meet diverse customer needs. In 2005, the government of Canada reorganized its diverse set of service providers under one umbrella service organization, called Service Canada, that offers citizens a single point of access to a wide range of government services and seeks to make access to services easier, quicker, and more convenient. As one part of its efforts to improve service delivery, Service Canada crafted a segmentation strategy centered around seven subpopulations: workers, seniors, people with disabilities, Aboriginal people, newcomers to Canada, youth, and families. Service Canada then tailored its service delivery processes to the needs of each subpopulation. Specifically, Service Canada outlined priority service issues for each population segment, as well as marketing approaches tailored to better inform customers about available services. For instance, to improve its outreach to Aboriginal populations, Service Canada created fact sheets in 11 Aboriginal languages, hired staff that spoke Aboriginal languages and modified the distribution of office locations to better serve remote and northern communities. Service Canada attributes improved results in the areas of citizen satisfaction, access to services and efficiency of service delivery to the implementation of its segmentation strategy. Organizations can also use social media to better understand and engage their customers. Social media can facilitate low effort communication between customers and service providers. It allows service providers to disseminate up to date and relevant information that may lead to improved customer decision making, while also allowing customers to provide feedback on their experiences. In order to encourage the use of social media by federal agencies, the General Services Administration (GSA) sponsors a Web site, http://www.usa.gov/webcontent/, that is managed by the Federal Web Managers Council, an interagency group of Web managers from every cabinet-level agency and numerous independent agencies. The Web site contains information on the benefits of using social media for customers of government agencies, as well as detailed advice on social media best practices. For example, it states that agencies can use microblogs to both provide timely information to citizens and improve understanding of customer needs by searching microblogs for references to their agency name or acronym. Additionally, GSA has negotiated terms of service with several social media vendors to make it easier for agencies to employ these tools on their Web sites. TSA is an example of an agency that has been using social media to engage and communicate with customers. TSA communicates policy changes and other relevant information via articles on its blog, http://blog.tsa.gov/, and allows customers to post comments and complaints. Using the feedback obtained from their blog, TSA learned from customers when policies weren’t being implemented appropriately in various airports or regions and made changes when appropriate. For example, a blog comment prompted the TSA Blog Team to investigate and ultimately stop a local airport policy that required passengers to remove all small electronics for individual screening. Additionally, social media may reduce the effort required of customers to complete service transactions. San Francisco enables city residents to sign up as followers of the San Francisco’s 3-1-1 customer service center on Twitter and send short messages containing service requests and complaints, rather than trying to reach city customer service representatives by phone. Providing this additional channel of communication has the potential to reduce the customer effort required and improve customer service. However, service transactions may require customers to communicate more detailed information than microblogs generally allow. Providing customers autonomy and control by allowing them to serve themselves can also be an effective strategy to improve customer service. Self service options may reduce customer effort and provide customers with information that allows them to make informed decisions. In addition, customers often report a preference to self-serve over speaking to a representative. Several U.S. federal agencies allow customers to self-serve via the Web. For example, customers can make reservations online for National Park Service, Forest Service, and Bureau of Land Management campsites. Similarly, FSA allows students to apply for financial aid online, manage their accounts and make payments over the Internet. Officials from Service Canada told us that they have started a campaign entitled “Why Wait in Line When You Can Go Online” to encourage the use of online services; they have also been working to ensure that self-service tasks are easy to complete. Agencies such as the Social Security Administration and the Department of Veterans Affairs now use their Web sites to help customers identify benefits that may be available to them and to develop a personalized estimate of those benefits. The Social Security Administration provides a Benefit Eligibility Screening Tool that enables customers to determine benefits they may be eligible for, as well as calculators that estimate future retirement, disability, and survivor benefits based on current law and the citizen’s earnings record, and the next steps to apply. In 2002, 10 federal agencies partnered to develop a Web site, http://www.GovBenefits.gov. The Web site helps customers find government benefit programs for which they may be eligible and then provides information on the next steps to learn more about and apply for those eligible benefits. As of fiscal year 2010, the initiative included 1,000 programs provided by 17 federal partners. Providing redress to customers when standards are not met can enhance the effectiveness of standards. For example, Service Ontario, a provincial partner of Service Canada, refunds the birth certificate fee a customer pays if the certificate is not issued within the established timeliness standard. In the United Kingdom, London TravelWatch, a local consumer watchdog organization established by Parliament, investigates complaints made by travelers using transportation services in London, including the London Underground and London’s buses, and makes recommendations for recompense when appropriate. In its literature, TravelWatch cited examples of Transportation providers dispensing compensation, such as ticket refunds, when they agreed with TravelWatch’s findings. California State University at Long Beach (CSULB) established a one year pledge relating to the performance of its College of Education graduates. If first year teachers experience problems at their school of employment, CSULB will assist them in areas related to their credential preparation. Including customer service measures in performance evaluations of frontline employees may be an effective strategy for improving customer service. As previously discussed, most of the services we surveyed already consider customer service measures in employee performance appraisals, though the extent and weight varied widely by job and service. However, as noted earlier, there can be challenges to creating effective performance standards for frontline employees. A primary concern is that some performance standards, such as call duration, are easy to measure but ignore the tradeoff between efficiency and quality customer service. Call center employees being judged solely on call duration might sacrifice the quality of the customer service they provide in order to end calls more quickly. One solution is to create performance metrics that attempt to balance operational efficiency and quality service. For example, telephone agents at the nonemergency services call center in the city of Denver, Colorado, are graded using a balanced scorecard which takes into account both call duration and whether or not the agent resolved a customer’s issue without having to transfer the caller to another employee. Because this measure emphasizes service quality, it serves to counterbalance the incentive to rush through service calls. A representative from the Georgia Office of Consumer Affairs (OCA) raised the additional concern that including customer service measures in performance appraisals may be perceived as a mechanism for placing blame on frontline employees. In order to gain employee support, the OCA representative recommended that agencies engage employees in the conversation about customer service management and seek employee input about how management can help them provide better customer service. The representative from the OCA reported that involving employees in the development of new processes has led to increased employee trust in management and openness to setting common goals and performance standards. Starting early in 2010, the Office of Management and Budget (OMB) began planning several initiatives to promote federal agencies’ responsibility for quality customer service to their customers. On September 14, 2010, President Obama issued a presidential memorandum to the Senior Executive Service (SES) on the Accountable Government Initiative. The memorandum was accompanied by a memorandum from OMB’s Deputy Director for Management to the SES outlining, among other things, the steps OMB is taking on customer service. An OMB representative told us they have begun working with GSA’s Office of Citizen Services and other agencies to generate and share ideas and improve customer service. OMB plans to accomplish this by holding agency discussion groups, one-on-one meetings with private sector CEOs who participated in a forum on modernizing government on January 14, 2010, and meetings with officials who were part of the National Performance Review. An OMB representative stated that they have already begun holding some meetings. Planned topics for discussion include: offering agency services online, coordinating services provided across multiple points of contact and examining how agencies gather and use customer feedback. The OMB representative told us this will involve looking at the Paperwork Reduction Act clearance process, which entails OMB approval of agency data collection prior to accumulating feedback from customers. As a part of its broader initiative with GSA’s Office of Citizen Services, OMB is also developing a pilot dashboard which contains agency standards and some related measures, with links to agency Web sites where customers can track their individual transaction status, where available. OMB has asked agencies participating in the pilot to identify metrics that are drivers of customer satisfaction, such as wait time, processing time, and first call resolution, and is currently reviewing their proposals. OMB expects the pilot dashboard to launch publicly in late fall 2010. On April 19, 2010, Presidential Executive Order 13538 established the President’s Management Advisory Board within the General Services Administration, to be chaired by the administration’s Deputy Director for Management. The board’s mission is to provide advice and recommendations on effective strategies for the implementation of best business practices related to federal government management and operation. OMB is selecting members and it expects to hold the first meeting of the board by the end of 2010. The board is expected to focus on improving productivity, the application of technology, and customer service. In addition, in its fiscal year 2011 Budget planning memorandum on June 11, 2009, OMB outlined its guidance to agencies to identify a limited number of high-priority performance goals for the next 12 to 24 months. These goals are intended to foster accountability and the chances that the federal government will deliver results on what matters most by making agencies’ top leaders responsible for specific goals that they themselves have named as most important. Although the guidance did not require agencies to create goals specifically related to customer service, more than half of the services we surveyed had at least one goal related to customer service in their agencies’ high-priority performance goals. For example, as part of its goal to improve customers’ service experience on the telephone, in field offices, and online, the Social Security Administration has a goal to increase the percentage of customers who rate service as “excellent,” “very good, “ or “good” from 81 percent to 83.5 percent. Also, the Department of Veterans Affairs has a goal implement a 21st Century paperless claims processing system by 2012 to ultimately reduce the average disability claims processing time to 125 days. See appendix IV for more examples of high-priority performance goals that relate to customer service. The elements of a customer-centered approach to delivery of federal service are common among those services with the most widespread contact with the public that we surveyed, as well as those we interviewed with less direct contact. All 13 government services we surveyed report they had established customer service standards, measured service results, and shared the results with customers. In a number of instances, the services report improvements in the quality of service delivered and customer satisfaction. Further, the fact that more than half of the services we surveyed had a specific goal related to customer service among their agencies’ High Priority Performance Goals indicates that these services recognize the importance of customer service. While a number of services have not encountered problems with survey clearances, some services that obtain customer input through surveys and other methods, which is critical to understanding the level of customer satisfaction, reported challenges related to obtaining PRA clearance for these activities. These challenges can lead to missed opportunities to involve customers in decision making. OMB has recently issued clarifying guidance on the PRA clearance process, including guidance on obtaining generic clearances, though it remains to be seen whether the guidance will reduce agency challenges or increase effective agency use of the generic clearance process. Communicating customer service standards and results in a way that is useful and readily available to customers is important in enabling them to hold government accountable and to inform customer decision making. Most services we contacted do make customer service standards and results available to customers, but many do so through documents that serve larger purposes, such as Performance and Accountability Reports and Budget Justifications which, while not excluding customers, are targeted to a much broader audience. On the other hand, some services make standards and results readily available to customers in documents, websites or government offices specifically targeted to customers to better deliver service and achieve higher levels of customer satisfaction. The OMB pilot dashboard initiative has the potential to facilitate agency efforts to make customer service standards and results readily available to customers, but has not yet been launched. Most services reported they base employee performance ratings from SES to frontline employees, in part, on customer service measures, but the manner and objective weight that attaches to this varied. There seems to be widespread agreement among services, management, and labor officials that customer service is an important factor in assessing employee performance, although there is also some reservation relative to how conclusively service measures can and should be applied at the individual employee level. But while labor officials and others expressed concerns about lack of employee control over variables that may affect the quality, accuracy, processing time, and level of a customer’s satisfaction, agency officials did not report recent issues. For example, IRS customer service is included in IRS’s performance appraisal system and IRS and NTEU officials stated that there had not been any recent issues with the system relating to customer service. While the experience of several of the surveyed services suggests that the use of service measures in performance appraisals can be effective and appropriate, as current and former researchers and practitioners pointed out, they need to be developed with care, particularly balancing all dimensions of customer service and involving employees in their selection and application. Tools and practices identified in our review, such as using social media to engage customers and segmenting customer groups to provide tailored services based on particular needs, could lead to potential benefits for customers of federal agencies. Some of these are already being used by some federal agencies, but OMB’s initiative to gather and share customer service ideas through the President’s Management Advisory Board, established in April 2010, and meetings with GSA’s Office of Citizen Services, agencies, and other groups offers an opportunity to evaluate the benefits of applying these tools and practices on a more widespread basis and to share those that are found to be beneficial. We recommend that the Director of OMB take the following two actions, building on the progress OMB has already made as part of its customer service initiative: Direct agencies to consider options to make their customer service standards and results more readily available to customers using documents or Web pages specifically intended for customers, or the dashboard once it is more fully developed. Collaborate with the President’s Management Advisory Board and agencies to evaluate the benefits and costs of applying the tools and practices related to understanding customers’ needs, facilitating improved customer decision making, and providing citizens with the information necessary to hold government accountable for customer service, and include those that are found beneficial to the federal government in the initiative on gathering and sharing customer service ideas. We provided a draft of this report for review to OMB and the departments of Agriculture, Education, Health and Human Services, Homeland Security, the Interior, State, the Treasury, Veterans Affairs, and the General Services Administration, the National Science Foundation, the Nuclear Regulatory Commission, and the Social Security Administration. OMB provided comments and additional information regarding the PRA review process. We made changes as appropriate to describe the process more fully. OMB had no comments on the recommendations. The departments of Health and Human Services, Homeland Security, the Treasury, and Veterans Affairs and the Social Security Administration provided technical comments, which we incorporated where appropriate. The departments of Agriculture, Education, the Interior, and State and the General Services Administration, National Science Foundation and Nuclear Regulatory Commission had no comments. We are sending copies of this report to the appropriate congressional committees; the Secretaries of Agriculture, Education, Health and Human Services, Homeland Security, the Interior, State, and Veterans Affairs; the Commissioners of the Internal Revenue Service and the Social Security Administration; the Administrator of the General Services Administration; the Directors of the National Science Foundation and OMB; the Chairman of the Nuclear Regulatory Commission; and other interested parties. The report will also be available at no charge on the GAO Web site at http://www.gao.gov. If you have any questions concerning this report, please contact Bernice Steinhardt at (202) 512-6543 or steinhardtb@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix V. The objectives of our study were to (1) assess the extent to which federal agencies are setting customer service standards and measuring results against these standards, (2) assess the extent to which federal agencies are reporting standards and results to customers and using the results to improve service, and (3) identify some customer service management tools and practices used by local, state, federal, and non-U.S. national governments. In addition, we examined steps the Office of Management and Budget (OMB) is taking to facilitate federal agency use of tools and practices to improve customer service. To assess whether and how federal agencies are setting customer service standards, measuring results, reporting those results and using them to improve services, we conducted a survey, based on the requirements of Executive Order 12862 and the related presidential memorandum “Improving Customer Service.” We surveyed 13 services provided by federal agencies that are among those with the most widespread contact with the public. We selected a nonrepresentative sample of government services because there is no single list of government services that could be used to pull a representative sample. We selected the sample based on a set of criteria including: services provided by a federal agency; 1 million customers served annually; customers are primarily U.S. individuals; and primary customers are not employees of a government agency receiving benefits related to employment. We compiled the list of potential services by starting with a list of Vanguard agencies (agencies having the most contact with the public) that was developed as part of the National Performance Review in the late 1990s and identifying services within those agencies. To this list we added agencies and respective services that were suggested by at least two of the five knowledgeable individuals in the area of customer service whom we consulted, or by one of the knowledgeable individuals in the area of customer service and OMB. The final list of services surveyed can be found in appendix II. The sample of services surveyed was not a representative sample of services provided by the federal government, meaning that results from our survey cannot be generalized to apply to any other services provided by the federal government. To minimize errors that might occur from respondents interpreting our questions differently than we intended, we pretested our questionnaire with four officials who were in positions similar to the respondents who would complete our actual survey. During these pretests, we asked the officials to complete the questionnaire as we observed the process. We then interviewed the respondents to check whether (1) the questions were clear and unambiguous, (2) the terms used were precise, (3) the questionnaire was unbiased, and (4) the questionnaire did not place an undue burden on the officials completing it. We also submitted the questionnaire for review by a GAO survey methodology expert and four external reviewers who were experts on the topic of the survey (selected based on their experience managing or designing government performance improvement initiatives). We modified the questions based on feedback from the pretests and reviews, as appropriate. We sent the questionnaire by e-mail to the individual identified by the service as the lead respondent. We asked services to complete the questionnaire within the electronic form and return it as an e-mail attachment. All 13 services completed the questionnaire. We reviewed all questionnaire responses and followed up by phone and e-mail to clarify the responses as appropriate. We analyzed responses to closed-ended questions by counting and summarizing the type and frequency of response for each of the 13 services. For responses to open-ended narrative questions, we coded the responses from each service and created categories for the purpose of organizing and summarizing the response of each service (e.g., methods used to gather input from customers regarding their level of satisfaction with the service). The practical difficulties of conducting any survey may introduce nonsampling errors. For example, differences in how a particular question is interpreted, the sources of information available to respondents, or the types of respondents who do not respond to a question can introduce errors into the survey results. We included steps in both the data collection and data analysis stages to minimize such nonsampling errors. As indicated above, we collaborated with GAO survey specialists to design and review draft questionnaires, versions of the questionnaire were pretested with four officials from services not included in our survey but who were in positions similar to the respondents who would complete our survey, we asked several external experts to review and comment on a draft of the questionnaire, and we revised the questionnaire as necessary to reduce the likelihood of nonresponse and reporting errors on our questions. We examined the survey results and performed computer analyses to identify inconsistencies and other indications of error, and addressed such issues as necessary. A second, independent analyst checked the accuracy of all computer analyses to minimize the likelihood of errors in data processing. In addition, GAO analysts answered respondent questions and resolved difficulties respondents had answering our questions. For questions that asked respondents to provide a narrative answer, we created content categories and had one analyst code each response into one of the categories, and another analyst verify the coding. Any discrepancies in the coding were resolved through discussion by the analysts. We did not evaluate the overall effectiveness of or level of customer service provided by any of the services reviewed. To gain a fuller understanding of the survey responses, we selected five of the services with varying answers to key questions on the survey for follow-up interviews to discuss their responses. The five services selected were the Forest Service, National Park Service, Veterans Benefits Administration, Customs and Border Protection, and Federal Student Aid. In addition, because the scope of the survey was limited to services that have widespread direct interaction with the public, we selected and interviewed five additional services with a lower volume of contact with the public and different missions or goals. These five services were selected from a listing of independent agencies of the United States government, and were chosen to ensure at least two services were from each of the following categories: Services whose direct customers are individuals, but have fewer contacts with the public than those in our survey Services that serve government customers and are not subject to the requirements of Executive Order 12862 Services that benefit the public as a whole rather than individuals directly, such as agencies that make policy or regulate businesses The sample of additional services interviewed was not a representative sample of services provided by the federal government, meaning that results from the interviews cannot be generalized to apply to any other services provided by the federal government. A list of the five additional services we interviewed can be found in appendix II. The key topics presented in the survey formed the basis of the interviews with the five additional services. To prepare for analyses of the open- ended interview questions, we created content categories and had one analyst code each response into one of the categories (e.g., measures related to customer satisfaction), and another analyst verify the coding. Any discrepancies in the coding were resolved through discussion by the analysts. In order to gain insight into the current application of leading customer service tools and practices, challenges to implementing them and strategies to overcome these challenges, we first reviewed relevant literature, such as industry, academic, and management journals dealing with customer service practices and an annual evaluation of customer service provided by national governments begun in 2000. Based on the literature review, we selected and interviewed six knowledgeable individuals in the area of customer service. We selected these individuals based on their having one or more of the following characteristics: (1) experience managing or designing government performance improvement initiatives, such as the National Performance Review and the Government Performance and Results Act at the federal level or similar initiatives at the state and local level; (2) experience implementing, in a government setting, one or more of the customer service management tools and practices we identified in our literature review; and (3) published in peer reviewed journals, books, or frequently referenced publications in the field of public sector performance improvement. Based on suggestions from these individuals and the literature, we identified several customer service tools and practices. We then identified local, state and foreign organizations to interview that were either implementing tools and practices we had identified, were suggested by the individuals we contacted, or were highly ranked in the 2007 edition of the annual evaluation of customer service provided by national governments, which was the most recent edition that contained rankings. We obtained input from the following governmental organizations: Cabinet Office, United Kingdom Centerlink, Australia Central Provident Fund, Singapore Customer Service Group, New York City Office of Consumer Affairs, Georgia, USA Service Canada, Canada State Services Commission, New Zealand We used the input from these organizations to refine and provide context for the list of tools and practices we had identified. We conducted this performance audit in Washington, D.C. between August 2009 to October 2010 in accordance with generally accepted government auditing standards, which require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. As part of our methodology we asked all services involved in the survey to provide us with a copy of their customer service standards. Below are examples of the standards we received from each service. Customs and Border Protection (CBP) pledges to: Cordially greet and welcome you to the United States Treat you with courtesy, dignity and respect Explain the CBP process to you Have a supervisor listen to your comments Accept and respond to your comments in written, verbal or electronic form. Provide all reservation transaction processing with 100 percent accuracy including advance reservations, walk-ins, cancellations, and transfers. A fiscal year 2011 target of 83.5 percent has been established for percent of individuals who do business with SSA rating the overall services as “excellent,” “very good,” or “good” A target of 5 percent has been established for the percentage of abandoned calls to Veterans Benefits Administration Public Contact Representatives in the National Call Center A target of 70.0 percent has been established for the agent availability A target of 75.0 percent has been established for overall quality, includes measures of technical proficiency, client contact behaviors, and effective call management. The President’s budget for fiscal year 2011 included high -priority performance goals that agencies had committed to achieve within 18 to 24 months. The identification of high-priority performance goals, coupled with measures and targets, will be used by the President to evaluate agency progress in meetings with cabinet officers. Several agencies have crafted specific goals to improve customer service. These goals include targets and measures to improve customer satisfaction, as well as objective performance measures such as wait times, operational capacity, citizen engagement, and call center efficiency. This appendix documents examples of customer service oriented, high-priority performance goals drafted by federal agencies. Simplified Student Aid: All participating higher education institutions and loan servicers operationally ready to originate and service Federal Direct Student Loans through an efficient and effective student aid delivery system with simplified applications and minimal disruption to students. Improve security screening of transportation passengers, baggage, and employees while expediting the movement of the traveling public (aviation security). Wait times for aviation passengers (Target: Less than 20 minutes by 2012). Improve security screening of transportation passengers, baggage, and employees while expediting the movement of the traveling public (surface transportation security). Strengthen disaster preparedness and response by improving FEMA’s operational capabilities and strengthening State, local and private citizen preparedness. Improve to 90 percent the percentage of shipments arriving with the requested materials at the requested location by the validated/agreed upon delivery date. Improve to 95 percent the percentage of respondents reporting they are better prepared to deal with disasters and emergencies as a result of training. Increase individual income tax filers’ American Customer Satisfaction Index score to 69. Improve telephone level of service to at least 75 percent by the end of 2011. By the end of 2011, reduce the average number of days to complete original Post-9/11 GI Bill education benefit claims to 18 days. Implement a 21st-century paperless claims processing system by 2012 to ultimately reduce the average disability claims processing time to 125 days. Deploy a Veterans Relationship Management (VRM) Program to improve access for all Veterans to the full range of Department of Veterans Affairs services and benefits by June 2011. By the end of 2010, implement call recording, national queue, transfer of calls and directed voice and self help. By the end of 2010, enhance transfers of calls among all Veterans Benefits Administration lines of business with capability to simultaneously transfer callers’ data. By the end of 2010, pilot the Unified Desktop within Veterans Benefits Administration lines of businesses to improve call center efficiency. Increase the Number of Online Applications: By 2012, achieve an online filing rate of 50 percent for retirement applications. In 2011, the Social Security Administration’s (SSA) goal is to: Achieve 44 percent of total retirement claims filed online. Achieve 27 percent of total initial disability claims filed online. Issue More Decisions for People Who File for Disability: SSA will work towards achieving the Agency’s long-term outcomes of lowering the disability backlogs and accurately processing claims. SSA will also ensure that clearly disabled individuals will receive an initial claims decision within 20 days. Finally, the agency will reduce the time it takes an individual to receive a hearing decision to an average of 270 days by 2013. In order to efficiently issue decisions in 2011, SSA’s goal is to: Process 3.317 million out of a universe of 4.316 million initial disability claims. Achieve 6.5 percent of initial disability cases identified as a Quick Disability Determination or a Compassionate Allowance. Process 799,000 out of a universe of 1.456 million hearing requests. Improve SSA’s Customers’ Service Experience on the telephone, in field offices, and online: To alleviate field office workloads and to provide the variety of services the public expects, SSA will improve telephone service on the national 800-number and in the field offices. By fiscal year 2011, SSA’s goal is to: Achieve an average speed of answer rate of 264 seconds by the national 800-number. Lower the busy rate for national 800-number calls from 8 percent to 7 percent. Raise our overall rating of “excellent,” “very good,” or “good” given by individuals who do business with SSA from 81 percent reflected in 2009 to 83.5 percent. Provide agile technologies and expertise for citizen-to-government interaction that will achieve unprecedented transparency and build innovative solutions for a more effective, citizen-driven government. Create three readiness assessments and criteria based tool selection guidance by April 15, 2010. Provide assistance to other federal agencies in conducting six dialogs by September 30, 2010. Realize 136 million touch points (citizen engagements) through Internet, phone, print, and social media channels by September 30, 2010. Successfully complete three agency dialogs with the public to better advance successful use of public engagements by September 30, 2010. Train 100 government employees on citizen engagement in forums, classes, and/or Webinars that are rated highly successful by participants and linked to agency capability building and successful engagement outcomes by September 30, 2010. Hiring Reform: 80 percent of departments and major agencies meet agreed upon targeted improvements to: Improve hiring manager satisfaction with applicant quality. Improve applicant satisfaction. Reduce the time it takes to hire. Disaster Assistance: Process 85 percent of home loan applications within 14 days and 85 percent of business and Economic Injury Disaster Loan applications within 18 days. In addition to the individual named above, key contributions to this report were made by William Doherty, Assistant Director; Charlesetta Bailey; Mason Calhoun; Martin De Alteriis; Justin Dunleavy; Karin Fangman; Robert Gebhart; Colin Morse; Kelly Rubin; Michael Silver; Eugene Stewman; and Ethan Wozniak. | The federal government has set a goal of providing service to the public that matches or exceeds that of the private sector. Executive Order 12862 (September 11, 1993) and a related 1995 memorandum require agencies to post customer service standards and report results to customers. As requested, this report (1) assesses the extent to which federal agencies are setting customer service standards and measuring related results, (2) assesses the extent to which agencies are reporting standards and results to customers and using the results to improve service, and (3) identifies some customer service management tools and practices used by various governments. The report also examines the steps the Office of Management and Budget (OMB) is taking to facilitate agency use of tools and practices. GAO surveyed 13 federal services among those with the most contact with the public, reviewed literature and interviewed agency officials as well as knowledgeable individuals in the area of customer service. All 13 government services GAO surveyed had established customer service standards, which varied in their form from quantitative standards based on hourly, daily, monthly or annual averages to general commitments to qualitative standards. All 13 services reported having measures of customer service, such as measures of wait times or accuracy of service and 11 services had measures of customer satisfaction. For example, the National Park Service surveys visitors at over 320 points of service through a survey card program. All services had methods to receive customer complaints, and all had methods of gathering ideas from front line employees to improve customer service. Although standards exist, GAO found that the surveyed services' standards were often made available in a way that would not be easy for customers to find and access or, in the case of two services, were not made available to the public at all. For example, five services made standards available in long, detailed documents mostly focused on other topics, such as annual performance plans, performance and accountability reports, and budget justifications. About half of the services reported customer service results in similar types of documents. All services reported comparing customer service results to the standards and using the results to improve internal processes. For example, Customs and Border Protection officials told GAO that after they studied wait times at land borders and airports, they made facility enhancements and staff assignment changes. At one port of entry, these changes reduced wait times by more than half. However, some services have not compared performance to the private sector, as required by the Executive Order. Most services reported considering customer service measures in employee performance appraisals. For example, according to IRS officials, the performance appraisals for all employees who provide taxpayer assistance are based in part on critical job elements related to customer satisfaction. GAO identified several customer service tools and practices government agencies have used to improve customer service, such as engaging customers through social media, providing self service options and offering redress for unmet standards. Additionally, OMB has begun an initiative to identify and share private sector best practices among federal agencies and to develop a dashboard where agencies can make customer service standards available. Building on the progress made under this initiative, OMB could evaluate the benefits and costs of applying these tools and practices on a more widespread basis and share those that are found to be beneficial. GAO recommends that OMB (1) direct agencies to consider options to make customer service standards and results more readily available and (2) collaborate with the President's Management Advisory Board and agencies to evaluate the benefits and costs of applying the tools and practices identified in this report, and include those found beneficial in its related initiative. OMB had no comments on the recommendations. |
There is no single definition for financial literacy, but it has previously been described as the ability to make informed judgments and to take effective actions regarding current and future use and management of money. Financial literacy encompasses both financial education and consumers’ behavior as it relates to their ability to make informed judgments. Financial education refers to the processes whereby individuals improve their knowledge and understanding of financial products, services, and concepts. However, being financially literate refers to more than simply being knowledgeable about financial matters—it also entails utilizing that knowledge to make informed decisions, avoid pitfalls, and take other actions to improve one’s present and long-term financial well-being. Evidence indicates that many U.S. consumers could benefit from improved financial literacy efforts. In a 2010 survey of U.S. consumers prepared for the National Foundation for Credit Counseling, a majority of consumers reported they did not have a budget and about one-third were not saving for retirement. In a 2009 survey of U.S. consumers by the FINRA Investor Education Foundation, a majority believed themselves to be good at dealing with day-to-day financial matters, but the survey also revealed that many had difficulty with basic financial concepts. Further, about 25 percent of U.S. households either have no checking or savings account or rely on alternative financial products or services that are likely to have less favorable terms or conditions, such as nonbank money orders, nonbank check-cashing services, or payday loans. As a result of this situation, many Americans may not be planning their finances in the most effective manner for maintaining or improving their financial well-being. In addition, individuals today have more responsibility for their own retirement savings because traditional defined-benefit pension plans have declined substantially over the past two decades. As a result, financial skills are increasingly important for those individuals in or planning for retirement to help ensure that retirees can enjoy a comfortable standard of living. Federal financial literacy programs and resources are spread widely among many different federal agencies, raising concerns about fragmentation and potential duplication of effort. As we noted in our recent report on overlap, duplication, and fragmentation, in 2009, more than 20 different agencies had more than 50 financial literacy initiatives under way that covered a number of topics, used a variety of delivery mechanisms, and targeted a range of audiences. This distribution of federal financial literacy efforts across multiple agencies can have certain advantages. For example, different agencies can focus their efforts on particular subject matter or target specific audiences for which they have expertise. However, this fragmentation also increases the risk of inefficiency and redundancy and highlights the need for strong coordination of these efforts. Further, fragmentation of programs across many federal agencies can make it difficult to develop a coherent overall approach for meeting needs, identifying gaps, and rationally allocating overall resources. Because of the fragmentation of federal financial literacy efforts, coordination among agencies is essential to avoid inefficient, uncoordinated, or redundant use of resources. Identifying potential inefficiencies can be challenging because federal financial literacy efforts have numerous different funding streams and there are little good data on the amount of federal funds devoted to financial literacy. Financial literacy efforts are not necessarily organized as separate budget line items or cost centers within federal agencies and there is no estimate of overall federal spending for financial literacy and education, according to the Department of the Treasury. In part to encourage a more coordinated response to financial literacy, in 2003 Congress created the multiagency Financial Literacy and Education Commission and mandated that the Commission develop a national strategy. We conducted a review of the Commission in 2006 and made recommendations related to enhancing public-private partnerships, conducting independent reviews of duplication and effectiveness, and conducting usability testing of the Commission’s MyMoney.gov Web site. We subsequently reported that the Commission had made progress in cultivating sustainable partnerships with states, localities, nonprofits, and private entities, and had acted on our recommendation to measure customer satisfaction with its Web site. The Commission and the Department of the Treasury also initiated two independent reviews, as we had recommended, addressing overlap in federal activities and the availability and impact of federal financial literacy materials. As we have noted in the past, the Commission faces significant challenges in its role as a centralized focal point: it is composed of many agencies, but it has no independent budget and no legal authority to compel member agencies to take any action. Our 2006 review also found that while the Commission’s initial national strategy was a useful first step in focusing attention on financial literacy, it was largely descriptive rather than strategic. In particular, the national strategy was comprehensive to the extent of discussing major issues and challenges in improving financial literacy and describing initiatives in government, nonprofit, and private sectors. However, it did not include a plan for implementation and only partially addressed some of the characteristics we had previously identified as desirable for any effective national strategy. For example, although it provided a clear purpose, scope, and methodology, it did not go far enough to provide a detailed discussion of problems and risks; establish specific goals, performance measures, and milestones; discuss the resources that would be needed to implement the strategy; or discuss, assign, or recommend roles and responsibilities for achieving its mission. However, in December 2010, the Commission released a new national strategy that identifies five action areas—policy, education, practice, research, and coordination––and clearly lays out a series of goals and related objectives intended to help guide financial literacy efforts over the next several years. To supplement this national strategy, the Commission has said it will be releasing an implementation plan for the strategy by the end of this fiscal year. While the new national strategy clearly identifies action areas and related goals and objectives, it still needs to incorporate specific provisions for performance measures, resource needs, and roles and responsibilities, which we believe to be essential for an effective strategy. The new strategy will benefit if the forthcoming implementation plan incorporates these elements, as well as addresses the fragmentation of federal financial literacy efforts. More recently, the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act) requires the establishment of an Office of Financial Education within the new Bureau of Consumer Financial Protection, further underscoring the need for coordination among federal agencies on this topic. The Dodd-Frank Act charges the new office within the bureau with developing and implementing a strategy to improve financial literacy through activities including opportunities for consumers to access, among other things, financial counseling; information to assist consumers with understanding credit products, histories, and scores; information about saving and borrowing tools; and assistance in developing long-term savings strategies. This new office presents an opportunity to further promote awareness, coordinate efforts, and fill gaps related to financial literacy. At the same time, the duties this office is charged with fulfilling are in some ways similar to those of the separate Office of Financial Education and Financial Access within the Department of the Treasury. As noted above, the Dodd-Frank Act charges the Bureau of Consumer Financial Protection with developing and implementing a strategy on improving the financial literacy of consumers—one that is consistent with, but separate from, the strategy required of the Commission. Thus, these entities will need to coordinate their roles and activities closely to avoid unnecessary overlap and make the most productive use of resources. Coordination and partnership among federal, state, nonprofit, and private sectors are also essential in addressing financial literacy, and there have been positive developments in these areas in recent years. For example, a recent partnership between the National Credit Union Administration, the Department of Education, and the Federal Deposit Insurance Corporation aims to improve the financial education of millions of students. These three agencies are coordinating to facilitate partnerships among schools, financial institutions, federal grantees, and other stakeholders to provide effective financial education. Additionally, the National Financial Education Network, the President’s Advisory Council on Financial Capability, and the Community Financial Access Pilot all represent examples of progress in fostering partnerships among participants in financial education. For example, our review in 2009 found that the establishment of the National Financial Education Network was a useful initial action to facilitate and advance financial education at the state and local levels. Similarly, the President’s Advisory Council on Financial Capability facilitates strategic alliances among federal, private, and nonprofit enterprises. Although numerous financial literacy initiatives are conducted by federal, state, local, nonprofit, and private entities throughout the country, there is little definitive evidence available on what specific programs and approaches are most effective. As part of ongoing work we are performing in response to a mandated study in the Dodd-Frank Act, we are conducting a review of studies that have evaluated the effectiveness of financial literacy efforts. More than 100 articles, papers, and studies have been published on the general topic of financial literacy since 2000, but our preliminary findings have identified only about 20 papers that constitute empirically based evaluations on the effectiveness of specific financial education programs. In addition, only about 10 of these studies actually measured the impact of a program on participants’ behavior rather than simply identifying a change in the consumer’s knowledge, understanding, or intent. This distinction is important because a change in behavior is typically the ultimate goal of any financial literacy program, and changes in behavior do not necessarily follow from changes in knowledge or understanding. We are currently in the process of analyzing the results of these studies and look forward to reporting more fully on our findings this summer. But in general, the consensus among a wide variety of stakeholders in the field of financial literacy is that relatively little is known about what financial literacy approaches are most effective in meaningfully changing consumers’ financial behavior. The limited number of rigorous, outcome-based evaluations of financial literacy programs is likely the result of several factors. Because the field of financial literacy is relatively new, many programs have not been in place long enough to allow for a long-term study of their effectiveness; many of the key federal financial literacy initiatives were created only within the past 10 years. In addition, experts in financial literacy and program evaluation have cited many significant challenges to conducting rigorous and definitive evaluations of financial literacy programs. For example, measuring a change in participant behavior is much more difficult than measuring a gain in knowledge, which can often be captured through a simple post-course survey. Similarly, financial literacy programs often seek to effect change over the long term, which means that effective evaluation can require ongoing follow up with participants—a complex and expensive process. In addition, discerning the impact of the financial literacy program as distinct from other influences, such as changes in the overall economy, can often be difficult. Nonetheless, given that federal agencies have limited resources, focusing federal financial literacy resources on initiatives that work is important. Some federal financial literacy programs, such as the Federal Deposit Insurance Corporation’s Money Smart, have included a strong evaluation component, while others have not. The Financial Literacy and Education Commission and many federal agencies have recognized the need for a greater understanding of which programs are most effective in improving financial literacy. The Commission’s original national strategy in 2006 noted, for example, that more research and program evaluation are needed so that organizations are able to validate or improve their efforts and measure the impact of their work. In response, in October 2008, the Department of the Treasury and the Department of Agriculture convened, on behalf of the Commission, the National Research Symposium on Financial Literacy and Education, which discussed academic research priorities related to financial literacy. Moreover, we are pleased to see that the Commission’s new 2011 national strategy sets as one of its four goals to “identify, enhance, and share effective practices.” The new strategy sets objectives for reaching this goal that include, among other things, (1) encouraging research on financial literacy strategies that affect consumer behavior, (2) establishing a clearinghouse for evidence-based research and evaluation studies, (3) developing and disseminating tools and strategies to encourage and support program evaluation, and (4) forming a network for sharing research and best practices. These measures are positive steps in helping ensure that, in the long term, scarce resources are focused efficiently and effectively. At the same time, as we have noted in the past, an effective national strategy goes beyond simply setting objectives; it also must describe the specific actions needed to accomplish goals, identify the resources required, and discuss appropriate roles and responsibilities for the players involved. We encourage the Commission and its participating agencies to incorporate these elements into the national strategy’s implementation plan, which is slated to be released later this year. In addition, it is important to note that financial education is not the only approach—or necessarily always the best approach—for improving consumers’ financial behavior. Alternative strategies or mechanisms, sometimes in conjunction with financial education, have also been successful in improving financial behavior. In particular, insights from behavioral economics that recognize the realities of human psychology have been used effectively to design strategies to assist consumers in reaching financial goals without compromising their ability to choose among different products or approaches. For example, one strategy has been to use what are referred to as commitment mechanisms, such as having individuals commit well in advance to allocating a portion of their future salary increases toward a savings plan. Another strategy for encouraging consumers to increase their savings has been to use incentives with tangible benefits, such as matching funds. In addition, changing the default option for enrollment in retirement plans—that is, automatically enrolling new employees while giving them the opportunity to opt out—has led to significant increases in plan participation rates among some organizations. The most effective approach to improving consumers’ financial decision making and behavior may be to use a variety of these types of strategies in conjunction with financial education. As I noted during my confirmation hearing, financial literacy is an area of priority for me as Comptroller General, and during my tenure, I hope to draw additional attention to this important issue. Improving financial literacy involves many stakeholders and must be a partnership between the federal government, state and local governments, the private and nonprofit sectors, and academia. My hope is that GAO can play a role in facilitating knowledge transfer among these different entities, as well as working with other organizations in the accountability community, such as the American Institute of Certified Public Accountants. Almost 7 years ago we hosted a forum on the role of the federal government in improving financial literacy. At that forum, public and private sector experts highlighted, among other things, the need for the federal government to serve as a leader in this area, but they also stressed the importance of public-private partnerships. We will host another forum on financial literacy later this year to bring together experts in financial literacy and education from federal and state agencies, nonprofit organizations representing consumers, educational and academic institutions, and private sector employers. This forum will address the gaps that exist in financial literacy efforts, challenges that federal agencies may face in addressing these gaps, and opportunities for improving the federal government’s approach to financial literacy. In addition, as part of our audit and oversight function, we will continue to conduct evaluations of the efficiency and effectiveness of federal financial literacy efforts. Financial literacy plays a role in a wide variety of areas that GAO regularly reviews—including student loans, retirement savings, banking and investment products, and homebuyer assistance programs, to name a few. For example, in work we have done on retirement savings, we have made recommendations intended to facilitate consumers’ understanding of retirement plans, disclosures, and any associated fees. Additionally, our reviews of financial products will continue to focus on consumer understanding of these products, as well as strategies for encouraging consumers to make sound decisions about them. Moreover, we will continue our body of work evaluating various consumer protections, which in conjunction with financial education are a key component in helping consumers avoid abusive or misleading financial products, services, or practices. Financial education has its limitations, of course, but it does represent an important tool that can benefit both individuals and our economy as a whole. On an individual level, better money management and financial decisions can play an important role in improving families’ standard of living and helping them achieve long-term financial goals. While personal financial decisions are made by individuals and their families, the federal government can play a role in helping ensure that its citizens have easy access to financial information and the tools they need to make sound decisions. Moreover, improving consumer financial literacy can be beneficial to our national economy as a whole. Financial markets function best when consumers understand how financial service providers and products work and know how to choose among them. Our income tax system requires citizens to have an adequate understanding of both the tax system itself and financial matters in general. Educated citizens are also important to well-functioning retirement systems—for example, workers should understand the benefit of saving for their retirement to supplement any benefits received from Social Security. Finally, our nation faces a challenging long-term fiscal outlook, and it is important that our citizens understand and are attentive to the fact that the federal government faces hard choices that will affect their own, and our nation’s, economic future. Chairman Akaka, Ranking Member Johnson, this completes my prepared statement. I would be happy to respond to any questions you or other Members of the Subcommittee may have at this time. For further information about this testimony, please contact Alicia Puente Cackley at (202) 512-8678 or at cackleya@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Individuals making key contributions to this testimony include Alicia Puente Cackley (Director), Jason Bromberg (Assistant Director), Tania Calhoun, Beth Ann Faraguna, Jennifer Schwartz, and Andrew Stavisky. Opportunities to Reduce Potential Duplication in Government Programs, Save Tax Dollars, and Enhance Revenue. GAO-11-318SP. Washington, D.C.: March 1, 2011. Consumer Finance: Factors Affecting the Financial Literacy of Individuals with Limited English Proficiency. GAO-10-518. Washington, D.C.: May 21, 2010. Financial Literacy and Education Commission: Progress Made in Fostering Partnerships, but National Strategy Remains Largely Descriptive Rather Than Strategic. GAO-09-638T. Washington, D.C.: April 29, 2009. Financial Literacy and Education Commission: Further Progress Needed to Ensure an Effective National Strategy. GAO-07-100. Washington, D.C.: December 4, 2006. Highlights of a GAO Forum: The Federal Government’s Role in Improving Financial Literacy. GAO-05-93SP. Washington, D.C.: November 15, 2004. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Financial literacy plays an important role in helping ensure the financial health and stability of individuals, families, and our broader national economy. Economic changes in recent years have highlighted the need to empower Americans to make informed financial decisions, yet evidence indicates that many U.S. consumers could benefit from a better understanding of financial matters. For example, recent surveys indicate that many consumers have difficulty with basic financial concepts and do not budget. This testimony discusses (1) the state of the federal government's approach to financial literacy, (2) observations on overall strategies for addressing financial literacy, and (3) the role GAO can play in addressing and raising awareness on this issue. This testimony is based largely on prior and ongoing work, for which GAO conducted a literature review; interviewed representatives of organizations that address financial literacy within the federal, state, private, nonprofit, and academic sectors; and reviewed materials of the Financial Literacy and Education Commission. While this statement includes no new recommendations, in the past GAO has made a number of recommendations aimed at improving financial literacy efforts.. Federal financial literacy efforts are spread among more than 20 different agencies and more than 50 different programs and initiatives, raising concerns about fragmentation and potential duplication of effort. The multiagency Financial Literacy and Education Commission, which coordinates federal efforts, has acted on recommendations GAO made in 2006 related to public-private partnerships, studies of duplication and effectiveness, and the Commission's MyMoney.gov Web site. While GAO's 2006 review of the Commission's initial national strategy for financial literacy found that it was a useful first step in focusing attention on financial literacy, it was largely descriptive rather than strategic. The Commission recently released a new strategy for 2011, which laid out clear goals and objectives, but it still needs to incorporate specific provisions for performance measures, resource needs, and roles and responsibilities, all of which GAO believes to be essential for an effective strategy. However, the Commission will be issuing an implementation plan to accompany the strategy later this year and the strategy will benefit if the plan incorporates these elements. The new Bureau of Consumer Financial Protection will also have a role in financial literacy, further underscoring the need for coordination among federal entities. Coordination and partnership among federal, state, nonprofit, and private sectors is also essential in addressing financial literacy, and there have been some positive developments in fostering such partnerships in recent years. There is little definitive evidence available on what specific programs and approaches are most effective in improving financial literacy, and relatively few rigorous studies have measured the impact of specific financial literacy programs on consumer behavior. Given that federal agencies have limited resources for financial literacy, it is important that these resources be focused on initiatives that are effective. To this end, the Commission's new national strategy on financial education sets as one of its four goals identifying, enhancing, and sharing effective practices. However, financial education is not the only approach for improving consumers' financial behavior. Several other mechanisms and strategies have also been shown to be effective, including financial incentives or changes in the default option, such as automatic enrollment in employer retirement plans. The most effective approach may involve a mix of financial education and these other strategies. GAO will continue to play a role in supporting and facilitating knowledge transfer on financial literacy. GAO will host a forum on financial literacy later this year to bring together experts from federal and state agencies and nonprofit, educational, and private sector organizations. The forum will address gaps, challenges, and opportunities related to federal financial literacy efforts. In addition, as part of GAO's audit and oversight function, GAO will continue to evaluate the effectiveness of federal financial literacy programs, as well as identify opportunities to improve the efficient and cost-effective use of these resources. |
The enterprises constitute one component of a range of federal initiatives that, since the 1930s, have facilitated the availability of mortgage credit and housing opportunities (see table 1). While these initiatives may involve differing missions, structures, and activities, they generally rely on federal support and subsidies to achieve their objectives. In some cases, these initiatives—such as the Federal Home Loan Bank System (FHLBank System), the enterprises’ general mortgage support activities, federal tax deductions for mortgage interest, and exemptions for capital gains—apply broadly and are designed generally to facilitate mortgage lending and homeownership. In other cases, the initiatives have been designed to facilitate home ownership and housing opportunities for targeted populations and groups. For example, programs administered by the Federal Housing Administration (FHA), Department of Veterans Affairs (VA), Department of Agriculture’s Rural Development Housing and Community Facilities Programs (USDA/RD), and HUD’s Office of Public and Indian Housing (PIH) are designed to facilitate homeownership and housing opportunities for moderate- and low-income persons, as well as first-time buyers, veterans, residents of rural areas, and Native Americans, respectively. In some cases, these federal housing initiatives also target similar populations and borrowers. For example, through their general business activities and affordable housing goal requirements, the enterprises, like FHA, provide mortgage credit to low-income borrowers and other targeted groups. During 2007 and the first half of 2008, Fannie Mae’s and Freddie Mac’s financial conditions deteriorated significantly, which FHFA officials said prompted the agency to establish the conservatorships. As later described in this report, the enterprises incurred substantial credit losses on their retained portfolios and their guarantees on MBS. These credit losses resulted from pervasive declines in housing prices, as well as specific enterprise actions such as their guarantees on MBS collateralized by questionable mortgages (mortgages with limited or no documentation of borrowers’ incomes), and investments in private-label MBS collateralized by subprime mortgages. In July 2008, Fannie Mae’s and Freddie Mac’s financial condition deteriorated, which prompted congressional and Executive Branch efforts to stabilize the enterprises and minimize associated risks to the financial system. In particular, Congress passed and the President signed the Housing and Economic Recovery Act of 2008 (HERA) which, among other things, established FHFA. HERA sets forth FHFA’s regulatory responsibilities and supervisory powers, which include expanded authority to place the enterprises in conservatorship or receivership, and provides Treasury with certain authorities to provide financial support to the enterprises, which are discussed below. While Treasury and other federal regulatory officials stated in July 2008 that the conservatorship or other major measures likely would not be necessary, the enterprises’ financial conditions continued to deteriorate. According to FHFA and Treasury officials, their ongoing financial analysis of Fannie Mae and Freddie Mac in August and early September 2008, as well as continued investor concerns about the financial condition of each enterprise, resulted in FHFA’s imposition of the conservatorships on September 6, 2008, to help ensure the enterprises’ viability, fulfill their housing missions, and stabilize financial markets. As conservator of the enterprises, FHFA has replaced their Chief Executive Officers, appointed new members of the boards of directors, assumed responsibility for overseeing key business decisions, and ceased the enterprises’ lobbying activities. While FHFA oversees key enterprise business decisions, agency officials said that they expect enterprise managers to continue to run day-to-day business activities. FHFA officials also said that the agency’s staff continues to oversee the enterprises’ safety and soundness and housing mission achievement. For example, FHFA officials said that agency examiners are located on-site at each enterprise to assess their ongoing financial performance and risk management. Since FHFA became conservator, the enterprises have been tasked by the federal government to help respond to the current housing and financial crisis. For example, in November 2008, the enterprises suspended the initiation of foreclosure proceedings on mortgages that they held in their portfolios or on which they had guaranteed principal and interest payments for MBS investors, and this initiative subsequently was extended through March 31, 2009. Furthermore, under the administration’s Homeowner Affordability and Stability Plan, which was announced on February 18, 2009, the enterprises are tasked to (1) provide access to low- cost refinancing for loans they own or guarantee to help homeowners avoid foreclosures and reduce monthly payments and (2) initiate a loan modification plan for at-risk homeowners that will lower their housing costs through a combination of interest rate reductions, maturity extensions, and principal forbearance or forgiveness. As authorized by HERA, the Secretary of the Treasury entered into agreements with Fannie Mae and Freddie Mac on September 7, 2008, to provide substantial financial support to the enterprises and thereby minimize potential systemic financial risks associated with their deteriorating financial condition. Specifically, Treasury has entered into agreements and announced the following initiatives: Enhance the enterprises’ financial solvency by purchasing their senior preferred stock and making funding available on a quarterly basis, to be recovered by redemption of the stock or by other means. While the initial funding commitment for each enterprise was capped at $100 billion, Treasury increased the cap to $200 billion per enterprise in February 2009 to maintain confidence in the enterprises. As of June 30, 2009, Treasury had purchased approximately $50.7 billion in Freddie Mac preferred stock and $34.2 billion in Fannie Mae preferred stock under the agreements. As part of the preferred stock purchase agreement, Treasury has received warrants to buy up to 79.9 percent of each enterprise’s common stock for $0.00001 per share. The warrants are exercisable at any time and should the enterprises’ financial conditions improve sufficiently, the warrants would help the government recover some of its investments in the enterprises. However, according to CBO, it is unlikely that the federal government will recover much of its massive financial investments in the enterprises. Treasury also is to receive dividends on the enterprises’ senior preferred stock at 10 percent per year and, beginning March 31, 2010, quarterly commitment fees from the enterprises. Purchase MBS until December 31, 2009, when the purchase authority expires. From September 2008 through July 2009, Treasury purchased $171.8 billion in the enterprises’ MBS. While Treasury’s authority under HERA to make such MBS purchases expires at the end of 2009, it may continue to hold previously purchased MBS in its portfolio beyond that date. Establish a temporary secured credit lending facility that allows the enterprises, as well as the FHLBank System, to borrow funds in the event they face difficulties issuing debt in financial markets. Under this Treasury program, the enterprises are to collateralize any borrowings with their MBS and the FHLBanks are to collateralize such borrowings with mortgage assets. To date, neither the enterprises nor any FHLBanks have used this borrowing authority, and Treasury’s authority for this program expires at the end of 2009. The Federal Reserve also has agreed to acquire substantial amounts of debt and MBS of the enterprises and other entities in order to reduce the cost and increase the availability of credit for the purchase of homes, and to foster improved conditions in financial markets. In November 2008, the Federal Reserve announced it would purchase up to $100 billion of debt issued by Fannie Mae, Freddie Mac, and the FHLBank System, and up to $500 billion in MBS guaranteed by Fannie Mae, Freddie Mac, and Ginnie Mae. On March 18, 2009, the Federal Reserve announced that during the current year it would purchase an additional $100 billion of the enterprises’ debt up to a total of $200 billion and an additional $750 billion of enterprise MBS up to a total of $1.25 trillion. As of August 19, 2009, the Federal Reserve had purchased $111.8 billion in federal agency housing debt securities and $609.5 billion in guaranteed MBS. To help inform the forthcoming congressional consideration of the enterprises’ future purposes and structures, this section discusses key aspects of their histories and performance in achieving key housing mission and safety and soundness objectives. Specifically, in this section, we discuss (1) the enterprises’ changing roles, structures, and activities over the years; (2) their performance in supporting mortgage finance consistent with charter obligations; (3) the extent to which the numeric housing goals may have materially benefited homeownership opportunities for targeted groups; and (4) the effect of the enterprises’ risk-management practices on their safety and soundness. As discussed below, the enterprises underwent important structural changes over the decades and accrued diverse missions and activities relating to the public and for-profit aspects of their structures and functions. Table 2 provides a time line that summarizes the key events in the federal housing finance system related to the enterprises over the past 77 years. Prior to the 1930s, the federal government did not play a direct role in supporting housing finance. Typically lenders—mainly savings and loans (thrifts), but also banks—originated short-term mortgages (from 3 to 10 years). Since thrifts and banks primarily served local markets, regional differences in the demand for and supply of mortgage credit resulted in regional disparities in mortgage interest rates and credit availability. During the Great Depression, thousands of thrifts and banks failed due to their credit losses, and housing finance generally became unavailable. In response, the federal government established institutions and initiatives to revive the housing finance market. In 1932, Congress established the FHLBank System—the first housing GSE—to provide short-term loans (called advances) to member savings and loans institutions that would use them to fund home mortgages. Additionally, Congress established FHA in 1934 in part to promote and insure long-term housing mortgages (up to 20 years) that called for borrowers to pay off the principal and interest of loans over a specified number of years. Fannie Mae was established by FHA under authority provided in 1938 as a government-held association to buy and hold mortgages insured by FHA, thereby providing additional liquidity to the mortgage market. During the 1940s, Congress authorized Fannie Mae to purchase VA-guaranteed mortgages to facilitate the efforts of veterans to purchase homes. The Housing Act of 1954 instituted Fannie Mae as a mixed-ownership corporation and specified in its federal charter the entity’s role and requirements that subsequently served as some of the enterprises’ key housing mission objectives. Among its provisions, the act required Fannie Mae to (1) provide liquidity for mortgage investments to improve the availability of capital for home mortgage financing and (2) support t mortgage market when there was a threat to the stability of the economy. The 1954 act also provided Fannie Mae with certain financial benefits thought necessary to carry out its objectives, such as exemptions from all local taxes except property taxes. Lenders that sold mortgages to Fannie Mae were required to purchase stock in it, but the federal government remained the enterprise’s majority owner. Throughout the late 1950s and 1960s, Fannie Mae’s purchases of FHA- insured and VA-guaranteed mortgages increased substantially. During this period, limits on interest rates that banks and thrifts could offer for deposits and restrictions on their ability to branch across state lines contributed to liquidity constraints and continuing regional disparities in mortgage interest rates. By operating across the nation, Fannie Mae could help alleviate such scarcities and disparities. In 1968, the Housing and Urban Development Act (the 1968 act) reorganized Fannie Mae as a for-profit, shareholder-owned company with government sponsorship and established Ginnie Mae as an independent government corporation in HUD. Ginnie Mae’s primary function was to guarantee the timely payment of principal and interest from pools of FHA-, USDA/RD-, and PIH-insured and VA-guaranteed mortgages. Ginnie Mae has the full faith and credit backing of the federal government. Although now a for-profit, shareholder-owned company, Fannie Mae continued its activities, which were mainly purchasing FHA and VA mortgages. According to some financial analysts, Congress largely reorganized Fannie Mae as a private company for budgetary purposes (that is, to remove its financial obligations from the federal budget). The 1968 act also gave the HUD Secretary general regulatory authority over Fannie Mae, as well as authority to require that a reasonable portion of its mortgage purchases serve low- and moderate-income families. The Secretary subsequently established numeric housing goals for Fannie Mae that essentially required that at least 30 percent of its purchases serve low- and moderate-income families, and at least 30 percent serve families living in central cities. However, HUD was not given authority to collect data that would be necessary to determine compliance with the goals. In the Emergency Home Finance Act of 1970, Congress chartered Freddie Mac as a housing GSE to help mitigate business challenges facing the thrift industry. Increasing interest rates had undermined thrifts’ capacity to finance long-term mortgages held in their portfolios. Freddie Mac was to purchase long-term mortgages from thrifts and thereby help stabilize the industry and enhance its capacity to fund additional mortgages. As a result, Freddie Mac was the first enterprise to develop products to facilitate securitization of mortgage loans. Freddie Mac was first owned by the Federal Home Loan Bank Board, which regulated the thrift industry. Freddie Mac did not become a shareholder-owned company like Fannie Mae until it was reorganized in 1989. While subject to HUD’s general regulatory oversight under the 1989 legislation, Freddie Mac initially was not subject to the same mortgage purchase goals as Fannie Mae. Although both Freddie Mac and Fannie Mae were to provide a secondary market for conventional mortgages, they pursued markedly different business strategies in the 1970s and 1980s. Freddie Mac focused its business activities on purchasing conforming, conventional mortgages from thrifts and issuing MBS rather than holding mortgages in its portfolio. According to a Freddie Mac official, this business strategy was intended to help the thrift industry manage interest-rate risk by passing such risk to the MBS investors. In contrast, Fannie Mae followed its traditional business strategy by purchasing mortgages and holding them in its portfolio. During the early 1980s, Fannie Mae experienced substantial losses, as did the thrift industry, due to sharply rising interest rates while Freddie Mac’s financial performance generally was unaffected. During this period, the federal government provided certain financial benefits to Fannie Mae, such as regulatory forbearance and tax benefits, to help it recover. After Freddie Mac was turned into a for-profit, shareholder- owned corporation in 1989, it began to hold more mortgages in its retained portfolio, similar to Fannie Mae. By 1992, Congress concluded that the enterprises posed potential safety and soundness risks, and regulations that had been in place since 1968 were inadequate to manage such risks. Over the years, the enterprises had become large and complex organizations, and Fannie Mae’s financial difficulties in the early 1980s indicated that they posed risks to taxpayers and financial stability. Furthermore, HUD had not fulfilled its statutory responsibility to monitor the enterprises’ financial operations and risks. For example, HUD did not routinely examine the enterprises’ financial activities or promulgate regulations necessary to help ensure their safe and sound operations. There was also a concern that the enterprises were not adequately serving the mortgage credit needs of low- and moderate- income borrowers and other targeted groups due to their potentially higher default risks. In a 1996 report, we noted that, in 1992, there was a perception that the enterprises’ distribution of conventional, conforming loan funding to low- and moderate-income borrowers was lagging behind the primary mortgage market, and a Federal Reserve study was consistent with this perception. Moreover, HUD did not enforce the housing goals— which at that time applied only to Fannie Mae—or collect the data necessary to do so. In enacting the Federal Housing Enterprises Safety and Soundness Act of 1992 (the 1992 Act), Congress fundamentally revised regulation of the enterprises and took steps to clarify Fannie Mae’s and Freddie Mac’s roles within the housing finance system and better define their public housing mission responsibilities. For example, the 1992 Act reiterated the enterprises’ long-standing obligations to support mortgage finance through secondary market activities, including during stressful economic periods, and clarified and expanded the enterprises’ charter obligations to facilitate the flow of mortgage credit serving targeted groups. Moreover, the 1992 Act set forth oversight authority and mechanisms to better manage potential conflicts between the enterprises’ profit motivations and housing missions. First, it established OFHEO as an independent agency in HUD responsible for the enterprises’ safety and soundness. Among other things, OFHEO was given supervisory authority to establish and monitor compliance with minimum and risk-based capital standards and conduct routine safety and soundness examinations. In so doing, Congress established a safety and soundness regulatory framework that resembled the supervisory framework for insured depository institutions such as banks and thrifts, although OFHEO’s authority was less extensive. Second, the 1992 Act expanded the enterprises’ previous housing mission responsibilities by requiring them to meet specific annual goals for the purchase of mortgages serving targeted groups. Specifically, it directed the HUD Secretary to promulgate regulations setting annual housing goals for both Fannie Mae and Freddie Mac for the purchase of mortgages serving low- and moderate-income families; special affordable housing for families (i.e., low-income families in low-income areas, and very low-income families); and housing located in central city, rural, and other underserved areas. The 1992 Act also provided HUD with the authority to collect data necessary to monitor the enterprises’ compliance with the goals and to enforce such compliance. It should be noted that the enterprises’ affordable housing goals required them to compete with other federal initiatives to support housing, particularly FHA’s mortgage insurance programs that also primarily serve low- and moderate-income borrowers and first-time homeowners. This issue is discussed in more detail later in this report. Third, the 1992 Act set forth HUD’s regulatory authority over the enterprises and specified procedures that HUD must follow when reviewing and approving new mortgage program proposals by the enterprises. That is, it directed the HUD Secretary to approve any new program that an enterprise proposed, unless the Secretary determined that the program violated the enterprise’s charter or would not be in the public interest. Additionally, the 1992 Act required the HUD Secretary, for a specified transition period, to reject a new program proposal if the Director of OFHEO determined that the proposal would risk a significant financial deterioration of the enterprise. While the 1992 Act enhanced the enterprises’ regulatory structure in several important respects, it still had important limitations in its capacity to ensure the enterprises’ safety and soundness and housing mission compliance. First, federal oversight of the enterprises and the FHLBank System was divided among OFHEO, HUD, and FHFB—which was the safety and soundness and housing mission regulator for the FHLBank System. However, OFHEO and FHFB were small agencies that lacked the resources necessary to monitor large and complex financial organizations from the standpoint of safety and soundness, as well as mission goals. Furthermore, as compared with federal bank regulators, both OFHEO and FHFB lacked key authorities—such as authority to take enforcement actions based on declining capital levels and unsound financial practices— that were available to federal bank regulators. Enterprise regulation also had limited capacity to address potential conflicts between the enterprises’ profit motivations and their federally mandated housing missions. In particular, we noted that, due to the financial benefits derived from their federal charters and their dominant position within the mortgage finance system, the enterprises had financial incentives to engage in potentially profitable activities that were not fully consistent with their charter obligations and restrictions. For example, Freddie Mac, during the mid-1990s, had invested in nonmortgage assets, such as long-term corporate bonds, that potentially allowed the enterprise to earn higher returns based on the enterprises’ funding advantage. Freddie Mac argued that its investments in nonmortgage assets were permissible and necessary to help manage the liquidity of its investment portfolio. Although HUD had general regulatory and new mortgage program authorities, it was not clear if HUD was well-positioned to assess such arguments or the extent to which the enterprises may have been straying from their charter obligations and restrictions. At that time, HUD officials said that they lacked staff with the expertise necessary to oversee large and complex financial institutions and determine if the enterprises’ activities were consistent with their charters and housing finance missions. By retaining the enterprises’ off-budget status as GSEs, the 1992 Act permitted a continuation of the lack of transparency about the enterprises’ risks and potential costs to taxpayers. Under the Federal Credit Reform Act of 1990, the potential costs associated with many direct federal loan and loan guarantee programs have to be disclosed in the federal budget. Congress and the Executive Branch can use such disclosures to assess the potential costs and future risks of such programs and take steps on a timely basis to potentially mitigate such costs and risks (for example, tightening eligibility criteria). Despite the implied federal guarantee of their obligations, the government’s exposure in connection with the enterprises is not disclosed in the federal budget because GSE activities were excluded from the federal budget totals. The 1992 Act did not change the status of the enterprises as off-budget entities. However, it should be noted that such financial disclosures could have involved an offsetting risk. Such treatment might have increased the perception that, despite the enterprises’ statements to the contrary, the federal government would provide financial support to them in an emergency, which may have further reduced market discipline and enterprise actions to mitigate risks. Congress substantially revised the enterprises’ regulatory structure with the passage of HERA in 2008. In HERA, Congress abolished OFHEO and FHFB and established FHFA as the regulator of the enterprises and the FHLBank System. HERA charges FHFA with responsibility for housing GSE safety and soundness. In this regard, HERA augments the safety and soundness responsibilities and authorities administered by the predecessor agencies. Additionally, HERA transferred responsibility for the enterprises’ mission oversight, including their satisfaction of numeric goals for purchases of mortgages to low- and moderate-income borrowers and the review and approval of enterprise new mortgage programs, from HUD to FHFA. FHFA’s supervisory authority over safety and soundness matters includes specific authority to place the housing GSEs into conservatorship or receivership based on grounds set forth in HERA. Since placing Fannie Mae and Freddie Mac into conservatorship in September 2008, FHFA has appointed new Chief Executive Officers and boards of directors at each enterprise and stands in lieu of shareholders in matters of corporate governance. In contrast, FHFA’s role with respect to the FHLBank System has remained solely that of an independent regulator. It is generally accepted that the enterprises have been successful in enhancing liquidity in the mortgage finance system as directed in their charters. We have reported that the enterprises established a viable secondary mortgage market for conventional loans that enabled capital to flow to areas with the greatest demand for mortgage credit. This free flow of capital tended to equalize interest rates across regions for mortgages with similar risk characteristics. However, the removal of restrictions on the ability of banks and thrifts to pay market rates for deposits and to operate across state lines also have contributed to mortgage liquidity and the establishment of an integrated national mortgage finance system. The enterprises’ activities also have been credited with achieving other benefits consistent with their charter obligations to support mortgage finance, which include the following: Lowering mortgage interest rates on qualifying mortgages below what they otherwise would be. GAO and others have stated that the advantageous borrowing rates that the enterprises derived from the implied federal guarantee on their financial obligations were passed on to borrowers to some degree, although estimates vary. However, we also have noted that these benefits were not entirely passed along to homebuyers. Rather, the enterprises’ shareholders and senior management also benefited for many years from the relatively higher profits that the companies achieved due to cost savings associated with the implied guarantee. Establishing standard underwriting practices and forms for conventional mortgages. Due to the enterprises’ large purchases of conventional mortgages each year, their underwriting guidelines and forms became the industry standard. GAO and others have found that standardization facilitated the efficiency of the mortgage underwriting process and resulted in cost savings for lenders and borrowers. The enterprises’ efforts to standardize mortgage underwriting likely also helped develop the MBS market, as consistent standards are viewed as critical for helping investors evaluate risks. However, the extent to which the enterprises have been able to support a stable and liquid secondary mortgage market during periods of economic stress, which are key charter and statutory obligations, is not clear. In 1996, we attempted to determine the extent to which the enterprises’ activities would support mortgage finance during stressful economic periods by analyzing Fannie Mae’s mortgage activities in some states, including oil producing states such as Texas and Louisiana, beginning in the 1980s. Specifically, we analyzed state-level data on Fannie Mae’s market shares and housing price indexes for the years 1980–1994. We did not find sufficient evidence that Fannie Mae provided an economic cushion to mortgage markets in those states during the period analyzed. During the current financial crisis, the enterprises have provided critical support to mortgage finance, but only with the benefit of substantial financial assistance provided by Treasury and the Federal Reserve during the conservatorships. As shown in figure 1, the enterprises and Ginnie Mae accounted for nearly 60 percent of MBS issuances in 2006, while private- label issuances, such as MBS collateralized by pools of subprime and jumbo mortgages, accounted for nearly 40 percent. By the end of 2008, the enterprises and Ginnie Mae accounted for about 97 percent of MBS issuances, while private-label issuances stood at about 3 percent due to the collapse of many subprime lenders and the associated reduction in nonconforming mortgage origination and precipitous downturn in securitization markets. According to FHFA’s former Director, one of the reasons that the agency established the conservatorships in September 2008 is that the financial challenges the enterprises were facing as independent entities compromised their capacity to support mortgage finance. For example, the enterprises’ mortgage purchases slowed in 2008, and they planned to raise certain fees to help offset their losses. While the enterprises are now a critical component of the federal government’s response to the housing crisis, such support would not be possible without Treasury’s financial support and the Federal Reserve’s plans to purchase almost $1.45 trillion of their MBS and debt obligations as well as those of other entities. While the enterprises’ numeric housing goal mortgage purchase program has been in place for more than 10 years, its effectiveness in supporting homeownership opportunities for targeted groups and areas is not clear. Pursuant to the 1992 Act, HUD established interim goals for 1993 and 1994, which were extended through 1995, and final goals for the period from 1996 through 1999. In 1998, we found these were conservative goals, which placed a high priority on maintaining the enterprises’ financial soundness. For example, according to research conducted by HUD and OFHEO, the additional mortgage purchases required under the goals were modest and would not materially affect the enterprises’ financial condition. HUD also established housing goals in 2000 (covering 2001– 2004) and in 2004 (covering 2005–2008). According to a speech by the FHA Commissioner in 2005, the 2004 goals established significantly higher requirements than the 2000 goals. According to HUD data, the enterprises generally have met the numeric housing goals since the beginning of the program. For example, table 3 shows that Fannie Mae and Freddie Mac met the low- and moderate- income housing goals in place from 2002 through 2007. However, the enterprises failed to meet this goal in 2008, and, according to HUD, did not meet certain subgoals in 2007. Although the enterprises generally satisfied the numeric purchase goals through 2007, HUD and independent researchers have had difficulty identifying tangible benefits for targeted groups associated with the enterprises’ purchase program. In setting higher housing goals beginning in 2005, HUD stated that the intent was to encourage the enterprises to facilitate greater financing and homeownership opportunities for the groups targeted by the goals. HUD concluded that, although the enterprises had complied with previous goals, they continued to serve less of the affordable housing market than was served by conventional conforming primary market lenders during those years. Furthermore, recent research indicates that, although the enterprises have enhanced their product offerings to meet the housing goals, the effects of the housing goals on affordability and opportunities for target groups have been limited. For example, a 2003 study that modeled the impacts of the housing goals found that the enterprises likely increased credit in specified areas in only 1 of the 5 years included in the model. A 2006 study concluded that the enterprises’ purchases of mortgages in certain targeted low- and moderate-income areas (census tracts in California during the 1990s with depressed housing markets) generally did not increase homeownership rates as compared with other low- and moderate- income areas that were not specifically targeted by the numeric housing goals. The research found only one low-income target area (in San Francisco) that showed improvements in homeownership rates as a result of the enterprises’ activities. Another study suggested that enterprise-FHA interactions in the same areas may help explain why the program’s benefits were limited. While the enterprises’ numeric mortgage purchase program and FHA’s mortgage insurance were intended to benefit similar targeted groups, such as low- income and minority borrowers, the study suggested that the programs may have offset each other. That is, as the enterprises increased their mortgage purchases in areas with concentrations of targeted groups, FHA activity declined in those areas. According to the study, while the relatively lower costs of conventional loans compared with FHA-insured loans provided benefits for those households able to switch to a conventional loan, this cost differential also permitted the enterprises to attract an increasing share of the most creditworthy targeted borrowers in these areas, which FHA had served before. In response to losing its more creditworthy borrowers, FHA could have retained market share by reaching for borrowers that represented greater credit risks and either (1) accepted the riskier portfolio wholesale or (2) increased premiums to insure itself against expected losses. However, the study concluded that FHA applied stricter underwriting standards and reduced its loan volume. Therefore, the overall impact of the two programs on promoting homeownership opportunities in these areas was limited. After 2002, both the enterprises’ and FHA’s market share declined in areas with concentrations of low-income and minority groups as subprime lending grew in size, which may have limited the impact of both the enterprises’ housing goal program and FHA’s mortgage insurance activities. Earlier research sponsored by HUD in 2001 largely discounted the alleged benefits for affordable multifamily finance resulting from the enterprises’ numeric mortgage goals. According to the research, the enterprises generally did not play a leading role in affordable multifamily mortgage finance because their underwriting standards were considered conservative and fairly inflexible, compared with other multifamily mortgage providers. In contrast, representatives from mortgage finance, housing construction, and consumer groups we contacted said that the benefits from the enterprises’ purchases of affordable multifamily mortgages pursuant to their goals have been significant. The representatives said that the enterprises’ involvement and, in some cases, guarantees on the financing of affordable multifamily projects, which may be complex and involve a variety of government and private-sector entities, were crucial to their successful completion. In addition, the representatives said that the enterprises were the only source of funding for multifamily projects because many other traditional providers, such as banks and insurance companies, largely have withdrawn from the market during the current financial crisis. While housing finance may have derived some benefits from the enterprises’ activities over the years, GAO, federal regulators, researchers, and others long have argued that the enterprises had financial incentives to engage in risky business practices to strengthen their profitability partly because of the financial benefits derived from the implied federal guarantee on their financial obligations. For example, during the late 1990s and early 2000s, we raised concerns about the rapid growth of the enterprises’ retained mortgage portfolios, which reached about $1.6 trillion by 2005 (see fig. 2). Although increasing the size of their mortgage portfolios may have been more profitable than issuing MBS, it also exposed the enterprises to significant interest-rate risk. We reported that the rapid increase in the enterprises’ mortgage portfolios and the associated interest-rate risk did not result in a corresponding benefit to the achievement of their housing missions. For example, the rapid growth in the enterprises’ retained mortgage portfolios in the late 1990s, and in 2003 through 2004, occurred during periods of strong economic growth when mortgage markets did not necessarily require the enterprises to be robust portfolio lenders. In 2003 and 2004, OFHEO found that Freddie Mac and Fannie Mae manipulated accounting rules so that their public financial statements would show steadily increasing profits over many years and thereby increase their attractiveness to potential investors. The misapplication of accounting rules generally involved standards for reporting on derivatives, which the enterprises used to help manage the interest-rate risks associated with their large retained mortgage portfolios. According to investigative reports, the enterprises also may have manipulated their financial reports to show consistently increasing profits to help ensure senior executives would receive bonuses. OFHEO also found that the enterprises lacked key operational capacities, such as information systems and personnel, necessary to manage large mortgage portfolios and account for them correctly. The enterprises were required to restate their financial statements and adjust their earnings reports by billions of dollars. While the enterprises were subject to increased OFHEO scrutiny because of these accounting and operational deficiencies in 2004 and 2005, they still embarked on aggressive strategies to purchase mortgages and mortgage assets with questionable underwriting standards. For example, they purchased a large volume of what are known as Alt-A mortgages, which typically did not have documentation of borrowers’ incomes and had higher loan-to-value ratio or debt-to-income ratios. Furthermore, as shown in figure 3, enterprise purchases of private-label MBS increased rapidly as a percentage of retained mortgage portfolios from 2003 through 2006. By the end of 2007, the enterprises collectively held more than $313 billion in private-label MBS, of which $94.8 billion was held by Fannie Mae and $218.9 billion held by Freddie Mac. According to some commenters, the 2004 increase in housing goals provided the enterprises with incentives to purchase mortgage assets, such as Alt-A mortgages and private-label MBS collateralized by subprime and Alt-A mortgages,that in large degree served targeted groups. However, former FHFA Director Lockhart stated that the enterprises’ primary motivation in purchasing such assets was to restore their share of the mortgage market, which declined substantially from 2004 through 2007 as the “nontraditional” (for example, subprime) mortgage market rapidly increased in size. FHFA further stated that the enterprises viewed such mortgage assets as offering attractive risk- adjusted returns. According to FHFA, while these questionable mortgage assets accounted for less than 20 percent of the enterprises’ total assets, they represented a disproportionate share of credit-related losses in 2007 and 2008. For example, by the end of 2008, Fannie Mae held approximately $295 billion in Alt-A loans, which accounted for about 10 percent of the total single- family mortgage book of business (mortgage assets held in portfolio and mortgages that served as collateral for MBS held by investors). Similarly, Alt-A mortgages accounted for nearly half of Fannie Mae’s $27.1 billion in credit losses of its single-family guarantee book of business in 2008. At a June 2009 congressional hearing, Lockhart said that 60 percent of the AAA-rated, private-label MBS purchased by the enterprises have since been downgraded to below investment grade. He also stated that investor concerns about the extent of the enterprises’ holdings of such assets and the potential associated losses compromised their capacity to raise needed capital and issue debt at acceptable rates. The enterprises’ mixed records in achieving their housing mission objectives and the losses and weaknesses that resulted in the conservatorships reinforce the need for Congress and the Executive Branch to fundamentally reevaluate the enterprises’ roles, structures, and business activities in mortgage finance. Researchers and others believe that there is a range of options available to better achieve housing mission objectives (in some cases through other federal entities such as FHA), help ensure safe and sound operations, and minimize risks to financial stability. These options generally fall along a continuum with some overlap among key features and advocate (1) establishing a government corporation or agency, (2) reconstituting the enterprises as for-profit GSEs in some form, or (3) privatizing or terminating them (see table 4). This section discusses some of the key principles associated with each option and provides details on how each could be designed to support housing objectives. Some proposals advocate that, after the FHFA conservatorships are terminated, consideration should be given to establishing a government corporation or agency to assume responsibility for key enterprise business activities. Supporters of these proposals maintain that the combination of the implied federal guarantee on the enterprises’ financial obligations, and their need to respond to shareholder demands to maximize profitability, encouraged excessive risk-taking and ultimately resulted in their failures. Accordingly, they also believe that a government corporation or agency, which would not be concerned about maximizing shareholder value, would be the best way to ensure the availability of mortgage credit for primary lenders, while minimizing the risks associated with a for-profit structure with government sponsorship. Establishing a government corporation or agency also would help ensure transparency in the federal government’s efforts through appropriate disclosures of risks and costs in the federal budget. Under one proposal, a government corporation would assume responsibility for purchasing conventional mortgages from primary lenders and issuing MBS. However, under this proposal, the enterprises’ retained mortgage portfolios would be eliminated over time because of their interest-rate risk and associated safety and soundness concerns. Taxpayer protections would come from sound underwriting standards and risk-sharing arrangements with the private sector. The government corporation also would be required to establish financial and accountability requirements for lenders and institute consumer protection standards for borrowers as appropriate. While this proposal advocates the establishment of a government corporation to replace Fannie Mae and Freddie Mac, it states that there are risks associated with doing so. For example, a government corporation might face challenges retaining capable staff or become overly bureaucratic and unreceptive to market developments. Accordingly, the proposal includes a provision that the government corporation should be carefully reevaluated to ensure that it does not “ossify” over time. The proposal also concludes that the new government corporation either “sunset” (terminate) after 5 years if the market has stabilized or be allowed to continue under a renewable charter that would require periodic reviews. Under a second proposal, a government corporation or agency also would focus on issuing MBS rather than maintaining a retained mortgage portfolio. Borrowers would be charged actuarially based premiums to help offset the risks associated with the government corporation’s or agency’s activities. For example, mortgages with a 10 percent or lower down payment would be subject to a higher premium than mortgages with a 20 percent down payment. The government corporation or agency also would focus its activities on middle-income borrowers, and the mortgage credit needs of targeted groups would be served by an expansion of FHA’s mortgage insurance programs. The proposal suggests that specific appropriations to FHA represent a more efficient means to assist low- income borrowers than seeking to assist such borrowers through the enterprises’ activities. A third proposal advocates that the government provide funding directly to targeted borrowers though down-payment assistance rather than relying on the enterprises’ mortgage purchase program. For purposes of comparison, we note that Ginnie Mae is an existing government corporation that performs important functions in the secondary markets for government guaranteed and insured mortgage loans. Specifically, Ginnie Mae guarantees the timely payment of principal and interest on MBS that are collateralized by pools of mortgages that are insured or guaranteed by FHA, VA, PIH, and USDA/RD. However, Ginnie Mae does not perform functions that are envisioned for a government corporation or agency that might replace Fannie Mae and Freddie Mac. In particular, Ginnie Mae does not issue MBS, as do Fannie Mae and Freddie Mac. Moreover, Ginnie Mae is not responsible for monitoring the underwriting or the credit risk associated with the mortgages that collateralize the MBS pools but instead relies on FHA, VA, PIH, and USDA/RD to do so. While many of the enterprises’ critics view the for-profit GSE structure as precipitating the enterprises’ financial crises that led to conservatorship, market participants and commenters, trade groups representing the banking and home construction industries, as well as community and housing advocates we contacted, believe that the for-profit GSE structure generally remains superior to the alternatives. They assert that continuing the enterprises as for-profit GSEs would help ensure that they would remain responsive to market developments, continue to produce innovations in mortgage finance, and be less bureaucratic than a government agency or corporation. But, they also generally advocate additional regulations and ownership structures to help offset the financial risks inherent in the for-profit GSE structure. While this option generally envisions that the enterprises would focus on issuing MBS, as is the case with proposals to establish government corporations or agencies, several proponents believe they should be permitted to maintain a mortgage portfolio to meet certain key responsibilities. For example, home construction, small bank, and community and housing advocates noted that the enterprises may need to maintain portfolios to support multifamily and rural housing finance. Representatives from the home building industry said that the enterprises generally have held the majority of their affordable multifamily mortgage assets in their portfolios. Fannie Mae officials also said that issuing MBS collateralized by multifamily mortgages can be difficult compared with issuing MBS collateralized by single-family properties for several reasons. A variation of this option involves breaking up the enterprises into multiple GSEs. For example, the Congressional Research Service (CRS) has stated that the enterprises could be converted into 10 or so GSEs, which could mitigate safety and soundness risks. That is, rather than having the failure of two large GSEs threaten financial stability, the failure of a smaller GSE likely would have a more limited impact on the financial system. CRS also has stated that creating multiple GSEs could enhance competition and benefit homebuyers. A potential regulatory action to limit the risks associated with reconstituting the enterprises as GSEs would be to establish executive compensation limits as deemed appropriate. As discussed previously, OFHEO investigative reports in 2003 and 2004 concluded that the enterprises manipulated their financial statements in part to help ensure that senior executives would receive bonuses. In June 2009, FHFA published proposed rules to implement sections of HERA that give FHFA authority over executive compensation at the enterprises. It has also been suggested that the enterprises be converted from publicly traded companies into cooperatives owned by lenders similar to the FHLBank structure. For example, one commenter suggested that, by having lenders assume some of the risks associated with the enterprises’ activities, mortgage underwriting standards could be enhanced. A mortgage lending group stated in a recent analysis of options to revise the secondary mortgage markets that, under a cooperative structure for enterprises, lenders would need to post as collateral a portion of their loan-sale proceeds to cover some initial level of potential losses. This collateral would be refundable to the lenders as loans age and that rights to the collateral could be sold to third parties. The trade group also noted that while the cooperatives would determine pricing, credit standards, and eligibility requirements, they still would need to be subject to safety and soundness oversight by the federal government. However, representatives from a trade group that represents smaller banks said that it might be difficult to convince such banks to participate in a cooperative. They said that many smaller banks suffered substantial losses on the preferred stock they held in Fannie Mae and Freddie Mac before their conservatorships and would be very reluctant to make such investments in the future. It also has been suggested that the reconstituted enterprises be subject to public utility-type regulation. Traditionally, such regulation has been cial used at the federal and state level to oversee and control the finan performance of monopolies or near monopolies, such as electric, telephone, and gas companies. To help prevent disadvantages to ratepayers, federal and state governments traditionally have imposed limits on such public utilities’ rate of return and required that their rate structures be fair and equitable. It has been suggested that the enterprises’ historically dominant positions in the mortgage markets, and their cost advantages associated with the implied guarantee, among other advantages, potentially make them candidates for public utility-type regulation. Former Treasury Secretary Paulson advocated keeping the enterprises as corporations, because of the private sector’s capacity to assess credit risk compared with government entities, with substantial government support but with a variety of controls on their activities. First, Paulson suggested that the corporations purchase mortgages with a credit guarantee backed by the federal government and not retain mortgage portfolios. Second, Paulson recommended that the corporations be subject to public utility- type regulation. Specifically, he recommended that a public utility-type commission be established with the authority to set appropriate targets for the enterprises’ rate of return and review and approve underwriting decisions and new mortgage products. Paulson also recommended that the enterprises pay a fee to help offset the value of their federal support and thereby also provide incentives for depository institutions to fund mortgages, either as competitors to a newly established government structure or as a substitute for government funding. Some analysts and financial commenters contend that privatizing, significantly reducing, or eliminating the enterprises’ presence in the mortgage markets represents the best public policy option. Advocates of this proposal believe that it would result in mortgage decisions more closely aligned with market factors and reduce safety and soundness risks. That is, sources of mortgage credit and risk would not be concentrated in two large and complex organizations that might take excessive risks because of the implied federal guarantee on their financial obligations. Instead, mortgage credit and risk would be diversified throughout the financial system. Federal Reserve Chairman Ben S. Bernanke has suggested that privatized entities may be more innovative and efficient than government entities, and operate with less interference from political interests. Proposals to privatize, minimize, or eliminate the enterprises’ presence in the mortgage markets may involve a transition period to mitigate any potential market disruptions and facilitate the development of a new mortgage finance system. For example, one proposal would freeze the enterprises’ mortgage purchase activities, which would permit banks and other lenders to assume a greater role in the financial system. Some researchers and financial commenters also have suggested that private- sector entities, such as consortiums or cooperatives of large banks, would have a financial incentive to assume responsibility for key enterprise activities, such as purchasing mortgages and issuing MBS. Given the substantial financial assistance that Treasury and the Federal Reserve have provided to the enterprises during their conservatorships, it may be very difficult to credibly privatize them as largely intact entities. That is, the financial markets likely would continue to perceive that the federal government would provide substantial financial support to the enterprises, if privatized as largely intact entities, in a financial emergency. Consequently, such privatized entities may continue to derive financial benefits, such as lowered borrowing costs, resulting from the markets’ perceptions. In exploring various options for restructuring the enterprises, Bernanke has noted that some privatization proposals involve breaking the enterprises into smaller units to eliminate the perception of federal guarantees. Bernanke also has questioned whether fully privatized enterprises would be able to issue MBS during highly stressful economic conditions. He pointed out that, during the current financial crisis, private-sector mortgage lending largely stopped functioning. Bernanke cited a study by Federal Reserve economists that advocated the creation of an insurer, similar to the Federal Deposit Insurance Corporation, to support mortgage finance under the privatization proposal. The new agency would offer premium-supported, government-backed insurance for any form of bond financing used to provide funding to mortgage markets. Bernanke and Paulson also have discussed using covered bonds as a potential means to enhance private-sector mortgage finance in the United States. According to Bernanke, covered bonds are debt obligations issued by financial institutions and secured by a pool of high-quality mortgages or other assets. Bernanke stated that covered bonds are the primary source of mortgage funding for European banks, with about $3 trillion outstanding. However, Bernanke concluded that there are a number of challenges to implementing a viable covered bond market in the United States. For example, as a source of financing, he said covered bonds generally are not competitive with financing provided by the FHLBanks or the enterprises, which have lower financing costs due to their association with the federal government. Each of the options to revise the enterprises’ structures involves important trade-offs in terms of their capacity to achieve key housing mission and safety and soundness objectives (see table 5). This section examines the three options in terms of their ability to (1) provide ongoing liquidity and support to mortgage markets, (2) support housing opportunities for targeted groups, and (3) ensure safe and sound operations. Furthermore, it identifies potential regulatory and oversight structures that might help ensure that the implementation of any of the options achieves their intended housing mission and safety and soundness objectives. With many of its activities funded directly through Treasury debt issuances, a government corporation or agency (a government entity) could help provide liquidity to mortgage markets during good economic times through the purchase of large volumes of mortgages that meet specified underwriting criteria and issue MBS collateralized by such mortgages. In the process, a government entity also could help ensure standardization in the mortgage underwriting process. Additionally, a government entity might have a structural advantage over private entities—such as reconstituted GSEs, banks, or other private lenders—in providing liquidity to mortgage markets during periods of economic stress. That is, a government entity may be able to continue to fund its activities through government debt issuances. In contrast, for-profit entities face potential conflicts in supporting mortgage finance during stressful economic periods because they also must be concerned about maintaining shareholder value, which may mean substantially reducing their activities or withdrawing from markets entirely as has occurred during the recent economic downturn. However, to the extent that a retained mortgage portfolio may be necessary to help respond to a financial crisis, a government entity without such a portfolio may face challenges in supporting mortgage finance, particularly if investor demand for its MBS were to become limited. Other federal entities, such as Treasury and the Federal Reserve, may have to step in to purchase and hold mortgage assets on their balance sheets (as has been the case during the current financial crisis) if such a situation existed. The absence of a retained mortgage portfolio for a government entity also could affect the traditional conventional, conforming mortgage market. Over the past 10 years, the enterprises, as discussed earlier, have maintained large mortgage portfolios. If this option is no longer available, lenders may find it more challenging to find buyers for these mortgages in the secondary market. It is not clear the extent to which a government entity could maintain the same general level of mortgage purchases as the enterprises if it were confined to assembling all such mortgages into MBS. A government entity most likely would be expected to support homeownership opportunities for targeted groups given its status as a public organization. This option also would resolve any structural conflicts that the enterprises faced over the years as for-profit, publicly-traded, shareholder-owned corporations in supporting homeownership opportunities for targeted groups. A government corporation or agency would be a public entity without the responsibility to maximize shareholder value. However, if a government entity were not permitted to have a retained mortgage portfolio, as some researchers have proposed, it likely would face challenges in implementing a numeric mortgage purchase program similar to that of the enterprises. As discussed previously, the enterprises tended to hold a significant portion of multifamily mortgages that were purchased pursuant to the numeric mortgage purchase programs in their retained portfolios. That is because it may be difficult to convert multifamily mortgage assets into MBS compared with single-family mortgages. There might be several different ways to address this challenge. For example, fees or assessments could be imposed on the activities of the government entity, and such revenues could be used to directly support the construction of affordable housing or provide down payment assistance to targeted homebuyers. Under HERA, the enterprises were to pay assessments to fund a Housing Trust Fund for the purposes of providing grants to states to increase and preserve the supply of rental housing and increase homeownership for extremely low- and very low-income families, but FHFA has suspended this program due to the enterprises’ financial difficulties. Alternatively, FHA could be expanded to assume responsibility for the enterprises’ ongoing efforts to support homeownership opportunities as one researcher has suggested. However, as is discussed later, FHA’s current operational capacity to manage a large increase in its business may be limited, which could increase taxpayer risks. In some respects, a government entity that focused its activities on purchasing mortgages and issuing MBS might pose lower safety and soundness risks than has been the case with the enterprises. For example, a government entity would not be motivated by an implied federal guarantee to engage in risky business practices to achieve profitability targets and thereby maintain shareholder value as was the case with the enterprises. Furthermore, if a government entity were not to retain a mortgage portfolio, as has been proposed, then it would be less complex and potentially less risky than the enterprises’ current structure. As already discussed, the enterprises’ large retained portfolios exposed them to significant interest-rate risk, and they misapplied accounting rules that governed the hedging techniques necessary to manage such risks. Nevertheless, successfully managing a large conventional mortgage purchase and MBS issuance business still may be a complex and challenging activity for a government entity, and the failure to adequately manage the associated risks could result in significant losses that could be the direct responsibility of taxpayers. For example, the enterprises’ substantial losses in recent years have been credit-related (due to mortgage defaults), including substantial losses in their MBS guarantee business. This risk may be heightened if a government entity were expected to continue purchasing mortgages and issuing MBS during stressful economic periods when the potential for losses may be greater than would otherwise be the case. As discussed previously, Ginnie Mae provides only a limited model for the establishment of such a government corporation or agency. Ginnie Mae guarantees the timely payment of principal and interest on MBS collateralized by mortgages, but federal agencies insure or guarantee such mortgages and lenders issue the MBS. Furthermore, Ginnie Mae is not responsible for establishing credit underwriting standards or monitoring lenders’ adherence to them; rather, these functions are carried out by FHA, VA, PIH, and USDA/RD. In contrast, a government entity that issued MBS collateralized by conforming mortgages as is the case with the enterprises would be responsible for managing credit risk, including setting appropriate guarantee fees to offset such risk. As described in our previous work on FHA, government entities may lack the financial resources necessary to attract the highly skilled employees needed to manage complex business activities or the information technology necessary to help do so. Furthermore, government entities sometimes are subject to laws and regulations that may limit their capacity to respond to market developments. In a range of recent reports, HUD’s Office of Inspector General expressed concerns about whether FHA had the staffing expertise and information technology needed to manage the rapid increase in its mortgage insurance business since 2008. To help ensure that a government entity could achieve its housing mission objectives while operating in a safe and sound manner, an appropriate oversight framework would need to be established. While such a framework would need to clearly define the roles and objectives of a government entity in the mortgage finance system, it would need to afford the entity sufficient flexibility to acquire adequate resources and manage its activities to fulfill its mission. The establishment of risk-sharing arrangements with the private sector, such as requirements that lenders that sell mortgages to the government entity retain some exposure to potential credit losses or that private mortgage insurance is obtained on such mortgages, could help mitigate the risk of potential losses. Such a government entity could be expected to reflect its risk and liabilities in the federal budget to help ensure financial transparency of its operations. Finally, robust congressional oversight of any such entity would be needed to help ensure that the entity was fulfilling its objectives. When mortgage credit markets stabilize, the enterprises as reconstituted GSEs might be expected to perform functions that they have performed for many years, such as purchasing conventional mortgages, issuing MBS, and perhaps managing a relatively small retained mortgage portfolio under some proposals. Through such activities, the enterprises might be expected to provide liquidity to mortgage markets during good economic periods, as well as provide standardization to the mortgage underwriting process and certain technical and procedural innovations. However, as for-profit corporations, significant concerns remain about how well the reconstituted enterprises would be able to support financial markets during stressful economic periods without substantial financial support from Treasury or the Federal Reserve. Moreover, the reconstituted GSEs, like government corporations or agencies, might face challenges in their ability to support mortgage finance if their mortgage portfolios were substantially downsized or eliminated as envisioned under some proposals. For example, with substantially downsized or eliminated mortgage portfolios, the reconstituted GSEs might further limit their capacity to respond to financial crisis, in which case, the likelihood that Treasury or the Federal Reserve would need to respond by buying and holding mortgage assets on their balance sheets would be increased. In addition, substantially downsizing or eliminating the reconstituted GSEs mortgage portfolios could limit lenders’ ability to sell conventional, conforming mortgages on the secondary market. Permitting the enterprises as reconstituted GSEs to maintain mortgage portfolios, albeit at lower levels than prior to their conservatorships, could help address these potential concerns. Similarly, decisions about the size of the reconstituted GSEs’ mortgage portfolios would likely affect their capacity to support mortgage purchase programs to facilitate the flow of mortgage credit to targeted groups. If the reconstituted GSEs’ mortgage portfolios were substantially decreased or eliminated, then their ability to purchase and hold multifamily mortgages that serve targeted groups might be limited. On the other hand, permitting the reconstituted GSEs to maintain a mortgage portfolio of an appropriate size could mitigate this potential concern. Another consideration associated with establishing the enterprises as for-profit GSEs is the potential conflict with a requirement to facilitate the flow of mortgage credit serving targeted groups. Alternatives to establishing a numeric mortgage purchase program to support homeownership opportunities for targeted groups could include assessing fees on the reconstituted GSEs to directly fund such programs, expanding FHA’s mortgage insurance programs, or providing direct down-payment assistance to targeted borrowers. Continuing the enterprises as GSEs could present significant safety and soundness concerns as well as systemic risks to the financial system. In particular, the potential that the enterprises would enjoy explicit federal guarantees of their financial obligations, rather than the implied guarantees of the past, might serve as incentives for them to engage in risky business practices to meet profitability objectives. Treasury guarantees on their financial obligations also might provide the enterprises with significant advantages over potential competitors. Furthermore, FHFA’s capacity to monitor and control such potentially risky business practices has not been tested. Since its establishment in July 2008, FHFA has acted both as the enterprises’ conservator and safety and soundness and housing mission regulator. Under the conservatorship, FHFA has significant control over the enterprises’ operations. For example, the FHFA Director has appointed the enterprises’ chief executive officers and boards of directors and can remove them as well. But FHFA officials said that agency staff also have been monitoring the enterprises’ business risks. It remains to be seen how effectively FHFA would carry out its oversight responsibilities solely as an independent regulator if the enterprises were reconstituted as for-profit GSEs. While converting the enterprises into multiple GSEs could mitigate safety and soundness and systemic risk concerns by minimizing concentration risks, it also likely would involve trade-offs. For example, multiple GSEs, due to their potentially small size, may not be able to achieve economies of scale and generate certain efficiencies for mortgage markets as has been the case with Fannie Mae and Freddie Mac. As discussed earlier, the enterprises, through their secondary mortgage market activities, have been credited with facilitating the development of a liquid national mortgage market and establishing standardized underwriting practices for mortgage lending. Alternatively, converting the enterprises from their current structure as publicly owned corporations into cooperatives owned by the lenders that sell mortgages to them may offer certain advantages in terms of their safety and soundness. For example, as the owners of the enterprises, lenders may have financial incentives to ensure that the mortgages that they sell to the enterprises are properly underwritten so as to minimize potential losses that would affect the value of their investments. As discussed previously, Freddie Mac, as a cooperative, generally managed to avoid the financial problems that Fannie Mae, which was a publicly owned corporation, faced during the early 1980s. However, it should also be noted that the cooperative structure may also have limitations. For example, some FHLBanks, which are members of the cooperative FHLBank System, have faced losses recently due to their investments in private-label mortgage assets. While the public utility model of regulation has been proposed as a means to help mitigate the risks associated with reestablishing the enterprises as GSEs, this proposal involves complexities and trade-offs. For example, it is not clear that the public utility model is an appropriate regulatory structure because, unlike natural monopolies such as electric utilities, the enterprises have faced significant competition from other providers of mortgage credit over the years. For example, as discussed previously, the enterprises’ market share declined substantially from 2004 through 2006 due to the rapid growth of the private-label MBS market. Public utility-type regulation also has been criticized as inefficient and many states have sought to deregulate their electric and other markets. Furthermore, these proposals may have offsetting effects on the GSEs’ financial viability. For example, former Treasury Secretary Paulson’s proposal would subject their credit decisions and rate of return to preapproval by a public utility- type board, and impose fees on them to offset any benefits derived from government sponsorship of their activities. It is not clear if an entity subject to such business activity restrictions, regulations, and fees, even with Treasury guarantees on its financial obligations, would be able to raise sufficient capital from investors or purchase mortgages on terms that mortgage lenders would find cost-effective. A range of potential options exist for developing an appropriate regulatory structure to help ensure that any reconstituted GSEs would operate in a safe and sound manner while achieving housing mission objectives. For example, if maintaining safety and soundness is viewed as a priority over a numeric housing goal program, then eliminating retained mortgage portfolios may be viewed as appropriate. Alternatively, the retained mortgage portfolios could be substantially smaller and restricted to certain types of assets to help ensure safety and soundness while promoting housing mission achievement. Other steps that may be deemed appropriate could include establishing capital standards for the enterprises commensurate with their risk, additional restrictions on their activities, executive compensation limits, public utility regulation, appropriate financial disclosures of risks and liabilities in the federal budget, and strong congressional oversight of the enterprises’ and FHFA’s performance. As with the preceding two sets of options, proposals that involve the privatization or ultimate termination of the enterprises involve a number of trade-offs. For example, if a consortium of large banks assumed responsibility for key activities (such as mortgage purchases and MBS issuances) of the enterprises, during good economic times it might be able to provide liquidity to the mortgage finance system, help ensure consistency through uniform underwriting standards, and potentially promote innovation in mortgage finance. However, the ability of private lenders to provide support to mortgage markets during stressful economic periods is questionable. As discussed previously, many private-sector lenders have failed or withdrawn from mortgage markets during the current economic downturn. The establishment of a federal mortgage debt insurer, as has been proposed, may facilitate private lenders’ capacity to support mortgage markets during stressful periods. Privatizing or terminating the enterprises also could affect the structure of mortgage lending that has evolved over the years. For example, lenders might be less willing to originate 30-year, fixed-rate mortgages, due to the associated interest-rate risk of holding them in portfolio, if any ensuing private-sector secondary market alternatives (such as a consortium of private-sector lenders) were less willing to purchase such mortgages than the enterprises had been. Additionally, privatization or termination could result in a relative increase in mortgage interest rates, because private- sector lenders might not have the funding advantages that the enterprises derived from their federal sponsorship over the years. This option also could eliminate the traditional legislative basis for requiring that they facilitate the flow of mortgage credit serving targeted groups, particularly through the numeric mortgage purchase program. That is, the enterprises currently have a responsibility to help meet the mortgage credit needs of all potential borrowers due to the financial benefits associated with federal sponsorship, which would not be the case for private-sector lenders under the termination and privatization proposals. However, if new federal organizations were established, such as a mortgage insurer, to facilitate the transition to a mortgage finance system in which the enterprises no longer exist, then they could be required to assume responsibility for facilitating the flow of mortgage credit to targeted groups. For example, the Community Reinvestment Act’s requirements that insured depositories, such as banks and thrifts, serve the credit needs of the communities in which they operate could be extended to nondepository lenders, such as independent mortgage lenders, which would obtain mortgage insurance from a new federal mortgage insurer. Moreover, if a consortium of large lenders or other financial institutions assumed responsibility for key enterprise functions (like MBS issuances), or purchased a substantial share of their assets, then requirements that such institutions serve the credit needs of targeted groups also might be justified. For example, such institutions could be perceived as benefiting from implied federal guarantees on their debt or being too big to fail. We note that Treasury and the Federal Reserve have provided direct financial assistance to a range of financial and other institutions during the current financial crisis, which may create the perception in financial markets that the federal government is more likely to intervene in a future crisis. The extent to which privatizing or terminating the enterprises mitigates current safety and soundness and financial stability risks is difficult to determine. Under one scenario, such risks would be mitigated because large and complex enterprises, which might engage in risky business practices due to the implied federal guarantee on their financial obligations, would not exist. Instead, private lenders would be subject to market discipline and consequently would be more likely to make credit decisions largely on the basis of credit risk and other market factors. However, this scenario would be complicated if a federal entity were established to insure mortgage debt. If the federal mortgage insurer did not set appropriate premiums to reflect the risks of its activities, then lenders might have incentives to engage in riskier business practices than otherwise would be the case. In similar situations, such as the National Flood Insurance Program, federal agencies have faced challenges in establishing appropriate premiums to compensate for the risks that they underwrite. If large private-sector financial institutions assumed responsibility for key enterprise activities or purchased a significant portion of their assets, the perception could arise that the failure of such an institution would involve unacceptable systemic financial risks. Therefore, markets’ perceptions that the federal government would provide financial assistance to such financial institutions could undermine market discipline. Moreover, limitations in the structure of the current U.S. financial regulatory system could heighten concerns about the potential safety and soundness risks associated with large financial institutions assuming responsibility for key enterprise financial activities or becoming larger due to the purchase of their assets. In a recent report, we stated that the current fragmented regulatory system for banks, securities firms, insurance companies, and other providers evolved over many years and often in response to financial crises. We stated that the large and complex financial conglomerates that have emerged in the past decades often operate globally across financial sectors and that federal regulators have faced significant challenges in monitoring and overseeing their operations. For example, the report noted that a Federal Reserve official recently acknowledged that, under the current structure, which consists of multiple supervisory agencies, challenges can arise in assessing risk profiles of these institutions, particularly because of the growth in the use of sophisticated financial products that can generate risks across various legal entities. Privatizing or terminating the enterprises also could increase the relative prominence of other federal programs designed to promote homeownership and housing opportunities, which also may have safety and soundness implications. Due to the diminished presence of the enterprises in mortgage finance under these proposals, market participants, such as banks and thrifts, increasingly might turn to the FHLBank System as a source of funding for their operations and lending activities. The FHLBank System could enjoy an advantage over other potential competitors in filling the void left by the enterprises because, as a GSE benefiting from an implied guarantee, it may be able to issue debt to fund its activities at relatively advantageous rates. However, some FHLBanks have recently reported losses due to investments in private- label MBS. Similarly, FHA’s mortgage insurance programs might increase if the enterprises’ diminished role limits the availability of mortgage credit in the conforming market. The development of an appropriate regulatory structure to help ensure housing mission achievement and safety and soundness, as deemed appropriate, would depend on the outcome of a range of contingencies associated with options to privatize or terminate the enterprises. For example, should Congress choose to establish a new public entity to insure all mortgage debt, with the federal government guaranteeing the insurance, then a new regulatory and oversight structure would be needed to oversee the operations of such an insurer. As with other options to reform the enterprises’ structures, an appropriate structure for such an entity might involve a regulatory agency with authorities to carry out its activities and capital standards that reflect the risk of the entity’s activities, disclosures of risks and liabilities in the federal budget to help ensure financial transparency, and robust congressional oversight. Furthermore, revisions may be necessary to help ensure that the U.S. financial regulatory system can better oversee the risks associated with large and complex financial institutions, which may assume responsibility for key enterprise activities or become larger over time through the acquisition of their assets. Our recent report identified a series of principles, such as establishing clear regulatory goals, ensuring a focus on systemwide financial risks, and mitigating taxpayer risks, for Congress to consider in deciding on the most appropriate regulatory system. Since the beginning of the FHFA conservatorships, the enterprises have been tasked to initiate a range of programs, such as assisting homeowners struggling to make their mortgage payments to refinance or modify their mortgage terms, to respond to the current crisis in housing markets. These initiatives could benefit housing markets and, in so doing, potentially benefit the enterprises’ financial condition. However, the initiatives also may involve additional risks and costs for the enterprises, which could increase the costs and challenges associated with transitioning to new structures over time. Similarly, certain provisions in the Treasury agreements with the enterprises may affect their long-term financial viability and complicate a transition to a new structure. Finally, any transition to a new structure would need to take into consideration the enterprises’ dominant position within housing finance, even during the conservatorships, and, therefore, should be carefully implemented— perhaps in phases—to help ensure its success. The following points summarize several of the initiatives that the enterprises have undertaken in response to the substantial downturn in housing markets: Under the Home Affordable Refinance Program (HARP), which was initiated in March 2009, borrowers that have current payment histories can refinance and reduce their monthly mortgage payments at loan-to-value ratios of up to 105 percent without obtaining mortgage insurance. On July 1, 2009, the program was extended to apply to mortgage loans with loan- to-value ratios of up to 125 percent. Under the Home Affordable Modification Program (HAMP), certain borrowers who are delinquent, or in imminent danger of default, on their mortgage payments may have the terms of their existing mortgages modified in order to make payments more affordable. Specifically, the program allows for interest rate reductions (down to 2 percent), term extension (up to 480 months), principal forbearance, and principal forgiveness. Under the program, the enterprises will provide up to $25 billion in incentives to borrowers and servicers for program participation and a successful payment history. In November 2008, the enterprises suspended the initiation of foreclosure proceedings on mortgages that they held in their portfolios or on which they had guaranteed principal and interest payments for MBS investors. This initiative subsequently was extended through March 31, 2009. In March 2009, the enterprises also suspended foreclosure sales on mortgages that may be eligible under HAMP until borrowers’ eligibility for HAMP has been verified. While these federal initiatives were designed to benefit homebuyers, in recent financial filings, both Freddie Mac and Fannie Mae have stated that the initiative to offer refinancing and loan modifications to at-risk borrowers could have substantial and adverse financial consequences for them. For example, Freddie Mac stated that the costs associated with large numbers of its servicers and borrowers participating in loan- modification programs may be substantial and could conflict with the objective of minimizing the costs associated with the conservatorships. Freddie Mac further stated that loss-mitigation programs, such as loan modifications, can increase expenses due to the costs associated with contacting eligible borrowers and processing loan modifications. Additionally, Freddie Mac stated that loan modifications involve significant concessions to borrowers who are behind in their mortgage payment, and that modified loans may return to delinquent status due to the severity of economic conditions affecting such borrowers. Fannie Mae also has stated that, while the impact of recent initiatives to assist homeowners is difficult to predict, the participation of large numbers of its servicers and borrowers could increase the enterprise’s costs substantially. According to Fannie Mae, the programs could have a materially adverse effect on its business, financial condition, and net worth. However, FHFA officials said that they strongly believe the recent initiatives to support the housing markets, on balance, represent the best means available for the enterprises to preserve their assets and fulfill their housing missions. For example, FHFA officials said that, to the extent that their initiatives are successful in stabilizing housing markets, the enterprises will be the major beneficiaries as the number of delinquent mortgages and foreclosures is reduced. FHFA officials also commented that recent modification programs, such as HAMP, are more likely to be successful than modification initiatives dating to 2008, which had high redefault rates. FHFA officials said that the more recent loan-modification initiatives were more likely to reduce borrowers’ monthly payments. According to an FHFA report, the traditional approaches to loan modifications (allowing borrowers to bring loans current by reamortizing past due payments over the remaining life of the loan) increased monthly payments and, therefore, often resulted in high redefault rates. Furthermore, FHFA officials stated that recent loan-to-value ratio refinance programs only apply to mortgages that the enterprises already guarantee, so they already are exposed to the credit risks on these loans. FHFA officials said that such refinancings also should lower borrowers’ payments and thereby further reduce the enterprises’ existing credit exposure. While FHFA’s positions are plausible, it is too early to reach any conclusion about the effects that the initiatives will have on the enterprises’ financial condition and preliminary data raise potential concerns. According to a report by the Office of the Comptroller of the Currency (OCC) and the Office of Thrift Supervision (OTS), loan modifications initiated in 2008 that reduced borrowers’ monthly payments by 20 percent or more had significantly lower redefault rates after 1 year than modifications that left monthly payments unchanged or higher. Specifically, the study found that, of the modifications that involved reductions of 20 percent or more, 38 percent were 60 or more days past due after 1 year, whereas the rate was nearly 60 percent for modifications that left monthly payments unchanged or higher. However, the fact that nearly 40 percent of loan modifications that substantially reduced monthly payments were already 60 or more days past due after 1 year raises concerns about whether the additional costs that enterprises incur in administering such programs will be effective. Furthermore, it is also not clear whether initiatives to suspend foreclosure proceedings will benefit the enterprises’ financial condition. Our previous work has found that, for mortgage providers such as Fannie Mae and Freddie Mac, foreclosure costs may increase the longer it takes to maintain and sell foreclosed properties. A potential risk of suspending pending foreclosure sales is that many borrowers facing foreclosure will not be able to obtain funds necessary to make their mortgage loan payments current. As a result of delays in foreclosing on such properties, the potential exists that the properties will not be maintained or will become vacant, which could increase the enterprises’ associated costs. Treasury’s agreements with Fannie Mae and Freddie Mac, which specify terms under which the department is to provide certain types of financial support to them, also may have long-term financial consequences. In connection with the agreements, quarterly dividends declared by the enterprises are to be paid to holders of the senior preferred stock (Treasury). These dividends accrue at 10 percent per year and increase to 12 percent if, in any quarter, they are not paid in cash. If either enterprise cannot pay the required dividends, then Treasury has a claim against the assets of the enterprise for the unpaid balance in a liquidation proceeding. Available financial data suggest that the enterprises, while in conservatorship and over the longer term, will face significant financial challenges in paying the required dividends to Treasury. For example, Treasury already purchased $50 billion in preferred stock in Freddie Mac, which translates into an annual dividend of $5 billion, and CBO estimated that the department will invest substantially more in the enterprise’s preferred shares in coming quarters (up to the guarantee limit of $200 billion). Prior to the conservatorship, Freddie Mac’s reported annual net income twice came close to or exceeded $5 billion, and the dividends that it distributed to shareholders in those years likely were substantially lower. In addition, the agreements require that, beginning on March 31, 2010, the enterprises pay a commitment fee to Treasury to compensate the department for the ongoing financial support that it is providing to them. While the size of the commitment fee is subject to negotiation, it represents another potential long-term challenge to the enterprises’ financial viability. For example, like the dividend requirements, any unpaid commitment fees become a claim by Treasury against the assets of the enterprises in a liquidation proceeding, unless Treasury waives the fee. Although it is not possible to predict what effects federal initiatives to respond the housing crisis and the Treasury agreements with the enterprises could have on the transition to a new structure, they could be substantial. For example, under the proposal to reconstitute the enterprises as for-profit GSEs, potential investors might not be willing to invest their capital if the reconstituted GSEs had a substantial volume of nonperforming mortgage assets or substantial financial obligations to Treasury. To minimize this risk, the federal government could arrange a transition process in which the government would retain nonperforming assets in a “bad bank” and spin off the performing assets of the enterprises to a “good bank” and key functions, such as issuing MBS, to investors in a reconstituted GSE. Or, the federal government could establish such a process as a means to terminate or privatize the enterprises. However, to the extent that the enterprises engage in activities during their conservatorships or incur financial obligations inconsistent with maintaining their long-term financial viability, the level of nonperforming mortgage assets and long-term costs to taxpayers ultimately may be higher than otherwise would be the case. Finally, regardless of what changes are implemented, policymakers should pay careful attention to how a potential transition is managed to mitigate potential risks to the housing finance system. The enterprises evolved over many years to become dominant participants in housing finance and, in some respects, their roles have expanded during the conservatorships. Therefore, transitioning to a new structure could have significant consequences for housing finance and should be managed carefully and perhaps implemented in phases with periodic evaluations to determine if any corrective actions would be necessary. For example, any changes likely would require regulators and institutions to make system changes and undertake other activities that would take extensive time to complete. Our previous work also has identified other key issues that likely would be critical components of any transition process. In particular, an effective communication strategy would be necessary to help ensure that all mortgage market participants, including lenders, investors, and borrowers, have sufficient information to understand what changes are being made and how and when they will be implemented. Moreover, it will be important to put effective strategies in place to help ensure that, under whichever reform strategy is chosen, the new financial institutions and their regulators will have the staffing, information technology, and other resources necessary to carry out their missions. We provided a draft of this report to FHFA, the Federal Reserve, HUD, and Treasury for their review and comment. While he was still the FHFA Director, James B. Lockhart III provided us with written comments, which are summarized below and reproduced in appendix II, as well as technical comments, which we incorporated as appropriate. Federal Reserve staff, HUD’s Assistant Secretary for Housing-Federal Housing Commissioner, and Treasury also provided technical comments, which were incorporated as appropriate. We also provided excerpts of a draft of this report to seven researchers whose studies we cited to help ensure the accuracy of our analysis. Six of the researchers responded and said that the draft report accurately described their research, while one researcher did not respond. In his comment letter, Lockhart stated that the report is timely and does a good job of summarizing the dominant proposals for restructuring the enterprises and summarizing their strengths and weaknesses. Lockhart also stated that initial attention should be to the role of mortgage finance in our society and how the government wants the institutions and markets that supply it to function and perform. In particular, he said this includes determining the most appropriate roles for private and public entities, competition and competitiveness, risks and risk management, and the appropriate channels and mechanisms for targeting the underserviced and protecting consumers. Further, he identified key questions and principles that he believes should be included in the debate on restructuring the enterprises. These principles include (1) deciding what the secondary market should look like, before considering specific institutions; (2) ensuring that the enterprises or any successors have well-defined and internally consistent missions; (3) ensuring that there is a clear demarcation of the federal government and the private sector in the secondary market; (4) establishing a regulatory and governance structure that ensures prudent risk-taking; and (5) ensuring that housing finance is subject to systematically prudent supervision that incorporates countercyclical capital to limit booms and busts. We concur with the thrust of the view that revising the enterprises’ structures should take place in a measured way and in the context of a broader assessment of the housing finance system. As discussed in our report, the enterprises have been key components of the housing finance system for many years and, therefore, any changes to their structures are likely to have broad implications for that system and market participants. In this regard, we stated that it will be important for Congress to reevaluate the enterprises’ roles, structures, and performance, and consider structural reform options to facilitate mortgage finance while mitigating safety and soundness concerns. These options, under certain scenarios, envision very different approaches to structuring the secondary market for mortgage loans and facilitating housing opportunities for targeted groups, and we believe the broad implications of these various options need to be carefully considered before any final decisions are made. For this reason, our report addresses implications for various participants in the mortgage markets, including FHA. Further, we discussed that a carefully managed and potentially lengthy transition process needs to be established to help ensure the successful implementation of whatever structural reform option for the enterprises is chosen by Congress and the Executive Branch. Additionally, Lockhart said that FHFA, in its role as the enterprises’ conservator, as well as their mission and safety and soundness regulator, is working diligently with the Treasury and Federal Reserve to maintain or restore safe, sound, liquid, and vibrant mortgage markets. He said a principal focus of FHFA’s efforts has been facilitating the enterprises’ participation in the Home Affordable Modification and Refinance Programs. While he said the enterprises’ participation in these programs may result in near-term costs, he believes the programs will result in stronger and more stable housing markets, which will also benefit the enterprises. Finally, he made several suggestions regarding certain aspects of the draft report. These suggestions and our responses are described below: The draft report should make clear that the structural reform options presented in the report are not exhaustive or mutually exclusive and that hybrids of these options also are possible and may prove to be the most appealing. We agree and, as the report notes, the options for revising the enterprises’ long-term structures generally fall along a continuum with some overlap between key features. For example, as Lockhart noted, options for privatizing or terminating the enterprises may involve establishing a government entity to insure mortgages originated by private lenders. In addition, the government entity and reconstituted GSE options generally involve focusing enterprise activities on issuing MBS while downsizing or eliminating their mortgage portfolios. The draft report should mention the enterprises’ performance in providing liquidity to mortgage markets. We agree that any discussion of the future roles of the GSEs should include consideration of their roles in providing securities that support an active and liquid mortgage market. As the report notes, providing liquidity to mortgage markets has been a key housing mission objective of the enterprises and that, while their secondary market activities have been credited with helping to establish a national and liquid mortgage market, their performance in providing support to mortgage markets during stressful economic periods is not clear. While the draft report’s discussion of the safety and soundness concerns related to the government entity option is reasonably balanced and fair, it is short on negative details. In particular, (1) the draft report is organized in such a way that makes it easy for the reader to conclude that the safety and soundness benefits of the government entity option outweigh the added risks; (2) the table in the draft report’s Highlights page states that the lack of a “profit motive” for a government entity may mitigate risk should be rephrased to state that the option “addresses the conflict between private profits and public sector risk bearing;” and (3) the discussion in the draft report on the potential elimination of the enterprises’ mortgage portfolios fails to recognize that such an action is a component of some but not all proposals to reconstitute the enterprises as GSEs or to establish a government entity, and therefore, mentioning the benefit of doing so under one option (the government entity option) and not the other (the reconstituted GSE option) is a significant inconsistency. Regarding (1), we do not agree that the order of the text in the draft report implied that the benefits of the government entity option outweigh its risks. While this option offers potential safety and soundness advantages such as addressing deficiencies in the traditional enterprise structures and eliminating their mortgage portfolios, it also has potentially significant drawbacks, which need to be considered. In particular, we stated that managing the enterprises’ ongoing MBS business may be complicated and challenging, and government entities may lack the resources and expertise necessary to manage such challenges and risks effectively. Regarding (2), we agree with the thrust of this comment and have modified the text in the table in the Highlights page to make it consistent with the related figure and text in the body of the report. Regarding (3), we agree that any assessment of the options for revising the housing enterprises’ long-term structure should include discussion of the implications of retaining mortgage portfolios. As described in the report, the government entity options we identified advocated the elimination of the enterprises’ portfolios. In contrast, the options we identified for reconstituting the enterprises as GSEs generally called for reducing the enterprises’ portfolios while one proposal called for their complete elimination. The report includes analysis of the potential implications of taking such steps regarding the enterprises’ mortgage portfolios under both options, as well as the possible elements of regulatory and oversight structures that could mitigate any potential safety and soundness and systemic stability risks. We are sending copies of this report to interested congressional committees and members. In addition, we are sending copies to FHFA, Treasury, the Federal Reserve, HUD, financial industry participants, and other interested parties. The report also is available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff have any questions about this report, please contact William B. Shear at (202) 512-8678 or shearw@gao.gov or Richard J. Hillman at (202) 512-8678 or hillmanr@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix III. The objectives of our report were to (1) discuss how the enterprises’ roles, structures, and activities have changed over time and their performance in achieving key housing mission objectives; (2) identify various options for revising the enterprises’ eventual structure; (3) analyze these options in terms of their potential capacity to achieve key housing mission and safety and soundness objectives; and (4) discuss how the federal government’s management of the conservatorships and response to the housing crisis could affect any transition. To address the first objective, we reviewed reports and studies on the enterprises and their regulation, including GAO reports, as well as reports from the Department of Housing and Urban Development (HUD), the Federal Housing Finance Agency (FHFA), the Office of Federal Housing Enterprise Oversight (OFHEO), the Congressional Budget Office (CBO), the Congressional Research Service (CRS), and independent researchers. We also reviewed legislative and charter documents, as well as an internal history of Fannie Mae, and financial performance data from a variety of sources. Through this research, we sought to identify key housing mission, safety and soundness, and other objectives that have been associated with the enterprises over the years, as well as their performance in meeting such objectives. In doing so, we identified and summarized recent literature that addressed the impact of the enterprises on affordability and opportunities for target groups. While GAO reviewed these studies and included those that were sufficiently methodologically sound for our limited purposes, users of this report should note that these studies are based on data prior to 2001 and contain limitations. Finally, we used data from the Securities Industry and Financial Markets Association (SIFMA) and OFHEO. As SIFMA’s data on mortgage-related issuance were consistent with other data sources and highlight well-established trends in mortgage-backed securities (MBS) and collateralized mortgage obligation activity, we found them and the OFHEO data on the balance sheets of the enterprises sufficiently reliable for our purposes. To address the second objective, we reviewed a variety of studies and proposals that have been made prior to and during the conservatorships to revise the enterprises’ structures. The inclusion of these studies and proposals is purely for research purposes and does not imply that we deem them definitive or without limitations. We also met with the authors of many of these studies and with researchers who have knowledge about housing finance, the operations of the enterprises, or who have made proposals to revise the enterprises’ structures. We met with representatives from FHFA, the Department of the Treasury (Treasury), the Federal Reserve, HUD, the Government National Mortgage Association (Ginnie Mae), CBO, the enterprises, bank and mortgage organizations, and trade and community groups. These interviews provided us with the different viewpoints about the proposals. For the third objective, we analyzed the proposed options for restructuring the enterprises in terms of the potential each proposal offered to achieve key housing mission and safety and soundness objectives. In our analysis, we also relied on principles associated with effective regulatory oversight. While it is not possible to conclusively determine the potential implications of the various proposals, we grounded our analysis of likely outcomes on previous research and evaluations. We also sought to include, where appropriate, assessments of how recent developments in financial markets (particularly actions by federal agencies to provide financial support to troubled banks and other institutions) could affect the various options. We recognize that a variety of factors, such as the condition of credit markets and the financial performance of the enterprises while in conservatorship, could change over time and affect our analysis of the options. For the final objective, which discusses how the federal government’s management of the conservatorships and response to the housing crisis could affect the transition of the enterprises to a new structure, we reviewed the actions undertaken by FHFA, Treasury, and the Federal Reserve, as authorized by the Housing Economic and Recovery Act of 2008. We also reviewed financial data from Fannie Mae and Freddie Mac, including their quarterly 10Q and annual 10K filings. We reviewed and considered the future impact on the enterprises’ financial condition from recent initiatives such as the Homeowner Affordability and Stability Act and foreclosure initiation suspensions. We also discussed relevant issues with Treasury and enterprise representatives. We conducted this performance audit from October 2008 to September 2009, from Washington, D.C., and in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. In addition to the individual named above, Wesley M. Phillips, Assistant Director; Triana Bash; Martha Chow; Lawrance Evans, Jr.; Marc Molino; Robert Pollard; Barbara Roesmann; Stacy Spence; Paul Thompson; and Barbara Williams made key contributions to this report. | Congress established Fannie Mae and Freddie Mac (the enterprises) with two key housing missions: (1) provide stability in the secondary market for residential mortgages (also in periods of economic stress) and (2) serve the mortgage credit needs of targeted groups such as low-income borrowers. To accomplish these goals, the enterprises issued debt and stock, purchased mortgages from lenders with the proceeds, and retained them in portfolio or pooled them into mortgage-backed securities (MBS) sold to investors. On September 6, 2008, the Federal Housing Finance Agency (FHFA) placed the enterprises into conservatorship out of concern that their deteriorating financial condition ($5.4 trillion in outstanding obligations) would destabilize the financial system. With estimates that the conservatorship will cost taxpayers nearly $400 billion, GAO initiated this report under the Comptroller General's authority to help inform the forthcoming congressional debate on the enterprises' future structures. It discusses the enterprises' performance in meeting mission requirements, identifies and analyzes options to revise their structures, and discusses key transition issues. GAO reviewed studies and data, and interviewed housing finance experts and officials from the enterprises, FHFA, Departments of the Treasury (Treasury) and Housing and Urban Development (HUD), the Federal Reserve, lenders, and community groups. The enterprises have a mixed record in meeting their housing mission objectives, and both capital and risk management deficiencies have compromised their safety and soundness as follows: (1) The enterprises' secondary market activities are credited with helping create a liquid national mortgage market, lowering mortgage rates somewhat, and standardizing mortgage underwriting processes. However, their capacity to support housing finance during periods of economic stress has not been established, and they only have been able to do so during the current recession with substantial financial assistance from Treasury and the Federal Reserve. (2)There is limited evidence that a program established in 1992 that required the enterprises to meet annual goals for purchasing mortgages serving targeted groups materially benefited such groups. (3) The enterprises' structures (for-profit corporations with government sponsorship) undermined market discipline and provided them with incentives to engage in potentially profitable business practices that were risky and not necessarily supportive of their public missions. For example, the enterprises' retained mortgage portfolios are complex to manage and expose them to losses resulting from changes in interest rates. Further, the enterprises' substantial investments in assets collateralized by subprime and other questionable mortgages in recent years generated losses that likely precipitated the conservatorship. It will be necessary for Congress to reevaluate the roles, structures, and performance of the enterprises, and to consider options to facilitate mortgage finance while mitigating safety and soundness and systemic risk concerns. These options generally fall along a continuum with some overlap in key areas: (1)Reconstitute the enterprises as for-profit corporations with government sponsorship but place additional restrictions on them. While restoring the enterprises to their previous status, this option would add controls to minimize risk. As examples, it would eliminate or reduce mortgage portfolios, establish executive compensation limits, or convert the enterprises from shareholder-owned corporations to associations owned by lenders. (2) Establish the enterprises as government corporations or agencies. Under this option, the enterprises would focus on purchasing qualifying mortgages and issuing MBS but eliminate their mortgage portfolios. The Federal Housing Administration (FHA), which insures mortgages for low-income and first-time borrowers, could assume additional responsibilities for promoting homeownership for targeted groups. (3) Privatize or terminate them. This option would abolish the enterprises in their current form and disperse mortgage lending and risk management throughout the private sector. Some proposals involve the establishment of a federal mortgage insurer to help protect mortgage lenders against catastrophic mortgage losses. During the conservatorship, the federal government has tasked the enterprises to implement a variety of programs designed to help respond to the current housing crisis, such as helping borrowers forestall foreclosures. While these efforts may be necessary to help mitigate the effects of the housing crisis, they also might significantly affect the costs of the conservatorship and transition to a new structure. For example, investors might be unwilling to invest capital in reconstituted enterprises unless Treasury assumed responsibility for losses incurred during their conservatorship. Finally, any transition to a new structure would need to consider the enterprises' still-dominant position in housing finance and be implemented carefully (perhaps in phases) to ensure its success. |
CMS and Palmetto GBA administer and implement the CBP and its bidding phases. In each CBP bidding phase, suppliers that want to participate in the CBP may submit a bid in one or more product categories in one or more designated competitive bidding areas. To be offered a CBP contract, suppliers must be qualified, meaning they have met general Medicare enrollment and quality standards as well as CBP’s financial and applicable licensing standards; be eligible to bill Medicare for DME items; have a DME surety bond; and be accredited and licensed. After CMS and Palmetto GBA review qualified suppliers’ bids, the bids are ordered by lowest to highest price, and CMS then makes offers to all those needed to meet or exceed CMS’s estimated beneficiary demand who submitted the lowest prices. If a bidding supplier accepts an offer to furnish a specific product category in a specific competitive bidding area, it must agree to furnish all of the items included in the product category to all eligible Medicare beneficiaries residing in the competitive bidding area at the applicable single payment amounts. Because the single payment amount is the median of the winning bid price offers for each DME item in a product category for each competitive bidding area, the payment can be less or more than a particular winning supplier’s actual bid for an item. Furthermore, because each competitive bidding area and product category combination is a separate competition, the same DME item may have a different single payment amount in each area. For example, the round 2 single payment amounts for a new foam rubber mattress ranged between $111.38 in the Palm Bay and Deltona, Florida, competitive bidding areas to $178.62 in the Honolulu, Hawaii, competitive bidding area. Contracts are generally awarded for 3 years and can include one or many competitions. While generally only contract suppliers that accepted a contract for a given product category and competitive bidding area are eligible to furnish those items, there are circumstances in which suppliers not awarded a CBP contract—referred to as non-contract suppliers—may be grandfathered to continue to furnish some CBP-covered items to certain beneficiaries for a limited time. In addition, physicians, treating practitioners, and hospitals can furnish walkers, folding manual wheelchairs, or external infusion pumps to their own patients as part of their professional services or during hospital admission or discharge without a CBP contract. When offering contracts, CMS takes steps that it believes will ensure beneficiary access and choice. For example, CMS’s goal is to award at least five contracts for each product category and competitive bidding area competition. To help meet this goal, CMS caps the estimated projected capacity of any single supplier at 20 percent of the total projected beneficiary demand for each product category, in each competitive bidding area, regardless of the capacity estimated by the supplier in its bid submission. CMS also tries to ensure that small suppliers are awarded CBP contracts by setting a target that 30 percent of the qualified suppliers awarded a contract in each product category and competitive bidding area competition are small. CMS is required by the MMA to conduct a new CBP phase at least once every 3 years, and begins the bid submission and award process for the new contracts before the current contracts expire. Beginning with the first round of CBP in 2008, CMS and Palmetto GBA have continued to phase- in CBP through additional rounds and programs. (See fig. 1 for a timeline summarizing the phase-in of the CBP and app. II for additional information about the CBP phases.) As of October 2015, there were 822 round 2 contract suppliers—520 of which were small suppliers—to furnish DME items and services in eight product categories and 100 competitive bidding areas. (See table 1 for a list of round 2 product categories and app. I for a list of the 100 competitive bidding areas.) Contracts were effective beginning July 1, 2013, and expired June 30, 2016. There were 19 national mail-order contract suppliers as of October 2015—6 of which were small suppliers—to furnish the eight mail-order diabetes testing supply HCPCS codes included in the national mail-order program. Contracts were effective beginning July 1, 2013, and expired June 30, 2016. The program operates in the 50 states, the District of Columbia, Puerto Rico, the U.S. Virgin Islands, Guam, and American Samoa, and includes the same eight mail-order diabetes testing supply HCPCS codes as the round 1 rebid. For this round and future national mail-order program rounds, a supplier’s diabetes testing supply bid must demonstrate that the supplier’s bid would cover at least 50 percent, by sales volume, of all types of diabetes test strips on the market. Medicare payments for items included in the national mail-order program are the same regardless of whether they are furnished via mail order or by retail, though out-of-pocket costs for beneficiaries may be higher when they receive these supplies through retail outlets. Medicare beneficiaries residing in competitive bidding areas have several sources available to help them locate contract suppliers and receive assistance for CBP-related issues, questions, or complaints. Medicare Supplier Directory. To locate a CBP contract supplier, beneficiaries can use the CMS online supplier directory tool on CMS’s Medicare website. The Medicare Supplier Directory contains the names of the contract suppliers in each competitive bidding area as well as the product categories for which they furnish CBP-covered items. Contract suppliers are responsible for submitting information to CMS each quarter regarding the specific brands of items they plan to furnish in the upcoming quarter, and CMS uses this information to update the supplier directory tool. 1-800-MEDICARE inquiries. CMS directs beneficiaries to call its 1- 800-MEDICARE beneficiary help line for assistance with CBP-related questions. Customer service representatives are trained to assist CBP beneficiaries and use several scripts to respond to questions and assist beneficiaries in locating contract suppliers. If a beneficiary’s inquiry cannot be addressed by the customer service representatives, the inquiry is forwarded to an advanced-level customer service representative, who researches and responds to the beneficiary’s inquiry. Palmetto GBA and CMS regional offices. Palmetto GBA provides CBP-related information and updates through its website and works with CMS regional office staff to monitor CBP activities and provide educational outreach. Competitive Acquisition Ombudsman. The Competitive Acquisition Ombudsman was created to respond to CBP-related complaints and inquiries made by suppliers and individuals, and works with CMS officials and contractors and Palmetto GBA to resolve them. CMS has implemented several activities to monitor whether beneficiary access or satisfaction have been affected by the implementation of CBP and to ensure that contract suppliers are meeting their contract obligations. Inquiries and complaints to 1-800-MEDICARE. CMS tracks all CBP-related inquiries to 1-800-MEDICARE. All calls are first classified as inquiries, and CMS defines as a CBP complaint only those inquiries that cannot be resolved by any 1-800-MEDICARE customer service representative and are elevated to another entity, such as Palmetto GBA, CMS’s regional offices, or the Competitive Acquisition Ombudsman for resolution. Secret shopping calls. According to CMS officials, CMS utilizes shopping calls, in which Palmetto GBA representatives pose as referral agents or family members acting on behalf of beneficiaries and call contract suppliers to request items, such as specific diabetes testing supplies, to determine whether the suppliers offer the supplies covered under their contracts. Some calls are conducted on a random basis and are intended to reach contract suppliers from across all product categories and competitive bidding areas, while others are directed at particular suppliers. According to CMS officials, the agency uses both random and targeted calls to monitor contract suppliers that have not billed for items included in their contracts over a period of time. In addition, CMS officials said that a third type of secret shopping call, focused calls, is used to investigate specific complaints and verify contract supplier compliance with contract requirements after any necessary education to the supplier(s) is provided. According to CMS, if contract suppliers remain noncompliant following education, the agency will start the contract termination process. Beneficiary satisfaction surveys. CMS conducted both pre- and post-round 2 and national mail-order program surveys to measure beneficiary satisfaction with CBP. The round 2 pre-implementation survey was conducted January 3 through March 18, 2013, and the post-implementation survey was conducted from March 5 through April 10, 2014. The national mail-order program pre-implementation survey was conducted April 10 through April 24, 2013, and the post- implementation survey was conducted April 17 through April 27, 2014. Health Status Monitoring Tool. CMS analyzes Medicare claims data to monitor health measures, including encounters with the health care system (such as hospitalizations, emergency room visits, and physician visits) and one outcome (death) for beneficiaries in both CBP-covered areas and non-CBP areas for both round 2 and the national mail-order program. CMS posts quarterly reports on its website to show historical and regional trends in health measures for specific groups of beneficiaries. The number of beneficiaries receiving DME items covered under CBP round 2 generally decreased after the implementation of round 2, and these utilization decreases generally were larger than the decreases for items or areas that were not included in CBP. The number of beneficiaries receiving diabetes testing supplies covered under the national mail-order program also generally decreased after the implementation of the program, although there was an increase in the utilization of some items through retail outlets. The number of beneficiaries receiving at least one DME item included in CBP round 2 decreased after the implementation of round 2, and these decreases were larger than those for items or areas that were not included in CBP. That is, the decrease was largest among items and areas for which CMS established competitively set rates beginning in July 2013. Specifically, the percentage decrease in the number of beneficiaries receiving at least one round 2 DME item between 2012 and 2014 was 17 percent. In contrast, the decreases in utilization were 6 to 7 percent for the same items in non-CBP areas and for non-CBP items. (See fig. 2.) Numerous factors could have contributed to the decrease in the number of beneficiaries receiving DME, and the decreases do not necessarily indicate that beneficiaries did not receive needed DME. Some stakeholders have expressed concern that lower payment rates and a smaller number of suppliers may have caused some beneficiaries to not receive needed DME. However, CMS stated that CBP has helped limit fraud and abuse and may have curbed unnecessary utilization of some CBP-covered items in competitive bidding areas. Additionally, in recent years CMS began implementing several broader antifraud efforts that were not limited to CBP-covered items or areas, such as taking additional steps to identify aberrant or suspicious billing patterns among all Medicare FFS claims before making payments, and implementing new safeguards to better screen existing and new Medicare suppliers. Looking at each of the eight product categories individually, the number of beneficiaries receiving covered items generally decreased after implementation of round 2, and these decreases were generally larger than those for the same items in non-CBP areas. Specifically, utilization of seven of the eight round 2 product categories decreased between 2012 and 2014, with decreases that were larger than in non-CBP areas for six of these seven categories. While there was substantial variation in the magnitude of the decrease in beneficiaries receiving these items, the percentage decreases in the number of beneficiaries receiving items and in the number of items received were generally similar, indicating that the average number of items beneficiaries received was relatively constant. Continuous positive airway pressure (CPAP) was the only product category for which the number of beneficiaries receiving items increased between 2012 and 2014, as well as the category with the largest difference between the percentage change in beneficiaries and in items. (See fig. 3.) Similarly, almost all round 2 areas experienced a decrease in the number of beneficiaries receiving CBP round 2 items after the implementation of round 2, though the magnitude of the decrease varied. Specifically, 95 of the 100 round 2 areas experienced a decrease between 2012 and 2014 in the number of beneficiaries receiving CBP round 2 items, ranging from 2 to 42 percent. Of the 15 areas with the largest percentage decreases in the number of beneficiaries receiving these items, 12 were in California and Texas. Nine of these 12 areas also had the largest relative decreases in the number of beneficiaries receiving round 2 items compared with the number receiving non-CBP-covered DME items. CMS officials told us that the relatively large decreases in California and Texas attributed to the implementation of CBP round 2 were likely because these states historically had high rates of potential fraud and abuse. (See fig. 4.) The total number of beneficiaries receiving at least one diabetes testing supply item covered under the national mail-order program decreased after implementation of the program, although there was an increase in the number of beneficiaries receiving some of these items through retail outlets. Specifically, there was an overall 12 percent decrease between 2012 and 2014 in the number of beneficiaries receiving at least one diabetes testing supply item through any acquisition method. However, during the same time period there was a 39 percent decrease in the number of beneficiaries receiving these items through mail order and a 13 percent increase in the number of beneficiaries receiving these items through retail outlets such as stores or pharmacies. The net result was a switch from a majority of beneficiaries (57 percent) receiving these supplies through mail order in 2012 to a majority receiving supplies through retail in 2014 (63 percent). (See fig. 5.) These utilization results were driven by changes in the two most commonly received items covered under the national mail-order program: diabetes test strips and lancets. For the other six covered diabetes testing supplies, the number of beneficiaries receiving these items through retail outlets decreased. Although CMS officials said that they could not speculate about reasons for this switch, a diabetes advocacy group we interviewed speculated that beneficiaries may have decided to switch to retail because they had difficulty finding contract suppliers that would provide the specific brand of test strips they requested. The use of retail outlets could result in higher out-of-pocket payments for beneficiaries because unlike mail-order contract suppliers, retail outlets are not required to accept assignment for diabetes testing supplies covered under the national mail-order program. Based on its monitoring of health measures, CMS reported that the implementation of the CBP has not resulted in widespread beneficiary access issues. The number of round 2 and national mail-order program inquiries and complaints to 1-800-MEDICARE generally decreased throughout the first 2 years of the programs’ implementation. CMS told us that it investigates complaints using secret shopping calls and that its post-implementation beneficiary satisfaction surveys remained positive. Nevertheless, several stakeholder groups we interviewed reported specific concerns such as delays in delivery of CBP-covered DME items and beneficiaries having difficulty locating a contract supplier. CMS reported that its health status monitoring tool showed similar trends in health measures in both CBP and non-CBP areas, both before and after the implementation of round 2. CMS has reported no changes in health measures attributable to the implementation of any CBP round since the agency began monitoring measures in 2011, which CMS officials told us is an indication that the CBP has not caused widespread problems with beneficiary access. Officials told us that CMS investigated aberrant trends in health measures identified by the monitoring tool and concluded that all such trends were the result of issues unrelated to CBP. For example, CMS officials told us that they investigated an increased mortality rate in one competitive bidding area and concluded it was due to an influenza outbreak. Similarly, they found that other issues resulted from changes in DME policy not specific to CBP, such as new requirements regarding prior authorization for wheelchairs. CMS’s health status monitoring tool uses Medicare claims data to track seven health measures—deaths, hospitalizations, emergency room visits, physician visits, admissions to skilled nursing facilities, average number of days spent hospitalized in a month, and average number of days in a skilled nursing facility in a month. CMS monitors these health measures for three groups of Medicare FFS beneficiaries residing in both CBP and non-CBP areas: (1) all beneficiaries enrolled in FFS, (2) beneficiaries likely to use one of the competitively bid products on the basis of related health conditions, and (3) beneficiaries for whom Medicare has paid a claim for one of the competitively bid products. CMS’s tool considers historical and regional trends in health status to monitor health measures in all CBP and non-CBP areas. CMS publishes aggregated results for four U.S. geographic regions quarterly on its website. CMS uses the tool to review Medicare claims data on a bi-weekly basis. According to a monitoring tool user guide that CMS provided us, the tool uses a statistical scoring algorithm to identify potential changes in health measures in every competition. The goal of the scoring algorithm is to identify persistent aberrant trends in health measures in individual competitive bidding area and product category competitions that are not mirrored in other regions. These aberrant trends can then be investigated further, through steps such as beneficiary and supplier outreach and contract compliance reviews. Based on our analysis, CMS’s methodologies and scoring algorithm used to evaluate health measure trends among CBP areas appear to be sound. However, we did not examine individual investigations that CMS conducted to assess aberrant changes in trends in particular competitive bidding areas and product categories and whether these trends could be attributed to CBP. The number of inquiries and complaints to 1-800-MEDICARE regarding CBP round 2 or the national mail-order program represented less than 1 percent of all calls to 1-800-MEDICARE, and inquiries and complaints generally decreased throughout the programs’ first 2 years of implementation. During the first quarter of implementation (July 1, 2013, to September 30, 2013), 1-800-MEDICARE received almost 100,000 inquiries—including almost 300 complaints—related to CBP round 2 or the national mail-order program. However, the number of inquiries and complaints dropped sharply during the second quarter of implementation, and then generally continued to decrease slowly or remained relatively consistent over subsequent quarters. (See fig. 6.) After implementation of the CBP, CMS began to track CBP-related inquiries to 1-800-MEDICARE and to classify them as either general CBP inquiries, such as requests for general information, or assistance determining whether a beneficiary resided in a competitive bidding area and whether the inquiries were about a specific product category. The highest number of general CBP inquiries occurred in the first quarter after implementation—61,457—which decreased to 8,741 by the last quarter of the 2-year period. The number of category-specific inquiries for each product category generally decreased from the first quarter to the last quarter. The mail-order diabetes testing supplies product category had the highest number of specific inquiries during the first quarter—15,953— which fell to 1,610 inquiries in the last quarter. The standard wheelchairs product category had the next-highest number of inquiries during the first quarter—6,438—which fell to 2,967 inquiries in the last quarter. The total number of complaints was small compared to the total number of inquiries. However, the number of complaints may not fully capture the number of people who expressed problems or dissatisfaction with the CBP program, because CMS defines a complaint as an inquiry to 1-800- MEDICARE that needs to be referred to an outside entity for resolution, such as Palmetto GBA. The 1,156 complaints recorded in the first 2 years occurred across multiple round 2 product categories, rather than being concentrated in a few specific categories. The complaints varied, but included complaints that contract suppliers refused to serve beneficiaries residing in their competitive bidding areas or had provided poor customer service, that beneficiaries had experienced delays in receiving DME or had received the wrong DME, and that beneficiaries had difficulty obtaining specific brands and models of DME items that had been prescribed by their physicians. For example, Palmetto GBA received several complaints indicating that contract suppliers refused to provide the U-Step walker, a specific brand of walker, furnished under a specific HCPCS code. According to CMS, U-Step walkers are infrequently prescribed, and Palmetto GBA worked with beneficiaries on a case-by- case basis to obtain them from contract suppliers. CMS told us that in every instance in which Palmetto GBA intervened, Palmetto GBA was able to locate a contract supplier to provide the item for each beneficiary with a prescription for the U-Step walker. CMS officials told us that the agency monitors contract suppliers to ensure that they are complying with the terms of their contracts, such as servicing all beneficiaries that reside in their competitive bidding areas and furnishing the same items to Medicare beneficiaries that they make available to other customers. Officials told us that when CMS receives complaints regarding contract suppliers’ noncompliance with contract terms, CMS may contact the supplier and/or beneficiary to obtain information necessary for investigation and may also work with other CMS contractors or deploy secret shopper calls, if deemed appropriate to do so. CMS told us that the reason for the secret shopping call helps the agency determine the strategy chosen—that is, whether it is appropriate to conduct the secret shopping calls with one or multiple supplier locations, or with multiple contract suppliers for particular competitive bidding areas or product category competitions. According to CMS, Palmetto GBA may conduct dozens or hundreds of focused secret shopping calls to investigate a single complaint to thoroughly investigate an allegation or verify contract supplier compliance. CMS told us that during secret shopping calls, Palmetto GBA representatives may pose as referral agents on behalf of beneficiaries and request items, such as specific diabetes testing supplies from contract suppliers, to determine whether the suppliers offer the supplies they claim to furnish. According to CMS, if those calls find that the contract supplier refused to provide DME items according to the terms of its CBP contract, Palmetto GBA describes the results of the call to the supplier’s representative and sends an educational letter reiterating contract obligations. Officials said that for a contract supplier who remains noncompliant, CMS sends a letter terminating the supplier’s CBP contract, and suppliers can submit a corrective action plan, accept the termination, or request a hearing. According to CMS, between July 2013 and June 2015, Palmetto GBA made a total of 3,953 focused secret shopping calls related to round 2 complaints and a total of 254 focused secret shopping calls related to the national mail-order program. CMS officials told us that as a result of those secret shopping calls, the agency issued 43 termination notices to contract suppliers during this time frame. Of those, 37 contract suppliers came into compliance and 6 suppliers’ contracts were terminated. CMS reported that its pre- and post-CBP implementation beneficiary satisfaction surveys found that the majority of respondents rated their experiences positively in both time periods. For both the pre- and post- implementation surveys, CMS obtained telephone responses from a random sample of approximately 400 beneficiaries in each of the 100 competitive bidding areas who received at least one CBP-covered DME item. The surveys asked beneficiaries to record their satisfaction ratings on a five-point scale for six questions: the beneficiary’s initial interaction with DME suppliers, the training received regarding the DME item, the delivery of the DME item, the quality of the item provided by the supplier, the customer service provided by the supplier, and the supplier’s overall complaint handling. In both the pre- and post-implementation surveys, a majority of the beneficiaries who responded to each of the six questions rated their experiences as either “good” or “very good,” although there were slightly fewer positive responses in the post-implementation survey. The percentage of positive responses after implementation decreased for each question, ranging between 85 and 92 percent for pre- implementation questions, and between 84 and 89 percent for the same questions post-implementation. (See fig. 7.) CMS also conducted pre- and post-implementation surveys for the national mail-order program using the same methodology. Those surveys also found that post-implementation satisfaction levels remained high. CMS obtained responses from a random sample of 2,086 beneficiaries for the 2013 pre-implementation survey and 2,000 for the 2014 post- implementation survey. For each of the six questions in both surveys, results indicated that between 82 and 92 percent of beneficiaries surveyed rated their experiences as either “good” or “very good,” although the exact percentage decreased slightly for some questions and increased slightly for others. (See fig. 8.) As we reported previously, CMS’s survey design has some limitations. For example, the survey design did not capture responses from beneficiaries who may have needed, but did not obtain, DME during the period. That is, if a beneficiary did not receive a DME item, the beneficiary’s response regarding his or her potential access issue to the DME item would not be included in the survey. Additionally, the distribution of beneficiaries who received items across the different product categories could have changed between the pre- and post- implementation surveys, and the results were not analyzed separately by product category. The stakeholders we interviewed reported varied experiences with CBP and the national mail-order program. Some stakeholders reported that they had no beneficiary access issues following the implementation of round 2 and the national mail-order program and others reported numerous beneficiary access concerns. In general, stakeholders from one of the five beneficiary advocacy groups and one of four state hospital associations (California, Florida, Illinois, and New York) reported no or very few specific concerns with CBP beneficiary access. However, stakeholders from four beneficiary advocacy groups reported that their members experienced several access issues with the implementation of CBP round 2 and the national mail-order program. These issues included delays in delivery of CBP-covered DME items, such as walkers, and trouble locating contract suppliers to provide specific DME items, such as liquid oxygen and specific brands of diabetes testing strips, or to repair wheelchairs. In addition, discharge planners and other stakeholders from three state hospital associations reported that beneficiaries also experienced delays in delivery of CBP-covered items, such as walkers and wheelchairs, and had difficulty locating contract suppliers to provide oxygen or service DME items when the beneficiaries were visiting in a competitive bidding area but resided elsewhere. Discharge planners and referral agents from the Florida and California hospital associations told us that the delays in delivery of needed DME resulted in an increase in the length of hospital stays for some beneficiaries. For example, individuals we spoke with from the Florida Hospital Association told us that prior to CBP, DME suppliers typically delivered DME within 24 hours of the request, 7 days a week. However, since CBP round 2 was implemented, contract suppliers in Florida’s competitive bidding areas no longer delivered on weekends and also delivered only during certain hours Monday through Friday. The planners and agents from the Florida Hospital Association told us that delays in delivery can extend the length of beneficiaries’ hospital stays when the DME is necessary prior to discharge, such as DME for orthopedic-related injuries. In some cases, they told us that the hospital loans or provides certain DME, such as walkers, in order to discharge beneficiaries on time. As previously described, CMS’s health status monitoring tool did not indicate changes in the average length of hospital stays resulting from the implementation of CBP. We conducted a limited analysis to compare the length of hospital stays in 2012 (pre-CBP) and 2014 (post-CBP) for beneficiaries who had a hip or knee replacement, both nationally and in California and Florida. We found that the average length of stay was slightly lower in 2014 than in 2012, both for the two selected states and nationally. This was true for both beneficiaries who received DME within a day of discharge and those who did not, within as well as outside competitive bidding areas. (See fig. 9.) This limited analysis did not indicate an increase in the average length of stay among beneficiaries as a result of the implementation of CBP, although we did not expand our analysis to determine if results were consistent across other reasons for hospitalization. Most round 2 and national mail-order program competitions had at least five active contract suppliers in 2014, but other competitions had just one or a few active suppliers. In addition, some competitions had a single contract supplier that accounted for the majority of the market share. Eleven percent of contract suppliers were inactive during 2014 in all the competitions in which they were awarded and accepted a contract, and about half of contract suppliers were inactive in at least one of the competitions in their contract. CMS told us that it monitors the contract supplier market and has its contractor conduct secret shopping calls of inactive suppliers on a quarterly basis to confirm that the suppliers are not in breach of their contracts. A large majority of the possible 801 individual competitions in which a contract supplier could be awarded a contract had at least five contract suppliers that were active in 2014, which we defined as having furnished at least one covered item during 2014 in at least one of the competitive bidding areas and product category competitions in which they were awarded and accepted a contract. We found that 84 percent of the competitions had at least five active contract suppliers; however, 11 percent of competitions had three or fewer active suppliers and 1 percent had just one active supplier. (See table 2.) During the bid evaluation and contract award processes, CMS tries to ensure beneficiary access and choice by awarding contracts to at least five contract suppliers in each competition. To ensure that there is sufficient capacity to satisfy beneficiary demand, CMS considers a competition with only one qualified supplier to be nonviable. According to CMS officials, once contracts are awarded, contract suppliers are required to furnish items and services upon request, but are not otherwise required to furnish a certain amount of DME items as part of their contract obligations. The number of active contract suppliers across the 800 round 2 competitions varied substantially by product category. The round 2 product category with the largest number of active contract suppliers was oxygen, in which 352 suppliers accounted for over $362 million in charges during 2014. Furthermore, the majority (86) of the 100 competitions for the oxygen product category had at least 11 active suppliers. In contrast, the product category with the fewest active contract suppliers was negative pressure wound therapy (NPWT), with 60 suppliers that accounted for over $48 million in charges. Over half (69) of the 100 competitions for NPWT had 4 or fewer active contract suppliers and 4 had a single active supplier. For the national mail-order program, all 19 contract suppliers were active in 2014. While multiple suppliers had substantial shares of the market for most competitions, in some competitions a single supplier had a majority. In 72 percent of the 801 competitions, no contract supplier had more than half of the market. However, in 14 percent (113) of the competitions, a single contract supplier had at least three quarters of the market, and in 6 percent (48) of the competitions, one contract supplier had 90 percent or more of the market. (See table 3.) Market concentration varied substantially by product category. The round 2 product category with the least market concentration was wheelchairs, for which the top contract supplier had less than 25 percent of the market in 43 of the 100 total competitions. One reason for this was the relatively high market share of non-contract suppliers, as physicians and hospitals are allowed to provide folding manual wheelchairs to their patients directly in certain circumstances. In addition, other suppliers chose to be grandfathered and continued furnishing capped rental wheelchairs to beneficiaries who were their customers when CBP round 2 began. For example, we found that non-contract suppliers had at least 25 percent of the market share in 56 of the 100 total wheelchair competitions. In contrast, the product category with the largest market concentration was NPWT, for which the top contract supplier had at least 90 percent of the market in 47 of the 100 competitions. NPWT was also the only round 2 product category for which non-contract suppliers accounted for less than 10 percent of market share in all 100 competitions. In addition, some contract suppliers had a substantial proportion of the total round 2 and national mail-order program market—which was over $1 billion in 2014—as well as even higher market share for individual product categories. For example, the contract supplier with the highest market share based on charges for CBP-covered items, Lincare, had $157 million in charges, or 13 percent of the round 2 and mail-order program market in 2014. It also had 25 percent of the oxygen market and the highest market share in three other product categories (CPAP, hospital beds, and walkers). Arriva, the contract supplier with the second highest overall market share, had $85 million in charges, or 7 percent of total 2014 charges. However, these charges were for mail-order diabetes testing supplies only; Arriva had 50 percent of the $170 million national mail-order program market. Apria, the contract supplier with the third highest overall market share, had $68 million in charges, or 6 percent of the total market, and accounted for over 10 percent of the CPAP and oxygen markets. KCI, with the fourth highest market share, was only active in the NPWT category, and had $42 million in charges during 2014, which accounted for 86 percent of the NPWT market. Eleven percent of contract suppliers (94 of 834) were inactive during 2014 in all of the competitions in their contract—that is, they did not furnish any CBP-covered items during 2014 in any of the competitive bidding areas and product categories in which they were awarded and accepted a contract. The completely inactive suppliers varied in their categories, areas, and duration of participation in CBP. The 94 completely inactive suppliers’ contracts included over half of all competitions (446 of 801), and spanned all 8 round 2 product categories and 99 of the 100 round 2 areas. In addition, about half of the 94 completely inactive contract suppliers stopped participating in CBP prior to the end of 2014—19 were terminated by CMS and 23 closed or otherwise voluntarily withdrew from the CBP. However, the majority of completely inactive contract suppliers had higher composite bids than other contract suppliers. Specifically, 82 of the 94 completely inactive contract suppliers’ bids (estimates of the price at which they could profitably provide those items) were above the median bid of all suppliers in the same competition, on average, and for 56 of the contract suppliers this difference was more than 10 percent. In addition, 61 of the 94 completely inactive contract suppliers were designated as small bidders by CMS, and 59 had only one or two competitions in their contract. In 2014, an additional 39 percent of contract suppliers (324 suppliers) were inactive in at least one of the competitions in their contract, and 3 percent (28) had less than $1,000 in total Medicare charges for CBP- covered items. These partially inactive suppliers included contract suppliers that were very active in other markets. For example, Lincare, the supplier with the largest overall market share, was inactive in 144 of the 689 competitions in its contract. Similar to the completely inactive suppliers, the 28 barely active contract suppliers’ contracts spanned all eight round 2 product categories and 96 of the 100 round 2 areas, and 24 of the 28 barely active contract suppliers bid above the median bid, on average. To help ensure beneficiary access and choice, CMS officials said that CMS monitors all competitions, but applies greater focus to those where there is just one or a few active contract suppliers and assesses whether there is a need for CMS to award subsequent contract offers to existing contract suppliers. CMS officials told us that the agency does not believe that having competitions with just one or a few active contract suppliers has decreased beneficiary access and choice because its routine monitoring of beneficiary access has not identified access issues. CMS also told us that the agency monitors suppliers that are inactive by having Palmetto GBA conduct secret shopping calls quarterly. According to CMS, most of the secret shopping calls to contract suppliers that were inactive resulted in finding that the suppliers were not in breach of contract, and in some instances, they reported that they had not received requests from beneficiaries or referral agents for CBP-covered items. However, CMS reported that Palmetto GBA found that eight inactive contract suppliers were in breach of their contracts because they were not providing items that had been requested. Of those, three contract suppliers were brought into compliance and the other five contract suppliers’ contracts were terminated. In addition, CMS officials told us that the agency also works with the Competitive Acquisition Ombudsman and uses local competitive bidding liaisons to provide an on-the-ground physical presence and investigate and address any potential issues. We provided a draft of this product to HHS for comment. HHS provided technical comments, which were incorporated as appropriate. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies of this report to the Secretary of Health and Human Services and appropriate congressional committees. The report will also be available at no charge on our website at http://www.gao.gov. If you or your staffs have any questions about this report, please contact me at (202) 512-7114 or kingk@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix III. Appendix I: The 100 Competitive Bidding Areas Included in the Centers for Medicare & Medicaid Services’ Competitive Bidding Program Round 2 by specific ZIP codes related to a metropolitan statistical area, and they may be the same size as, larger than, or smaller than the related metropolitan statistical area, depending on a variety of considerations. The competitive bidding area is the area wherein only contract suppliers may furnish competitively bid items to beneficiaries unless an exception is permitted by law. Metropolitan statistical areas with populations over 8 million may be subdivided into multiple competitive bidding areas. Most round 2 metropolitan statistical areas have only one competitive bidding area. However, the three largest metropolitan statistical areas (Chicago, Los Angeles, and New York) are subdivided into multiple competitive bidding areas, so there are a total of 100 competitive bidding areas. Since the durable medical equipment (DME) competitive bidding program (CBP) was implemented in 2008, the Centers for Medicare & Medicaid Services (CMS) has phased in several additional CBP rounds and programs. Round 1. CMS awarded contracts to 329 contract suppliers—208 of which were designated by CMS as small suppliers—to furnish DME items and services in 10 product categories in 10 competitive bidding areas. According to CMS, the round 1 competitive bidding areas were selected, in part, because they may have had prior instances of unnecessary DME utilization. Contracts were intended to be effective for a 3-year period, July 1, 2008, through June 30, 2011; however, round 1 was stopped on July 15, 2008, through the enactment of the Medicare Improvements for Patients and Providers Act of 2008 (MIPPA). Among other provisions, MIPPA terminated the contracts already awarded by CMS to suppliers in round 1 and required CMS to repeat the competition for round 1—referred to as the CBP round 1 rebid. Round 1 rebid. CMS awarded contracts to 356 contract suppliers—219 of which were small suppliers—to furnish DME items and services in nine product categories in nine competitive bidding areas. Contracts were effective January 1, 2011, and expired after 3 years on December 31, 2013, except for the contracts for the mail-order diabetes testing supplies product category, which expired on December 31, 2012. Round 1 recompete. CMS awarded contracts to 282 contract suppliers— 163 of which were small suppliers—to furnish DME items and services in six product categories and nine competitive bidding areas. The recompete contracts were effective January 1, 2014, and expire December 31, 2016. Round 1 2017. Round 1 2017 will include seven product categories, with the same items as were bid in the round 1 recompete except for the external infusion pumps and supplies product category, which is not included in round 1 2017. Round 1 2017 will be implemented in the same 9 competitive bidding areas as the round 1 rebid and round 1 recompete, but because competitive bidding areas that included more than one state have been redefined so that there are no longer any multistate areas, the total number of competitive bidding areas is 13. These contracts become effective January 1, 2017, and expire December 31, 2018. Round 2 and national mail-order program. CMS awarded contracts to 822 contract suppliers—520 of which were small suppliers—to furnish DME items and services in eight product categories and 100 competitive bidding areas. CMS also awarded 19 contracts—6 of which were to small suppliers—to furnish mail-order diabetes testing supplies included in the national mail-order program. The round 2 and national mail-order program contracts were effective July 1, 2013, and expired June 30, 2016. Round 2 recompete and national mail-order program recompete. According to CMS officials, the agency awarded contracts to 586 contract suppliers—334 of which were small suppliers—to furnish DME items for seven product categories with most of the same items bid in other rounds, although CMS grouped certain items into different product categories. Round 2 recompete will be implemented in the same 100 competitive bidding areas as round 2, but because competitive bidding areas that included more than one state have been redefined so that there are no longer any multistate areas, the total number of competitive bidding areas is 117. CMS officials told us that the agency also awarded contracts to 9 contract suppliers—2 of which were small suppliers—to furnish the same diabetes testing supplies in the same areas that were included in the first national mail-order program. The round 2 recompete and national mail- order program recompete contracts were effective July 1, 2016, and expire December 31, 2018. Kathleen M. King, (202) 512-7114 or kingk@gao.gov. In addition to the contact named above, Martin T. Gahart, Assistant Director; Michelle Paluga, Analyst-in-Charge; Todd Anderson; Alison Binkowski; Elizabeth T. Morrison; and Emily Wilson made key contributions to this report. Medicare: Bidding Results from CMS’s Durable Medical Equipment Competitive Bidding Program. GAO-15-63. Washington, D.C.: November 7, 2014. Medicare: Second Year Update for CMS’s Durable Medical Equipment Competitive Bidding Program Round 1 Rebid. GAO-14-156. Washington, D.C.: March 7, 2014. Medicare: Review of the First Year of CMS’s Durable Medical Equipment Competitive Bidding Program’s Round 1 Rebid. GAO-12-693. Washington, D.C.: May 9, 2012. Medicare: The First Year of the Durable Medical Equipment Competitive Bidding Program Round 1 Rebid. GAO-12-733T. Washington, D.C.: May 9, 2012. Medicare: Issues for Manufacturer-level Competitive Bidding for Durable Medical Equipment. GAO-11-337R. Washington, D.C.: May 31, 2011. Medicare: CMS Has Addressed Some Implementation Problems from Round 1 of the Durable Medical Equipment Competitive Bidding Program for the Round 1 Rebid. GAO-10-1057T. Washington, D.C.: September 15, 2010. Medicare: CMS Working to Address Problems from Round 1 of the Durable Medical Equipment Competitive Bidding Program. GAO-10-27. Washington, D.C.: November 6, 2009. Medicare: Covert Testing Exposes Weaknesses in the Durable Medical Equipment Supplier Screening Process. GAO-08-955. Washington, D.C.: July 3, 2008. Medicare: Competitive Bidding for Medical Equipment and Supplies Could Reduce Program Payments, but Adequate Oversight Is Critical. GAO-08-767T. Washington, D.C.: May 6, 2008. Medicare: Improvements Needed to Address Improper Payments for Medical Equipment and Supplies. GAO-07-59. Washington, D.C.: January 31, 2007. Medicare Payment: CMS Methodology Adequate to Estimate National Error Rate. GAO-06-300. Washington, D.C.: March 24, 2006. Medicare Durable Medical Equipment: Class III Devices Do Not Warrant a Distinct Annual Payment Update. GAO-06-62. Washington, D.C.: March 1, 2006. Medicare: More Effective Screening and Stronger Enrollment Standards Needed for Medical Equipment Suppliers. GAO-05-656. Washington, D.C.: September 22, 2005. Medicare: CMS’s Program Safeguards Did Not Deter Growth in Spending for Power Wheelchairs. GAO-05-43. Washington, D.C.: November 17, 2004. Medicare: Past Experience Can Guide Future Competitive Bidding for Medical Equipment and Supplies. GAO-04-765. Washington, D.C.: September 7, 2004. | To achieve Medicare savings for DME, the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 required that CMS implement the CBP for certain DME, such as wheelchairs and oxygen, in phases, or rounds. Round 1 started in 2008, and round 2 and the national mail-order program started in 2013. CMS estimated that the first 2 years of round 2 and the national mail-order program saved Medicare approximately $3.6 billion. GAO has reported on several prior CBP rounds. GAO was asked to continue to review the implementation of the CBP. In this report, GAO examines the extent to which round 2 and the national mail-order program have affected (1) utilization of CBP-covered DME items, and (2) beneficiaries' access to DME items. This report also (3) describes the number and market shares of the round 2 and mail-order program suppliers. To examine the effect of CBP on utilization, GAO used Medicare DME claims data from 2012 and 2014—the year before and the year after implementation of round 2—to compare the number of beneficiaries who received CBP-covered DME items. To examine the effect of CBP on beneficiary access, GAO reviewed information about CMS's efforts to monitor the effects of the CBP, and interviewed selected Medicare beneficiary organizations and state hospital associations. To describe the supplier markets, GAO analyzed 2014 Medicare claims data, the latest year with complete available data when GAO began this engagement. The number of beneficiaries receiving durable medical equipment (DME) items covered under the competitive bidding program (CBP) generally decreased after implementation of two CBP phases that began July 1, 2013—round 2 and the national mail-order program for diabetes testing supplies. Under the CBP, (administered by the Centers for Medicare & Medicaid Services (CMS)), only competitively selected contract suppliers can furnish certain DME items at competitively determined prices to beneficiaries in designated competitive bidding areas. From the year before (2012) to the year after (2014) implementation, the number of beneficiaries receiving covered items in round 2 areas decreased 17 percent, compared with a 6 percent decrease for beneficiaries in non-CBP areas. The number of beneficiaries that received diabetes testing supplies through the national mail-order program also decreased 39 percent between 2012 and 2014, with a corresponding 13 percent increase in the number of beneficiaries receiving these items through retail outlets. CMS officials stated that CBP has helped limit fraud and abuse and may have curbed unnecessary utilization of some CBP-covered items in competitive bidding areas. CMS reports that available evidence from the agency's monitoring efforts indicates that the implementation of round 2 and the national mail-order program have had no widespread effects on beneficiary access. In particular, CMS has reported that its health status monitoring tool has not detected any changes in health measures attributable to the CBP, and the results of its 2014 post-CBP beneficiary satisfaction surveys remained positive. In addition, the number of CBP inquiries and complaints generally decreased throughout the first 2 years of round 2 and the national mail-order program. CMS officials told GAO that CMS took measures to ensure that contract suppliers met their contract obligations, such as investigating complaints using secret shopping calls, and terminating contracts of suppliers that remained noncompliant after receiving targeted education. However, some beneficiary advocacy groups and state hospital associations reported specific access issues, such as difficulty locating contract suppliers that will furnish certain items and delays in delivery of DME items. Round 2 and the national mail-order program included 801 separate competitive bidding area and product category competitions. Most of these competitions had at least five active contract suppliers in 2014. However, 11 percent of the competitions had three or fewer active contract suppliers and 1 percent had just one active contract supplier. In addition, while multiple suppliers had substantial shares of the market for most competitions, in some competitions a single supplier had a majority. For example, in 6 percent of the competitions, one contract supplier had at least 90 percent of the market. Conversely, 11 percent of contract suppliers did not furnish any CBP-covered items for any competitions in their contract. CMS officials told GAO that CMS monitors these suppliers to help ensure that they are meeting their contractual obligations, such as being willing to service all beneficiaries in their areas and to furnish the same items to Medicare beneficiaries that they make available to other customers. The Department of Health and Human Services provided technical comments on a draft of this report, which were incorporated as appropriate. |
Regulation B imposes a general prohibition on collecting data on personal characteristics for nonmortgage loan applicants. But in 2003, FRB expanded its exceptions to this prohibition to include permitting lenders to collect data on race, gender, and other personal characteristics in connection with a self-test for the purpose of determining the effectiveness of the lender’s compliance with ECOA and Regulation B. A self-test is any program, practice, or study that is designed and used by creditors to determine the effectiveness of the creditor’s compliance with ECOA and Regulation B. The results of a self-test are privileged—that is, they cannot be obtained by any government agency in an examination or investigation in any lawsuit alleging a violation of ECOA. Although Regulation B prohibits creditors, except in limited circumstances such as conducting a self-test, from collecting data on personal characteristics for nonmortgage loan applicants, creditors are required to collect such data for mortgage loan applicants. Specifically, HMDA, as amended in 1989, requires certain financial institutions to collect and publicly report information on the racial characteristics, gender, and income level of mortgage loan applicants. In 2002, FRB, pursuant to its regulatory authority under HMDA, required financial institutions to report certain mortgage loan pricing data in response to concerns that minority and other targeted groups were being charged excessively high interest rates for mortgage loans. Authority for enforcing compliance with ECOA with respect to depository institutions, such as Federal Reserve System member banks, national banks, state-chartered banks, saving associations, and credit unions, lies with the five federal regulators—FRB, the Office of the Comptroller of the Currency (OCC), the Federal Deposit Insurance Corporation (FDIC), the Office of Thrift Supervision (OTS), and the National Credit Union Administration (NCUA). To carry out their responsibilities, the agencies may conduct periodic compliance examinations of depository institutions. These compliance exams generally assess depository institutions’ loan underwriting guidelines and credit decisions to detect possible discrimination in both mortgage and nonmortgage lending. FRB’s Survey of Small Business Finances (SSBF) is one of the principal sources of information available on the factors that affect the availability of credit for small businesses. FRB has conducted the SSBF about every 5 years from 1987 through 2003 from a nationwide sample of small businesses of varying sizes, locations, and ownership characteristics. In 2007, FRB decided to discontinue the SSBF due to its cost and other considerations. However, according to FRB officials, FRB plans to include elements of the SSBF in another survey, the Survey of Consumer Finances (SCF), starting in 2010. The limited number of studies on nonmortgage lending that met our criteria for selection in our June report focused primarily on the small business sector, and suggested that certain minority-owned businesses may be denied loans more often or be offered higher interest rates than similar white-owned businesses. However, the key data source for most of these studies, FRB’s SSBF, has certain limitations compared with HMDA data, and this may limit the data’s usefulness as an analytical tool. The few studies we identified that addressed possible discrimination in automobile and credit card lending relied on SCF data, which has certain limitations similar to those of the SSBF data. Further, our report found that data limitations may also impede the relative efficiency of the bank regulators’ fair lending examination process for the nonmortgage sector as compared with the mortgage sector. Primarily using data obtained from FRB’s SSBF, all eight studies we identified on minority business lending generally found that lenders denied loans to minority-owned businesses (seven of the eight specifically refer to African-American-owned businesses) or required them to pay higher interest rates for loans significantly more often than white-owned small businesses. This finding generally remained consistent after considering a variety of risk factors, such as borrower creditworthiness, industry sector, and other firm characteristics (e.g., business location, assets, and profits). In addition, studies have found that Hispanic-owned businesses were denied credit or charged higher interest rates more often when compared with white-owned businesses with similar risk characteristics. On the other hand, some studies we reviewed did not identify evidence that women-owned businesses face credit denials or higher rates significantly more often than male, white-owned businesses. While studies using SSBF data have provided important insights into possible discrimination in small business lending, researchers and FRB officials also pointed out a number of limitations: SSBF data are collected from individual small business borrowers rather than lenders, which limit their analytical value. For example, SSBF data do not allow researchers to assess the overall small business lending underwriting standards or lenders’ performance by type of institution, by size, or by geographic or metropolitan region. SSBF survey data are self-reported and are not verified by FRB. For example, FRB relies upon survey respondents to accurately report their race, gender, and other characteristics, as well as requested information on their business and their financing. Since the survey may be conducted long after the survey respondent applied for credit, the timing of the SSBF increases the risk that respondents may not accurately recall and report information from the time when the credit decision was made. FRB conducts the SSBF about every 5 years rather than annually and, therefore, the survey results may not be timely. To illustrate, most of the studies that we reviewed were based on data that are about 10 years old from surveys conducted in 1993 and 1998. Researchers and FRB officials that we spoke with said it may also take FRB a significant period of time to review and process the SSBF data prior to releasing it to the public. In contrast, HMDA data offer certain advantages over SSBF data as a research tool to assess possible discrimination in mortgage lending. In particular, HMDA data are collected directly from a large and identified population of mortgage lenders on a consistent and annual basis. Researchers have used HMDA data to conduct analyses of possible discrimination by type of lending institution, size of the institution, and geographic or metropolitan area. FRB also requires that lenders help verify the HMDA data they report, such as applicant data on personal characteristics and the interest rates charged on certain types of mortgages. Despite these advantages, we noted that analyses of HMDA data as a basis for conducting research on possible discrimination in mortgage lending have been criticized for not including key loan underwriting variables, such as the borrowers’ credit scores or mortgages’ loan-to-value ratios. Some argue that such underwriting variables may account for many apparent discrepancies between minority and white mortgage borrowers. To compensate for the lack of underwriting variables in the HMDA data, several researchers have collected such data from proprietary sources and matched it with HMDA data. According to a study on auto lending, racial discrimination could play a role in differences between the treatment of minority and white borrowers. The study relied on data from FRB’s SCF, which asks a nationwide sample of about 4,500 U.S. consumers to provide detailed information on the finances of their families and on their relationships with financial institutions. Because SCF data is also collected from borrowers rather than lenders, like SSBF data, it cannot be used as a basis for assessing individual lenders’ lending practices or lending practices industrywide (i.e., by type of institution, size of institution, or geographic or metropolitan area). The two studies we identified that also relied on SCF data had mixed results with respect to possible discrimination in credit card lending. One study found that minorities were likely to pay higher interest rates on credit card debt than white credit cardholders even after considering the payment history and financial wealth of each group. Another study did not find that minority credit cardholders paid higher interest rates as compared with white credit cardholders after controlling for creditworthiness factors. These studies showed the strength of the SCF as a data source (e.g., the ability to consider data on personal characteristics and loan underwriting factors), as well as its limitations (e.g., the data are collected from borrowers rather than lenders). Representatives from the four federal bank regulatory agencies we contacted (FRB, OCC, FDIC, and OTS) said that the availability of HMDA data has facilitated the fair lending law examination process. In particular, agency staff said that the analysis of HMDA data provided insights into lenders that might be at high risk of engaging in potentially discriminatory practices in mortgage lending. While agency staff said that HMDA data were only a first start in the investigative process (because they must evaluate a range of underwriting criteria and practices that may help explain disparities in a lender’s mortgage lending patterns), HMDA data allowed them to prioritize their examination resources. We found that in the absence of similar race, gender, and other data on personal characteristics for nonmortgage loan applicants, examiners may rely on time-consuming and possibly unreliable techniques to assess lenders’ compliance with fair lending laws. Under the Interagency Fair Lending Examination Procedures, examiners can use established “surrogates” to make educated guesses as to the personal characteristics, such as race or gender, of nonmortgage loan applicants to help determine whether the lenders they regulate are complying with established laws and regulations in extending credit to minority and other individuals targeted for loan applicants. For example, examination guidance allows examiners, after consulting with their agency’s supervisory staff, to assume that an applicant is Hispanic based on the last name, female based on the first name, or likely to be an African-American based on the census tract of the address. While these techniques may help identify the racial or gender characteristics of loan applicants, they have potential for error (e.g., certain first names are gender neutral, and not all residents of a particular census tract may actually be African-American). As a result of the limitations of the data on personal characteristics for nonmortgage loan applicants, as well as regulatory guidance directing examiners to consider using surrogates, federal oversight of lenders’ fair lending law compliance in this area may be less efficient than it is for mortgage lending. According to a comment letter submitted by a Federal Reserve Bank to FRB as it considered amending Regulation B in 1999, its examiners were unable to conduct thorough fair lending examinations or review consumer complaints alleging discrimination for nonmortgage products due to the lack of available data. Moreover, our reviews of agency fair lending examination guidance and discussions with some agency staff (OCC, FDIC, and OTS) suggest that, due in part to HMDA data availability, agencies focus most of their resources on possible discrimination in mortgage lending rather than nonmortgage lending. We plan to further explore the issue of fair lending enforcement in future work, including the impact of potential data limitations on regulatory agencies’ oversight and enforcement of the fair lending laws for mortgage and nonmortgage lending. While some individuals we contacted generally agreed with FRB’s 2003 conclusion that permitting lenders to voluntarily collect data on personal characteristics for nonmortgage loan applicants could create some risk of discrimination, many other individuals we contacted expressed skepticism about this argument. Even so, a range of researchers, regulatory staff, and representatives from both consumer and banking groups we contacted generally concurred with FRB that voluntarily collected data might not be useful or reliable and that very few banks would choose to collect it. Consequently, the benefits of permitting lenders to voluntarily collect data on personal characteristics as a means for researchers, regulators, and others to better understand possible discrimination in nonmortgage lending would likely be limited. Some researchers, staff from a bank regulatory agency, and representatives from banking and business trade groups we contacted generally agreed with FRB that permitting voluntary data collection on personal characteristics could create a risk that the information would be used for discriminatory purposes. These officials told us that the best way to protect borrowers against discrimination is to minimize the availability of information to lenders about their personal characteristics. However, many other researchers, staff from some regulatory agencies, and officials from consumer groups expressed skepticism on this conclusion. First, a staff member from a regulatory agency, several researchers, and representatives from consumer groups said that, in certain cases, lenders were already aware of the race and gender or other information on personal characteristics of nonmortgage loan applicants. Therefore, simply collecting data on personal characteristics on applicants in such cases would not necessarily create a risk of discrimination. Other researchers and officials from banking institutions disagreed. They noted that, in some cases, lending decisions may be made by officials who do not interact directly with loan applicants. Second, lenders’ voluntary collection and use of data on personal characteristics for nonmortgage loan applicants, outside of the ECOA self- test privilege, would also be subject to varying degrees of regulatory scrutiny, which could serve to deter lenders from using such data for discriminatory purposes. Similarly, all lenders that chose to collect and use such data for discriminatory purposes would face the risk of public disclosure of such practices through litigation. Further, according to a variety of researchers and officials we contacted, as well as FRB documents we reviewed, there is no evidence that lenders have used HMDA data for discriminatory purposes. These officials generally attributed the transparency of the HMDA program, through regulatory reviews and public reporting requirements, as serving to help deter lenders from using the data to discriminate in mortgage lending. Finally, FRB could potentially have mitigated some of its concerns that voluntarily collected data could be used for discriminatory purposes by including, as part of its 1999 proposal, minimum procedures for the collection and use of such data. FRB established such procedures for federally regulated lenders that choose to conduct a self-test. These procedures include developing written policies describing the methodology for data collection and keeping data on personal characteristics separate from loan underwriting data that are used to make credit decisions. Imposing such minimum procedures and requirements for a voluntary program could serve to enhance regulators’ oversight of lenders’ data collection, processes, practices, and uses of the data, and further deter possibly discriminatory practices. Even so, many researchers, regulatory staff, and representatives from consumer groups and banking trade groups agreed with FRB’s conclusion that the reliability of voluntarily collected data may be limited in identifying possible discrimination in nonmortgage lending. In particular, they agreed with FRB that, due to potentially inconsistent data collection standards, it would be difficult to use voluntarily collected data to compare fair lending performance across different lenders. Additionally, there may be data inconsistency problems for any given lender that chooses to collect data on personal characteristics for nonmortgage loan applicants. For example, a lender could “cherry pick,” or collect racial, gender, and other data on personal characteristics on applicants only for certain loan products that they felt would reflect favorably on their fair lending practices and not collect data for other products. Just as FRB could potentially have mitigated some of its concerns about the possibility that lenders would use voluntarily collected data for discriminatory purposes by adopting minimum procedures, as mentioned previously, it could also potentially have considered adopting data collection standards. Such standards could have served to better ensure the consistency of the data and enabled regulators and others to use the data to assess individual lender performance and compare lending practices across different financial institutions. However, according to a senior FRB official, a researcher, and a bank industry trade association official, the imposition of such standards would have undermined the voluntary nature of the data collection proposal. For example, FRB could be required to conduct examinations to help ensure that federally regulated lenders were collecting the data in a manner consistent with any such standards. Moreover, the establishment of such data collection standards might also have further diminished lender interest in a voluntary program, which researchers, FRB officials, and others said was already limited due to the potential for increased regulatory and public scrutiny of their lending practices. According to bank regulators and banking trade groups, very few, if any, lenders choose to conduct self-tests out of concern that the results of such tests would be subject to regulatory review even though they are privileged. Finally, while some officials we contacted and documents we reviewed said that any data that was collected and potentially reported by lenders would provide important insights into nonmortgage lending practices that are not currently available, other researchers and researchers suggested that such data would be prone to substantial selection bias. That is, the data would likely be skewed by the possibility that only lenders with good fair lending compliance records would choose to collect such data. Consequently, although voluntarily collected data on personal characteristics could provide some benefits, it would not likely materially assist the capacity of researchers, regulators, and others to better understand possible discrimination in nonmortgage lending. In concept, a requirement that lenders collect and publicly report data on the personal characteristics of nonmortgage loan applicants, similar to HMDA requirements, could help address some of the existing data limitations that complicate efforts by researchers, federal bank regulators, and others to identify possible discrimination. However, mandatory data collection and reporting would impose some additional costs on the lending industry, although opinions differed on how burdensome these costs might be. While options exist to potentially mitigate some of these costs, such as limiting data collection and reporting to specific types of lending, these options also involve additional complexities and costs that must be considered. Required data collection and reporting for nonmortgage loan applicants, similar to HMDA’s requirements, could help address some of the existing limitations of available data and facilitate the efficiency of the fair lending examination process for nonmortgage lending. Such data would be more timely than SSBF data, and the implementation of data collection standards could help ensure its reliability. For example, researchers and financial regulators would be able to analyze the practices of specific lenders and compare practices across lenders, assessing lending practices by type, size, and location of the institutions, similar to analyses done currently with HMDA data. While such analyses would represent only the first step in determining whether or not particular lenders were engaging in discriminatory practices, they could potentially help regulators prioritize their examinations and better utilize existing staff and other resources. While it is not possible to quantify the potential costs associated with a reporting requirement, in part because the requirements could vary, banking organizations and banks that we contacted identified a variety of additional costs that lenders might face. These officials also said that they were concerned about such costs and that the additional expenses associated with data collection and reporting would, in part, be passed on to borrowers. According to the officials, most of the costs associated with a reporting requirement would involve developing the information technology necessary to capture and report the data, including system integration, software development, and employee training. Moreover, the officials said that, as with HMDA data, verifying, any reported data would also entail costs, including expenses associated with conducting internal audits. The regulatory agency responsible for assembling, verifying, and reporting the data to the public would also accrue costs for these activities. Some researchers and representatives from consumer groups we contacted said that they did not think that the costs associated with required collection and reporting of data on personal characteristics of nonmortgage loan applicants would be significant because many lenders already collect and report data on personal characteristics under HMDA. But representatives from banks and banking organizations, along with one researcher, said that lending information systems and personnel were not integrated in many mortgage and nonmortgage organizations. For this reason, they reiterated that a data collection and reporting requirement would involve additional system integration and employee training costs, among others. One potential option to mitigate the costs associated with a requirement that regulated lenders collect and report data on the personal characteristics of those seeking nonmortgage loans would be to limit the requirement to certain types of loans, such as small business and/or automobile loans. Similar to mortgage loan applications, small business and automobile loan applications are often made on a face-to-face basis, which could enhance the ability of lenders to help verify the race, gender, or other personal characteristics of the applicants. In contrast, lenders’ capacity to record data on personal characteristics for other types of nonmortgage applicants, such as applicants for credit card loans, may be limited by the fact that credit card loan applications and credit decisions are typically done by mail or over the Internet. However, researchers, federal bank regulatory staff responsible for fair lending oversight, banking officials, and representatives from some consumer groups we contacted cautioned that there were still significant complexities and potential costs associated with a data collection and reporting requirement that was limited to small business lending. Unlike mortgage and automobile lending, which have relatively uniform underwriting criteria, these officials said that small business loan underwriting is heterogeneous and more complex. For example, the types of financing that small businesses typically seek can vary widely, ranging from revolving lines of credit to term loans, and the risk of the collateral pledged against these loans may also vary widely (i.e., from relatively secure real estate to inventory). As discussed previously, studies of possible discrimination in small business lending that use SSBF data consider a variety of other indicators of creditworthiness, such as applicants’ credit scores, personal wealth, and history of bankruptcy. Without information on key underwriting variables, the officials said, research based on the reported data could be subject to significant controversy and potential misinterpretation, much like research based on HMDA data, which lacks information on these variables. At the same time, costs for the necessary technology, employee training, and data verification would likely increase as the range of data that lenders were required to collect and report increases. One option to potentially enhance federal oversight of the fair lending laws, while mitigating lender cost concerns, would be to require lenders to collect data on personal characteristics for small business loan applicants, and perhaps other types of nonmortgage lending like automobile lending, and make the data available to regulators but not require public reporting of such data or any other information. This approach could facilitate federal bank regulators’ ability to prioritize fair lending examinations for regulated lenders because the agencies currently do not have ready access to data on personal characteristics for nonmortgage loan applicants. It could also limit lender costs because they would not have to collect, publicly report, and verify data on a range of underwriting variables because regulators already have access to this information. However, due to the lack of a public data reporting requirement, such an option would not enhance the capacity of researchers, Congress, and the public to better understand the possibility of discrimination in nonmortgage lending. In closing, assessing the potential for discrimination in nonmortgage lending is an important and complex issue. While current data sources, primarily FRB’s SSBF and SCF provide important insights into possible discrimination in certain types of lending, they both have limitations that may impede the ability of researchers, regulators, Congress, and the public to further assess lender compliance with the fair lending laws. It is also not yet clear how FRB’s decision to discontinue the SSBF and incorporate elements of the survey into an expanded SCF beginning in 2010 will impact the already limited amount of information about possible discrimination in nonmortgage lending. Therefore, from a public policy perspective, now may be the time to consider whether the benefits of additional data for research and regulatory purposes outweigh the costs of collecting the data, as well as the trade-offs of various options to enhance available data, from a purely voluntary program to a data collection and reporting requirement, and decide whether such a requirement is warranted. Mr. Chairman, this concludes my prepared statement. I would be pleased to respond to any questions you or other Members of the Subcommittee may have. For further information about this testimony, please contact Orice M. Williams on (202) 512-8678, or at williamso@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Individuals making key contributions to this testimony include Wesley M. Phillips, Assistant Director; Benjamin Bolitzer; Emily Chalmers; Kimberly Cutright; John Forrester; Simin Ho; Omyra Ramsingh; Robert Pollard; Carl Ramirez; and Ethan Wozniak. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The Federal Reserve Board's (FRB) Regulation B, which implements the Equal Credit Opportunity Act of 1974 (ECOA), generally prohibits lenders from collecting certain data from loan applicants, such as their race or gender, for nonmortgage loans (e.g., small business loans). FRB has stated that this provision of Regulation B minimizes the chances that lenders would use such data in an unlawful and discriminatory manner. However, others argue that the prohibition limits the capacity of researchers and regulators to identify possible discrimination in nonmortgage lending. This testimony is based on the GAO report, Fair Lending: Race and Gender Data Are Limited for Nonmortgage Lending ( GAO-08-698 , June 27, 2008). Specifically, GAO analyzes (1) studies on possible discrimination in nonmortgage lending and the data used in them, (2) FRB's 2003 decision to retain the prohibition of voluntary data collection, and (3) the benefits and costs of a data collection and reporting requirement. For this work, GAO conducted a literature review; reviewed FRB documents; analyzed issues involving the Home Mortgage Disclosure Act (HMDA), which requires lenders to collect and publicly report data on personal characteristics for mortgage loan applicants; and interviewed FRB and others. FRB did not take a position on this report's analysis. In addition to restating its rationale for retaining the prohibition of voluntary data collection, FRB summarized GAO's findings, including the potential benefits and costs of additional data for fair lending enforcement. GAO's June 2008 report found that most research suggests that discrimination may play a role in certain types of nonmortgage lending, but data limitations complicate efforts by researchers and regulators to better understand this issue. For example, available studies indicate that African-American owned small businesses are denied loans more often or pay higher interest rates than white-owned businesses with similar risk characteristics. While the primary data source for these studies, a periodic FRB small business survey, provides important insights into possible discrimination, it also has limits compared to HMDA data. For example, the FRB survey data are collected from borrowers rather than lenders, which limit their usefulness as a means to assess lending practices. In addition, federal bank regulators that enforce ECOA said that HMDA data facilitates the identification of lenders that may be engaging in discriminatory mortgage lending. In the absence of such data for nonmortgage loans, regulators may rely on time-consuming and less reliable approaches to identify possible discrimination, such as assuming a loan applicant is Hispanic based on his or her last name. While testimony from researchers and other information GAO collected did not fully agree with all aspects of FRB's 2003 rationale for retaining the prohibition of voluntary data collection, there was general agreement that such voluntary data would have limited benefits. FRB did not adopt a proposal that would have allowed lenders to collect data, without any standards, because it said the proposal would have (1) created an opportunity for lenders to use the data for discriminatory purposes and (2) such data would not be useful since lenders may use different collection approaches. While some researchers and others agreed with FRB's first rationale, others said that data collection alone would not necessarily create the risk for discrimination because, in some cases (e.g., small business lending), lenders may already be aware of applicants' personal characteristics as such lending is often done on a face-to-face basis. Even so, a range of researchers, regulatory staff, and others agreed that voluntarily collected data would not likely materially benefit efforts to better understand possible discrimination because the data would be collected on an inconsistent basis or few lenders would participate out of concern for additional regulatory scrutiny of their nonmortgage lending practices and the potential for litigation. Requiring lenders to collect and publicly report data on personal characteristics for nonmortgage loan applicants could help address current data limitations that complicate efforts to better assess possible discrimination. However, such a requirement would impose additional costs on lenders that could be partially passed on to borrowers. These potential costs include those associated with information system integration, software development, data storage and verification, and employee training. Limiting a requirement to certain types of loans could help mitigate such costs but may also involve complexities that would need to be carefully considered. For example, to the extent that small business lending is more complicated than other types of lending, lenders may need to collect and report additional information on a range of underwriting standards in addition to data on personal characteristics so that informed judgments can be made about their lending practices. |
ONDCP expects that disrupting the illicit drug market will reduce the availability of illicit drugs, increase their cost, and, eventually, reduce the rate of drug usage. To disrupt the cocaine market, the U.S. National Drug Control Strategy calls for, among other things, seizing “enormous and unsustainable” amounts of cocaine from traffickers. One part of the strategy to disrupt the market focuses U.S. interdiction efforts on seizing cocaine and other illicit drugs bound for the United States from South America in the transit zone. The transit zone is a six million square mile area that encompasses Central America, Mexico, the Caribbean Sea, the Gulf of Mexico, and the eastern Pacific Ocean (see fig. 1). Typically, drug traffickers use go-fast boats and fishing vessels to smuggle cocaine from Colombia to Central America and Mexico en route to the United States. Go- fast boats are capable of traveling over 40 knots and are difficult to detect in open water. Moreover, the go-fast boats often travel at night. When they travel in daylight, the boats are often painted blue, or the crew can cover the boat with a blue tarpaulin, thereby becoming virtually impossible to see. Even when detected, go-fast boats can often outrun conventional ships deployed in the transit zone. The primary departments supporting interdiction operations in the transit zone are Defense, Justice, and Homeland Security. Defense provides surveillance aircraft, helicopters, and maritime vessels; Justice provides prosecutorial and law enforcement assistance; and Homeland Security— primarily, the Coast Guard and CBP—provides surveillance aircraft, maritime vessels, and law enforcement assistance. JIATF-South also receives some operational support from various countries in the transit zone; and France, the Netherlands, and the United Kingdom provide air and maritime detection and monitoring assistance in the eastern Caribbean Sea. Justice’s Drug Enforcement Administration (DEA), Federal Bureau of Investigation, and U.S. Attorney’s Office provide intelligence for operations and investigate and prosecute drug traffickers. Other organizations help guide and support interdiction efforts in the transit zone: ONDCP oversees and coordinates implementation of the U.S. National Drug Control Strategy and reviews department and agency budget proposals for anti-drug programs, including interdiction efforts in the transit zone. The U.S. Interdiction Coordinator reports to the Director, ONDCP, and provides strategic advice and oversight of U.S. agencies’ interdiction efforts in the transit zone. It also manages the Consolidated Counterdrug Database that records drug trafficking events, including detections, seizures, and disruptions. The database is vetted quarterly by members of the interagency counterdrug community to minimize duplicate or questionable reported drug movements. JIATF-South, under the U.S. Southern Command, is the primary operations center and coordinator for detecting and monitoring suspect air and maritime drug trafficking events in the transit zone. JIATF-South includes representatives from Defense, Justice, Homeland Security, and others; nations such as France, the Netherlands, and the United Kingdom; and several nations in the transit and source zones. The Coast Guard, CBP, and Defense do not routinely track funds obligated and expended, ship days, or flight hours provided for drug interdiction in the transit zone. However, during fiscal years 2000-2005, the United States provided about $6.2 billion to support counternarcotics and related programs in the source and transit zones (see table 1). In the source zone, U.S. assistance supports eradication and interdiction efforts and related programs for alternative development and judicial reform, primarily in Bolivia, Colombia, and Peru. In the transit zone, the United States provided about $365 million in assistance—primarily to El Salvador, Guatemala, Haiti, and Mexico —to support interdiction and other law enforcement programs. Cocaine seizures and disruptions in the transit zone have increased over two-thirds from calendar year 2000 to calendar year 2004. JIATF-South and other cognizant agency officials pointed to a number of factors that contributed to the increases, such as better intelligence on cocaine movements that allow JIATF-South to target specific cocaine shipments; the introduction of armed helicopters increasing the capability to interdict cocaine shipments on go-fast boats; and increased cooperation from nations in the region, which has led to more efficient use of resources. Cocaine seizures and disruptions have increased about 68 percent since calendar year 2000. As shown in table 2, the United States and its allies seized or disrupted 157 metric tons of cocaine in calendar year 2003 and 196 metric tons in 2004, a record amount. During this period, about two- thirds of all the cocaine seized or disrupted was in the eastern Pacific Ocean and western Caribbean Sea, where according to JIATF-South officials, the United States has positioned most of its ships and aircraft for interdiction and other purposes. Improved intelligence on cocaine shipments gained from law enforcement operations has allowed JIATF-South to more often target suspected drug shipments rather than searching wide expanses of the transit zone for suspicious movements of boats and small aircraft. In addition, in calendar year 2000, the Coast Guard began deploying armed helicopters, which has greatly increased the U.S. capability to stop go-fast boats. Further, several nations in the region have increased their cooperation with U.S. efforts through bilateral maritime agreements, as well as agreements to allow the United States to deploy and re-supply U.S. aircraft and ships on their territory. An interagency law enforcement investigation operation known as “Panama Express” has increased the amount of intelligence on drug trafficking activities. Panama Express is a multi-agency investigation that was initiated in 1995 and is jointly managed by the Federal Bureau of Investigation, DEA, and Immigration and Customs Enforcement but also includes officials from several federal, state, and local law enforcement agencies. The U.S. Attorney’s Office in Tampa, Florida, provides prosecution support for all Panama Express investigations. According to the U.S. Attorney’s office, Panama Express investigations have contributed to the successful prosecution of around 700 individuals, and the seizure and disruption of about 380 metric tons of cocaine since 2000 (or about 51 percent of all the cocaine seized and disrupted). In addition, a JIATF-South official told us that, as of September 2005, 16 embassies and consulates throughout the region had tactical analysis teams. These teams analyze information on drug trafficking activities and provide the information to JIATF-South, DEA, and other operational organizations for use in interdiction missions. Interdiction efforts have been strengthened by the addition of armed helicopters. Beginning in calendar year 2000, the Coast Guard began deploying MH-68A helicopters equipped with 7.62 millimeter machine guns and 50 caliber sniper rifles from its base in Jacksonville, Florida. The helicopters’ main value is their capability to pursue and stop go-fast boats, which can travel over 40 knots—often outrunning U.S. and allied surface ships. According to Coast Guard officials, since November 2002 the armed helicopters have successfully stopped every go-fast boat that they have engaged and contributed to 77 maritime drug interdictions and the seizure of approximately 94 metric tons of cocaine. In addition, the U.S. Navy plans to deploy armed helicopters to pursue and stop go-fast boats in fiscal year 2006. According to Defense officials, these helicopters will have specialized equipment similar to the Coast Guard helicopters. According to the Coast Guard, it will have personnel on board for law enforcement purposes. Since 2000, the number of countries in the transit zone that have signed all or parts of bilateral maritime agreements increased from 21 to 25. These agreements permit, among other things, ship boarding rights for U.S. law enforcement officials and pursuit and entry into these countries’ territorial waters to interdict drug traffickers. In addition, several countries have agreed to additional provisions in the maritime agreements, in particular, the International Maritime Interdiction Support Clause, which allows the United States to fly suspected drug traffickers detained on U.S. ships directly to the United States for prosecution. As a result, these surface ships can resume patrols in the transit zone rather than transporting the suspects directly to the United States, or waiting in a foreign port for the suspects to be extradited. Coast Guard officials told us that, in many cases, this saves up to a week of ship time. (See app. II for a list of countries that have signed all or parts of the bilateral maritime agreements and the interdiction support clause.) Moreover, to offset the closing of Howard Air Force Base in Panama in 1999, the United States has reached agreements with other nations in the region to station U.S. military and civilian aircraft. In return, the United States has upgraded the facilities at these airbases. The facilities at Manta, Ecuador; Comalapa, El Salvador; and Curacao are used by U.S. civilian and military aircraft, while the facilities in Aruba; Liberia; Costa Rica; and Panama are used by the Coast Guard and CBP. Access to these facilities allows aircraft to spend more time monitoring and interdicting drug trafficking activities and also allows them to range further west into the Pacific Ocean. Overall, since fiscal year 2000, the availability of U.S. and allied assets provided for detecting and disrupting drug trafficking activities in the transit zone has varied. On-station ship days peaked in fiscal year 2001 and flight hours peaked in fiscal year 2002, but both have declined since then, primarily because Defense has provided fewer assets. Declines in Defense assets in recent years were mostly offset by additional ship days and aircraft hours provided by the Coast Guard, CBP, and allied nations. However, the ship days and flight hours provided by the allied nations are primarily in the eastern Caribbean Sea where the United States does not usually conduct monitoring and detection activities. Nevertheless, according to JIATF-South, it cannot detect many of the known maritime cocaine movements reported in the western Caribbean Sea and the eastern Pacific Ocean because it cannot get ships or aircraft to the suspected movement in time. Despite a rise in fiscal year 2001, the total number of ship days spent on interdiction missions in the transit zone has generally decreased since fiscal year 2000 from about 3,600 ship days to about 3,300 ship days (or about 8 percent) in fiscal year 2005. U.S. Navy ship days declined from about 1,980 in fiscal year 2000 to about 890 in fiscal year 2005. Defense officials said that the Navy reduced its ship days because of other national security concerns; in particular, the need to provide support for the armed conflicts in Afghanistan and Iraq. The reduction in U.S. Navy ship days was partly offset by increases in ship days by the Coast Guard. After a decline to about 1,160 ship days in fiscal year 2003 (compared to 1,330 days in 2002), the Coast Guard increased its ship days to about 1,660 in 2004 and almost 1,700 in 2005—becoming the primary provider of maritime surface vessels in the transit zone. Figure 2 illustrates the total on-station ship days in the transit zone by provider for fiscal years 2000-2005. On-station flight hours for counternarcotics activities in the transit zone increased about 35 percent between fiscal years 2000-2002—from about 10,500 hours to about 14,200—but have generally declined since then to about 12,870 hours in 2005. Declines in the availability of Defense maritime patrol aircraft over the period were offset by increased support from the Coast Guard, CBP, and several allied nations—France, the Netherlands, and the United Kingdom. The Coast Guard’s on-station flight hours in the transit zone decreased from about 2,440 in fiscal year 2000 to less than 1,300 hours in fiscal year 2002. According to Coast Guard officials, the decline was due, in part, to resources being diverted to other Coast Guard missions. But the Coast Guard has since increased its flight hours to over 2,780 hours in fiscal year 2005. CBP has increased its on-station flight hours for interdiction operations in the transit zone since fiscal year 2000—from about 180 hours in fiscal year 2000 to nearly 4,385 hours in fiscal year 2005. However, in fiscal year 2003, its flight hours declined to about 2,040 due to the diversion of resources for other homeland security missions. Defense on-station flight hours for interdiction in the transit zone gradually declined from about 6,860 in fiscal year 2000 to 6,500 in fiscal year 2002, but declined more rapidly since then to about 2,940 flight hours in fiscal year 2005. JIATF-South officials attribute the recent declines primarily to the reduced availability of U.S. Navy P-3 maritime patrol aircraft because of structural problems. In addition, U.S. Air Force Airborne Warning and Control System (AWACS) aircraft were diverted to support homeland security missions and the armed conflicts in Afghanistan and Iraq. Allied nations steadily increased their flight hours in support of interdiction operations from about 1,000 hours in fiscal year 2000 to 4,070 hours in fiscal year 2004. However, in fiscal year 2005, the allies’ total hours declined to about 2,760. According to JIATF-South and U.S. Interdiction Coordinator officials, most of the allies’ surveillance operations are in the eastern Caribbean Sea where the United States does not usually conduct interdiction operations. Figure 3 illustrates the total on-station flight hours by provider for fiscal years 2000-2005. Since calendar year 2000, JIATF-South officials report that they had information about more maritime drug movements than they could detect (make visual contact with). The number of “known actionable” maritime events in the western Caribbean Sea and the eastern Pacific Ocean more than doubled from 154 in 2000 to 330 in 2004. According to JIATF-South officials, in many cases, the maritime event is too far away for available ships and aircraft to go to the area and visually locate the suspected drug movement. However, once JIATF-South locates a suspect movement, the disruption rate has significantly increased since 2000—from less than 60 percent in 2000 and 2001 to over 80 percent in 2003 to 2005 (see table 3). While the United States has increased the number of seizures and disruptions in the transit zone since 2000, the Coast Guard, CBP, and Defense face several challenges in maintaining the current level of assets provided for transit zone interdiction operations. JIATF-South officials expressed concern that continued declines in U.S. on-station ship days and on-station flight hours will limit their ability to monitor the transit zone and detect illicit drug trafficking. Specifically, according to JIATF-South and other Defense officials, the reduced availability of the U.S. Navy P-3 maritime patrol aircraft will degrade JIATF-South’s ability to detect maritime movements. In addition, the readiness rates of older Coast Guard ships, which support interdiction operations in the transit zone, have declined since fiscal year 2000, and the surface radar system on its long- range surveillance aircraft is often inoperable. JIATF-South, Coast Guard, CBP, and U.S. Interdiction Coordinator officials stated that, while some short-term fixes have been made, the longer-term implications of the likely continued declines in monitoring and interdiction assets for the transit zone have not been addressed. Moreover, in response to increases in cocaine seizures and disruptions, drug traffickers have adjusted their tactics for transporting cocaine through the transit zone. Finally, nations in the region lack the resources to offset any decline in assets. According to cognizant officials with JIATF-South and U.S. Interdiction Coordinator officials, compensating for the reduced availability of P-3 maritime patrol aircraft is the most critical challenge to the future success of interdiction operations. According to these officials, because of its longer range, the P-3 aircraft can monitor a much larger surface area than other maritime patrol aircraft and can provide covert surveillance until other assets arrive. The availability of the P-3 aircraft has declined for several reasons. In fiscal years 2000-2003, the U.S. Navy provided the majority of P-3 maritime patrol flying hours in support of interdiction efforts—about 60 percent of the on- station flight hours (or over 3,900 hours per year). However, in fiscal year 2004, the Navy began limiting the use of its P-3 maritime patrol aircraft for transit zone interdiction missions because of structural problems in its wings and other worldwide commitments. Since fiscal year 2000, the number of hours flown by U.S. Navy P-3s has decreased nearly 60 percent to about 1,500 hours in fiscal year 2005. In addition, in December 2004, the Netherlands removed the P-3 aircraft it used to fly interdiction missions in the transit zone. According to the U.S. Interdiction Coordinator, the P-3s flown by the Netherlands were vital to interdiction efforts in the Caribbean Sea, averaging over 1,300 flight hours per year—or about 20 percent of all P-3 flight hours—during fiscal years 2000-2004. In April 2005, the Netherlands began using the Fokker F-60, a shorter-range twin engine aircraft, to fly interdiction missions; but, according to Defense officials, these aircraft are less capable than the P-3. To help compensate for the reduction in the P-3 availability, CBP has increased its P-3 maritime patrol on-station flight hours in the transit zone—from about 180 flight hours in fiscal year 2000 to over 4,300 hours in 2005. However, CBP officials told us that their aircraft cannot totally compensate for the loss of the Navy’s and Netherlands’ P-3 flying hours. Further, these officials are unsure how long CBP will be able to continue to provide the additional flying hours, which has been more than its allotted budget, because of other homeland security priorities. Figure 4 illustrates the recent decline in P-3 flight hours. In August 2004, the U.S. Interdiction Coordinator established a working group to evaluate the problem of the reduction in maritime patrol aircraft flying hours and identify possible solutions. The group identified several possible alternatives, including: redeploying AWACS to support an aerial interdiction program in substituting other Defense aircraft for the Navy P-3s, extending the amount of time aircraft are deployed to forward operating locations throughout the region, increasing the number of hours flown by the Coast Guard HC-130 surveillance aircraft in support of interdiction, upgrading the sensors on existing aircraft to improve their capabilities, deploying aircraft to locations closer to the suspected trafficking routes, requesting additional funding to support the deployment of United Kingdom maritime patrol aircraft in the transit zone. According to JIATF-South officials, several of these proposals have been implemented. AWACS have begun flying missions in support of the aerial interdiction program in Colombia (also known as the “Air Bridge Denial” program)—relieving CBP of much of this burden. The Coast Guard and CBP have deployed aircraft to locations in Costa Rica and Ecuador, and the Coast Guard has received funding to upgrade some of its detection and monitoring equipment with improved sensors. Finally, Defense is providing housing for United Kingdom aircrews (relieving them of the financial burden) that provide surveillance in the Caribbean Sea while flying training missions. But, JIATF-South, the U.S. Interdiction Coordinator, and other cognizant officials noted that the longer term prospect of further declines in U.S. P-3 or other maritime patrol assets for interdiction operations has not been addressed. During fiscal years 2000 through 2004, the readiness rates of the Coast Guard’s older ships and aircraft showed a general decline, although the rates fluctuated from year to year. For example, ships used to monitor drug trafficking activities and carry the helicopters that disable and stop go-fast boats were below their target levels for time free of major deficiencies or loss of at least one primary mission. Further, the percentage of time that HC-130 surveillance aircraft were available to perform missions was below the target level in fiscal year 2004, and the surface radar system on the aircraft is subject to frequent failures. In some instances, mission flight crews had to look out the windows of the aircraft for targets because the radar systems were inoperable. The Coast Guard has taken several actions to keep its assets operational. These include establishing a compendium of information for making decisions regarding maintenance and upgrades; performing more extensive maintenance between deployments; and exploring strategies for prioritizing the maintenance and capability enhancement projects needed to provide more objective data on where to spend budget dollars and enhance mission capabilities. However, these additional efforts, while helpful in preventing a more rapid decline in the condition of existing assets, are unlikely to solve the problem. For fiscal year 2004, the Coast Guard’s estimated cost of deferred maintenance for these assets totaled approximately $28 million. Interdiction operations are further challenged by the changing tactics of the drug traffickers. In the eastern Pacific Ocean, JIATF-South in recent years has detected suspect fishing vessels and go-fast boats traveling further south and about 300 miles southwest of the Galapagos Islands before turning towards Central America and Mexico. This change in tactics has greatly increased the ocean surface area that must be monitored and, because of the distances to travel, has made it more difficult for U.S. surface ships and aircraft to respond to suspected cocaine movements. In addition: Drug traffickers have begun using larger go-fast boats. In the past, go- fast boats typically transported 2 to 4 metric tons of cocaine, but some are now capable of carrying up to 8 metric tons. Drug traffickers in the eastern Pacific Ocean have increasingly used vessels registered to countries, such as Ecuador and Mexico, with which the United States has no maritime boarding agreements. The use of these vessels complicates the process of boarding and searching for illicit drugs. Moreover, Ecuador has declared that its territorial waters extend 200 miles from its coastline and around the Galapagos Islands (most nations, including the United States set a 12-mile territorial limit). Although the United States does not recognize Ecuador’s claim, U.S. ships and aircraft generally do not enter this area for interdiction missions to avoid the potentially politically sensitive issue of having to deal with the Government of Ecuador. Although most nations in the transit zone—especially countries in Central America and the eastern Caribbean Sea—cooperate with U.S. interdiction efforts, JIATF-South and other officials told us that these countries lack the capability and resources to compensate for any decline in U.S. and allies’ assets. Most countries in the transit zone are able to provide only a few vessels or aircraft to intercept drug traffickers or lack secure communications equipment needed to coordinate with U.S. law enforcement counterparts during interdiction missions. Also, many of these countries do not have a civilian law enforcement or military presence in the coastal areas where drug traffickers work with local criminal organizations to offload cocaine shipments and deliver them to the United States. In addition, U.S. agencies are reluctant to work with law enforcement officials in some countries because of widespread corruption. From fiscal year 2000 through 2005, the United States provided about $365 million in assistance to countries in the transit zone. Of this, Mexico received approximately $115 million to support its efforts to eradicate opium poppy and marijuana, and improve surveillance and intelligence capabilities. The transit zone countries in Central America and the Caribbean received the remainder, with most of this assistance for programs to assist civilian law enforcement and military agencies in El Salvador, Guatemala, Jamaica, and Panama. Assistance provided to transit zone countries in Central America and the Caribbean from State’s Bureau for International Narcotics and Law Enforcement Affairs increased from approximately $18 million in fiscal year 2000 to about $25 million in fiscal year 2002 but has since declined to about $7 million in fiscal year 2005. A State official attributed the decline in assistance to commitments in Afghanistan and Iraq. State assistance has been used for helicopter airlift support to Guatemala’s police counternarcotics units and interceptor boats for the Bahamas, among other things. Assistance for transit zone countries in Central America and the Caribbean provided through the Foreign Military Finance and International Military Education and Training programs has increased from about $19 million in fiscal year 2000 to about $36 million in fiscal year 2005. This assistance has been used to train host country personnel and provide equipment such as helicopters, spare parts, and fuel to support counternarcotics training and missions. The Government Performance and Results Act of 1993 requires federal agencies to develop performance measures to assess progress in achieving their goals, and to communicate their results to the Congress. The act requires agencies to set multiyear strategic goals in their strategic plans and corresponding annual goals in their performance plans, measure performance toward the achievement of those goals, and report on their progress in their annual performance reports. These reports are intended to provide important information to agency managers, policymakers, and the public on what each agency accomplished with the resources it was given. Moreover, the act calls for agencies to develop performance goals that are objective, quantifiable, and measurable, and to establish performance measures that adequately indicate progress toward achieving those goals. Our previous work has noted that the lack of clear measurable goals makes it difficult for program managers and staff to link their day-to- day efforts to achieving the agency’s intended mission. When multiple agencies are involved in achieving the desired results, as with transit zone interdiction operations, agencies should coordinate the development of performance measures to ensure that they are complementary. In the case of the transit zone, performance measures and approaches for assessing transit zone operations vary among the Coast Guard, CBP, and Defense. In addition, basic information about cocaine production and trafficking and cocaine usage in the United States that would help in assessing transit zone interdiction operations is problematic. Further, ONDCP and other agencies involved have not fully addressed recommendations for improving illicit drug data collection and analysis. According to the 2005 U.S. National Drug Control Strategy update, the 2002 U.S. National Drug Control Strategy “clearly laid out a plan for accountable results in achieving a single goal—reducing drug use.” Yet specific measures of performance in the transit zone linking interdiction operations to the Strategy’s priority of disrupting the illicit drug market were not included in the 2002 strategy, nor subsequently developed across the agencies in conjunction with ONDCP. The performance measures developed by the Coast Guard, and the measures being developed by CBP and Defense, vary in their emphasis and may not be helpful in assessing progress in disrupting the illicit drug market. The Coast Guard’s performance measures related to the transit zone set specific goals to reduce the flow of cocaine. For fiscal year 2004, the goal was to remove 15 percent of the cocaine flowing through the transit zone; according to Coast Guard officials, the removal goal increases to 35 percent in 2011. CBP is developing performance measures related to operational readiness rates (a measure of its ability to respond when requested), but these rates are not specific to the transit zone or to counternarcotics activities and do not measure results. Defense is developing performance measures that focus on the number of disruptions of cocaine trafficking events, but it has not yet set any targets or goals to assess its progress. Data for assessing U.S. interdiction operations in the transit zone and relating the results to the U.S. National Drug Control Strategy’s priority of disrupting the illicit drug market and to its overall goal of reducing drug usage in the United States are problematic. Specifically, the data for estimating the cocaine flowing through the transit zone towards the United States has been called into question, and data to demonstrate progress in reducing drug use are difficult to obtain in ways that can be generalized to the United States. The principal source of information about cocaine flow in the transit zone is ONDCP’s Interagency Assessment of Cocaine Movement (IACM). The IACM is prepared annually for ONDCP by an interagency group representing departments and agencies involved in U.S. counternarcotics efforts. The assessment is intended to advise policymakers and resource planners whose responsibilities involve detecting, monitoring, and interdicting illicit drug shipments. It draws on other studies and several datasets to estimate cocaine supply (that is, how much cocaine is produced and available for export to the United States) and demand (that is, how much cocaine is used in the United States and elsewhere). In the past, the IACM’s estimates of cocaine supply were close to the estimated demand in the United States and other world markets plus estimated losses in transit, source, and arrival zones. But for calendar years 2003 and 2004, according to the interagency group, the IACM’s estimate of the amount of cocaine available for export was too low in relation to estimated U.S. and non-U.S. demand for cocaine after taking into account seizures and disruptions. The interagency assessment noted that the increasing difference between supply and demand estimates may be the result of several different scenarios, a combination of which points to potential shortfalls in some or many of the data sets. For example: Production and consumption estimates could be widely off the mark. Production estimates are not designed to capture dynamic activities such as eradication, replanting, or changing processing efficiencies. Worldwide consumption estimates are several years old (for example, the U.S. estimate was developed in 2002) and may not yet show the effects of record eradication and interdiction efforts in recent years. Time lags in estimating aspects inherent to the cocaine trade—including price, purity, and demand—could delay the apparent effects and documentation of a shortage. As a result of the disparity between the estimated cocaine supply and demand, the interagency group stated that a precise estimate of cocaine flow was not possible for 2004, and that “a range of possible amounts was more intellectually and analytically honest.” The group estimated that between 325 and 675 metric tons of cocaine flowed towards the United States in 2004, but such a wide range is not useful for assessing transit zone interdiction efforts. ONDCP and other interagency group officials agreed that the wide range reflects the difficulty in obtaining specific information about the production of cocaine and how it gets to the United States. According to ONDCP, the Director of ONDCP has established a working group to identify and, where possible, quantify the uncertainties with the IACM, and identify ways to improve the data and reduce uncertainty. ONDCP uses several surveys and databases to help it assess the availability and rate of illicit drug use in the United States and changes in those measures over time, but a variety of issues affect their reliability and validity. According to several studies and ONDCP officials, a large portion of major cocaine (and other drug) users are members of generally hard-to- survey populations, such as the homeless or incarcerated, and those who are questioned about illicit drug use may be inclined to provide a socially acceptable, and legal, response to survey questions. In addition, typical survey response problems, such as low response rates and failure of respondents to accurately recall past events, also apply. Moreover, according to ONDCP officials, obtaining information from and about persons engaged in an illegal activity is difficult at best, much of the available information about illicit drug use cannot be generalized to the United States, and the logistics of collecting meaningful data means that a time lag will always exist between the data collection period and the time when the data are available for public policy purposes. Thus, under the best of circumstances, the effect of a drug policy change may take a number of years to demonstrate. Table 4 lists the primary sources of information ONDCP uses to track illicit drug trends in the United States and highlights the latest date of data collection, ONDCP’s primary purpose, and selected issues about the information source. In some cases, the survey or database was designed for other purposes and ONDCP officials have adapted it for their use. In a 2001 report prepared for ONDCP, the National Research Council concluded that the United States had neither the data systems nor the research infrastructure needed to assess the effectiveness of drug control enforcement policies. In particular, the Council highlighted the absence of adequate, reliable data on illicit drug prices and use. The Council made numerous recommendations to improve data collection and analysis— eight were addressed in full or in part to ONDCP. Regarding interdiction, the Council recommended that research be done to address the following: to what extent traffickers can limit the effect of interdiction operations by shifting their routes and modes of transportation; how the deterrent effects of supply-reduction programs can be measured and the size of these effects; and how quickly drug production and trafficking adapt to supply-reduction activities, what happens to supply and price during the period of adaptation, and how long the deterrent effects of supply-reduction operations last before new supply sources emerge. ONDCP officials told us these recommendations have not been fully addressed. They noted that the illicit and secretive nature of the illicit drug market precludes the systematic collection of cultivation, production, and trafficking information. Furthermore, these officials emphasized that the economics of cocaine production and trafficking may not follow typical supply and demand relationships—profit margins likely remain high despite record coca eradiation and cocaine interdiction efforts, and drug traffickers can quickly react to changing interdiction tactics and circumstances. Since fiscal year 2000, the United States has provided over $6 billion for counternarcotics and related programs in South and Central America and throughout the transit zone, primarily to reduce the amount of illicit drugs produced and transported to the United States. While the number of on- station ship days and on-station flight hours provided for monitoring and interdiction operations in the transit zone has varied since 2000, they have generally declined in recent years. JIATF-South, Coast Guard, and CBP officials are concerned that the current level of operations cannot be sustained. Defense on-station ship days and flight hours have already declined due to operational priorities in other parts of the world, primarily Afghanistan and Iraq, and structural limitations on the U.S. Navy’s primary maritime patrol aircraft—the P-3. The reduced availability of P-3 maritime patrol aircraft will degrade JIATF-South’s ability to detect and monitor go- fast boats and other vessels suspected of transporting illicit drugs. JIATF- South, Coast Guard, and CBP officials are concerned that this problem is likely to worsen as budget constraints and other homeland security priorities arise that limit the assets available for interdiction operations. While some short-term fixes have been taken, the longer-term implications of further declines in the availability of monitoring and interdiction assets have not been addressed. Developing performance measures—linking interdiction operations in the transit zone to disrupting the illicit drug market, specifically the cocaine market—is difficult. Nevertheless, the Coast Guard, CBP, and Defense, in conjunction with ONDCP, should develop and coordinate performance measures that more directly relate to transit zone operations. Data compiled by the U.S. Interdiction Coordinator and vetted by the principal agencies involved in transit zone interdiction efforts —data primarily documenting detections of drug movements and seizures and disruptions of illicit drugs—could serve as benchmarks for assessing progress until data more directly related to disrupting the illicit drug market can be developed. Collecting relevant data to assess the effect of interdiction operations on the cocaine market and drug usage in the United States is problematic. Assessing how much cocaine is produced and moves towards the United States is not easy—as the 2004 IACM demonstrates. Most other readily available data on cocaine price, purity, and availability—indicators of U.S. demand—cannot be generalized to the United States. More systematic surveys on drug usage take time—sometimes several years—to complete. Thus, a number of years is often needed to show the effect of a drug policy change. The National Research Council reported similar issues in 2001. However, until ONDCP and other cognizant agencies fully address the Council’s recommendations, data to help assess U.S. drug usage will remain problematic. We recommend that the Secretaries of Defense and Homeland Security plan for the likely decline in the future availability of ships and aircraft for transit zone interdiction operations and, specifically, determine how they will compensate for the decline in P-3 maritime patrol aircraft availability. We recommend that the Secretaries of Defense and Homeland Security develop and coordinate, in conjunction with the Director of ONDCP, performance measures for transit zone interdiction operations that take advantage of available drug interdiction data (such as detections, seizures, and disruptions) to provide a basis for (1) assessing transit zone interdiction performance and (2) deciding how to deploy increasingly limited assets, such as the P-3 maritime patrol aircraft. We also recommend that the Director of ONDCP address each of the recommendations made by the National Research Council and report to the Congress what departments and agencies need to take action, what remains to be done, and when action is expected to be completed. In those instances where ONDCP reports that action is not necessary, we recommend that it document the reasons why. Defense, Homeland Security, and ONDCP provided written comments on a draft of this report. See appendixes III, IV, and V, respectively. Justice and State did not provide written comments. However, we discussed the draft report with cognizant officials at each of the departments. Overall, the departments and ONDCP stated that they generally concurred with the recommendations that applied to them, but none detailed when and how they will address them. Defense and Homeland Security noted that they already have taken or are in the process of taking appropriate action regarding planning for drug interdiction asset requirements and developing relevant performance measures. Defense specifically noted that the U.S. Navy has developed a Fleet Response Plan designed to transition from the P-3 to its replacement aircraft—the Multi-Mission Maritime Aircraft, which is scheduled to be introduced into the fleet in fiscal year 2011. Defense added that it will continue to coordinate with all maritime patrol asset providers in the transit zone. Regarding performance measures, Defense stated that it has appropriate mechanisms in place to assess its efforts. Yet, Defense also noted that during the past year it has coordinated with its commands to develop performance goals, measures, and targets for the transit zone and continues to work with them as they prepare their submissions. During the course of this engagement, cognizant Defense officials stated on several occasions that they had not finalized performance measures. Homeland Security stated it develops asset requirements as part of an interagency process where asset commitment is based on threat level and the availability of funding. While it remains committed to “robust support” of maritime transit zone interdiction efforts, unforeseen events—Hurricane Katrina, for example—can affect asset availability. Regarding performance measures, Homeland Security noted that as a result of the Intelligence Reform and Terrorism Prevention Act of 2004, it has undertaken an effort to develop and coordinate a performance measurement system to better assess Homeland Security’s counternarcotics’ activities, including transit zone interdiction operations. Homeland Security did not state when it will complete this effort. ONDCP did not comment directly on interdiction asset availability, but overall, ONDCP agreed that developing performance measures that link interdiction operations to disrupting the cocaine market is difficult but necessary, and stated that it will work with Defense and Homeland Security within the framework of the Government Performance and Results Act to develop appropriate measures. Our recommendation regarding planning for the likely decline in interdiction asset availability was intended to have Defense and Homeland Security work together to address asset availability; in particular, the decline in maritime patrol aircraft hours. However, the departments do not address in their comments when or how their planning efforts have been coordinated nor when their on-going efforts will be completed. We continue to believe that such a coordinated planning effort is both appropriate and necessary to ensure that the departments’ limited assets are used in the most effective manner, and encourage the departments to follow through with one another. Concerning the National Research Council’s recommendations, ONDCP did not specifically address our recommendation to report to Congress on what actions have been taken or still need to be taken. Rather, ONDCP referred to a 2001 study—Measuring the Deterrent Effect of Enforcement Operations on Drug Smuggling, 1991-1999—that it published as one of several steps it has taken to address the Council’s recommendations. We note that the ONDCP study was published the same year the Council published its report, and that it does not refer to the Council’s report or its recommendations. The ONDCP study’s primary objective was to “measure the impact of drug enforcement operations on the cocaine smuggling industry.” Among other things and related to this report, the study concludes that better data was available on what drug traffickers were doing than on the activities of U.S. drug interdiction assets. According to the study’s authors, “this was most notably the case with the interdiction activities of the Department of Defense.” In addition, to address the Council’s recommendations, ONDCP also noted that it has established two operational priorities focused on (a) understanding the drug market and (b) enhancing the data sets ONDCP relies on. These priorities were initiated during the course of our engagement. According to ONDCP officials, they have not been completed. As we note in this report, collecting relevant data for assessing the effect of interdiction operations on the cocaine market and U.S. illicit drug usage is problematic. But more than three years after the National Research Council’s report addressing the illicit drug data issue was finalized, ONDCP cannot point to any specific action it has completed to address the Council’s recommendations for improving the collection of illicit drug trafficking and usage data. Given the importance of the illicit drug problem and the on-going controversy in the United States about how best to confront it, better data for evaluating alternative drug control policies is paramount. We continue to urge ONDCP to follow up on the Council’s findings and recommendations and take the steps necessary to address the continuing shortcomings in its illicit drug data and report to Congress the status of its efforts. In commenting on the IACM, Homeland Security stated that the decision by the interagency group to report a range for cocaine movement was a wise one because it acknowledged the lack of precision inherent in the data that is currently available. We agree, and did not intend to imply anything else. We also agree with Homeland Security that greater efforts need to be made to improve the methodology and integrity of data instruments in order to gain a better and more useful understanding of drug production and consumption. Finally, all the departments and ONDCP provided additional information on various aspects of their roles regarding drug interdiction in the transit zone. Defense, in particular, emphasized that it is not authorized to conduct law enforcement operations but can support them. In addition, all the departments and ONDCP provided us technical comments and updates that we have incorporated throughout the report, as appropriate. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copies of this report to interested congressional committees and the Secretaries of Defense, Homeland Security, Justice, and State, and the Director of ONDCP. We will also make copies available to others upon request. In addition, this report will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff have any questions concerning this report, please contact me at (202) 512-4268 or FordJ@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report were Al Huntington, Joe Carney, J.J. Marzullo, José Peña, and Jim Strus. Overall, to examine the status of U.S. interdiction assets and progress made in disrupting drug trafficking in the transit zone, we focused on U.S. (1) efforts to interdict cocaine because nearly all the cocaine entering the United States comes from South America and (2) operations in the western Caribbean Sea and the eastern Pacific Ocean because the United States has positioned most of its interdiction assets in these areas. We also agreed with the requesters to limit the scope of this engagement by examining U.S. operations to interdict cocaine before it reaches intermediate staging points in the Caribbean Sea, Central America, and Mexico on the way to the United States. We further narrowed our examination to U.S. efforts to detect and disrupt unscheduled maritime movements of cocaine— primarily go-fast boats, but also fishing and other types of ocean-going vessels. Although drug traffickers use small aircraft to transport cocaine and other drugs, the United States essentially relies on the law enforcement authorities in the countries where they land to interdict them. To track the changes in the amounts of cocaine seized and disrupted since calendar year 2000, we relied on the Interagency Assessment of Cocaine Movement (IACM). While the data sets used to prepare the IACM have been called into question by the interagency group that prepares it, the IACM data on cocaine seizures and disruptions are based on the Consolidated Counterdrug Database, which is managed by the Office of the U.S. Interdiction Coordinator. Beginning with 2004, the Coordinator’s Office began using stricter rules to vet the data on drug movements, detections, seizures, and disruptions in quarterly meetings with the interagency drug community. The intent was to minimize duplicate or questionable reported drug movements. According to the database manager, however, the more careful review of the data did not materially affect the categorization or number of maritime drug movements—our primary emphasis. Rather, many suspect aircraft are no longer counted as drug movement events unless corroborating information strongly suggests drugs were aboard the aircraft. We observed a quarterly vetting session and discussed the process for determining whether reported drug events can be “confirmed and substantiated” based on intelligence and other sources with several members of the interagency group that participate, including the Defense Intelligence Agency (DIA); the U.S. Director of Central Intelligence, Crime and Narcotics Center; the Drug Enforcement Administration (DEA); the Joint Interagency Task Force-South (JIATF-South); the Coast Guard; and Customs and Border Protection (CBP). Based on the foregoing, we determined that the cocaine seizure and disruption data provided to us were sufficiently reliable for describing and documenting trends in the transit zone. To determine the factors that have contributed to the changes in cocaine seizures and disruptions, we reviewed the IACMs for calendar years 2000 through 2004. We discussed the observations made in the IACMs and reported operational changes since 2000 with cognizant officials in the Office of National Drug Control Policy (ONDCP), the Department of Defense’s (Defense) Office of Drug Enforcement Policy and Support, and other members of the interagency counternarcotics community—namely, DIA, the Crime and Narcotics Center, DEA, JIATF-South, and the U.S. Interdiction Coordinator. To analyze the trends in the assets provided by the United States and its allies since fiscal year 2000 to support illicit drug interdiction in the transit zone, we relied on data provided by JIATF-South and JIATF-West that tabulates on-station ship days and flight hours spent on interdiction missions by the provider and the type of ship or aircraft. Until fiscal year 2004, JIATF-West had responsibility for U.S. drug detection and monitoring activities in the eastern Pacific Ocean off the coast of Mexico. We analyzed data for fiscal years 2000 through 2005, although JIATF-West could not provide information about on-station ship days for the first quarter of fiscal year 2000 (September-December 1999). Compared to the first quarter of 2001, JIAFT-West’s contribution was about 5.4 percent of the total ship days on station. To determine the reliability of the JIATF-South and JIATF-West data, we discussed how the respective data was compiled with cognizant officials and compared it to available—albeit, less specific—data from the Coast Guard, CBP, and Defense. Through these efforts, we determined that the data provided to us were sufficiently reliable for describing the trends in asset availability during fiscal years 2000 through 2005. To determine the challenges facing the departments and agencies in maintaining current levels of transit zone interdiction operations, we reviewed issues raised in the IACM and other department and agency planning and budgeting documents. We discussed these matters with cognizant law enforcement and military authorities at Defense, Department of Homeland Security, Department of Justice, and headquarters in Washington, D.C. We also discussed operational challenges with Coast Guard officials at their base in Jacksonville, Florida, where the Coast Guard stations armed helicopters used in interdiction operations; CBP officials at their base in Jacksonville, Florida, where CPB stations P-3 maritime patrol aircraft used in interdiction operations; DEA officials in Miami, Florida, and Nassau, Bahamas, who manage an ongoing drug interdiction program in the waters around the Bahamas called “Operation Bahamas, and Turks and Caicos;” JIATF-South officials in Key West, Florida, including the Commander; representatives from the Coast Guard, CBP, Defense, and other U.S. agencies; as well as the liaison officers from France, the Netherlands, and the United Kingdom; and Justice officials with the U.S. Attorney’s offices in Tampa and Sarasota, Florida, directly involved in the Panama Express operation. To determine how the Coast Guard, CBP, and Defense assess their performance, we met with cognizant officials in their respective planning and budgeting offices and reviewed relevant reports and related documents. Since only the Coast Guard had completed developing performance measures, we relied on discussions with the key officials to document what CBP and Defense were developing. To determine what data are available for assessing transit zone performance, we met with the principals involved in preparing the IACM. We discussed issues that have arisen concerning cultivation and production estimates (primarily, officials with the Crime and Narcotics Center and DEA), as well as interdiction amounts (primarily, the U.S. Interdiction Coordinator’s Consolidated Counterdrug Database manager). In addressing usage and demand estimates in the United States, we asked ONDCP officials what they used to measure progress against the National Drug Control Strategy’s goal of reducing drug usage in the United States. These officials cited the data sources we refer to in the body of the report (see table 4) and described how they used the information. We also reviewed the scope and methodology statements for each data source, and in the cases of the National Survey on Drug Use and Health and Monitoring the Future, we spoke with the current project directors to obtain their views. The United States has signed Counternarcotics Maritime Law Enforcement agreements with 25 countries in the transit and source zones—three since 2000 after the United States closed its military installations in Panama. According to Coast Guard officials, these agreements have improved cooperation with nations in the region and increased U.S. and, in particular, the Coast Guard’s capability to board suspect vessels and detain suspected drug traffickers. These officials added that these agreements are one of the primary reasons for increased interdictions of cocaine shipments in the transit zone. These bilateral agreements typically have six provisions to them. The United States and the countries negotiate each provision separately, which means that some countries may agree to some provisions and not others. The six parts provide for the following: Shipboarding provisions allow U.S. agencies under certain conditions to stop, board, and search suspicious vessels registered in that country without having specific permission. Shiprider provisions permit countries to place law enforcement officials on another’s vessels. Pursuit provisions allow U.S. law enforcement agencies, under very limited circumstances, to pursue aircraft and vessels in a country’s airspace and territorial waters. In particular, the provisions permit U.S. law enforcement agencies to board and search a suspect vessel if the country does not have a vessel or aircraft available to respond immediately. Entry-to-investigate provisions allow the U.S. law enforcement agencies, under very limited circumstances, to enter a country’s airspace or territorial waters to investigate aircraft or vessels suspected of illicit drug trafficking. Specifically, the provisions permit U.S. law enforcement agencies to board and search a suspect vessel if the country does not have a vessel or aircraft available to respond immediately. Overflight provisions permit the U.S. law enforcement aircraft to fly over the country’s territorial waters, with appropriate notice to the country’s coastal authorities. Relay order-to-land provisions allow U.S. law enforcement agencies to relay an order to land from the host country to the suspect aircraft. Moreover, an additional International Maritime Interdiction Support clause permits U.S. law enforcement agencies, principally the Coast Guard, to transport suspected drug traffickers through that country to the United States for prosecution and provides for expedited access to that country’s dockside facility to search suspect vessels. Since 2000, the United States has entered into support clauses with eight additional countries. The following table lists the law enforcement agreements, including the international maritime interdiction support clause that the United States has negotiated with countries in the transit and source zones. | One of the U.S. National Drug Control Strategy's priorities is to disrupt the illicit drug market. To this end, the Departments of Defense and Homeland Security provide ships and aircraft to disrupt the flow of illicit drugs, primarily cocaine, shipped from South America through the Caribbean Sea and eastern Pacific Ocean--an area known as the transit zone. The Office of National Drug Control Policy (ONDCP) oversees the U.S. anti-drug strategy. The Joint Interagency Task Force-South (JIATF-South) directs most transit zone operations. We examined U.S. efforts to interdict maritime movements of cocaine. We analyzed the (1) changes in cocaine seizures and disruptions since calendar year 2000, (2) trends in interdiction assets provided since fiscal year 2000, (3) challenges to maintaining transit zone interdiction operations, and (4) performance measures the agencies use to assess their progress. Cocaine seizures and disruptions in the transit zone have increased about 68 percent since calendar year 2000--from 117 metric tons in 2000 to 196 metric tons in 2004. About two-thirds of the disruptions were in the western Caribbean Sea and eastern Pacific Ocean where the United States has most of its interdiction assets. JIATF-South and other cognizant officials attribute the increase to improved interagency cooperation and intelligence, the introduction of armed helicopters to stop go-fast boats, and increased cooperation from nations in the region. Since fiscal year 2000, the availability of assets--ships and aircraft--to disrupt drug trafficking in the transit zone have varied. On-station ship days peaked in fiscal year 2001 and flight hours peaked in 2002, but both have generally declined since then, primarily because the Department of Defense has provided fewer assets. Declines in Defense assets have been largely offset by the Coast Guard, Customs and Border Protection (CBP), and certain allied nations. Nevertheless, in recent years, JIATF-South has detected less than one-third of the "known and actionable" maritime illicit drug movements in the western Caribbean Sea and eastern Pacific Ocean. Yet, once detected, over 80 percent of the drug movements were disrupted. Various factors pose challenges to maintaining the current level of transit zone interdiction operations. The reduced availability of the U.S. Navy's P-3 maritime patrol aircraft due to structural problems will degrade the U.S. capability to detect suspect maritime movements, readiness rates of older Coast Guard ships have declined since fiscal year 2000, and the surface radar system on the Coast Guard's long-range surveillance aircraft is often inoperable. Coast Guard and CBP officials also noted that they may not be able to sustain their level of assets in light of budget constraints and other homeland security priorities that may arise. These officials expressed concern that the long-term implications of likely declines in transit zone assets have not been addressed. The Government Performance and Results Act of 1993 requires agencies to develop performance measures to assess progress in achieving their goals. The Coast Guard's measures relate to reducing cocaine flow through the transit zone, CBP's planned measures are not specific to the transit zone, and Defense's planned measures focus on the number of disruptions of cocaine movements. But data that would help in assessing transit zone interdiction operations are problematic. For instance, in its assessment for 2004, ONDCP reported that between 325 metric tons and 675 metric tons of cocaine may be moving towards the United States. Such a wide range is not useful for assessing transit zone interdiction operations. In addition, data on U.S. drug usage are difficult to obtain and often cannot be generalized to the United States. In a 2001 report for ONDCP, the National Research Council made similar observations and recommended ways to improve the collection and analysis of illicit drug data, but ONDCP has not fully addressed them. |
The National Wildlife Refuge System (NWRS)—the only system of federal lands protected specifically for wildlife conservation—provides more than 96 million acres of habitat for over 700 species of birds, hundreds of threatened or endangered species, and a wide variety of other species. Each year, millions of birds stop to rest at refuges strategically located along their migration routes. In 1903, President Theodore Roosevelt established what is now recognized as the first refuge, the Pelican Island National Wildlife Refuge (NWR) in Florida. During the more than 100 years since, the refuge system has grown to include 548 wildlife refuges and 37 wetland management districts that address a variety of wildlife purposes. For example, the Merced NWR in California was established in 1951 with the broad purpose of serving as a sanctuary for migratory birds, while the Antioch Dunes NWR, also in California, was established to protect three specific endangered species—Lange’s metalmark butterfly, Contra Costa wallflower, and the Antioch Dunes evening primrose. The refuge system employs more than 4,000 staff dispersed in its offices across the country. Individual refuges may report directly to a regional office, or may be grouped with other refuge units into a complex under the common management of a project leader. Complexes range in size from 2 to 19 refuges, and one of the refuges in each complex usually serves as the complex headquarters. Complexing has reorganized the 585 refuges into 126 complexes and 96 stand-alone refuges. Officials in headquarters serve as advisors to regional refuge chiefs and to refuge managers. Figure 1 shows the location of the 585 refuges comprising the NWRS. Until the passage of the National Wildlife Refuge System Improvement Act of 1997, the refuge system was the only major federal public lands network without a basic statute providing a mission for the system, policy direction, and management standards for all of its units. The Improvement Act gave the refuge system a unifying mission—to administer a national network of lands and waters for the conservation, management, and, where appropriate, restoration of the fish, wildlife, and plant resources and their habitats within the United States for the benefit of present and future generations of Americans. The legislation also called for FWS to plan and direct the continued growth of the system in a manner designed to accomplish this mission. In addition, the Improvement Act required that the biological integrity, diversity, and environmental health of the refuge system be preserved. The act generally requires refuges to complete comprehensive conservation plans—long-range plans for managing, among other things, habitats and providing visitor services—by 2012. An important component to the act was that it recognized six wildlife- dependent recreational uses of the refuge system—hunting, fishing, wildlife observation, wildlife photography, environmental education, and environmental interpretation—as appropriate uses that are consistent with the mission of the refuge system, when they are determined to be compatible with the purposes of individual refuges. While hunting and fishing have always been popular uses on refuges, wildlife observation is the most prevalent activity on refuges today, and attracted over 23 million visitors in 2006. For the most part, refuges generally perform similar activities that are compatible with the mission of protecting wildlife and habitat and providing visitor services: Habitat management. Refuges manage their lands to provide adequate habitat for the species they were established to conserve and to maintain biological diversity and integrity. Management activities may include (1) performing habitat management work such as maintaining water levels in water impoundments and ponds and treating invasive species; (2) performing fire management activities including conducting prescribed burns; (3) restoring habitat to make it more useful for wildlife purposes; (4) monitoring species and habitat through surveys and other studies; (5) managing volunteers doing habitat- or wildlife-related work; and (6) coordinating habitat management efforts with outside entities, such as private land owners, state agencies, and other groups. Visitor services. Nearly 40 million people visit refuges each year, and the vast majority of refuges provide visitors with the opportunity to participate in one or more of the six wildlife-dependent recreational uses outlined in the Improvement Act. To support these activities, refuges install roads, trails, docks, and boat ramps, and develop interpretive and educational exhibits, among other things. Refuges also perform work that supports both habitat management and visitor services: Maintenance. The refuge system maintains more than $18 billion in real property, including more than 41,000 facilities such as buildings, visitor infrastructure, and roads; more than 4,000 vehicles; and almost 4,000 pieces of heavy equipment. Refuge staff perform preventative maintenance on their refuges’ real property to achieve specific performance targets that are tied to the refuge system’s mission, such as restoring wetlands, monitoring wildlife, and providing recreation opportunities. The refuge system currently has a deferred maintenance backlog, which is described in appendix III. Appendix IV discusses the Refuge Operational Needs System, which maintains information on refuge operational requirements such as staff, equipment, and planned projects. Law enforcement. The refuge system employs law enforcement officers who are tasked with protecting refuges’ natural resources, infrastructure, and the visiting public. Officers also enforce conservation agreements with private landowners. Conservation planning. Refuges are required to complete comprehensive conservation plans that outline priorities for wildlife and habitat as well as visitor services. Wildfire suppression. The refuge system supports wildfire suppression needed on refuge lands as well as other federal lands. The refuge system receives most of its funding for core refuge operations and maintenance activities from FWS’s annual resource management appropriation; funds for fire management to restore and improve habitat as well as wildfire suppression come via a separate appropriation. Funding from several other sources supports other types of refuge system activities. For example, the refuge system receives annual allocations from FWS’s construction appropriation to construct, improve, acquire, or remove buildings and other facilities, and from FWS’s land acquisition appropriation to acquire interests in lands, including easements that provide important fish and wildlife habitat. Refuges also may apply for grants from federal, state, and local governments and nonprofit organizations, among others, to supplement their funding. The Department of Transportation’s Federal Highway Administration (FHWA), through its Public Lands Highway-Refuge Roads Program, provides funds to maintain and improve public roads that provide access to or within a refuge. In addition, the refuge system has a permanent appropriation authorizing refuges to use recreation fees they collect and to accept donations, voluntary services, and in-kind contributions from private conservation groups, such as land or equipment donations. The refuge system receives additional funding through other FWS programs, such as Endangered Species or Fisheries and Habitat Conservation. Like the refuge system, these programs also receive allocations from the resource management appropriation and may, in turn, obligate a portion of this funding to support projects occurring on refuge lands. Figure 2 shows the principal sources of funding for the refuge system. While most refuges carry out the same type of activities, key characteristics of refuges such as acreage, visitation levels, and the type of ecosystems they contain—and consequently the challenges they face—can vary. Sixteen refuges in Alaska account for approximately 85 percent— more than 76 million acres—of the refuge system’s total acreage, and these refuges generally operate somewhat differently than others. The Arctic NWR in northeastern Alaska, for example, contains 8 million acres of wilderness that is relatively undisturbed; as such, activities focus primarily on research, monitoring, and education. In contrast, the Tualatin River NWR—located 15 miles from Portland, Oregon—faces the challenge of protecting natural resources amid rapidly increasing visitation levels. Refuges along the southwest border of the United States, meanwhile, face unique law enforcement challenges as they support the Department of Homeland Security’s border control efforts. From fiscal years 2002 through 2007, the refuge system experienced fluctuations in funding and staffing levels, the introduction of several new refuge system policy initiatives, and increases in the influence of external factors such as extreme weather and development that affect refuge operations. Inflation-adjusted funding for core refuge system activities— measured as obligations for refuge operations, maintenance, and fire management—increased by 6.8 percent from fiscal year 2002 to fiscal year 2003 for the celebration of the refuge system’s centennial, then declined quickly to 4.7 percent below peak levels by fiscal year 2005, before increasing again to 2.3 percent below peak levels in fiscal year 2007, when adjusted for inflation (in 2002 dollars); it ended the period 4.3 percent above fiscal year 2002 levels. In nominal dollars, core funding increased each year over the time period from about $366 million in fiscal year 2002 to about $468 million in fiscal year 2007. Core refuge system staffing levels peaked in fiscal year 2004 after increasing 10.0 percent, and then declined more slowly than funding to 4.0 percent below this level by the end of fiscal year 2007; they ended the period 5.5 percent above fiscal year 2002 levels. During the same period, several refuge system policy initiatives were implemented to reduce staff levels and reprioritize funding among refuges, ensure the completion of required conservation plans, shift focus toward constructing a greater number of smaller visitor facilities, and increase the number of full-time law enforcement officers and associated training; other initiatives increased the administrative workload on refuges. Refuges also experienced an increase in the influence of various external factors that may complicate managers’ abilities to protect habitat and provide visitor services, such as extreme weather events and development on adjacent lands. Obligations for core refuge activities—operations, maintenance, and fire management—peaked in fiscal year 2003, then decreased and remained below peak levels through fiscal year 2007, when adjusted for inflation (in 2002 dollars), but above fiscal year 2002 levels. As shown in figure 3, total nominal obligations for core refuge system activities increased each year from about $366 million in fiscal year 2002 to about $468 million in fiscal year 2007—an average annual increase of 5.1 percent or about $18.5 million. However, when adjusted for inflation, total core obligations peaked in fiscal year 2003 at about $391 million for the wildlife refuge centennial—an increase of 6.8 percent over fiscal year 2002. Core inflation-adjusted obligations then quickly fell back to 4.7 percent below peak levels by fiscal year 2005. By fiscal year 2007, inflation-adjusted core obligations rebounded somewhat to about $382 million—still 2.3 percent below peak levels, but 4.3 percent above fiscal year 2002 levels. While the refuge system did receive an increase in the allocation from the resource management appropriation for fiscal year 2008, we did not include it in our analysis because the fiscal year was not yet complete. At the refuge level, the trends in inflation-adjusted core refuge obligations at the 222 complexes and stand-alone refuges varied considerably during our study period. Specifically, from fiscal year 2002 through fiscal year 2007, core inflation-adjusted obligations decreased for 96 complexes and stand-alone refuges, increased for 92, and stayed about the same for 34. The magnitude of the changes in core funding at the refuge level also were more pronounced than for the trend overall. For example, the refuge with the largest percentage inflation-adjusted decrease in funding was the Kootenai NWR in Idaho, where obligations fell from $957,506 in fiscal year 2002 to $324,283 in fiscal year 2007, a decrease of 66 percent. The refuge experiencing the largest inflation-adjusted dollar decrease was the Mid- Columbia NWR Complex in Washington state, where obligations fell from about $9.4 million in fiscal year 2002 to about $5.2 million in fiscal year 2007, a decrease of about $4.1 million. Moreover, from fiscal year 2002 through fiscal year 2007, core funding for 39 complexes and stand-alone refuges decreased by more than 25 percent. On the other hand, the refuge receiving the largest percentage increase in inflation-adjusted funding was the Caddo Lake NWR in Texas. Its obligations increased by 156 percent, from $95,255 in fiscal year 2002 to $244,094 in fiscal year 2007, largely reflecting an increase in operations at this refuge since it was established in October 2000. The refuge experiencing the largest dollar increase in inflation-adjusted funding was the Okefenokee NWR in Georgia, where obligations for core refuge activities increased from about $7.4 million in fiscal year 2002 to about $15.8 million in fiscal year 2007—an increase of about $8.4 million. However, almost 90 percent of this increase consisted of fire management funding provided largely to respond to the wildfires the refuge faced in April 2007. Appendix V presents obligations for core refuge activities for all 222 complexes and stand-alone refuges in both nominal and inflation-adjusted dollars for each of fiscal years 2002 through 2007. Total obligations, in nominal dollars, for the refuge system were about $816 million in fiscal year 2007. As illustrated in figure 4, core obligations comprised about 57 percent of this total, or about $468 million. Obligations of allocations from Interior’s construction and land acquisition appropriations added approximately 19 percent, or about $154 million. Additional obligations of funds received through recreation fees, donations, conservation funds, and all other sources, contributed approximately 18 percent, or about $148 million. Other funds were obligated from grants and allocations from the FHWA and from other FWS programs. Total: $816 million In contrast to the trend in core funding, total inflation-adjusted funding for the refuge system as a whole did not peak in fiscal year 2003, but instead steadily decreased from fiscal year 2002 levels until fiscal year 2005 and rebounded somewhat thereafter (see fig. 5). Even after rebounding somewhat in fiscal year 2005, however, inflation-adjusted total funding decreased to about $666 million in fiscal year 2007—an average annual decrease of 1.6 percent (about $11.2 million) or 7.5 percent below fiscal year 2002 levels. The main driver in the generally decreasing trend in total funding is a sharp drop in funding for land acquisition, which fell from about $101 million in fiscal year 2002 to about $38 million in fiscal year 2007. In nominal dollars, total obligations increased from about $720 million in fiscal year 2002 to about $816 million in fiscal year 2007—an average annual increase of 2.5 percent or about $18.3 million. Beyond receiving financial resources, refuges also receive in-kind donations from nonprofit groups, for-profit companies, and other organizations. From our survey, we obtained information on donations received by 246 individual refuges—67 percent of refuges responding to our survey. These donations typically consisted of equipment and other supplies that refuges used to help manage habitat or deliver visitor services, though larger donations included land and construction of visitor centers. Donations at most of the refuges totaled $500,000 or less over the entire 6-year time frame; however, several refuges reported that they received more than $1 million over this period. For example, a refuge in Washington state estimated that it received in-kind donations totaling about $20 million, consisting primarily of land donations from nonprofit organizations, bridge work, and habitat restoration projects. Another refuge in Michigan estimated that it received about $5 million worth of in- kind donations, including land donations from local industries, as well as shoreline restoration, fence removal, and tree removal projects performed by these industries. The declining trends in refuge system funding were comparable to general declines in obligations for FWS and Interior overall, although Interior fared somewhat better. FWS fared about the same as the refuge system ending in fiscal year 2007 at 7.5 percent below fiscal year 2002 levels, when adjusted for inflation. Interior overall fared somewhat better over the same period, declining 3.9 percent when adjusted for inflation. Figures 6 and 7 show the trends in nominal and inflation-adjusted obligations, respectively, made by the refuge system, FWS, and Interior from fiscal years 2002 through 2007. Staffing levels, as measured by FTEs the refuge system actually used, peaked later and declined more slowly than funding for both core refuge activities (core staffing) and all refuge activities (total staffing). FTEs for core staffing, which includes operations, maintenance, and fire management, increased from 3,283 in fiscal year 2002 to a peak of 3,610 in fiscal year 2004—an increase of 10.0 percent. Core staffing then fell back to 3,464 FTEs by fiscal year 2007—still 5.5 percent higher than the fiscal year 2002 level, but 4.0 percent below peak staffing levels. While operations and maintenance FTEs increased 3.6 percent overall during our study period, they ended the period down 6.9 percent from their 2004 peak. Fire management FTEs, on the other hand, increased 14.3 percent over fiscal year 2002 levels. Table 1 shows FTE trends for core refuge system activities from fiscal years 2002 through 2007. In contrast with funding, FTEs for noncore activities account for a relatively small portion of the total FTEs that support the refuge system. In fiscal year 2007, for example, 664 noncore FTEs supported the refuge system—about 16 percent of total FTEs—as illustrated in figure 8. Slightly more than 500 of these FTEs were allocated to the refuge system to manage construction projects, land acquisitions, grants, and donations, and to collect fees, among other refuge activities. FWS employees assigned to other agency programs accounted for about 157 of the 664 noncore FTEs for activities in support of the refuge system. For example, biologists from FWS’s Ecological Services program often monitor various species at refuges, supplementing the refuges’ habitat management activities while also furthering Ecological Services’ mission to conserve and restore threatened and endangered species. Similar to the trend in core FTEs, total FTEs used in support of the refuge system overall also peaked in fiscal year 2004 and then decreased through the remainder of the period. As table 2 illustrates, total FTEs increased 5.8 percent from fiscal year 2002 through fiscal year 2004, then declined through 2007, to close the period 0.9 percent higher than the fiscal year 2002 level. This amounted to a 4.7 percent drop from the peak staffing levels of fiscal year 2004. In addition to FTEs, the number of employees on board in refuge system positions also declined after peaking in fiscal year 2004. Through fiscal year 2007, nearly 375 employees were lost from the refuge system’s peak staffing levels, a reduction of 8.4 percent over this period (see table 3). About three-quarters of this loss came through a reduction in permanent employees. Refuge managers and regional and headquarters officials told us that the number of filled, permanent positions at refuges is a key measure of the effective strength of the workforce available to conduct core refuge activities because they represent employees on board indefinitely. Thus, the loss of 275 permanent employees (7.5 percent) since fiscal year 2004—generally through the elimination of vacant positions created by retirements and resignations—has reduced the number of staff available to conduct needed work. For the overall study period, total employees declined 4.0 percent below fiscal year 2002 levels and permanent employees declined 1.7 percent. The overall fluctuation in staffing levels and the reductions since fiscal year 2004 in particular have affected many refuges. During the first 2 years of our study period, from fiscal year 2002 through 2004, 114 complexes and stand-alone refuges increased their permanent staff by more than 5 percent, while only 49 lost more than 5 percent and 55 stayed about the same. However, over the final 3 years, the situation was reversed: from fiscal year 2004 through 2007, the number of complexes and stand-alone refuges that lost more than 5 percent of their permanent staff more than doubled to 122, while only 38 gained at least 5 percent and 58 stayed about the same. The refuge system implemented several policy initiatives from fiscal years 2002 through 2007, including efforts to achieve more sustainable staffing levels, ensure the completion of conservation plans, construct a greater number of small visitor facilities, and modify the refuge system’s law enforcement function. In addition, various refuge system, FWS, and Interior policies increased administrative work for nonadministrative refuge staff during this period. Because core staffing levels peaked later and declined more gradually than the system’s core inflation-adjusted funding, as shown in figure 9, rising salary and benefit costs for these staff began to account for an increasing share of refuge budgets after fiscal year 2003. In many cases, there was an existing imbalance in refuge budgets that meant that personnel costs already were putting pressure on refuges’ ability to operate. Generally, this reduced refuges’ management capability—that is, the percentage of a refuge’s budget available to pay for other operational costs that support its daily work, such as utilities, fuel, supplies, and seasonal labor. Although circumstances varied by refuge, some refuges’ management capability shrank to less than 5 percent of their total budget—a nearly unsustainable operational scenario, according to some refuge managers and regional and headquarters officials we interviewed. To attain a more sustainable balance between staffing costs and management capability, in fiscal year 2006, each regional office was directed to develop a workforce management plan. According to FWS guidance for these plans and interviews with senior refuge officials, regions were instructed to focus on doing “fewer things better,” that is, to allocate limited resources in such a way as to showcase selected refuges, rather than to allocate them across all refuges such that the level of habitat management and visitor services would be equally degraded. Although workforce plans differed by region, they generally proposed to increase management capability to a minimum of 25 percent of refuges’ operating budgets by reducing the share devoted to salaries and benefits to 75 percent or less; reduce staff costs by (1) abolishing staff positions that became vacant through retirements and resignations, and (2) further consolidating refuges into complexes to eliminate redundant positions and reduce administrative costs; categorize all refuges into one of three tiers—called focus refuges, targeted reduction refuges, or unstaffed satellite refuges—to prioritize them for funding and staffing increases or decreases; and realign some vacated positions by moving them from lower- to higher- tiered refuges. Although refuge system management did not intend for regions’ workforce plans to conform to a rigid national standard, program headquarters did provide criteria for regions to use when placing refuges into the following tiers: Focus refuges, where FWS would strive to maintain or enhance field operations, would be selected because of the significance of their natural resources, important opportunities for wildlife-dependent recreation, or other “highly significant” values. Targeted reduction refuges, where reductions in operations would occur, also would be selected on the basis of natural resources, recreation, and other values, but would be considered a lower priority than focus refuges. Unstaffed satellite refuges had no specific criteria, but would include both refuges that have never been staffed and those that were to be destaffed due to budget constraints. Refuge system documents, as well as our interviews and site visits, showed these refuges to be often smaller, more remote, and less complex to manage than those in the upper two tiers. According to refuge system officials, the process for determining staff reductions, realignments, and refuge tiers varied considerably across regions, and refuge managers disagreed over the appropriateness of the methods some regions used. For example, Region 7 (Alaska) designated all of its refuges as focus refuges, while all other regions placed their refuges into each of the three tiers. Of the 275 refuge managers who answered a survey question on this issue, 41 percent responded that the criteria for categorizing refuges into tiers were appropriate to distinguish among the competing priorities in their respective regions, another 37 percent responded that they were not appropriate, and the remainder said they had no basis to judge. While most of the respondents who disagreed with their region’s criteria were from lower-tiered refuges, 25 percent of those who responded this way managed at least one highest-priority, or focus, refuge. Refuge managers acknowledged that additional management capability was necessary for continued operations, and understood that workforce planning decisions that affected funding and staffing levels were inherently difficult. Still, according to a senior regional office official, refuge tiering added to the emotional strain of an already stressed workforce, establishing a “have” versus “have not” mentality that many staff took personally. Implementation of workforce plans shifted funds and staff from lower- priority refuges to higher-priority refuges or from the regional office to the field, and reduced the total number of positions located at refuges. In all, about 375 refuge and regional office positions were either abolished— through elimination of vacant positions—or moved, with managerial, biological, and maintenance positions among those most frequently targeted for reduction or realignment. These changes were responsible for most of the 275 permanent employees who were lost after fiscal year 2004. By design, lower-tiered refuges absorbed a heavier share of these staff cuts and realignments. According to refuge system officials, the $36 million increase in the fiscal year 2008 allocation for the refuge system from the resource management appropriation was being used in part to restore some of these lost positions and funding at targeted reduction refuges. Further, regional officials reported that management capability across each region had reached the desired margin of at least 25 percent of refuges’ operational budgets. However, given that the fiscal year was not complete before the end of our review, we did not obtain additional data on FTE or position changes at refuges. From fiscal years 2002 through 2007, refuge system officials implemented steps intended to ensure that comprehensive conservation plans mandated by the Improvement Act are completed on time. In early fiscal year 2004, refuge officials realized that they were not on track to complete the 554 conservation plans required by 2012—the due date mandated by the act. At that time, refuge officials assessed needs and goals with regard to completing the plans, and provided recommendations to encourage timely completion as well as a monitoring and evaluation strategy. At the end of fiscal year 2005, however, refuge officials noted that the completion of these plans was still behind schedule—only 19 percent of the plans were complete even though more than half of the 15-year time frame had elapsed. To help ensure that the plans would be completed on time, refuge system officials required, among other things, refuge managers to identify work that could be set aside to focus on completing conservation plans. They also required regions to develop completion dates and milestones for completing the plans and to use a central database for tracking milestones. To date, about half of the 554 plans have been completed and about one- third are underway. Refuge officials said that they believe that they can meet the deadlines; however, some plans are still behind schedule. In 2003, in response to discussions with Congress about how best to fulfill the requirement to provide the six wildlife-dependent activities described in the Improvement Act, the refuge system began an initiative to place greater emphasis on constructing small visitor facility structures, such as observation decks, informational kiosks, and restrooms, instead of larger visitor centers. These small structures are less expensive than visitor centers—which in 2007 were estimated to cost an average of $5.7 million each—and can be completed more quickly. Thus, a larger number of refuges can receive visitor facility funds, enabling refuge system investments to benefit a larger number of visitors. For fiscal years 2003 through 2007, the refuge system directed about $28 million toward these projects. Nevertheless, large visitor centers continue to be funded, and the refuge system was appropriated more than $51 million for visitor center construction from fiscal years 2002 through 2007. Figure 10 shows examples of visitor facility infrastructure. In July 2002, in response to safety concerns, the Secretary of the Interior directed the refuge system to begin an initiative to modify its law enforcement program by, among other things, increasing the training requirements for officers, reducing the system’s reliance on dual-function officers—staff with other primary duties who perform law enforcement duties part time—and creating an officer deployment model. Specifically, the refuge system increased the training requirements for law enforcement officers from 18 weeks to about 30 weeks from 2002 through 2007, with the vast majority of that increase coming from a new field training evaluation program. Refuge system officials also required some senior- level staff, and all dual-function officers who were performing law enforcement functions less than 25 percent of their time, to cease performing their law enforcement duties beginning in 2003. As a result, the refuge system reduced the number of dual-function officers from 495 to 164 and hired 76 full-time officers who serve a single refuge. FWS also created a “zone officer” position to serve multiple refuges and has hired 45 of these officers. In 2005, the International Association of Chiefs of Police released a law enforcement deployment model it developed with the refuge system to identify the level of law enforcement personnel needed to provide adequate protection of refuge resources and the public, and where those officers should be deployed. The model recommended a total of 845 law enforcement FTEs for the refuge system—about 600 FTEs more than the refuge system currently has on board. Given the refuge system’s current funding situation and the chances of the system attaining such a level, senior refuge officials have identified 450 positions as the minimum number they believe necessary to provide adequate protection. Refuge officials are hoping they can approach that minimum number by hiring an additional 200 officers in the near term. During our study period, refuge managers told us that they began feeling the burden of a myriad of new administrative work, especially work that applies to nonadministrative staff, resulting from refuge system, FWS, and Interior policies. Ninety-three percent of refuge managers who responded to our survey said administrative duties for nonadministrative staff have increased since 2002; less than 2 percent of refuge managers reported a decrease in this workload. For example, refuge managers expressed concerns that the number of national reporting requirements and extent of mandatory training classes, among other administrative tasks, were burdensome. Furthermore, the refuge system created a new maintenance database that required much more data entry than the previous system. In addition, managers indicated that they have been receiving an increasing number of data calls over the years. In 2003, refuge managers began an effort to address increasing administrative requirements and more than 300 refuge managers participated in discussions about the problem. Several managers formed an ad hoc committee in 2003 to address the issue and several officials from headquarters and regional offices joined the effort in 2004. Together, they drafted a white paper that provided several recommendations to reduce the burden. A headquarters team took the effort over and, in October 2007, released a report detailing their findings and 17 recommendations for reducing some requirements, such as reviewing national reporting requirements and eliminating those deemed unnecessary, as well as making Web-based training optional. According to refuge officials at headquarters, FWS is beginning to implement these 17 recommendations. A variety of factors that were generally outside the control of refuge system management became more influential between fiscal years 2002 and 2007. Some of these factors were natural occurrences, such as extreme weather, while others were due to the intensification of human activities, such as development. These factors added to refuge workload, complicating managers’ abilities to protect habitat quality and provide visitor services. One commonly cited external factor was extreme weather events such as droughts, floods, and severe winds. Survey results show that the contribution of extreme weather events to habitat problems increased at 52 percent of refuges; only 2 percent reported a decline. Storm damage also increased at many more refuges than it decreased: in particular, hurricanes Ivan, Katrina, and Rita in 2004 and 2005 damaged large parts of refuges in the southeastern United States. Eighteen refuges in three states were temporarily closed to the public as a result of these storms; a 19th refuge— Louisiana’s Sabine NWR—remains closed to public use. According to FWS, storm damage to the refuge system in 2005 alone exceeded $300 million. Development pressures caused by the expansion of urban areas and problems associated with the conversion of off-refuge land to agriculture or industrial use also increased during this period. Refuge managers reported that human settlement infrastructure such as roads, housing, and airports increasingly contributed to refuge habitat problems between 2002 and 2007—around 46 percent of refuges. These development pressures can contribute pollution to refuge lands and waters and make it more difficult to maintain viable, interconnected habitat in and around a refuge’s borders. Moreover, increasing development around refuges can be accompanied by an increase in the demand for recreational uses of nearby refuges, including some uses—such as recreational boating or rock- climbing—that may be incompatible with a refuge’s established purpose. In addition, the influence of off-refuge agricultural and industrial activities increased for many more refuges than it decreased. Because refuges do not exist in isolation, they must be managed in concert with adjacent lands to maintain healthy habitats, a reality that requires managers to allocate time to spend away from their refuges to develop working relationships with adjacent and upstream landowners. Other external factors affecting the refuge system include inadequate water rights and rights-of-way, such as public roads that divide refuge lands; the impacts associated with these factors worsened at more refuges than they improved, though they were stable for almost half of all respondents. Additionally, almost a quarter of the responses to our survey identified impacts associated with climate change as one of the biggest threats to habitat condition throughout the system. Managers reported that they already are seeing effects that they attribute to climate change, including drying of wetlands and wildfires of increased frequency and intensity. In addition to the obvious effects on habitat condition, these disturbances can affect wildlife-dependent visitor services, such as hunting or photography, to the extent that they change waterfowl migration patterns or the ranges of land- and water-based wildlife that historically are native to a given refuge. From fiscal years 2002 through 2007, several changes occurred in refuges’ habitat management and visitor services, creating concerns about the refuges’ abilities to maintain high-quality habitat and visitor services in the future. While 28 percent to 40 percent of habitats on refuges for several types of key species improved between fiscal years 2002 and 2007, conditions at some refuges worsened and 7 percent to 20 percent of habitats were in poor condition in 2007. Refuge habitats are facing growing pressure from increasing habitat problems and external factors, and although most refuges increased time spent on habitat management activities, there is increasing concern from managers that staffing and funding constraints will inhibit the ability of refuges to maintain quality habitat in the future in light of increasing habitat problems and resource constraints. The quality of visitor services improved on one-fifth to nearly one-half of refuges between fiscal years 2002 and 2007, but environmental education and interpretation programs were of poor quality at about one- third of refuges in 2007. While some refuges have been able to increase the time spent on visitor services, refuge managers are concerned about their ability to provide high-quality visitor services to the public given recent funding and staffing changes. While 28 percent to 40 percent of habitats on refuges for several types of key species improved between fiscal years 2002 and 2007, conditions of 11 percent to 18 percent of refuge habitats worsened and 7 percent to 20 percent of habitats were in poor condition in 2007. Habitat problems and external factors are increasing at refuges, and most refuges increased the time spent on habitat management activities. However, managers are concerned that staffing and funding constraints will inhibit the refuges’ ability to maintain quality habitat in the future. Refuge managers reported that habitats for five key types of species we surveyed refuges about improved between 2002 and 2007 about two times as often as they worsened (see table 4). Tualatin River NWR outside of Portland, Oregon, for example, saw a marked improvement in wetland habitat, according to the refuge manager, as the refuge has begun to address an invasive weed infestation over the past year. The refuge has been addressing two primary invasive plants— knotgrass and cocklebur—that had infested approximately one-third of the refuge’s 600 wetland acres since 2003, overtaking the native wetland plants that thousands of birds rely on for food during migration. Through herbicide application, mowing and discing, and water level manipulation, the refuge was able to cut infestations in half over the last year, bringing the habitat back up to sufficient quality for use by the migrating birds. We observed the results of some of these activities to remove knotgrass (see fig. 11). fig. 11). Refuge managers also reported that 11 percent to 18 percent of habitats on refuges for key species have worsened since 2002. Camas NWR in Idaho, for example, has faced a drought for the last several years. According to the refuge manager, the lack of water has negatively impacted a riparian zone of cottonwoods and willows that migrating birds, such as yellow warblers, use during migration. In addition, many of the trees in the riparian area are close to 100 years old and are dying. Currently, the refuge is working on a plan to restore the vegetation, relying in part on wells for irrigation, but expects it will take decades to restore. As might be expected, we found differences in changes in the quality of habitat for waterfowl and other migratory birds between focus and targeted reduction refuges when compared to unstaffed refuges (see figs. 12 and 13). Specifically, we found that managers at focus and targeted reduction refuges were significantly more likely to report that habitat quality for waterfowl improved between 2002 and 2007 than at unstaffed satellite refuges. For example, between fiscal years 2002 and 2007, more than twice as many focus refuges experienced improved waterfowl habitat (42 percent) as experienced worsened waterfowl habitat (20 percent). At unstaffed satellite refuges, by contrast, habitat for waterfowl worsened almost as frequently as it improved, with 20 percent of refuges experiencing improved quality and 16 percent experiencing worsened quality. We found a similar situation for other migratory birds. We also found that these relationships generally remain strong in statistical models that simultaneously account for the effects of the change in staff time and the change in external factors, such as extreme weather and agricultural activity, which can contribute to habitat problems. For example, based on these models, we estimate that focus refuges were 3.4 times more likely than unstaffed satellite refuges to experience improved rather than worsened habitat quality for other migratory birds and that targeted reduction refuges were 3.9 times more likely. In addition to analyzing the change in quality by tier, we analyzed changes in the quality of habitat as a function of the time spent by permanent staff at a refuge on habitat management activities. We found that refuge managers were more likely to report that habitat quality improved at refuges that increased the time spent on habitat management since 2002 than for those that reduced time, and were less likely to report that habitat quality worsened (see fig. 14). These results were consistent with our analysis of the change in quality of habitat as a function of staffing level changes at complexes and stand-alone refuges. The odds of habitat for waterfowl improving rather than worsening were significantly higher at refuges where staff time on habitat management activities increased rather than decreased between fiscal years 2002 and 2007. For example, among refuges where staff time increased, more than three times as many refuges experienced improved habitat for waterfowl (47 percent) as experienced worsened habitat (14 percent). In contrast, among refuges where staff time decreased, nearly the same number of refuges experienced improved habitat for waterfowl (30 percent) as experienced worsened habitat (27 percent). We found similar results when comparing change in staff time with the change in habitat quality for other migratory birds (see fig. 15). When we developed statistical models of habitat change, refuges where staff time on habitat management activities increased were about 3.0 times more likely than refuges where staff time decreased to report improved, rather than worsened, habitat for both waterfowl and other migratory birds, even after accounting for the effects of tier designation and the change in external factors, such as extreme weather and agricultural activity, that may cause habitat problems. Resource prioritization at refuges obviously influences the ability of refuges to maintain quality habitat. Refuge managers told us that decisions on how many resources to direct to refuges are based on a variety of factors. Some managers weigh the management needs of all the refuges within a complex, taking into account the relative importance of the habitats as well as the amount of time and resources needed for the management activities. Unstaffed satellite refuges generally are smaller and have lower-priority habitats and, in some cases, the refuges are limited in what management can do, according to managers. Because some unstaffed refuges are located some distance from equipment and supplies and from where refuge staff are located, these distances and associated costs are taken into account as well. Other managers told us that they will undertake efforts mainly in response to specific problems identified at these refuges, while otherwise they generally do not expend resources for habitat management. According to refuge managers, it often is difficult to know what needs to be done at unstaffed refuges because staff generally do not visit the refuges very frequently to monitor the habitats. Even though the condition of many habitats is improving, many of these are still not high quality. Specifically, 40 percent of waterfowl habitats that improved since 2002 were still of moderate quality or poorer in 2007, while 65 percent of habitats that stayed in the same condition were of moderate quality or poorer. Similarly, for other migratory birds, 40 percent of habitats that improved in condition since 2002 were of moderate or poorer quality in 2007 and 55 percent of habitats that stayed in the same condition were of moderate quality or poorer. Refuge managers reported that, on average, habitats on about 44 percent of refuges for each of several types of key species—waterfowl, other migratory birds, threatened and endangered species, candidate threatened and endangered species, and state species of concern—were of high quality in 2007. A similar percentage of refuges deemed their habitats to be of moderate quality, and 7 percent to 20 percent of refuges reported habitats to be of low quality, depending on the species type (see table 5). Habitat quality is determined by the availability of several key components, including fresh water, food sources, and nesting cover, among other things, and the absence of habitat problems, such as invasive species. High-quality habitat generally provides adequate amounts of each of these main habitat components and is not significantly affected by habitat problems, while low-quality habitat generally lacks these components and may have significant problems. Moderate-quality habitat has a mixture of good and bad attributes. For example, a habitat may have an excellent tree canopy that provides good nesting areas and protection, but the underlying vegetation may be inadequate as a food source due to an infestation of an invasive species that has driven out native plants. Some aspects of moderate-quality habitat are acceptable, but the problems must be addressed overall for these habitats to fully support the species that depend on them, according to managers with whom we spoke. Other migratory birds, threatened and endangered species, and state species of concern appear to be faring the best with 47 percent or more of habitats on refuges deemed to be of high quality and 13 percent or less of habitats of low quality. Habitats for waterfowl and species that are candidates for listing under the Endangered Species Act are doing somewhat worse. Refuge managers told us that these findings may in part reflect the difficulty in addressing the very specific habitat needs of these species and a lack of focus on addressing those needs because they are not yet listed under the act. We spoke with some managers who have areas of robust, high-quality habitat on their refuge. For instance, Cache River NWR in Arkansas has approximately 65,000 acres of bottomland hardwood—45,000 acres of which are in pristine condition, with the remaining 20,000 acres in the early stages of regrowth after being restored from prior agricultural use. This refuge serves as an annual wintering area for 250,000 to 500,000 waterfowl including mallards, pintail, widgeon, gadwall, teal, and wood ducks. The refuge eradicated invasive kudzu plants using herbicides and its high-quality habitat provides necessary food, water, and cover for these waterfowl and also supports a variety of other migratory birds including warblers, indigo buntings, bluebirds, shorebirds, and wading birds. However, we also spoke to managers who reported low-quality habitat on their refuges. At Bowdoin NWR in northern Montana, for example, habitat is compromised by water quality and quantity problems as a result of activities on nearby lands, including haying and cropland use. The quantity of water that the refuge receives is insufficient to allow adequate flow- through of the water supply and, as a result, the water available for the refuge contains high levels of residual salt as well as agricultural chemicals, which affect the composition of vegetation and the survival of invertebrates. While a variety of bird species uses the refuge, including waterfowl, shorebirds, bald eagle, peregrine falcons, and piping plover—a federally listed threatened and endangered species—some populations of these species have declined over time. In addition to habitat quality, whether a refuge’s habitat is meeting the needs of key species types is an important indicator as to a refuge’s effectiveness in meeting its conservation mission. Our survey found that refuge managers reported that habitats at a majority of refuges were meeting the needs of key species types to a moderate or large extent in 2007 (see table 6). Some refuge managers indicated that poor-quality habitat could still meet the needs of some species to a large extent, just as high-quality habitat could fail to meet the needs of some species to a large extent depending on the species’ needs. High-quality habitat could fail to meet the needs of a given species if, for instance, the species’ population was too large for the refuge to support or if other species were competing for the same refuge habitat, according to managers. In contrast, a habitat of moderate quality could meet the needs of a species if that species population was small. Species that are candidates for listing under the Endangered Species Act appear to be faring the worst, while other migratory birds appear to be faring the best with regard to how well the habitat is meeting species’ needs. It is important to note that wildlife refuges are not necessarily intended to provide habitat for all types of species—some refuges were established to serve the needs of specific species such as waterfowl or a particular endangered species, for example—and do not necessarily focus on providing habitat for other species. When managers were asked to rate the importance of their habitat for the different types of key species we asked about, some of the refuge managers that reported low-quality habitats also rated the habitat as having low importance or priority for the species in question. For instance, about 56 percent of waterfowl habitat that managers reported as low quality, they also considered that habitat on their refuge to be low-priority habitat or not a priority for waterfowl; they considered about 28 percent of low-quality habitat for other migratory birds to be low-priority habitat or not a priority. Refuge managers reported that many refuges were negatively affected by a number of problems and external factors, including invasive species, habitat fragmentation, water quantity and quality problems, and soil erosion, and that these problems and factors were increasing (see table 7). Invasive plant infestation was the most frequently reported problem, cited as a large problem on more than half of refuges and a moderate problem on nearly a quarter of refuges, and was reported to be increasing on more than half of refuges. This concern is consistent with information from the most recent refuge system performance report as well, which shows that more than 2.3 million acres of refuge lands are infested by invasive plants and more than 80 percent of refuges have at least some invasive plants present. Refuge managers with whom we met during site visits stressed that invasive plants have become a problem that affects the quality of their refuges’ habitats and threatens the quality of the refuge system as a whole. According to managers, these invading plants overtake native plant species that are used by animals for food and shelter, and have deleterious effects on biological diversity. For instance, the refuge manager at Merritt Island NWR in Florida told us that 30,000 acres of habitat on the refuge are infested with invasive plants including Australian pine, Brazilian pepper, Old World climbing fern, guinea grass, and cogongrass. In fact, the Brazilian pepper has infested every one of the 75 water level control structures on the refuge, with some impoundments more than 50 percent overtaken by the invasive plant. These invasive plants are eliminating native habitat and negatively impacting migrating birds such as rails, bitterns, sparrows, Florida scrub-jays, and other species, according to the refuge manager. We observed a common invasive plant, purple loosestrife, at several refuges we visited. Purple loosestrife crowds out native plants and can dramatically reduce food, shelter, and nesting sites for wetland-dependent species (see fig. 16). The refuge manager responsible for Antioch Dunes NWR north of San Francisco, described her refuge as being “in a constant uphill battle” against invasive plants, including vetch, thistles, and various grasses. The refuge is home to two endangered plants that depend on dune habitat on the refuge—the Contra Costa wallflower and the Antioch Dunes evening primrose—as well as the Lange’s Metalmark butterfly, a federal endangered butterfly species that occurs only on this refuge. However, the refuge has been inundated with a variety of invasive plants that, if not constantly addressed, threaten to overtake the native habitat, including the naked stem buckwheat on which the butterfly depends. The refuge manager told us that the butterfly population declined for 4 consecutive years, losing about 50 percent of the population each year, but increased in fiscal year 2007. In addition, refuge officials told us that invasive plants, like many other problems, can worsen if they are not dealt with swiftly. For example, a refuge may be able to completely eradicate an invasive plant if it addresses it early and thoroughly. In some cases this may require actions for several years in a row, as invasive plants frequently require consistent investment and treatment strategies from year to year. If not treated early, the infestation may spread exponentially and become a serious, long-term problem. Gains made in one year can be lost many times over if control efforts are not sustained. Invasive animals also are problematic for refuges and were reported to be a large problem on one-fifth of refuges. For example, nutria, a large rodent species from South America, has infested refuges in east, west, and Gulf coast states. Nutria can wreak havoc on water level control at refuges by burrowing into and destabilizing streambanks and damaging water control structures (see fig. 17). Habitat fragmentation was the second-most frequently identified problem for refuges, reported as a large problem at 44 percent of refuges, and increasing on 57 percent of refuges. Habitat fragmentation occurs when corridors of continuous habitat are disrupted, often by human development activities, which affects the refuge system’s ability to accomplish its wildlife conservation mission. The seriousness of this issue was highlighted at a recent meeting of the Western Governors’ Association by the release of a report on wildlife corridors. Specifically, the report discusses the rapid changes due to development across the United States—but in the West in particular—and how this adversely affects wildlife, and emphasizes the need for habitat connectivity for species survival. Some species, for example, require large areas of homogenous habitat for successful nesting, foraging, or movement. Managers at refuges close to urban centers showed us examples of development adjacent to their refuge that have cut off natural habitat corridors, which can lead to animals trying to cross busy roads or can cut them off from other members of their species leading to genetic homogeneity and inbreeding. For example, the refuge manager at Great Swamp NWR in New Jersey told us that increased development surrounding the refuge has fragmented or eliminated habitat. Valuable woodlots adjacent to refuge lands are decreasing in size or disappearing altogether around the refuge, limiting suitable nesting areas for species such as the red-shouldered hawk—a state threatened species. In addition, the manager said that movement by the bog turtle—a federal threatened species—has been constrained by fragmentation of its habitat. Managers of more rural refuges talked about increasing pressures to convert lands to agricultural uses, citing factors such as the increasing price of corn, or to industrial uses, such as oil and gas development. Habitat fragmentation sometimes occurs within a refuge’s “approved acquisition boundary.” A refuge’s approved acquisition boundary delineates an area that has been approved for inclusion in a national wildlife refuge but does not necessarily indicate that the entire area inside this boundary has been—or ever will be—acquired by FWS. An important conservation strategy for the refuge system that was codified in the Improvement Act is the ability to acquire important habitats, when possible. Thus, many refuges have acquisition plans for lands adjacent to or near existing refuge lands to complete or supplement current refuge habitat. For example, the acquisition plan for Nisqually NWR outside of Olympia, Washington, includes 4,470 acres for eventual purchase within its approved acquisition boundary (see fig. 18). The refuge manager at Nisqually reported that increasing urban development is one of the biggest problems facing the refuge. In addition to impacts such as reduced water quality and increased crime, the manager told us that quality habitat around the refuge is being lost to development despite an active refuge acquisition program, because the refuge cannot address all habitat and land protection needs at the pace necessary to offset habitat loss. Overall, the refuge system has purchased only limited amounts of land within the last 5 years, growing the system at approximately 0.25 percent per year. When asked about their ability to manage an even larger refuge system—a logical concern given the funding and staffing concerns currently facing the refuge system—several managers were quick to point out that simply protecting lands from development was a critical first step in conserving wildlife, even if they did not have the resources to actively manage the land. As noted previously, refuge managers also reported a number of external factors that contribute to habitat problems on refuges, including extreme weather and development, and that the contribution of these factors increased during our study period. Refuge managers told us that extreme weather has caused water levels to vary, which can result in the drying of wetlands, increased fire, and actual changes in the size and location of species ranges. Development activities also can increase air, soil, and water pollution to refuge lands and waters. For example, the refuge manager at Ridgefield NWR, which lies in a small watershed in southern Washington, told us increased urban development means more impervious surfaces such as roads, driveways, and sidewalks, thus increasing the amount of water polluted with oil, gasoline, yard chemicals, and animal waste, among other things, that runs directly into the refuge during rainfall, and decreases the amount of water that permeates the soil. In addition, development can increase visitation at refuges, which can negatively impact refuge resources. For instance, the refuge manager at the Upper Mississippi NWR told us that recreational boating has led to trash dumping, trampling of habitat, and excessive noise. Likewise, Ridgefield NWR must deal with increased litter, illegal dumping, increased trespassing, and damage to the habitat from increased refuge visitation levels. Agricultural activities near the refuges also can contribute pollutants to refuge lands and waters from runoff from animal waste and fertilizers, for example. Not surprisingly, managers reported that habitat was more likely to worsen at refuges where there was an increase in external factors that contribute to habitat problems, such as extreme weather and off-refuge agricultural activities (see fig. 19). For example, among refuges that reported no net increase in external factors, about 9.5 times more refuges reported improved waterfowl habitat (38 percent) than reported worsened waterfowl habitat (4 percent). By contrast, among refuges that experienced a net increase in external factors, the number of refuges that experienced improved waterfowl habitat (38 percent) was much closer to the number that experienced worsened habitat (22 percent). We found similar results for the change in other migratory bird habitat (see fig. 20). Based on our statistical models, which assess the effects of a change in external factors while adjusting for the effects of tier designation and the change in staff time, we estimate that refuges that experienced no net increase in the number of external factors were about 7.0 times more likely to experience improved, rather than worsened, waterfowl habitat quality and 5.1 times more likely to experience improved, rather than worsened, habitat quality for other migratory birds. Refuge managers reported increasing the time spent on a number of key habitat management activities on many refuges between fiscal years 2002 and 2007 (see table 8). Not surprisingly, given the number of refuges with invasive plant problems, refuge managers reported somewhat or greatly increasing time spent addressing invasive plant infestations on 61 percent of refuges, while somewhat or greatly decreasing time spent on this activity at only 9 percent of refuges. The refuge system’s national strategy for managing invasive species states that “invasive species are, collectively, the single greatest threat to native plants, fish, and wildlife with the potential to degrade entire ecosystems;” however, eradicating or effectively controlling invasive species through actions such as controlled burning, mowing, manual removal, and herbicide application is often resource intensive. For example, Antioch Dunes NWR is using a combination of managed grazing, herbicide application, manual removal, and tractors and other equipment to prevent several invasive species from harming two endangered plant species and an endangered butterfly species. In fiscal year 2007, about $9.8 million was budgeted for specific invasive species activities. For example, about $2.3 million was budgeted for invasive species strike teams in five specific areas of the country. These teams were designed to eradicate newly identified infestations before they become widespread with the goal of saving substantial funds in the long run. In addition, more than $500,000 was budgeted for mapping and tracking invasive plants on refuges and coordinating volunteer work for invasive species control activities. Eradication programs for specific species also were budgeted funds, such as $700,000 for nutria eradication at the Blackwater NWR and Eastern Neck NWR in eastern Maryland and the Southeast Louisiana Refuges Complex; over $1.3 million for spartina grass eradication at the Willapa NWR and Grays Harbor NWR in western Washington; and about $200,000 for an exotic rodent species at the Pacific Remote Islands NWR Complex. In addition to these specific programs, refuge managers also may spend resources on other invasive species- related activities on their refuges. Despite these investments, however, performance data from fiscal year 2007 show that only about 12 percent of the acreage identified as infested with invasive plants was treated in fiscal year 2007. The estimated cost of unfunded invasive species control projects found in the refuge system’s operational needs database was more than $150 million dollars at the end of 2007. The refuge system’s national strategy for invasive species control states that nearly half of all refuges report that invasive species infestations interfere significantly with their wildlife management objectives. Given that refuge managers reported that invasive plants were increasing at many refuges and that new invasive species gain a foothold in the United States every year, refuges will likely be constantly battling this problem. As might be expected given reported improvements in habitat quality during our study period, refuge managers reported increasing the time spent on basic habitat management activities such as haying, mowing, prescribed burning, or manipulating water levels on 43 percent of refuges and increasing time spent on habitat restoration activities, such as planting native grasses or trees, and creating water control structures, such as levees, at about 48 percent of refuges since fiscal year 2002. Basic management activities represent the day-to-day work that refuge staff perform to protect, conserve, and improve habitat conditions. Refuge managers, biologists, maintenance workers, and others routinely monitor water impoundments to ensure water levels are optimal for migrating birds, for instance. Restoration projects often have longer-term timelines and are meant to re-establish native habitats on the refuge. Such restoration projects—which may be as extensive as a restoration of a 100- acre grove of bottomland hardwood forest that could take nearly 50 years to complete, or as small as constructing a water impoundment to flood a small wetland area that could be constructed in a couple of months—are key to attracting and sustaining wildlife populations. Managers reported decreasing the time spent on basic management and restoration activities at 18 percent and 14 percent of refuges, respectively. Refuge managers’ responses on changes in the time spent on inventory and monitoring surveys of habitat conditions and wildlife populations— which are key activities that allow refuges to identify, report on, and manage wildlife populations and specific problems—were more mixed. Refuge managers reported increasing the time spent on these activities at 40 percent of refuges, while decreasing time spent on 20 percent of refuges since 2002. These surveys are an important way to understand how well wildlife populations and habitats are doing and whether a refuge is accomplishing its habitat management goals. Managers told us that having accurate data on habitat conditions and wildlife populations, among other things, is critical as they develop and deploy their comprehensive conservation plans. In addition, a number of refuge managers told us that their comprehensive conservation planning efforts led them to increase the amount of survey work they conducted, as the planning efforts require baseline data on habitat and wildlife conditions. For example, Rappahannock River Valley NWR in Virginia increased its inventory and monitoring surveys, partly to support development of its comprehensive conservation plan. In addition, accurate data from these surveys are important for correctly reporting data for FWS’s annual performance report and for early identification of problems affecting habitat. In the case of invasive species, for instance, a small infestation can spread exponentially over a very short time period. Some managers indicated that they are conducting fewer surveys, some of which are of lower quality. The managers said that they are depending more on volunteers or temporary workers to do the surveys, which can limit the survey quality because volunteers may not have the requisite background, experience, or training in biological survey methods. Some refuge managers told us that they have had to cut back on needed survey work due to staffing and funding shortfalls, while a few others told us that increasingly available and easy-to-use technologies have helped them increase the amount of survey work being done. Relatedly, refuge managers reported increasing time spent developing comprehensive conservation plans at nearly 60 percent of refuges since 2002—about two-thirds of refuges reported engaging in planning activities in fiscal year 2007. This is not surprising, given the requirement that all refuges must complete their plans by 2012, and the fact that less than 50 percent have been completed to date. The conservation planning process can be time consuming, given the need to hold public meetings, conduct environmental reviews, and coordinate with state and local entities. During our interviews, several refuge managers told us that they generally did not obtain additional staff to develop these plans; instead, they have had to shift responsibilities from existing staff or curtail other refuge management activities to devote time to the plans. For example, one refuge manager reported that the refuge set aside an invasive species eradication project after several years of implementation in order to work on the conservation plan for the refuge. To minimize the time such planning has taken away from other refuge activities, some refuge managers we interviewed told us that they or their staff worked on the plans on their own time. One refuge manager reported that he worked on the comprehensive conservation plan on weekends and converted the refuge’s biologist to a full-time planner on a temporary basis. Another manager stated that refuge staff attended fewer public meetings that were not related to the comprehensive conservation plan so they would not have to cut back on other refuge work. Consistent with the importance that FWS places on working with owners of lands adjacent to and near refuges, refuge managers reported coordinating with landowners at more than 85 percent of refuges in 2007 and increasing the time spent on this activity at 50 percent of refuges since 2002. This coordination with adjacent landowners is increasing as concerns about habitat fragmentation and off-refuge pollution grow, and FWS increasingly is considering the need to deal with the broader ecosystems of which the refuges are a part. In addition, two refuge managers noted that coordination with adjacent landowners has the added benefit of being a good outreach tool and of giving the refuge, and the refuge system, a better image and a more positive status with the broader public. Some refuges have employees specifically designated to undertake these efforts, while others depend on the work of the refuge manager for these efforts. Permanent staff and volunteers increased the time spent on habitat management activities at 48 percent and 45 percent of refuges, respectively (see table 9). However, consistent with shifts in resources as a result of workforce planning, permanent and temporary staff time spent on habitat management decreased at 29 percent and 19 percent of refuges, respectively. Refuge managers have discretion over the activities on which staff spend their time. For instance, individual refuges face different challenges—due to such things as natural weather cycles and increases or decreases in habitat problems or external factors affecting habitat—that managers need to address. Other influences over how to prioritize staff time may result from management decisions, such as the decision to focus more on working with adjacent landowners, or statutory requirements, such as the comprehensive conservation plans that must be completed. In discussing habitat management activities with refuge managers, managers indicated that unstaffed satellite refuges will generally only see habitat management work on an infrequent or “as-needed” basis. For instance, the manager for a refuge complex in southern Washington state told us that the complex’s unstaffed satellite refuge, which is not open to the public, gets management attention only during a few weeks per year when equipment and staff are available. Similarly, an unstaffed refuge in a complex in New Hampshire has been visited by refuge staff only five times in the past 7 years, according to the complex manager. Based on our site visits, we learned that some refuges were attempting to address reductions in permanent staff by relying on volunteers or contractors more heavily. However, our survey results indicated that refuges that increased permanent staff time on habitat management were more likely to increase time spent by volunteers and contractors than refuges that decreased permanent staff time (see table 10). While volunteers and contractors perform important functions, they cannot replace refuge staff because they must be managed, trained, and supervised. Across the board, if a refuge reported an increase in time spent on habitat management by permanent staff, they generally also spent more time on habitat management activities by all other type of worker; rarely did these refuges reduce the amount of time other workers spent on these activities, indicating the importance—as nearly every refuge manager we spoke with did—of permanent staff in order to carry out needed refuge work. For refuges that reduced permanent staff time on habitat management activities, the results were much more mixed. These refuges reported significantly more decreases in the amount of time that other workers spent on habitat management activities, and it does not appear to hold true that refuges that reduce staff necessarily rely more heavily on other types of workers. In particular, refuges where permanent staff time on habitat management activities decreased were significantly more likely to report decreases in time spent by volunteers at refuges and less likely to report increases in volunteer time, as compared to refuges where permanent staff increased. This likely reflects the catch-22 that refuge managers face with regard to volunteers—while having people interested in helping the refuge can help relieve their workload, volunteers still need direction, oversight, and sometimes training by refuge staff. While some refuge managers indicated that they would not be able to accomplish their habitat management objectives without their volunteer corps, there also is concern about the over-reliance on volunteers to assist with these activities because their availability over the long term is not guaranteed and volunteerism levels can fluctuate greatly. In light of increasing problems and threats affecting refuge conditions, as well as recent funding and staffing constraints, refuge managers and regional and headquarters officials expressed concern about refuges’ abilities to sustain or improve current habitat conditions for wildlife into the future; our survey results corroborate these concerns. While each refuge operates under unique circumstances and faces unique habitat challenges, refuge managers across the system are concerned about a variety of specific threats to their individual refuge habitats and to the refuge system as a whole. Although our survey results do not indicate major declines in habitat quality since fiscal year 2002, many managers are concerned about their ability to maintain quality conditions in the future. Even though our survey showed that a large number of refuges increased staff time on habitat management activities, some refuge managers we interviewed explained that staff were simply working longer hours to get the work done. Several refuge managers repeatedly indicated that they are still trying to do everything possible to maintain adequate habitat, especially habitats for key species, such as waterfowl, other migratory birds, and threatened and endangered species, despite growing habitat problems and other factors affecting refuge habitats and an increasing administrative workload that reduces the amount of time refuge staff can spend performing habitat management work. Several managers said that attention to key habitats is the last thing that will stop receiving management attention in the event of declining funding. They told us that refuge staff are very dedicated to the purpose and mission of the refuge system, but that they fear employee burnout. Several managers even said that they have to limit the amount of time staff spend at the refuge, as these employees are working overtime without extra pay. The quality of visitor services improved on one-fifth to nearly one-half of refuges between fiscal years 2002 and 2007, but environmental education and interpretation programs worsened at some refuges and were of poor quality at about one-third of refuges in 2007. Although some refuges have increased the time spent on these services, refuge managers are concerned about their continued ability to provide high-quality visitor services to the public given recent funding and staffing changes. Encouragingly, visitor services quality was reported as staying stable or improving since 2002 by the vast majority of refuge managers responding to our survey. Most notably, environmental education and interpretation programs showed the largest percentage of refuges reporting improvement, although these programs also showed the largest percentage reporting declines as well, as compared to other visitor services (see table 11). Our survey found that four of the six key visitor services provided to the public were of moderate or better quality at most refuges in 2007, but environmental education and interpretation were reported to be low quality at about one-third of refuges (see table 12). Visitor services deemed moderate quality did not invoke the same level of concern from refuge managers as did habitat deemed moderate quality. While managers would prefer high-quality programs, moderate quality does not jeopardize the survival of certain species, as moderate-quality habitat may. Hunting and wildlife observation programs topped the list of visitor services in quality, with high-quality programs at more than half of refuges and just about 10 percent of refuges with low-quality programs. Some managers told us that there is a focus on ensuring that hunting programs are successful because of the significant public demand for hunting on refuges. They also noted that it is fairly easy to support wildlife observation, as well as photography, via regular refuge infrastructure such as roads and trails; therefore, it is not resource intensive for refuges to implement and manage high-quality programs for these activities. Environmental education and interpretation programs received the lowest marks, with about one-third of refuges with low-quality programs; about the same percentage of refuges had programs deemed high quality. Managers told us that education and interpretation are among the most resource-intensive visitor service programs because they require staff time for developing and delivering educational supplies, as well as infrastructure, such as classrooms. Refuge managers we met with told us that, for these reasons, environmental education and interpretation programs often are among the first areas to be cut when a refuge faces competing demands. A major factor influencing the quality of visitor services—beyond the abundance of fish and wildlife populations—is the amount and quality of refuge infrastructure and the availability of supplies. For example, the availability of trails and tour routes is essential to providing the public with access to what refuges have to offer and is generally important for supporting any type of visitor service activity. Hunting and fishing infrastructure depends largely on physical structures such as duck blinds, boat launches, and fishing platforms. Providing wildlife observation and photography opportunities simply requires adequate access to the refuge, but can be enhanced through observation platforms and photography blinds. Figure 21 shows examples of infrastructure for wildlife observation and photography. Environmental education depends on physical infrastructure, such as classrooms, and supplies, such as workbooks, handouts, and microscopes. Environmental interpretation also depends on physical infrastructure such as informational kiosks and interpretive signs along trails. Figure 22 shows examples of infrastructure for environmental education and interpretation. The amount and quality of visitor services infrastructure stayed about the same or increased on the vast majority of refuges since 2002 (see table 13). According to refuge managers and regional officials, improvements at least partly reflect the initiative to focus funding on small-scale visitor services infrastructure, such as improvements to parking lots and construction of informational kiosks and restrooms. The increased ability of refuges to implement such projects likely is responsible for refuge managers’ assessments of improvements in environmental interpretation. However, infrastructure conditions worsened at between 12 percent and 26 percent of refuges. According to refuge managers, this reflects that, in many cases, refuges have insufficient staff and funding to keep up with necessary infrastructure maintenance and repairs. In addition, many managers we interviewed reported that they still lack enough infrastructure to deliver quality visitor services and meet the demand for these services. Refuge managers reported that insufficient infrastructure has negatively impacted the quality of education and interpretation programs to at least a moderate extent on about 60 percent of refuges, photography and observation programs on about 40 percent of refuges, and hunting and fishing programs on about 20 percent of refuges. Some refuge managers reported that there is no infrastructure at all on their refuges for the visiting public. In addition, some refuge managers reported not being able to meet public demand for some programs. For example, one refuge manager told us that while many local schools request environmental education programs, the refuge must turn them down because they have no facilities to accommodate school groups. Some refuge managers we spoke with indicated that they would be able to stimulate additional demand for all visitor services if they could improve the amount and quality of infrastructure, including trails, hunting blinds, boat launches, photography blinds, and observation platforms. Consistent with the improvements in program quality noted for environmental education and interpretation (at 40 percent and 47 percent of refuges, respectively), managers reported increases in the time spent on these programs at 44 percent of refuges. These programs, however, also received less time on 29 percent and 27 percent of refuges, respectively. Overall, at least one in five refuges reported a decrease in staff time for each visitor service area (see table 14). In some cases, according to several refuge managers, the changes in time spent may reflect a shift in staffing due to workforce planning, while in other cases, it may reflect the prerogative of refuge managers to move staff away from visitor services in favor of needed habitat management activities. Not surprisingly, more than half of refuge managers reported increasing the amount of time spent on visitor services by volunteers (see table 15). Refuge managers said that volunteers frequently are relied upon to help manage visitor centers and deliver education programs. Given staff reductions due to workforce planning and comments from managers that visitor services are the first to be cut when resources are constrained, it also is not surprising to see that time spent by permanent staff on visitor services had been reduced at more than one-third of refuges. When comparing these results to those for habitat management, more refuges increased permanent, temporary, and contractor staff time on habitat management activities than increased their time on visitor services activities. Conversely, more refuges increased time spent by volunteers and cooperators on visitor services than on habitat management. We found that the time spent by various types of workers on visitor service activities as a function of increases or decreases in time spent by permanent staff were similar to the results for the time spent on habitat management. Specifically, as with time spent on habitat management, refuges that reported an increase in permanent staff time for visitor services were more likely than those reporting a decrease also to report an increase in other staff time. For example, refuges that reported an increase in permanent staff time spent on visitor services were eight times more likely to report an increase in the time spent by temporary staff on these programs than refuges that reported a decrease in permanent staff time. Refuges where permanent staff time increased also were twice as likely to increase volunteer time (see table 16). This analysis also is consistent with our analysis of changes in staff time as a function of staff increases or decreases at a stand-alone refuge or complex. In addition, those refuges that increased time spent on habitat management also tended to be the refuges that spent more time on visitor services. Specifically, refuges that reported spending more permanent staff time doing habitat management work were almost five times more likely to report spending permanent staff time providing visitor services, and refuges that reported spending less permanent staff time doing habitat management work were about four times more likely to report a decrease in permanent staff time providing visitor services. Again, this suggests that some refuges are seeing an overall gain in staff—most likely at focus refuges—while others are seeing an overall loss—likely at targeted- reduction and unstaffed satellite refuges. As with habitat management activities, a lack of staff was identified as a key factor hindering the quality of some visitor service programs. Refuge managers identified staffing as a key factor negatively affecting the quality of environmental education and environmental interpretation programs at 85 percent of refuges; staffing was cited as negatively affecting hunting programs at more than 50 percent of refuges. According to refuge managers and regional documentation, some visitor service programs that were active in the past have had to be cut back due to staffing. For instance, plans to renovate a building used for environmental education at Wallkill River NWR in New Jersey were halted when staff reductions made it impossible to continue the refuge’s emphasis on environmental education. Minnesota Valley NWR reported a 13 percent drop in the number of students participating in environmental education after the loss of park ranger staff, and Kodiak NWR reported that it curtailed its educational programs due to the elimination of an environmental education specialist position. In other cases, refuge managers told us that they do not seek out groups that would be interested in programs at the refuge because there simply are not enough refuge employees to provide the additional education and interpretation services that would be needed if more visitors were to come to the refuge. Similarly, some refuges do not have adequate staff to administer check stations full time during hunting seasons or adequate law enforcement personnel to enforce permitting requirements and take limits. The refuge manager at Cape Romain NWR in South Carolina reported an approximate 20 percent reduction in participation of the refuge’s hunting program after the refuge lost a park ranger. Some stakeholder groups have voiced serious concerns with the deteriorating condition of visitor services and public access to refuges due to recent funding trends and assert that the refuge system needs substantially more funding to fulfill the requirements of the Improvement Act. Some have noted concerns with refuges reducing hours or closing refuges to the public because of staff constraints. According to our survey of refuge managers, however, these concerns do not reflect widespread conditions. A very small number—about 4 percent—of refuges have decreased the hours they are open to the public, while slightly more than 12 percent indicated that they have actually increased the hours they are open. However, the change in the number of hours that visitor centers are open to the public varied quite substantially. Specifically, for those refuges that reported having visitor centers, 20 percent indicated that visitor center hours increased, while about 27 percent reported that hours decreased. For example, the visitor contact station at Occoquan Bay NWR has been closed for several years due to staffing shortages; instead, the refuge relies on an “honor fee” system (see fig. 23). Survey results do not show a significant difference among the three refuge tiers with regard to changes in the hours refuges are open to the public. However, according to our survey, targeted reduction refuges were almost twice as likely as focus refuges to have decreased the hours that their visitor center is open. This is most likely due to staffing changes resulting from workforce planning, which were targeted at targeted-reduction refuges. While some refuges have visitor services staff and sufficient infrastructure, refuge managers indicated that staffing changes, partially resulting from workforce planning, and a lack of resources for increasing and maintaining infrastructure, raise concerns about their ability to provide quality visitor services into the future. In fact, in response to an open- ended question concerning the biggest threats to visitor services on refuges, a large majority of managers cited either funding, staffing, and infrastructure as primary threats to the quality of visitor service delivery; almost 75 percent of refuge managers singled out staffing alone as the key problem affecting visitor services. As noted previously, refuge managers tend to focus resources on maintaining habitat conditions in times of tight budgets, at the expense of visitor services. Refuge managers also are concerned about the impact that the increasing administrative workload incurred by nonadministrative refuge staff is having on the refuges’ ability to deliver visitor services—also noted as a major concern for refuges’ ability to maintain habitat management. Managers also expressed concern about a continued, and in some cases increasing, dependence on volunteers to keep up with public demand for visitor services. Volunteers help refuges with a large variety of visitor services activities including operating visitor centers, providing education and interpretive services, and building and maintaining interpretative kiosks and other infrastructure. However, managers told us that although volunteers provide valuable services, they cannot fully replace lost refuge staff. In addition, volunteerism levels are unpredictable, and many of the refuge managers we met with indicated that volunteer levels have generally been declining. Furthermore, as noted previously, even if a refuge has a good supply of volunteers, it will still need to devote employee time to training, supervising, and coordinating volunteers. Refuge managers and regional and headquarters officials expressed concern about the long-term implications of declining and low-quality visitor services occurring at some refuges. Many refuge managers cited the importance of providing opportunities for the public to utilize refuge resources—in particular, ensuring they have positive outdoor experiences and providing them with meaningful educational and interpretative services—to the future of the refuge system. This helps ensure that refuges have visibility in the community, that the public understands the purpose and importance of what the refuge system does, and that refuges are thought of as a vital community resource. Refuge managers told us that positive recreational and educational experiences help ensure public support for refuge operations, and outreach and public education help bolster the number of people interested in volunteer opportunities on refuges. These activities also are important, according to refuge managers, because the refuge system increasingly is turning toward partnerships with private landowners in an effort to maintain and improve ecosystems both on and around refuges; public education about the refuge system can increase the viability of important refuge partnerships with nonprofit environmental and land management organizations who work with adjacent landowners, other federal and state land management agencies, and others on conservation efforts. Refuge officials also told us that public perception of land management work assists with land acquisitions, inasmuch as more private landowners will be willing to work with the agency on land transactions. In addition, refuge managers cited the availability of visitor services as a way to get young people interested in future careers with the refuge system and instill in children an appreciation for wildlife and the outdoors and an interest in maintaining these resources. In light of continuing federal fiscal constraints and an ever-expanding list of challenges facing refuges, maintaining the refuge system as envisioned in law—where the biological integrity, diversity, and environmental health of the refuge system are maintained; priority visitor services are provided; and the strategic growth of the system is continued—may be difficult. While some refuges have high-quality habitat and visitor service programs and others have seen improvements since 2002, refuge managers are concerned about their ability to sustain high-quality refuge conditions and continue to improve conditions where needed because of expected continuing increases in external threats and habitat problems affecting refuges. Already, FWS has had to make trade-offs among refuges with regard to which habitat will be monitored and maintained, which visitor services will be offered, and which refuges will receive adequate law enforcement coverage. FWS’s efforts to prioritize its use of funding and staff through workforce planning have restored some balance between refuge budgets and their associated staff costs. If threats and problems afflicting refuges continue to grow as expected, it will be important for the refuge system to monitor how these shifts in resources are affecting refuge conditions. GAO provided Interior with a draft of this report for its review and comment. The department provided technical comments that we have incorporated as appropriate. The department’s comments are presented in appendix VI. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 2 days from the report date. At that time, we will send copies of this report to interested congressional committees, the Secretary of the Interior, and other interested parties. We will also make copies of this report available to others upon request. In addition, the report will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your offices have questions about this report, please contact me at (202) 512-3841 or nazzaror@gao.gov. Contact points for our offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix VII. The objectives of this study were to (1) describe changing factors that the National Wildlife Refuge System experienced from fiscal years 2002 through 2007, including funding and staffing changes, and (2) examine how habitat management and visitor services changed during this period. To address these objectives, we reviewed relevant laws, regulations, and policies as well as numerous agency documents discussing the refuge system. We also reviewed other reports that related to refuge system operations such as reports on climate change and possible impacts on refuges published by the U.S. Environmental Protection Agency, on development and impacts on wildlife by the Western Governors’ Association, and on challenges facing the refuge system by the Cooperative Alliance for Refuge Enhancement—a consortium of groups interested in the refuge system. We obtained and analyzed funding and staffing data from the Federal Financial System and the Federal Personnel Payroll System, and refuge planning and performance data from the U.S. Fish and Wildlife Service’s (FWS) Refuge Annual Performance Planning System. We worked with national program officials to identify the 222 complexes and stand-alone refuges that existed during the time period we reviewed. We interviewed database technicians and their managers to understand how the information in these databases is compiled and maintained. Where necessary, we worked with database technicians to ensure their output files contained needed data elements, and we validated our resulting analyses with regional and national program officials. Our review of the data and our discussions with program officials indicated that the payroll and personnel databases were sufficiently reliable for the purposes of our review. We used the performance planning system to assist in site selection and determined that it was sufficiently reliable for this purpose. We analyzed the obligations data in both nominal and inflation-adjusted terms. To remove the effects of inflation, we adjusted nominal dollars using the Gross Domestic Product (GDP) Price Index for Government Consumption Expenditures and Gross Investment (federal nondefense sector), with 2002 as the base year. The price index reflects changes in the value of government output, measured by the cost of inputs, including compensation of employees and purchases of goods and services. Consistent with the proportion of FWS’s operating expenditures on personnel, this price index is more heavily weighted by changes in federal workers’ compensation than the overall GDP price index. We met with officials at refuge system headquarters, refuge offices at 4 FWS regions (Hadley, Massachusetts; Denver, Colorado; Portland, Oregon; and Minneapolis, Minnesota), and 19 refuges, and conducted phone interviews with officials at the other 4 regional offices and about 50 additional refuges. We selected refuges for site visits in order to see a range in geographic location, visitation level, refuge prioritization level, and type of management activities and challenges. Given the differences in the refuges across the system and the need to gather information on a range of topics, we surveyed all 585 units within the refuge system—including stand-alone refuges and refuges within complexes. Survey respondents primarily were refuge managers or project leaders of refuge complexes. Survey questions were crafted to obtain information on a variety of issues, including how the following changed between fiscal years 2002 and 2007: the quality of various types of habitat, the extent of various habitat problems, the extent to which external factors affected habitat problems, the amount of time spent on various management activities, the amount and quality of visitor services infrastructure, and the quality of visitor services. We also inquired about their activities between fiscal years 2002 and 2007, and their perspectives on the challenges facing the refuge system and refuge management and reporting. During our site visits and in interviews with 9 randomly selected refuge managers, we explored various potential survey questions to confirm we were eliciting the information we intended and whether managers could answer the questions in a clear, consistent manner with minimal difficulty in data recall, among other things. When it became clear that we needed to include “pick lists” of possible answers for certain questions, we utilized knowledge gained from prior interviews and obtained feedback from 12 refuge managers about what should be included in these lists. We conducted formal pretests with 8 refuge managers and used structured probes to determine: (1) if respondents had the information and knowledge necessary to answer the question, (2) if respondents interpreted the questions in the same way as other respondents, (3) if respondents interpreted the questions as we intended, (4) if respondents felt that the response categories offered the correct level of precision, and (5) if respondents felt that we used the terminology commonly used by refuge managers. After changes were incorporated to address pretesting concerns, we sent the survey to a random sample of 10 refuge managers to complete and “validate” the survey. No concerns were raised during the validations about respondent bias, response burden, relevancy of the questions, ability of the respondent to answer, or confidence in response accuracy. We also obtained comments from refuge system officials in headquarters. We then distributed the survey to all refuge units. Subsequently, we determined that 538 units should be included in our scope of analysis because FWS does not have full management responsibility for all of its units. We received an 81 percent response rate for this subset of refuges used in our analysis. We conducted follow-up interviews with 14 refuge managers to verify that we were correctly interpreting responses and to clarify certain points. See appendix II for a detailed discussion of the analysis of this survey. Below is a summary of the key questions we are reporting on from the survey. We also asked other questions that we do not specifically report on to obtain further context. Between fiscal year 2002 and fiscal year 2007, did the overall quality of pesticide runoff, soil erosion, and manure), off-refuge industry (such as energy development, mining, logging, and military activities), off-refuge human settlement (such as roads, construction, housing, septic systems, and airports), rights of way (from roads and utilities), on-refuge sources of pollution (such as energy production, grazing, and legacy waste), on- refuge activities (such as visitation and fire suppression), inadequate water rights, and other factors (respondents could write in)? In fiscal year 2007, did workers on your refuge (including permanent employees, temporary employees, contract workers, volunteers, and cooperators) conduct the following habitat management activities and did the amount of time spent on these activities increase, stay the same, or decrease between fiscal years 2002 and 2007: addressing invasive plants, addressing invasive animals, addressing water pollution, addressing soil contamination, addressing air pollution, addressing light pollution, addressing plant disease, addressing animal disease, conducting routine habitat management activities, conducting habitat restoration projects, conducting inventory and monitoring surveys of habitat condition, conducting inventory and monitoring of wildlife populations, addressing habitat fragmentation, addressing lack of water, addressing excessive water, addressing damage to habitat form wildfire, addressing damage to habitat from storms, addressing damage to habitat from recreational use, addressing damage to habitat from crime, doing conservation planning, and coordinating with nearby landowners? Between fiscal years 2002 and 2007, did the amount of time spent by the following types of workers to conduct habitat management activities on your refuge increase, stay the same, or decrease: permanent employees, temporary employees, contract workers, volunteers, and cooperators? In fiscal year 2007, did your refuge provide the following visitor services: hunting, fishing, wildlife observation, wildlife photography, environmental education, and environmental interpretation? If you did not provide some of the six visitor services, what were the reasons: not compatible with refuge, lack of resources to provide, or other reasons? Between fiscal years 2002 and 2007, did the hours of the refuge or the visitor center on your refuge (if there is one) increase, stay the same, or decrease? Between fiscal years 2002 and 2007, did the quantity and condition of the following types of visitor services infrastructure increase, stay the same, or decrease: trail and tour routes, hunting infrastructure (such as blinds and check stations), fishing infrastructure (such as boat launches, docks, and platforms), wildlife observation infrastructure (such as platforms and viewing areas), wildlife photography infrastructure (such as blinds, platforms, and viewing areas), education infrastructure (such as buildings and study locations), and interpretation infrastructure (such as signs, kiosks, and exhibits)? Between fiscal years 2002 and 2007, did the amount of time spent by all workers on your refuge (including permanent employees, temporary employees, contract workers, volunteers, and cooperators) to provide the following visitor services increase, stay the same, or decrease: hunting, fishing, wildlife observation, wildlife photography, environmental education, and environmental interpretation? Between fiscal years 2002 and 2007, did the amount of time spent by the following types of workers to provide visitor services on your refuge increase, stay the same, or decrease: permanent employees, temporary employees, contract workers, volunteers, and cooperators? In fiscal year 2007, how would you rate the overall quality of the following visitor services at your refuge and did the quality improve, stay the same, or worsen between fiscal years 2002 and 2007: hunting, fishing, wildlife observation, wildlife photography, environmental education, and environmental interpretation? In your professional judgment, what are the biggest threats to the condition of habitat and visitor services at your refuge? In your professional judgment, what are the biggest threats to the condition of habitat and visitor services for the National Wildlife Refuge System? Do you feel that the criteria for placing refuges into tiers were appropriate to distinguish among competing priorities of the refuges in your region? Between fiscal years 2002 and 2007, did the number of work days spent by your refuge’s nonadministrative staff on selected administrative activities increase, stay the same, or decrease? We conducted our work between July 2007 and September 2008 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. We also conducted reliability assessments of the data we obtained electronically and determined those data to be of sufficient quality to be used for the purposes of this report. During our site visits to wildlife refuges, during our interviews with refuge managers, and in written responses to our survey questions, refuge managers asserted that staff reductions threaten the quality of wildlife habitat. To test these assertions, we developed statistical models that assess whether staffing change is associated with changes in habitat quality. In particular, our models assess whether refuges where staff spent decreased time on habitat management activities between fiscal years 2002 and 2007 were more likely than other refuges to report that habitat conditions worsened rather than improved. In assessing this likelihood, our models account for refuge characteristics other than change in staff time that might impact habitat conditions, including tier designation, which indicates relative importance of a refuge compared to other refuges, and the change in external factors, such as extreme weather and off-refuge agriculture, that contribute to habitat problems. The results of our models show that habitat was more likely to worsen, rather than to improve, at refuges where staff time decreased between fiscal years 2002 and 2007 compared to those where staff time increased, even after adjusting for these other characteristics. Our analysis assesses the change in the quality of waterfowl and other migratory bird habitat as a function of three key refuge characteristics: (1) the change in staff time spent on habitat management, (2) tier designation, and (3) the change in external factors that contribute to habitat problems. Frequency counts for the key variables used in our analysis are presented in table 17. To measure habitat change, we used responses to a question on our survey, which asked whether the overall quality of habitat on a refuge improved, stayed the same, or worsened between fiscal years 2002 and 2007. The question was asked separately for both waterfowl and for other migratory birds, leading to two indicators of the change in habitat quality. We used these two indicators to develop separate statistical models for waterfowl and for other migratory birds. To measure the first key refuge characteristic, the change in staff time spent on habitat management activities, we used data from a question on our survey, which asked whether the total amount of time that permanent staff spent conducting habitat management activities on a particular refuge between fiscal years 2002 and 2007 increased, stayed the same, or decreased. For the second key refuge characteristic, tier designation, we obtained data from FWS for each refuge. In 2006, the agency designated refuges as focus, targeted reduction, or unstaffed satellite refuges. In our analysis, tier designation represents the relative importance of a given refuge within a complex as the agency determined in 2006. To measure the third key refuge characteristic, the impact of external factors, we used responses to a question on our survey, which asked whether the contribution of various external factors to habitat problems on a refuge increased, stayed the same, or decreased between fiscal years 2002 and 2007. We used these data to classify refuges into two groups. Refuges in the first group, which we refer to as having a net increase in external factors, reported that the contribution to habitat problems increased for more of these factors than it decreased. Refuges in the second group, which we refer to as having no net increase in external factors, reported that either the contribution of these factors decreased for more factors than it increased or that the contribution of these factors increased for the same number that it decreased. Refuges were included in our analysis if they indicated that providing habitat for waterfowl or other migratory birds was at least somewhat of a priority and if they provided usable responses to each of the key survey questions described above. Of the 437 refuges that responded to our survey, 40 were excluded from our analysis of the change in waterfowl habitat because they reported that providing this type of habitat was not a priority at their refuge. Similarly, 14 refuges were excluded from our analysis of the change in other migratory bird habitat for the same reason. Of the remaining 397 refuges, 374 provided useable responses for all key survey questions and were included in our analysis of waterfowl habitat, while 400 of the remaining 423 refuges provided sufficient responses to be included in our analysis of other migratory bird habitat. Before developing statistical models, we cross-tabulated data on the change in habitat quality against data for the three key refuge characteristics described above (namely, the change in staff time, tier designation and the change in external factors). The results of these cross- tabulations show that, although habitat was more likely overall to improve than it was to worsen, the odds of improving rather than worsening, vary considerably depending upon the characteristics of a refuge. Complete results of these cross-tabulations, along with tests of the statistical significance of these associations, are presented in tables 18 and 19. These cross-tabulations show that the change in habitat quality is associated with the change in staff time. For example, as shown in table 18, among refuges where staff time increased, more than three times as many refuges experienced improved habitat for waterfowl (47 percent) as experienced worsened habitat (14 percent). By contrast, among refuges where staff time decreased, nearly the same number of refuges experienced improved habitat for waterfowl (30 percent) as experienced worsened habitat (27 percent). We found similar results when comparing change in staff time with the change in habitat quality for other migratory birds (table 19). The cross-tabulations also indicate that changes in habitat quality depend, in part, upon tier designation. Focus and targeted reduction refuges were more likely than unstaffed satellite refuges to experience improved rather than worsened habitat. For example, as shown in table 18, between fiscal years 2002 and 2007, more than twice as many focus refuges experienced improved waterfowl habitat (42 percent) as experienced worsened waterfowl habitat (20 percent). At unstaffed satellite refuges, by contrast, habitat for these birds worsened almost as frequently as it improved, with 20 percent of refuges experiencing improved quality and 16 percent experiencing worsened quality. The cross-tabulations show a similar disparity among tiers with regard to changes in the quality of habitat for other migratory birds (table 19). Finally, our cross-tabulations indicate that changes in habitat quality also depend upon the change in external factors that contribute to habitat problems. For example, among refuges that reported no net increase in external factors, about nine times more refuges reported improved waterfowl habitat (38 percent) than reported worsened waterfowl habitat (4 percent), as shown in table 18. By contrast, among refuges that experienced a net increase in external factors, the number of refuges that experienced improved waterfowl habitat (38 percent) was much closer to the number that experienced worsened habitat (22 percent). We found similar results for the change in other migratory bird habitat (table 19). Although the results of the cross-tabulations are strong and statistically significant, they provide only a partial assessment of the relationship between habitat change and refuge characteristics. This is because the cross-tabulations compare habitat change with each refuge characteristic individually without accounting for the influence of the other characteristics. For this reason, we developed statistical models that allow us to account for the effects of all of these characteristics simultaneously. Our statistical models, technically referred to as a multinomial logistic regression, were used to assess the effects of each refuge characteristic on the change in habitat quality while adjusting for the effects of the other characteristics. These models estimate the effects of each refuge characteristic on (1) the odds of habitat improving, rather than worsening, and (2) the odds of habitat staying the same, rather than worsening. For example, our models estimate the number of times more likely that habitat is to improve, rather than to worsen, at refuges where staff time increased compared to refuges where staff time decreased, controlling for the effects of tier designation and the change in external factors. Similarly, our models estimate effects for tier designation and for the change in external factors. In order to test the adequacy of our models, we verified that our data contained a sufficient number of refuges with each combination of characteristics, each model adequately fit the data based on chi-square goodness-of-fit tests, and high associations among the refuge characteristics would be unlikely to confound model estimates. We used robust regression techniques to adjust for the fact that refuges clustered within the same complex may have provided similar responses to survey questions. In order to be confident that the estimates from our regression models are robust to various specifications, we formulated and tested several alternative models to ensure that we obtained similar estimates for the effects of staffing change, tier designation, and change in external factors. The statistical analysis was performed by a senior research methodologist and was reviewed by a managing methodologist and a professional statistician. Their review assessed the model specification, model development, model results, and the conclusions derived from these results. In order to ensure that the analysis was free of programming errors, each line of the computer syntax used to develop the model was verified by a senior data analyst. Our regression models indicate that change in staff time and tier designation are associated with changes in habitat quality. These findings are true for both waterfowl and other migratory bird habitat. The models indicate that these associations are statistically significant even after adjusting for other refuge characteristics, including the change in external factors. The results of our regression models are presented in tables 20 and 21. In particular, highlights include: The odds of habitat improving, rather than worsening, were significantly greater at refuges where staff time increased compared to those where staff time decreased. Specifically, we estimate that refuges where staff time increased were about 3.0 times more likely than refuges where staff time decreased to report improved, rather than worsened, habitat for both waterfowl and other migratory birds. The odds of habitat improving, rather than worsening, were higher at focus and targeted reduction refuges as compared to unstaffed satellite refuges. For example, we estimate that focus refuges were 3.4 times more likely than unstaffed refuges to experience improved rather than worsened habitat quality for other migratory birds and that targeted reduction refuges were 3.9 times more likely. Targeted reduction refuges were also significantly more likely than unstaffed satellite refuges to experience improved rather than worsened waterfowl habitat, although focus refuges were not significantly different from unstaffed refuges in this regard. Change in external factors is strongly associated with habitat change. For example, refuges that experienced no net increase in the number of external factors were about 7.0 times more likely to experience improved, rather than worsened, waterfowl habitat quality and 5.1 times more likely to experience improved, rather than worsened habitat quality for other migratory birds. Our models also found that the change in staff time and the change in external factors were associated with an increased likelihood of habitat quality staying the same rather than worsening. In particular, we found the following: The odds of habitat staying the same rather than worsening were higher at refuges where staff time stayed the same. Specifically, we estimate that refuges where staff time stayed the same were 3.9 times more likely than refuges where staff time decreased to report that habitat for other migratory birds stayed the same rather than worsened. This effect was only marginally significant for waterfowl habitat. The odds of habitat staying the same rather than worsening were higher at refuges that did not experience a net increase in external factors that contribute to habitat problems. Specifically, we estimate that refuges that experienced no net increase in external factors were 8.0 times more likely to report that waterfowl habitat stayed the same, rather than worsened, and 5.8 times more likely to report that other migratory bird habitat stayed the same rather than worsened. The key results of our models are robust to alternative specifications. We tested for the presence of interaction effects between staff time and tier designation, that is, whether the effect of a change in staff time depends upon tier, but found no evidence of such an effect. We fit models that accounted for habitat priority using data from a question on our survey that asked: according to the purpose of each refuge, how much of a priority is providing habitat for waterfowl and other migratory birds. The results of these models show that the estimated effects of staff change and tier designation did not diminish after accounting for habitat priority, indicating that the effects of staff change and tier are not limited to lower priority habitats. We also fit models that used data on the change in the number of full-time equivalent staff at the complex level, rather than survey data about changes in staff time at the refuge level, as a measure of staffing change between fiscal years 2002 and 2007. These models found results similar to those reported above: namely, that refuges that were part of complexes that gained staff were significantly more likely than those that were part of complexes that lost staff to report that habitat conditions improved rather than worsened. significantly associated with increased staff time, and that worsened habitat quality is significantly associated with decreased staff time, it is subject to certain limitations. First, our data on habitat change are based on the perception of refuge managers rather than on direct measurements of habitat conditions. To minimize this limitation, we conducted more than two dozen pretests to ascertain, before administering our survey, that land managers could provide valid responses to these questions. Second, while our model identifies a statistical association between the change in habitat condition with the change in staffing, it is not able to assess whether staffing changes actually caused changes in habitat quality. Statistical correlation is necessary but not sufficient evidence to demonstrate causation. Third, because of the dynamic nature of wildlife habitat, the effects of staffing changes may not appear immediately. Staffing cuts may result in a reduction in habitat management activities, which may not find their effect on habitat for several years. Conversely, restoration projects may take several years to see an effect. Fourth, we do not have data on the initial quality of habitat; while we were confident that refuge managers could assess trends at their refuge, we were not confident that they could accurately rate the quality of habitat in fiscal year 2002, 6 years prior to the administration of our survey. As a result, we are unable to determine, for example, whether refuges that improved in quality were already of high quality or whether refuges that worsened were already of low quality. In spite of these limitations, our models are consistent with the assertions made by refuge managers in showing that decreases in staff are strongly associated with worsening habitat, although the actual size of the effect might be somewhat higher or lower than we estimate it to be. Deferred maintenance is maintenance to repair, rehabilitate, dispose of, or replace buildings and other facilities. Deferred maintenance projects are monitored in the Service Asset and Maintenance Management System (SAMMS), which tracks, among other things, asset maintenance and capital improvement needs. SAMMS replaced the Maintenance Management System (MMS) database in 2005 because refuge system management wanted to move toward a more comprehensive system, according to a refuge system official. Refuge managers identify maintenance needs and document the needs in SAMMS via work orders. If these needs have not been addressed after 1 year, they are eligible to become deferred maintenance projects. In addition, regional facility managers conduct condition assessments of refuge buildings, grounds, equipment, and infrastructure, during which they may find assets that should have been entered into SAMMS but were not. In this case the asset may be identified as deferred maintenance without the 1-year waiting period. Furthermore, deferred maintenance projects are limited to those projects that require less than 25 percent capital improvement; will be completed within 2 years; cost at least $5,000, but less than $750,000; and have repair costs that do not exceed the asset’s current replacement value. The deferred maintenance backlog appears to be increasing, but the actual change in the backlog cannot be determined because the refuge system implemented several recordkeeping changes between fiscal years 2002 and 2007. Specifically, refuge system officials stated that the following recordkeeping changes occurred during this time period. 1. Each refuge asset is now entered into SAMMS as a single project, whereas multiple assets could have been entered as a single project in MMS. 2. All assets must now be entered into SAMMS, including inexpensive assets such as signs and fencing, whereas only the more expensive refuge assets, such as buildings, were typically entered into MMS. 3. The refuge system implemented comprehensive condition assessments—which will take place once every 5 years—to better determine maintenance deficiencies, and the data were entered into SAMMS between fiscal years 2006 and 2007. 4. The Federal Highway Administration completed an assessment of the maintenance needed on refuges’ public-use roads, one of the most expensive assets to maintain, and the newly identified roads projects were entered into SAMMS between fiscal years 2006 and 2007. Taken together, these recordkeeping changes limit our ability to assess yearly trends in the refuge system’s deferred maintenance backlog. However, refuge system officials reported that the changes allow them to maintain more complete and accurate information about the condition of refuge assets. For example, the requirement that all assets be entered as separate projects in SAMMS was made to allow refuge system officials to determine if the cost of maintenance was greater than the current replacement value of each individual asset. According to refuge management, it was too difficult to assess the maintenance cost relative to the current replacement value when multiple assets were listed as one project in MMS. Table 22 presents the refuge system’s deferred maintenance backlog by region for fiscal years 2002 through 2007. The refuge system received increased funding from fiscal year 2002 through fiscal year 2004 to address its deferred maintenance backlog, although the funding decreased from 2004 through 2007. For example, Congress funded 2 years worth of deferred maintenance in fiscal year 2001, and this combined total became the new base funding amount in fiscal year 2002, according to refuge system management. Although we cannot determine the extent to which the additional funding helped the refuge system address the backlog, for the reasons outlined above, several senior refuge system officials asserted that the funding has had a positive impact and that the refuge system’s assets appear to be in better condition now than before the funding increase. The Refuge Operating Needs System (RONS) is the refuge system database for cataloging operational requirements such as staff, equipment, and planned projects at refuges throughout the system. According to refuge system management, it is intended to be a full inventory of funding needs for annual operations at field stations. Refuge managers determine their total needs on the basis of operational plans, congressional direction, and departmental priorities, and enter these needs as projects into the electronic RONS database. Since its inception in the early 1990s, RONS has been one of the primary tools refuge system officials use to develop estimates for additional funding needed to address high-priority projects at certain refuges. These project estimates are included in budget requests submitted annually to congressional appropriators. In reports that accompany the refuge system’s annual appropriation, Congress indicates whether the program’s RONS project requests are fully funded. In fiscal year 2002, 108 RONS projects were funded; 5 years later, there were 4 projects included in the 2007 budget submission. Refuge system management told us that some RONS projects are given a higher priority than others. During the fiscal year 1998 congressional appropriations process, appropriations staff asked the refuge system to set priorities within RONS, and also directed them to conduct an analysis to determine minimum staffing levels at refuges. To fulfill this requirement, refuge system officials divided existing RONS projects into two tiers: (1) tier 1, consisting of staffing projects as determined by the minimum staffing analysis as well as other high-priority projects that were deemed critical to the mission of the refuge, and (2) tier 2, comprising projects that were considered to be lesser priorities. At the end of fiscal year 2007, the total value of RONS projects awaiting funding was just over $1 billion. According to refuge system officials, these projects are sometimes referred to as the RONS backlog (adopting the language used to describe the system’s deferred maintenance). Tier 1 projects are for mission-critical needs and therefore would be comparable to backlogged maintenance projects; at the end of fiscal year 2007, tier 1 projects encompassed about 2,300 unfunded projects totaling about $300 million. Conversely, tier 2 projects describe activities associated with expanded capabilities, not unmet requirements, and thus refuge system officials do not consider these projects to be “behind schedule” in the same sense as those in the system’s deferred maintenance backlog. Table 23 presents the RONS project backlog, containing both tier 1 and tier 2 projects, for fiscal years 2002 through 2007. According to the refuge system’s RONS manager, the refuge system’s leadership targets tier 1 projects for funding unless explicitly directed to target tier 2. However, evolving national-level priorities can result in tier 2 projects receiving funding before their tier 1 counterparts. For example, the increased attention on law enforcement following the attacks of September 11, 2001, resulted in funding for law enforcement personnel as part of projects that were placed in tier 2. Overall, a quarter of RONS projects funded from fiscal years 2002 through 2007 were identified as tier 2 projects, as shown in table 24. Headquarters officials maintain oversight over the RONS inventory, periodically removing projects that are no longer needed, according to the RONS manager. Moreover, between 2000 and 2006, headquarters attempted to keep the number of projects in the RONS inventory stable in order to make progress in reducing the backlog. During this period, refuge managers were prevented from entering new project requests into the database. In fiscal year 2006, refuge system officials reviewed RONS’ second tier of projects, deleting entries that appeared to be out of date. This accounted for much of the 48 percent drop in total project cost between fiscal years 2005 and 2006, as seen in table 23. The database was then reopened in 2007 to allow for new project entries from refuge managers. However, several managers that we spoke with on our site visits were unaware that the database had been reopened, or had given up entering RONS projects altogether due to the many years that the database was closed to new entries. Other managers perceived little realistic chance that their projects would be funded, given the magnitude of the project backlog and the fact that the number of projects requested significantly outpaces the ability of the refuge system to secure RONS funding. Table 25 presents the ratio of projects in the RONS backlog to each project that received funding for fiscal years 2002 through 2007. As shown in the table, there were 2,119 projects in the RONS backlog for every 1 project that received funding in fiscal year 2007. Similarly, for every $1 in funding directed toward RONS projects in 2007, there were $1,469 worth of unfunded projects in the backlog. Sonny Bono Salton Sea NWR Complex – Coachella Valley NWR and Sonny Bono Salton Sea NWR Stillwater NWR – Anaho Island NWR, Fallon NWR, and Stillwater NWR Total, NWRS Headquarters (Region 9) uartered at the Des Lacs NWR. Since 2006, Des Lacs NWR has been a satellite of a complex headuartered at the J. Clark Salyer NWR. All of Des Lacs NWR’s former satellite refuges are now part of the Lostwood WMD complex. In addition to the individual named above, Trish McClure, Assistant Director; Mark Braza; David Brown; Stephen Cleary; Tim Guinane; Carol Henn; Richard Johnson; Michael Krafve; Alison O’Neill; George Quinn, Jr.; and Stephanie Toby made key contributions to this report. | The National Wildlife Refuge System, which is administered by the Fish and Wildlife Service in the Department of the Interior, comprises 585 refuges on more than 96 million acres of land and water that preserve habitat for waterfowl and other migratory birds, threatened and endangered species, and other wildlife. Refuges also provide wildlife-related activities such as hunting and fishing to nearly 40 million visitors every year. GAO was asked to (1) describe changing factors that the refuge system experienced from fiscal years 2002 through 2007, including funding and staffing changes, and (2) examine how habitat management and visitor services changed during this period. We surveyed all refuges; visited 19 refuges in 4 regions; and interviewed refuge, regional, and national officials. In commenting on a draft of this report, the Department of the Interior made technical comments that we have incorporated as appropriate. GAO is not making recommendations in this report. Between fiscal years 2002 and 2007, the refuge system experienced funding and staffing level fluctuations, the introduction of several new policy initiatives, and the increased influence of external factors such as extreme weather that threaten wildlife habitat and visitor infrastructure. Although core funding--measured as obligations for refuge operations, maintenance, and fire management--increased each year, inflation-adjusted core funding peaked in fiscal year 2003 at about $391 million--6.8 percent above fiscal year 2002 funding. Inflation-adjusted core funding ended the period 2.3 percent below peak levels, but 4.3 percent above fiscal year 2002 levels by fiscal year 2007. Core refuge staffing levels peaked in fiscal year 2004 at 3,610 full-time equivalents--10.0 percent above the fiscal year 2002 level--and then declined more slowly than funding levels. By fiscal year 2007, staffing levels fell to 4.0 percent below peak levels, but 5.5 percent above fiscal year 2002 levels. Through fiscal year 2007, the number of permanent employees utilized by the refuge system declined to 7.5 percent below peak levels. During this period, refuge system officials initiated new policies that: (1) reduced staff positions and reallocated funds and staff among refuges to better align staff levels with funding; (2) required refuge staff to focus on a legislative mandate to complete refuge conservation plans by 2012; (3) shifted to constructing a larger number of smaller visitor structures, such as informational kiosks, and fewer large visitor centers to spread visitor service funds across more refuges; (4) increased the number of full-time law enforcement officers and their associated training and experience requirements; and (5) resulted in additional administrative work. During this period, external factors that complicate refuge staffs' ability to protect and restore habitat quality also increased, including severe storms and development around refuges. Our survey showed that the quality of habitat management and visitor service programs varied across refuges during our study period. Habitat conditions for key types of species improved about two times more often than they worsened, but between 7 percent and 20 percent of habitats were of poor quality in 2007. Certain habitat problems increased at more than half of refuges during this period, and managers reported that they increased the time spent on certain habitat management activities, such as addressing invasive plants, despite declining staffing levels. However, several managers we interviewed told us that staff were working longer hours without extra pay to get work done, and managers expressed concern about their ability to sustain habitat conditions. While the quality of four key visitor service programs was reported to be stable or improving between fiscal years 2002 and 2007 at the vast majority of refuges, the other two key programs--environmental education and interpretation--were considered poor quality at one-third of refuges in 2007. Changes in the time spent on visitor services varied considerably across refuges, and managers noted that visitor services generally are cut before habitat management activities when resources are limited. Managers are concerned about their ability to provide high-quality visitor services in the future given staffing and funding constraints. |
Over the past decade, VA’s system of health care for veterans has undergone a dramatic transformation, shifting from predominantly hospital-based care to primary reliance on outpatient care. As VA increased its emphasis on outpatient care rather than inpatient care, it was left with an increasingly obsolete infrastructure, including many hospitals built or acquired more than 50 years ago in locations that are sometimes far from where veterans live. To address its obsolete infrastructure, VA initiated its CARES process— the first comprehensive, long-range assessment of its health care system’s capital asset requirements since 1981. CARES was designed to assess the appropriate function, size, and location of VA facilities in light of expected demand for VA inpatient and outpatient health care services through fiscal year 2022. Through CARES, VA sought to enhance outpatient and inpatient care, as well as special programs, such as spinal cord injury, through the appropriate sizing, upgrading, and locating of VA facilities. Table 2 lists key milestones of the CARES process. We have previously reported that a range of capital asset alignment alternatives were considered throughout the CARES process, which adheres to capital planning best practices. Moreover, there was relatively consistent agreement among the DNCP prepared by VA, the CARES Commission appointed by the VA Secretary to make alignment recommendations, and the Secretary as to which were the best alternatives to pursue. Although the Secretary tended to agree with the CARES Commission’s recommendations, the extent to which he agreed varied by alignment alternative. In particular, the Secretary always agreed with the commission’s recommendations to build new facilities, enter into enhanced use leases, and collaborate with the Department of Defense and universities, but was less likely to agree with the CARES Commission’s recommendations to contract out or close facilities. The decisions that emerged from the CARES process will result in an overall expansion of VA’s capital assets. According to VA officials, rather than show that VA should downsize its capital asset portfolio, the CARES process revealed service gaps and needed infrastructure improvements. We also reported that a number of factors shaped and in some cases limited the range of alternatives VA considered during the CARES process. These factors included competing stakeholder interests; facility condition and location; veterans’ access to facilities; established relationships between VA and health care partners, such as DOD and university medical affiliates; and legal restrictions. The challenge of misaligned infrastructure is not unique to VA. We identified federal real property management as a high-risk area in January 2003 because of the nationwide importance of this issue for all federal agencies. We did this to highlight the need for broad-based transformation in this area, which, if well implemented, will better position federal agencies to achieve mission effectiveness and reduce operating costs. But VA and other agencies face common challenges, such as competing stakeholder interests in real property decisions. In VA’s case, this involves achieving consensus among such stakeholders as veterans service organizations, affiliated medical schools, employee unions, and communities. We have previously reported that competing interests from local, state, and political stakeholders have often impeded federal agencies’ ability to make real property management decisions. As a result of competing stakeholder interests, decisions about real property often do not reflect the most cost-effective or efficient alternative that is in the interest of the agency or the government as a whole but instead reflect other priorities. In particular, this situation often arises when the federal government attempts to consolidate facilities or otherwise dispose of unneeded assets. Through the CARES process, VA gained the tools and information needed to plan capital investments. As part of the CARES process, VA modified an actuarial model that it used to project VA budgetary needs. According to VA, the modifications enabled the model to produce 20-year forecasts of the demand for services and provided for more accurate assessments of veterans’ reliance on VA services, capacity gaps, and market penetration rates. The information provided by the model allowed VA to identify service needs and infrastructure gaps, in part by comparing the expected location of veterans and demand for services in years 2012 through 2022 with the current location and capacity of VA health care services within each network. In addition to modifying the model, VA conducted facility condition assessments on all of its real property holdings as part of the CARES process. These assessments provided VA information about the condition of its facilities, including their infrastructure needs. VA continues to use the tools developed through CARES as part of its capital planning process. For example, VA conducts facility condition assessments for each real property holding every 3 years on a rotating basis. In addition, VA uses the modified actuarial model to update its workload projections each year, which are used to inform the annual capital budget process. The CARES process serves as the foundation for VHA’s capital planning efforts. The first step in VHA’s capital budget process is for networks to submit conceptual papers that identify capital projects that will address service or infrastructure gaps identified in the CARES process. The Capital Investment Panel, which consists of representatives from each VA administration and staff offices, reviews, scores, and ranks these papers. The Capital Investment Panel also identifies the proposals that will be sent forward for additional analysis and review, and may ultimately be included as part of VA’s budget request. According to VA officials, all capital projects must be based on the CARES planning model to advance through VHA’s capital planning process. On the basis of CARES-identified infrastructure needs and service gaps, VA identified more than 100 major capital projects in 37 states, the District of Columbia, and Puerto Rico. In addition to these projects, the CARES planning model identified service needs and infrastructure gaps at other locations throughout the VA system. The model is updated annually to reflect new information. VHA’s 5-year Capital Plan outlines CARES implementation and identifies priority projects that will improve the environment of care at VA medical facilities and ensure more effective operations by redirecting resources from the maintenance of vacant and underutilized buildings to investments in veterans’ health care. In VA’s fiscal year 2010 budget submission, VA requested about $1.1 billion to fund 12 VHA major construction projects and about $507 million for VHA minor construction projects. VA has begun implementing some CARES decisions. Specifically, VA is currently in varying stages (e.g., planning or construction) of implementing 34 of the major capital projects that were identified in the CARES process. Eight major capital CARES projects are complete. Although VA is moving forward with the implementation of some CARES decisions, we previously reported that a number of VA officials and stakeholders, including representatives from veteran service organizations and local community groups, view the implementation process as too lengthy and lacking transparency. For instance, stakeholders in Big Spring, Texas, noted that it took almost 2 years for the Secretary to decide whether to close the facility. During this period, there was a great deal of uncertainty about the future of the facility. As a result, there were problems in attracting and retaining staff at the facility, according to network and local VA officials. We also previously reported that a number of stakeholders we spoke with indicated that the implementation of CARES decisions has been influenced by competing stakeholders’ interests—thereby undermining the process. In its February 2004 report, the CARES Commission also noted that stakeholder and community pressure can act as a barrier to change, by pressuring VA to maintain specific services or facilities. In 2007, we reported that VA does not use, or in some cases does not have, performance measures to assess its progress in implementing CARES or whether CARES is achieving the intended results. Performance measures allow an agency to track its progress in achieving intended results. Performance measures can also help inform management decision making by, for example, indicating a need to redirect resources or shift priorities. In addition, performance measures can be used by stakeholders, such as veterans’ service organizations or local communities, to hold agencies accountable for results. Although VA has over 100 performance measures to monitor other agency programs and activities, these measures either do not directly link to the CARES goals or VA does not use them to centrally monitor the implementation and impact of CARES decisions. We also reported that VA lacked critical data, including data on the cost of and timelines for implementing CARES projects and the potential savings that can be generated by realigning resources. Given the importance of the CARES process, we previously recommended that VA develop performance measures for CARES. Such measures would allow VA officials to monitor the implementation and impact of CARES decisions as well as allow stakeholders to hold VA accountable for results. In responding to our recommendation, VA created the CARES Implementation Monitoring Working Group. This working group has identified performance measures for CARES and the group will monitor the implementation and impact of CARES decisions. VA has a variety of legal authorities available to help it manage real property. These authorities include enhanced-use leases (EUL), sharing agreements, and outleases. (See table 3 for descriptions of these authorities.) VA uses these authorities to help reduce underutilized and vacant property. For example, in 2005, in Lakeside (Chicago), Illinois, VA reduced its underutilized property at the medical center by nearly 600,000 square feet by using its EUL authority with Northwestern Memorial Hospital. VA also uses these authorities to generate financial benefits. For example, the VA Greater Los Angeles Healthcare System enters into a number of sharing agreements with the film industry. VA officials told us that these agreements are typically temporary arrangements—sometimes lasting a few days—during which film production companies use VA facilities to shoot television or movie scenes. According to VA officials, these agreements generate roughly $1 million to $2 million a year. However, legal restrictions associated with implementing some authorities affect VA’s ability to dispose of and reuse property in some locations. For example, legal restrictions limit VA’s ability to dispose of and reuse property in West Los Angeles and North Hills (Sepulveda) California. The Cranston Act of 1988 precluded VA from taking any action to dispose of 109 of 388 acres in the West Los Angeles medical center and 46 acres of the Sepulveda ambulatory care center. In 1991, when EUL authority was provided to VA, VA was prohibited from entering into any EUL relating to the 109 acres at West Los Angeles unless the lease was specifically authorized by law or for a childcare center. The Consolidated Appropriations Act of 2008 expanded the EUL restrictions to include the entire West Los Angeles medical center. The Consolidated Appropriations Act of 2008 also prohibits VA from declaring as excess or otherwise taking action to exchange, trade, auction, transfer, or otherwise dispose of any portion of the 388 acres within the VA West Los Angeles medical center. Budgetary and administrative disincentives associated with some of VA’s available authorities may also limit VA’s ability to use these authorities to reduce its inventory of underutilized and vacant property. For example: VA cannot retain revenue that it obtains from outleases, revocable licenses, or permits; such receipts must be deposited in the Department of the Treasury. VA has said that, except for EUL disposals, restrictions on retaining proceeds from disposal of properties are a disincentive for VA to dispose of property. In 2004, VA was authorized until 2011 to transfer real property under its jurisdiction or control and to retain the proceeds from the transfer in a capital asset fund for property transfer costs, including demolition, environmental remediation, and maintenance and repair costs. In our previous work, we reported several administrative and oversight challenges with using capital asset funds. Moreover, VA officials told us that this authority has significant limitations on the use of any funds generated by disposal. For example, VA officials we spoke with reported that the capital asset fund is too cumbersome to be used, and VA does not have immediate access to the funds because they have to be reappropriated before VA can use them. The maximum term for an outlease, according to VHA law, is 3 years; according to VA officials, this time limit can discourage potential lessees from investing in the property. Implementing an EUL agreement can take a long time. According to VA officials, EULs are a relatively new tool, and every EUL is unique and involves a learning process. In addition, VA officials commented that the EUL process can be complicated. According to VA officials, the average time it takes to implement an EUL can range generally from 9 months to 2 years. The officials noted that land due diligence requirements (such as environmental and historic reviews), public hearings, congressional notification, lease drafting, negotiation, and other phases contribute to the length of the overall process. VA has taken actions to reduce the time it takes to implement an EUL agreement, but despite changes to streamline the EUL process, some officials stated that it is still time consuming and cumbersome. VA can dispose of underutilized and vacant property under the McKinney- Vento Act to other federal agencies and programs for the homeless. However, VA officials stated that disposing of property under the McKinney-Vento Act also can be time-consuming and cumbersome. According to VA officials, the process can average 2 years. Under this law, all properties that the Department of Housing and Urban Development deems suitable for use by the homeless go through a 60-day holding period, during which the property is ineligible for disposal for any other purpose. Interested representatives of the homeless submit to the Department of Health and Human Services (HHS) a written notice of their intent to apply for a property for homeless use during the 60-day holding period. After applicants have given notice of their intent to apply, they have up to 90 days to submit their application to HHS, and HHS has the discretion to extend the time frame if necessary. Once HHS has received an application, it has 25 days to review, accept, or decline the application. Furthermore, according to VA officials, VA may not receive compensation from agreements entered into under the McKinney-Vento Act. Despite these challenges, VA has used these legal authorities to help reduce its inventory of unneeded space. In 2008, we reported that VA reduced underutilized space ( i.e., space not used to full capacity) in its buildings by approximately 64 percent from 15.4 million square feet in fiscal year 2005 to 5.6 million square feet in fiscal year 2007. Although the number of vacant buildings decreased over the period, the amount of vacant space remained relatively unchanged at 7.5 million square feet. We estimated VA spent $175 million in fiscal year 2007 operating underutilized or vacant space at its medical facilities. While VA’s use of various legal authorities, such as EULs and sharing agreements, likely contributed to VA’s overall reduction of underutilized space since fiscal year 2005, VA does not track the overall effect of using these authorities on its space reductions. Without such information, VA does not know what effect these authorities are having on its effort to reduce underutilized or vacant space or which types of authorities have the greatest effect. We concluded that further reductions in underutilized and vacant space will largely depend on VA developing a better understanding of why changes occurred and what impact these agreements had. Therefore, we recommended in our 2008 report that VA track, monitor, and evaluate square footage reductions and financial and nonfinancial benefits resulting from new agreements at the building level by fiscal year in order to better understand the usefulness of these authorities and their overall effect on VA’s inventory of underutilized and vacant property from year to year. The officials said that tracking financial benefits will require a real property cost accounting system which VA is in the process of developing. According to VA officials, VA will institute a system in June 2009 that will track square footage reductions at the building level, but the system will not track financial benefits at this level. Mr. Chairman, this concludes my prepared statement. I would be pleased to respond to questions from you or other Members of the Subcommittee. For further information on this statement, please contact Mark L. Goldstein at (202) 512-2834 or goldsteinm@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Individuals making key contributions to this testimony were Nikki Clowers, Hazel Gumbs, Edward Laughlin, Susan Michal-Smith, and John W. Shumann. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Through its Veterans Health Administration (VHA), the Department of Veterans Affairs (VA) operates one of the largest integrated health care systems in the country. In 1999, GAO reported that better management of VA's large inventory of aged capital assets could result in savings that could be used to enhance health care services for veterans. In response, VA initiated a process known as Capital Asset Realignment for Enhanced Services (CARES). Through CARES, VA sought to determine the future resources needed to provide health care to our nation's veterans. This testimony describes (1) how CARES contributes to VHA's capital planning process, (2) the extent to which VA has implemented CARES decisions, and (3) the types of legal authorities that VA has to manage its real property and the extent to which VA has used these authorities. The testimony is based on GAO's body of work on VA's management of its capital assets, including GAO's 2007 report on VA's implementation of CARES (GAO-07-408). The CARES process provides VA with a blueprint that drives VHA's capital planning efforts. As part of the CARES process, VA adapted a model to estimate demand for health care services and to determine the capacity of its current infrastructure to meet this demand. VA continues to use this model in its capital planning process. The CARES process resulted in capital alignment decisions intended to address gaps in services or infrastructure. These decisions serve as the foundation for VA's capital planning process. According to VA officials, all capital projects must be based on demand projections that use the planning model developed through CARES. VA has started implementing some CARES decisions, but does not centrally track their implementation or monitor the impact of their implementation on its mission. VA is in varying stages (e.g., planning or construction) of implementing 34 of the major capital projects that were identified in the CARES process and has completed 8 projects. Our past work found that, while VA had over 100 performance measures to monitor other agency programs and activities, these measures either did not directly link to the CARES goals or VA did not use them to centrally monitor the implementation and impact of CARES decisions. Without this information, VA could not readily assess the implementation status of CARES decisions, determine the impact of such decisions, or be held accountable for achieving the intended results of CARES. VA has recently created the CARES Implementation Working Group, which has identified performance measures for CARES and will monitor the implementation and impact of CARES decisions in the future. VA has a variety of legal authorities available, such as enhanced-use leases, sharing agreements, and others, to help it manage real property. However, legal restrictions and administrative- and budget-related disincentives associated with implementing some authorities affect VA's ability to dispose and reuse property in some locations. For example, legal restrictions limit VA's ability to dispose of and reuse property in West Los Angeles and Sepulveda. Despite these challenges, VA has used these legal authorities to help reduce underutilized space (i.e., space not used to full capacity). In 2008, we reported that VA reduced underutilized space in its buildings by approximately 64 percent from 15.4 million square feet in fiscal year 2005 to 5.6 million square feet in fiscal year 2007. While VA's use of various legal authorities likely contributed to VA's overall reduction of underutilized space since fiscal year 2005, VA does not track the overall effect of using these authorities on space reductions. Not having such information precludes VA from knowing what effect these authorities are having on reducing underutilized or vacant space or knowing which types of authorities have the greatest effect. According to VA officials, VA will institute a system in 2009 that will track square footage reductions at the building level. |
Established in 1965 as part of the Elementary and Secondary Education Act (ESEA), Title I provides grants to help schools establish and maintain programs to improve the educational opportunities of low-income and disadvantaged students. Most Title I funds are distributed by formula from Education to states. The states then pass through most of these funds to their school districts after retaining some funds—up to 1.5 percent—for state administration and state-level school improvement activities. The amount of Title I funds a school district gets is determined by a formula based on the number of students from low-income families and the state’s per pupil expenditures. Once LEAs receive funds from the state, they have flexibility in how they allocate Title I funds to individual schools and how each school delivers Title I services. As long as priority is given to schools with the highest concentration of children from low-income families, LEAs are generally free to designate which schools, among those eligible, receive funds and how much each should get. LEAs can also select the type of framework through which they deliver Title I services. Some districts have only targeted assistance in their Title I schools, some only have schoolwide programs, and others have a mixture of both. When Title I began, all schools administered targeted assistance programs. These programs targeted funds and services—such as teachers and materials—to specific qualified students who met Title I eligibility requirements. In 1978, a limited number of schools were allowed to deliver the services in the form of schoolwide programs if 75 percent or more of their student population was poor. Schools choosing to operate schoolwide programs can combine federal resources with other funds to improve the school as a whole and help all students achieve. In subsequent reauthorizations, the schoolwide option was made available to more schools by lowering the threshold percentage of low-income children required to operate a schoolwide program. NCLB allows schoolwide programs in schools with a poverty rate of 40 percent or more. During the mid-1990s there was a trend toward providing more flexibility to state and local recipients of federal grants so they may operate programs that best serve the needs of their communities. This trend towards flexibility is evident, not only in education programs such as Title I, but in many social service programs and health programs. In these circumstances, we have found that new approaches to ensuring accountability need to be designed to achieve a balance between flexibility and accountability for attaining certain national objectives. In 2001, we reported on some of the challenges in maintaining a federal-state fiscal partnership in welfare reform and concluded that a broad-based maintenance of effort requirement calling for states to maintain spending across a wide range of relevant programs might both limit substitution of state funds while at the same time preserve state and local flexibility better than a traditional supplement-non-supplant requirement. Specifically, we found that, once accountability shifts to the broad purposes of the grant, federal fiscal oversight needed to shift as well and focus not only on the specifics of welfare funding but also on how states used multiple funding streams—federal, state, and local—to accomplish the program’s broad goals. These findings could apply to the Title I program because schoolwide program goals are broader than the goals of a targeted assistance program and schoolwide programs combine funding streams. Title I contains three fiscal requirements that grantees must comply with in order to continue to receive Title I funds from one year to the next. If an SEA or LEA fails to comply with MOE, SNS, or Comparability provisions, it is required by law to return the amount of misused funds to Education. Maintenance of effort (MOE). An LEA may receive funds if the SEA finds that the LEA’s combined fiscal effort per student or the aggregate expenditures of the LEA from state and local funds for free public education for the preceding year is not less than 90 percent of the combined fiscal effort or aggregate expenditures for the second preceding year. Supplement-not-supplant (SNS). State and local education agencies must use federal funds to supplement, and not supplant, the amount of funds that would, in the absence of federal funds, be made available from nonfederal sources for the education of Title I students. Comparability. State and local funds must be used to provide services in Title I schools that are “at least comparable” to services provided by state and local funds in non-Title I schools within the same LEA. Each fiscal requirement is enforced at a different level. For example, the MOE requirement applies only to the LEA not to individual Title I schools. The Comparability requirement is evaluated at the school level because it seeks to weigh the services provided in Title I schools with those provided in non-Title I schools. In contrast, the SNS provision is applied differently depending on how Title I services are applied. It is applied to the program or the student in targeted assistance programs to ensure that those targeted programs are providing more services for a Title I student than non-Title I students receive, or it is applied to the school if it operates a schoolwide program. (See fig. 1.) There are a variety of approaches that officials at the federal, state, and local levels take to oversee state and local education agency compliance with the fiscal accountability provisions associated with Title I, including formal monitoring systems, such as the use of the single audit, and more informal monitoring systems, such as monitoring provided by interest groups. Education distributes Title I funds to the individual states and has primary responsibility for overseeing federal education programs and providing guidance and technical assistance to SEAs. Monitoring efforts focus on state compliance with both programmatic and fiscal requirements. Any issues of noncompliance reported at the state level are to be communicated to Education and are typically resolved through the development of an SEA corrective action plan, the implementation of which will be monitored by federal agency officials. States are considered the primary recipient, or grantee, of federal awards like Title I and are responsible for ensuring that their subrecipients comply with all federal laws and regulations governing the grant. Since SEAs pass through most of the federal funds to the LEAs, states must have the appropriate subrecipient monitoring systems in place to track Title I spending. Program monitoring systems typically include a review of funding applications, local budgets, self-assessment documents, scheduled on-site visits to schools, and technical assistance. The Single Audit Act replaced multiple audits of separate grant awards with one organizationwide audit. A single audit includes an audit of the federal grant recipient’s financial statements as well as an examination of its internal controls and its compliance with laws and regulations governing federal awards. It does not, however, cover every federal grant received by the organization. The objectives of the Single Audit Act are as follows: Promote sound financial management, including effective internal controls, with respect to federal awards administered by nonfederal entities. Establish uniform requirements for audits of federal awards administered by nonfederal entities. Promote the efficient and effective use of audit resources. Reduce burdens on state and local governments, Indian tribes, and nonprofit organizations. Ensure that federal departments and agencies rely on and use audit work done pursuant to the act. In 1994, we reported that state and local officials had reported that the single audit process had contributed to improving state and local government financial management practices. Guidance for conducting a single audit is found in the Office of Management and Budget’s (OMB) Circular A-133 and the accompanying compliance supplement. The guidance states that the scope of the audit shall include an examination of financial statements—to determine if they are presented fairly in all material respects in conformity with generally accepted accounting principles and whether the schedule of expenditures of federal awards is presented fairly; internal controls—to obtain an understanding of internal control over federal programs sufficient to plan the audit to support a low assessed level of control risk; compliance—to determine whether the auditee has complied with laws, regulations, and the provisions of contracts or grant agreements that may have a direct and material effect on the federal program on each of its major programs; and prior audit findings—to perform procedures to assess the reasonableness of the summary schedule of prior audit findings. Any single audit report should discuss the auditor’s analysis of these areas and include a section that specifically focuses on federal awards, including a schedule of findings and questioned costs. State and local governments and nonprofit organizations that spend $300,000 or more in federal awards in a fiscal year must undertake a single audit. In addition to formal monitoring systems, fiscal accountability is also monitored informally by interest groups, parents groups, individuals, and the media. The public nature and easy accessibility of school district budgets, financial reports, and other fiscal information promotes budget transparency and information sharing among people outside the school system. This informal system may promote grantee compliance with applicable laws and regulations and raise red flags for the attention of the formal monitoring system. In the states we visited, state program officials use three tools—the states’ annual financial reports, the single audit process, and limited program monitoring—to monitor Title I’s fiscal accountability requirements in their LEAs. In these states, enforcing the MOE provision is straightforward and objective. However, a number of factors made it difficult to ensure compliance with SNS. In the states we visited, verifying compliance with Title I’s MOE requirement was a straightforward mathematical exercise and relied on LEAs’ data gathered through statewide financial accounting systems. In part, monitoring and enforcing compliance with the MOE provision might have presented few challenges because until recently state and local revenues were increasing, and few grantees struggled to meet the MOE requirement. Many state and LEA officials told us that the robust economy and sound fiscal situation they experienced in the late 1990s allowed them to increase spending on education. Each of the six states we visited has strong vested interests in the integrity of its LEAs’ financial reports because each has a large stake in education finance in their states; all of these states mandate the level of effort the local education agencies must provide each year in order to receive state funds and impose their own financial reporting requirements on local education agencies. While we did not verify the quality of the data, it is the same data used to calculate the MOE requirements. Five of the six SEAs we visited use their annual financial reports to verify LEA compliance with the MOE requirements. In the sixth, Florida, the SEA relied on LEAs to submit a separate form verifying that they were in compliance with MOE requirements. A program official verified the form submitted against the previous year’s submission and other grant award documentation but did not independently check against the state’s accounting records. However, this check is done by auditors in separate compliance reviews. State officials in Florida said that they were considering changing their MOE verification process. They said that audited data were available from their annual financial reporting system, and they were considering streamlining the verification process to eliminate the separate reporting requirement. Two of the states we visited, Arizona and California, do not verify LEA compliance with MOE requirements until after the current year’s grant has been awarded. This practice is due to routine delays in year-end account reconciliation and timing of audits. The Arizona Office of the Auditor General cited the state’s department of education for failing to enforce the MOE provisions before the current year’s grant was awarded. State program officials acknowledged that they complete the grant award process before the audited financial data are available to verify compliance with MOE; however, they said that verification is finished well before the funds have been disbursed. Similarly, in California the audited data are not available until 9 months after the grant has been awarded. State program officials said that, despite these delays, there were few risks that they would be unable to collect the penalties against an LEA that was out of compliance with its MOE requirements. While verifying Title I’s MOE requirement was straightforward and objective, verifying compliance with the SNS provision was more challenging. SEA officials relied primarily on the single audit process to enforce the SNS provisions. But, many state and local program officials and auditors we spoke with cited a number of factors that made it difficult to ensure that grantees were in fact using federal funds to supplement and not supplant their own funds. These factors include difficulties applying the SNS provision to unique circumstances in their school districts, reliance on the single audit to ensure compliance without understanding its scope and methodology, and limited state and local program oversight of the fiscal accountability provisions. It can be difficult for auditors to establish a finding of supplantation. One of the challenges auditors face in evaluating compliance with the SNS requirement is determining the basis for the states’ and LEAs’ funding decisions. The SNS requirement generally prohibits replacing state funds with federal funds where the displaced state funds could continue to be available for their original purpose. However, under certain circumstances, where state funding is discontinued, grantees may be able to replace these eliminated funds with Title I dollars. For example, if an SEA or LEA discontinued its own support for a particular program in response to a potential budget deficit, the use of Title I funds may be permissible. The challenge for auditors in deciding if an LEA has improperly supplanted, is determining what the LEA would have done in the absence of federal funds. For example, where a state reduces its own financial support for a program and uses federal funds instead, an auditor may presume supplanting has occurred; but, a grantee could rebut that presumption by presenting evidence that fiscal stress required state budget cuts that might not have otherwise been considered. The statute also permits states to use Title I funds to replace state or local funds that had been expended for a program meeting a Title I purpose by allowing such supplemental state funds to be excluded from the SNS compliance determination. In other words, an LEA is allowed to shift funds from one state or locally funded program targeted to low-income children, substitute federal funds for that program, and move its own funds to other priorities for disadvantaged children. Even if auditors could determine what a grantee would have done if it had not received federal funds, the way that Title I services are delivered can also make it difficult to apply the SNS provisions. Table 1 summarizes the relationships between the different ways programs are delivered— targeted assistance, schoolwide programs, and districtwide reforms and the application of the fiscal accountability provisions. For schoolwide programs the distinction between state/local funds and federal funds—and hence the notion of supplantation—becomes unclear. In general, when services are delivered through schoolwide programs, federal, state, and local funds are pooled, making it impossible to distinguish among funding streams in an audit because a schoolwide school does not have to (1) show that Title I funds are paying for additional services, (2) demonstrate that Title I funds are used only for specific target populations, or (3) separately track federal program funds once they reach the school. While one can identify the separate funding sources going into a school one cannot identify what services they funded. Therefore, for schoolwide programs a test for SNS compliance could include either (1) a comparison from one year to the next of total— federal, state, and local—funds allocated to a Title I school or (2) a comparison of state and local funds spent in Title I schools and non-Title I schools. However, there are problems applying either test to schoolwide programs. For example, in the Glendale (Arizona) Elementary School District, every school is a Title I school and all schools operate schoolwide programs. District officials argued that because the district met its MOE requirement (in 2000-2001 it exceeded 100 percent of its preceding years expenditures), it did not need a separate internal control procedure to test for SNS. However, auditors cited the district for not having such a procedure. Our analysis shows that, for districts such as Glendale, in order to avoid supplanting funds, the district would have to maintain the same state and local funding from year to year. In other words, in districts where every school is a Title I school and all schools operate a schoolwide program, they would have to maintain a much higher MOE requirement, 100 percent, than districts that are not in this circumstance in order to avoid supplanting funds unless they otherwise would not have spent those funds. Moreover, comparing expenditures from one year to the next in school districts where there are both targeted assistance and schoolwide schools presents challenges. Since schoolwide program administrators can reallocate funds among programs in their schools, they can engage in budgetary practices that are not allowed in a targeted assistance school in the same district. Theoretically, the SNS provision imposes a higher expectation on schools operating schoolwide programs than it does on their targeted assistance counterparts because a year-to-year funding comparison essentially requires a schoolwide school to maintain 100 percent of its previous effort in order to comply with the SNS provision. Furthermore, comparing the allocation of state/local funds among schools in the same year also presents challenges. For example, in Duval County (Florida) all 72 of the district’s Title I schools operate schoolwide programs but not all schools were Title I schools. Duval’s auditors assessed compliance with SNS by comparing the per pupil expenditure of state and local funds in Title I schools to the allocation in non-Title I schools within the same year. They found that in 2001, 5 of the district’s 72 Title I schools received significantly less state and local funding per pupil than the average school received, resulting in questioned costs of $2.5 million. The auditors suspected that the district may have used federal Title I funds in place of state and local funds in these schools, but district officials claimed that many mitigating factors, such as higher teacher salaries for more experienced teachers, could explain variations in the per pupil expenditures among schools in the same district. The SEA determined that the information presented in the audit report was not sufficient to determine that supplanting occurred. SEA officials told us that auditors would need to have programmatic expertise to interpret the results of the per pupil cost comparisons in order to prove that supplantation had occurred. The auditors agreed that many factors could have contributed to the observed disparity in funding among the 5 Title I schools, but they said there were limitations to what could be expected of a single audit and pointed out that ultimately the SEA should use the audit findings as a basis for determining whether the LEA is in compliance or not. Finally, when a district engages in comprehensive districtwide reform resulting in programmatic changes, it can be difficult to make the types of comparisons necessary to determine compliance with SNS, particularly in the first year of reforms. For example, in 2000, San Diego City (California) Unified School District (USD) began school reform that entailed financing new initiatives. As these reforms were implemented, the district and its programs were restructured in such a way that there were no longer points of comparison for determining whether the district was in compliance with the SNS provision. In other words, funding for programs in the current school year could not be compared with funding for those in the previous year, because the programs had not previously existed. The Superintendent of the San Diego City USD told us that the reform plan could not have been implemented without the flexibility to reallocate resources within and among schools. While some auditors struggled to apply the SNS provision to the particular circumstance of districts and schools, SEA officials were frequently unaware of what the results of a single audit actually meant, potentially failing to cover key programs or provisions of law, and thus adding to the difficulty of enforcing the SNS provision. For example, numerous state and local program officials told us that they assumed that the single audit covered every LEA and every program, even though this is not what single audits are designed to do. As a result, these officials may not engage in other oversight activities that are warranted. Because only those grant recipients that spend more than $300,000 in federal awards in any given year must undertake a single audit, not all LEAs that receive Title I funds are required to undergo one. Furthermore, even if an LEA is audited, the Title I program may not be covered in the audit. The 1996 amendments to the Single Audit Act give auditors more freedom to determine which federal programs to include in their audit plan each year, allowing them to exclude some programs based on risk-based criteria and on expenditure- based criteria. Many of the auditors we spoke with assessed risk by determining whether or not there had been findings of noncompliance in recent audits. For example, if an LEA had a clean audit with respect to the Title I program for the last few years, an auditor might legitimately view the inherent risk of this program as low and exclude it from the audit in the next year. Auditors for both San Diego City (California) USD and Jefferson Parish (Louisiana) Public Schools told us that Title I probably would not be covered in the districts’ 2002 single audit since there have been no recent findings on the program. In addition, some officials thought that the single audit examined every transaction, even though it does not. As a result, officials may think that by fixing the instances reported they are solving all the problems, when in fact those problems may be more widespread. Generally accepted government auditing standards allow statistical sampling methods and auditors often use audit sampling to evaluate compliance with applicable requirements. This involves testing less than 100 percent of the items within a group of transactions for the purpose of evaluating compliance with applicable laws and regulations. Transactions could be randomly selected from all of the auditee’s financial transactions for the year under review. While this technique allows auditors to test the population of transactions for evidence of noncompliance and internal control weaknesses, it will not identify every specific instance of noncompliance. When auditors of Douglas (Arizona) USD identified significant internal control weaknesses based on analysis of a sample of the district’s financial they reported that their review of the district’s internal controls would not necessarily disclose all instances of non-compliance. However, the SEA resolved the issue by requiring the auditee to reimburse the Title I program for the amount of the transaction under question only and did not further investigate if there were other erroneous payments made. While there were problems with program officials understanding what single audits are and what results from the single audit meant, in general the auditors’ work plans we reviewed followed the guidance recommended by Education and OMB for single audits. However, some of the auditors could not document that they had followed their work plans. For example, audit workpapers for San Diego City (California) USD show that auditors held a discussion with a district budget official who told the auditors that they were in compliance with the SNS provision, but the workpapers did not indicate independent verification of these claims. In five of the six states we reviewed, the SEA had a procedure in place to resolve audit findings reported through the single audit process. However, in 2001 the Louisiana Legislative Auditor reported in its statewide single audit that the SEA did not have adequate internal controls to monitor subrecipients for compliance with many federal education programs, including Title I. The SEA concurred with the auditor’s finding and has implemented policies to address the deficiencies. All of the states we visited supplemented their reviews of LEAs’ single audit reports with additional monitoring activities. While program monitoring provided a depth of coverage that cannot be achieved in single audits, these efforts were limited. Furthermore, in all of the states we visited, the primary focus of any additional monitoring activities has now centered on addressing efforts to raise the level of student achievement with considerably less focus given to fiscal accountability requirements. Informal monitoring by individuals and groups augmented the formal monitoring process. Limited program monitoring also took place during the application review process. In all of the states we visited, the annual application process for Title I funding contains questions on historic and proposed program budget information that can be used by program officials in the SEA to oversee compliance with SNS and MOE. In addition, some states require LEAs to complete a self-assessment document in which they are asked to assess themselves on their compliance with federal program requirements. SEAs use these self-assessment documents for various purposes. For example, in Florida, annual self-assessments were used as a self- monitoring tool, but only one-quarter of the LEAs were required to actually send in the completed guide for review in any given year. In Arizona, the tool was also required annually and is designed to provide guidance in program development and to identify areas in which technical assistance may be needed. In California, LEAs must complete self- assessments once every 4 years, at which time SEA officials evaluate the self-assessment documents and use them to target their on-site monitoring activities to those LEAs that pose the highest risks. While four of the six states we visited followed up LEAs’ self-assessments with on-site visits, the extent to which they conducted such visits varied and, in some cases, was limited. Massachusetts and Arizona have scheduled on-site monitoring visits in the LEAs at least once every 6 years. In Louisiana, state officials said that each LEA is visited once every 3 years. In California, while each LEA must go through the review cycle every 4 years. In addition, in 2001 two of the six states we visited, Florida and Indiana, did not follow up on the self-assessments with periodic on- site program monitoring. Several state officials highlighted the importance of informal monitoring networks, such as parents groups, in raising issues of noncompliance. These watchdog groups play an informal role in questioning inappropriate spending and submitting complaints to the school boards and, if they feel their concerns are not addressed at this level, elevating the issues to the SEA. SEA officials in both Indiana and California discussed recent inquiries that were brought to their attention, not through single audit reports or even program monitoring efforts, but rather by informal watchdog groups. In Indiana, the issues raised dealt with unallowable costs and high administrative charges to federal programs in one school; the SEA is investigating and, according to state officials, the matter is still unresolved. In California, a parents group in San Diego filed a complaint with the California Department of Education citing issues relating to, among other things, the reallocation of state and federal funding, including Title I funds, by the San Diego City (California) USD to fund its districtwide school reform strategy which, the watchdog group claimed, no longer provides a comparable level of service to all students with state and local funding. The SEA concurred and ordered the district to develop a plan to allocate the state and local supplemental funds that complies with all the federal comparability provisions. However, Education granted the district a waiver in August 2002 which will allow the district to proceed with the reform strategy under its current budget plan for 1 year. We identified three key efforts the Department of Education made to help enforce the fiscal accountability provisions but each had limitations. First, Education developed guidance and provided technical assistance to state and local officials and their auditors; but, these officials have expressed confusion regarding application of the SNS provision to their particular circumstances. Second, Education conducted limited program monitoring of its own, but these efforts did not have fiscal accountability as a primary focus. Finally, Education reviewed states’ single audit reports conducted under the Single Audit Act. But, the Inspector General of Education found that many reviewers in the department lacked knowledge about the single audit process and compliance issues. As a consequence, Education’s monitors could fail to review key programs or provisions of law. In addition, we recently reported that Education could not demonstrate it consistently worked to resolve audit findings. Education developed guidance for its programs which appeared in the compliance supplement to OMB’s Circular A-133. This guidance was the basis for the audit plans for all the districts we visited. Education’s guidance itemizes the SNS, MOE, and the comparability requirements as separate statutory requirements. However, many state and local officials and auditors we spoke with thought the three requirements were related to each other and that, by meeting one or two of the requirements, they would automatically be in compliance with the others. Some auditors and program officials confuse the comparability requirement with the SNS provision. While comparability is primarily used to ensure that services—not funding—is comparable across schools in the LEA, the two issues are closely related and frequently confused. For example, guidance issued by the SEA in Arizona on the comparability requirement states that comparability is used to ensure that schools within an LEA do not supplant state and local funds with federal program funds. Operating under the same misconception, Indianapolis Public Schools incorrectly used the comparability test as the internal control to ensure compliance with SNS. Moreover, the district’s auditors failed to question the appropriateness of this test to ensure compliance with SNS. A similar confusion was evident when officials in the Duval County Public Schools told auditors that they could not understand how they failed to comply with the prohibition on supplantation, given that they had not cut back on their own overall spending thereby meeting their MOE requirements and had documented meeting their comparability requirement. Education recognizes that there is some confusion about the application of the provisions. Education officials acknowledge the challenges of writing guidance that can be understood and applied in every circumstance. Many federal program officials said that they frequently field questions from district officials and some auditors seeking technical assistance applying the provisions in local circumstances. In December 2002, Education issued new regulations that reorganized its guidance on schoolwide programs in a manner that might help address some of the confusion. State and local education agencies and their auditors told us that they also rely on nongovernmental sources of guidance, such as workshops and materials provided by consulting firms. For example, auditors in Arizona provided us with excerpts from handbooks and other guidance on the Title I program. School officials and auditors in other districts we visited also told us they supplement federal guidance with similar nongovernmental sources of guidance. The fiscal accountability provisions have not been the focus of Education’s own monitoring efforts. From 1995-2001, Education used an approach to program monitoring called an integrated review approach. Its primary focus was to see how all federal grant programs, working together, supported state and local reform efforts. The Title I program was included in these reviews. However, only 1 of the 9 indicators Education’s monitors used in integrated reviews focused on fiscal issues; the rest focused on program performance, such as whether the state supported and promoted high standards for all children, and whether states used education research findings to inform decision making. Education’s Inspector General criticized this approach to program reviews in 2001 because the integrated approach allotted insufficient time to monitor specific programs for compliance with federal laws and regulations. The Inspector General also found that the various teams of reviewers lacked knowledge of the single audit process, thereby taking inconsistent approaches to doing the reviews. In 2002 Education drafted guidelines for its monitors to use in a new approach to program monitoring, but we found that the new approach gives fiscal accountability requirements little emphasis and it does not even mention SNS. In 2002, Education developed a new monitoring strategy which it has named: Achievement Focused Monitoring (AFM). As its name implies, the AFM approach seeks to realign oversight and technical assistance for Title I to concentrate on student achievement. Education officials acknowledged that their program monitoring guide does not mention SNS and said they would provide additional guidance to their monitors on the provision for use in the future. Education’s AFM plan includes visits to 15 states—and at least one district in each state—in 2002 and 2003. By October 2002 Education had completed visits to four states. Education has responsibility for reviewing the audit reports of state education agencies. In 2002, we reported actions were needed to ensure that grantees correct findings identified in state single audit reports. Each state must undertake a single audit each year. Each year their auditors determine which federal programs to include in their audit plan and audit those programs for compliance with the federal laws and regulations covering those grant programs. Although Education had procedures for obtaining states’ single audit reports, distributing audit findings to appropriate audit offices, and assessing the seriousness of the findings, we found that reviewers did not demonstrate they consistently worked to resolve audit findings. Specifically, reviewers did not consistently follow- up with written management decisions on final audit resolution and did not communicate findings to senior department management. We found that few changes have occurred in the relative shares of federal, state’ and local funding for education for school years (SY) 1999-2000 and 2000-2001 (the most recent data available) in the six states we reviewed. (See fig. 2.) It is too soon to tell how recent increases in federal funds for Title I and other federal education programs and the fiscal pressures facing states will affect funding for education in general and how changes, if any, in state and local financing will affect the federal share. However, this information does provide a baseline against which we can compare the impact of increases in federal funds and state and local fiscal pressures in the future. In addition to the concern about the fiscal balance in education funding overall, questions have been raised about the federal share of operating SEAs. SEA operations include the administration of programs—primarily oversight, technical assistance, and training—related to specific federal programs operated at the local level. SEAs may also operate state-level programs, such as vocational rehabilitation. As we noted in a previous report, the level of federal support for SEA operations varied widely among states depending on the number and types of federal and state programs the SEA operates, ranging in fiscal year 1993 from about 10 to about 80 percent, with the average level of support being 41 percent. To update this information, we looked at the federal share of SEA funding in the six states we visited for school year 1999-2000. As in the past, we found the federal share varied, from 18 percent in Florida to 43 percent in Indiana. (See the shaded bars in fig. 3.) Another way to look at the federal share of SEA operating costs is through the number of full-time equivalent positions (FTEs) that are funded by federal funds. Some states operate federal programs at the state level, such as vocational rehabilitation and disability determination. These may require many more SEA FTEs than programs operated at the local level. For example, in 2000-2001, the Florida SEA assumed responsibility for the federal vocational rehabilitation programs that were previously housed in another state department, adding more than 1,000 positions to the SEA and raising its percent of federally funded FTEs from 43 percent to 63 percent. Finally, table 2 provides some additional context when making comparisons and contrasts among the states we visited. Per pupil expenditure calculations serve as a proxy reflecting the cost differences among states in providing education. Single audits are a valuable oversight tool but they cannot be regarded as the sole tool to use in enforcing the compliance requirements. Additional oversight is always necessary to ensure that grantees are in compliance with the laws and regulations governing specific programs and grant management in general. Single audits should inform, not substitute for program monitoring. However, as we have noted, many state officials told us that they relied primarily on the single audits to oversee compliance with federal laws and regulations. Because of this reliance, state program officials responsible for overseeing this program must have a better understanding of the scope and limitations of these audits and supplement the audits with more effective and frequent oversight activities. Instances of noncompliance found in the course of a single audit should trigger a broader search to determine whether the error is systemic. While NCLB emphasizes achieving higher student achievement levels, enforcing fiscal accountability is and will remain a critically important oversight activity. Resources for audit and evaluation activities will remain limited, and, as a result, these resources must be targeted where they will have the greatest impact. As we have noted, ensuring compliance with an MOE provision presents few challenges and requires few additional audit resources, whereas monitoring the SNS provision is very challenging and requires significant audit resources. Maintaining the intergovernmental fiscal partnership in the education of disadvantaged and low-income students presents many challenges. Title I’s two fiscal accountability provisions—the MOE and the SNS provisions— are intended to limit the extent that grantees can use federal funds to replace their own and thereby erode the fiscal partnership. But each provision helps to maintain the fiscal balance in very different ways and at different levels—schools versus districts. The primary effect of a nonsupplant provision is to prevent the reallocation of state and local resources within a Title I school; essentially, that means that expenditures paid for with state and local resources in a Title I school in one year cannot be paid for with federal funds the next year. On the other hand, the MOE provision’s primary effect is to limit the extent to which states and LEAs can use federal funds for general fiscal relief; that is, substituting federal funds for state and local funds generally, not just in Title I schools. As noted, in schoolwide programs grantees are not required to show that Title I funds are paying for additional services or are targeted to specific students, nor are they required to separately track federal program funds with other funds once they reach the school, thus “limiting the reallocation of resources” becomes unworkable in a schoolwide setting. An inherent tension exists between fostering a flexible grant environment and ensuring fiscal accountability. For broader purpose grants, such as schoolwide programs, the SNS provision can work to constrain local flexibility in the use of federal funds by preventing districts from reallocating the use of federal, state, and local funds. Moreover, the provision is difficult to apply and can be very challenging to monitor and enforce, primarily because it is not workable in those environments. As we have previously reported, in flexible grant environments a strong MOE provision may prove more useful than an SNS provision in limiting the degree to which grantees can use federal funds to simply reduce their overall fiscal commitments. That different parties would have different views of the value of the nonsupplant provisions is to be expected. Some argue that allowing supplantation of any kind increases the likelihood that states could weaken their commitment to educating disadvantaged children and diminish the fiscal impact of the federal grant. Potentially, supplantation allows the SEAs and LEAs to convert the federal Title I grant into a kind of revenue sharing program with very little incremental impact on education spending. Others would point to periodic changes to the Title I program allowing more schools to participate as schoolwide programs, suggesting that the Congress may be trying to encourage more flexible use of Title I funds to improve the quality of education for disadvantaged students and raise student achievement levels for all students, including low-income students. Furthermore, in times of fiscal stress and greater needs in educating the disadvantaged, the reallocation of resources within and among schools may be the only way to finance comprehensive districtwide reform efforts. A nonsupplant provision could stymie those districts that need more flexibility to attempt such reforms. To better align its expectations for accountability with Title I schoolwide program goals, the Congress should consider eliminating the SNS requirement for schoolwide programs. If Congress eliminates SNS in the context of schoolwide programs, Congress may want to consider strengthening the other fiscal accountability requirement, MOE. Currently, LEAs must maintain only 90 percent of their previous years’ expenditures in order to participate in the Title I program. For example, if this requirement were increased, it would impose a higher expectation on those districts to maintain the fiscal balance and it could represent a reasonable tradeoff for those districts that want to begin more comprehensive reform efforts. We recommend that the U.S. Department of Education enhance its technical assistance and training efforts to ensure that SEAs and Education program staff have a clearer understanding of the strengths and weaknesses of the single audit process and the role the audits can play in required oversight activities and encourage them to heighten the level of attention they give the fiscal requirements in their own monitoring efforts. In addition, we recommend that Education amend its guidance for grantees and oversight officers to address all of Title I’s fiscal requirements, including the SNS provision. We received comments from Education on a draft of this report, which are reprinted in Appendix II. Education generally agreed with our recommendations for executive action to enhance its technical assistance and training efforts on the single audit process and to amend its own guidance to address all of Title I’s fiscal accountability provisions. On the policy issue of whether to eliminate the SNS requirement for schoolwide programs, Education is not ready to take a position. However, Education questioned the basis for the matter for congressional consideration that we propose. Education acknowledges the difficulties enforcing the SNS provision in schoolwide programs and we found that none of the districts we visited were able to develop a test for SNS that could be applied in a schoolwide setting. Education cited recent supplanting violations found by Title I monitoring staff to show that it was possible to assess supplantation in a schoolwide setting. However, according to an Education official, these findings were not for schoolwide programs. Education says that the loss of the SNS requirement would not be completely offset by an enhanced MOE requirement because it would shift responsibility for fiscal accountability from the school to the district level. However, our review shows that the current requirement is unworkable in a schoolwide setting. As we said, while one can identify the separate funding sources going into a school, one cannot identify what services they funded in a schoolwide setting because federal, state, and local funds are pooled. In contrast, an MOE requirement is easier to measure, identify, and track, and therefore better promotes fiscal accountability in these settings. If Congress considers eliminating the SNS provision, we believe that enhancing the MOE requirement is a reasonable tradeoff. With regard to Education’s regulation governing the SNS requirement that Education said we did not discuss, we did discuss this on page 13. We have added a footnote to make the report more clear on that point. Education said that it did not agree that the level or scope of monitoring is inadequate. However, we found that Education’s efforts to enforce the fiscal provisions have some limitations. By design, Education’s current monitoring effort is directed at the provisions on accountability for academic results, but we found that the fiscal requirements were given little attention, and the materials developed by Education to guide monitoring efforts did not even mention SNS. Finally, with regard to Education’s review of single audit reports, this finding was published previously in our June 2002 report and specifically assessed the Title I program. The department concurred with our findings at that time and has provided us with a corrective action plan. Secretary Paige’s August 26, 2002, letter to GAO indicated that it planned to address these findings by February 28, 2003. In addition, we provided segments of this draft report to the states and school districts we visited. We have incorporated their comments in the report as appropriate. We are sending copies of this report to the Secretary of Education, appropriate congressional committees, and other interested parties. In addition, the report will be available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staff have any questions or wish to discuss this material further, please call Paul L. Posner at (202) 512-9573 or Marnie S. Shaul at (202) 512-7215. Other GAO contacts and staff acknowledgments are listed in appendix III. To determine how select states ensure compliance with maintenance of effort (MOE) and supplement-not-supplant (SNS), we interviewed state program and budget officials in six states: Arizona, California, Florida, Indiana, Louisiana, and Massachusetts. We also reviewed budgets and financial statements for school years 1999-2000 and 2000-2001, as well as state guidance on fiscal accountability requirements. We also spoke with state auditors and reviewed their audit plans and other relevant workpapers. In addition to meeting with state officials, we spoke with local program and budget officials and school district administrators in six local education agencies including Douglas Unified School District and Glendale Elementary School District in Arizona, San Diego City Unified School District in California, Duval County Public Schools in Florida, Indianapolis Public Schools in Indiana, and Jefferson Parish Public Schools in Louisiana. Again, we reviewed budgets and financial statements for school years 1999-2000 and 2000-2001 and local auditors’ audit plans and relevant workpapers. We selected two of the states and three local school districts based on our search of the Federal Audit Clearinghouse, which is a Web-based database that we searched to identify states and school districts found out of compliance with one or more of the Title I fiscal accountability requirements in 2001. Two of the six states we selected were out of compliance with one of the fiscal accountability requirements, while the other four were not. Likewise, three of the school districts we visited were found to be out of compliance with the SNS provisions; the other three were not. Those states and local school districts without audit findings were selected to ensure variation in enrollment size, ethnic composition, economic condition, and geographic location. To determine what efforts the U.S. Department of Education has taken to enforce the Title I fiscal accountability provisions and what limitations, if any, these efforts may have, we spoke with Education officials and reviewed Education guidance and documentation as well as recent GAO and OIG reports. To assess what changes occurred between school years 1999-2000 and 2000-01 in the federal share of education expenditures and to what extent federal funds were used to support state education agencies’ operating expenditures, we gathered information from state program and budget officials on federal, state, and local funding streams as well as full time equivalent (FTE) and operating expenditure data. We analyzed and summarized this information and presented it in a way that provides context and comparison across the six states. Due to the limited number of states and districts selected, our findings cannot be generalized to school districts nationwide. In addition to those named above Bill Keller, Jennifer Ashford, and Leah Nash made key contributions to this report. Patrick DiBattista provided exceptional editorial assistance on the content and message of the report. Behn Miller and Richard P. Burkard supplied legal advice on several complex aspects of our work. Thomas Broderick and Jacquelyn Hamilton provided technical assistance on the Single Audit Act. Education School Finance: Per-Pupil Spending Differences between Selected Inner City and Suburban Schools Varied by Metropolitan Area. GAO-03-234. Washington, D.C.: December 9, 2002. Title I: Education Needs to Monitor States’ Scoring of Assessments. GAO- 02-393. Washington, D.C.: April 1, 2002. Title I Funding: Poor Children Benefit Though Funding Per Poor Child Differs. GAO-02-242. Washington, D.C.: January 31, 2002. Title I Preschool Education: More Children Served, but Gauging Effect on School Readiness Difficult. GAO/HEHS-00-171. Washington, D.C.: September 20, 2000. Title I Program: Stronger Accountability Needed for Performance of Disadvantaged Students. GAO/HEHS-00-89. Washington, D.C.: June 1, 2000. Education Finance: Extent of Federal Funding in State Education Agencies. GAO/HEHS-95-3. Washington, D.C.: October 14, 1994. Single Audit: Single Audit Act Effectiveness Issues. GAO-02-877T. Washington, D.C.: June 26, 2002. Single Audit: Actions Needed to Ensure That Findings Are Corrected. GAO-02-705. Washington, D.C.: June 26, 2002. Single Audit: Survey of CFO Act Agencies. GAO-02-376. Washington, D.C.: March 15, 2002. Single Audit: Update on the Implementation of the Single Audit Act Amendments of 1996. GAO/AIMD-00-293. Washington, D.C.: September 29, 2000. Single Audit: Refinements Can Improve Usefulness. GAO/AIMD-94-133. Washington, D.C.: June 21, 1994. Welfare Reform: Challenges in Maintaining a Federal/State Fiscal Partnership. GAO-01-828. Washington, D.C.: August 10, 2001. Welfare Reform: Challenges in Saving for a Rainy Day. GAO-01-674T. Washington, D.C.: April 26, 2001. Welfare Reform: Early Fiscal Effects of the TANF Block Grant. GAO/AIMD-98-137. Washington, D.C.: August 18, 1998. Federal Grants: Design Improvements Could Help Federal Resources Go Further. GAO/AIMD-97-7. Washington, D.C.: December 18, 1996. Block Grants: Issues in Designing Accountability Provisions. GAO/AIMD-95-226. Washington, D.C.: September 1, 1995. Block Grants: Characteristics, Experience, and Lessons Learned. GAO/HEHS-95-74. Washington, D.C.: February 9, 1995. Proposed Changes in Federal Matching and Maintenance of Effort Requirements. GAO/GGD-81-7. Washington, D.C.: December 23, 1980. | New resources for education come at a time when states are struggling to address budget shortfalls. Two provisions in Title I--maintenance of effort (MOE) and supplement not supplant (SNS)--are designed to limit the extent to which federal funds could be used to replace state and local resources. To assess the quality of oversight of these provisions, GAO determined (1) how 6 states--Arizona, California, Florida, Indiana, Louisiana, and Massachusetts--conducted oversight of the MOE and SNS provisions and what factors affected their ability to do so; (2) what efforts were made by the U.S. Department of Education to enforce MOE and SNS; and (3) in the 6 states, what changes have occurred in the federal share of education funding from school year 1999-2000 to 2000-2001. In the states we visited, state program officials used three tools--the states' annual financial reports, the single audit process, and limited program monitoring--to oversee Title I's fiscal accountability requirements. While program officials had little difficulty in applying the MOE provision because it involves a straightforward calculation, state and local program officials and auditors we spoke with cited a number of factors that made it difficult to enforce the SNS provision under certain circumstances. One of the challenges auditors faced was determining whether a school district would have removed its own funds from a program and allocated them elsewhere even if federal funds had not been available--an action that is allowable. Another challenge was applying the SNS provision in circumstances where it is difficult to track federal dollars such as in schoolwide programs--where all funds are pooled--or in districts undergoing significant districtwide reforms--where comparisons to previous budgets are problematic. While some auditors struggled to apply the SNS provision to the particular circumstance of districts and schools, program officials relied primarily on the results of the single audits without being aware of some of these audit's limitations. For example, some officials did not understand that not all districts, programs, or transactions may be covered by the audit. While program monitoring adds a degree of depth to the efforts to oversee the SNS provision, most of the states in GAO's review conducted only limited program monitoring. We identified three key efforts Education made to guide, monitor, and enforce the fiscal accountability provisions, but each had limitations. First, Education provided guidance and technical assistance to state and local education agencies and auditors on how to interpret and apply Title I's fiscal accountability requirements. Despite the availability of this guidance, many of the auditors and program officials we spoke with expressed confusion regarding the application of these provisions to their particular circumstances, such as schoolwide programs. Second, Education conducted program monitoring of select state and local education programs each year; however, coverage was limited. Third, Education, reviewed the audit reports conducted under the Single Audit Act. However, Education's Office of Inspector General and GAO have criticized the review and audit follow up process. Few changes occurred in the federal/state/local fiscal partnership in financing education services between school year 1999-2000 and 2000-2001. It is too soon to tell how recent increases in federal funds and state and local fiscal pressures will affect funding for education and the federal share. |
Since the early 1990s, increasing computer interconnectivity—most notably growth in the use of the Internet—has revolutionized the way that our government, our nation, and much of the world communicate and conduct business. The benefits have been enormous, but without proper safeguards in the form of appropriate information security, this widespread interconnectivity also poses significant risks to the government’s computer systems and the critical operations and infrastructures they support. In prior reviews we have repeatedly identified weaknesses in almost all areas of information security controls at major federal agencies, including VA, and we have identified information security as a high risk area across the federal government since 1997. In July 2005, we reported that pervasive weaknesses in the 24 major agencies’ information security policies and practices threatened the integrity, confidentiality, and availability of federal information and information systems. As we reported, although federal agencies showed improvement in addressing information security, they also continued to have significant control weaknesses that put federal operations and assets at risk of inadvertent or deliberate misuse, financial information at risk of unauthorized modification or destruction, sensitive information at risk of inappropriate disclosure, and critical operations at risk of disruption. These weaknesses existed primarily because agencies had not yet fully implemented strong information security programs, as required by the Federal Information Security Management Act (FISMA). The significance of these weaknesses led us to conclude in the audit of the federal government’s fiscal year 2005 financial statements that information security was a material weakness. Our audits also identified instances of similar types of weaknesses in nonfinancial systems. Weaknesses continued to be reported in each of the major areas of general controls: that is, the policies, procedures, and technical controls that apply to all or a large segment of an entity’s information systems and help ensure their proper operation. To fully understand the significance of the weaknesses we identified, it is necessary to link them to the risks they present to federal operations and assets. Virtually all federal operations are supported by automated systems and electronic data, without which agencies would find it difficult, if not impossible, to carry out their missions and account for their resources. The following examples show the broad array of federal operations and assets placed at risk by information security weaknesses: ● Resources, such as federal payments and collections, could be lost or stolen. ● Computer resources could be used for unauthorized purposes or to launch attacks on others. ● Personal information, such as taxpayer data, social security records, and medical records, and proprietary business information could be inappropriately disclosed, browsed, or copied for purposes of identity theft, industrial espionage, or other types of crime. ● Critical operations, such as those supporting national defense and emergency services, could be disrupted. ● Data could be modified or destroyed for purposes of fraud, theft of assets, or disruption. ● Agency missions could be undermined by embarrassing incidents that result in diminished confidence in their ability to conduct operations and fulfill their fiduciary responsibilities. The potential disclosure of personal information raise identity theft and privacy concerns. Identity theft generally involves the fraudulent use of another person’s identifying information— such as Social Security number, date of birth, or mother’s maiden name—to establish credit, run up debt, or take over existing financial accounts. According to identity theft experts, individuals whose identities have been stolen can spend months or years and thousands of dollars clearing their names. Some individuals have lost job opportunities, been refused loans, or even been arrested fo crimes they did not commit as a result of identity theft. The Feder Trade Commission (FTC) reported in 2005 that identity theft represented about 40 percent of all the consumer fraud complaints it received during each of the last 3 calendar years. Beyond the serious issues surrounding identity theft, the unauthorized disclosure of personal information also represents a breach of individuals’ privacy rights to have control over their own information and to be aware of who has access to this information. Federal agencies are subject to security and privacy laws aimed in part at preventing security breaches, including breaches that could enable identity theft. FISMA is the primary l federal government; it also addresses the protection of personal n information in the context of securing federal agency informatio r and information systems. The act defines federal requirements fo securing information and information systems that support federalaw governing information security in the agency operations and assets. Under FISMA, agencies are requiredto provide sufficient safeguards to cost-effectively protect their information and information systems from unauthorized access, use disclosure, disruption, modification, or destruction, including controls necessary to preserve authorized restrictions on access and disclosure (and thus to protect personal privacy, among other things). The act requires each agency to develop, document, and implement an agencywide information security program to provide , security for the information and information systems that support the operations and assets of the agency, including those provided or managed by another agency, contractor, or other source. FISMA describes a comprehensive information security pr including the following elements: periodic assessments of the risk an result from the unauthorized access, use, disclosure, disruption, modification, or destruction of information or information systems; d magnitude of harm that could risk-based policies and procedures that cost-effectively reduce ris to an acceptable level and ensure that security is addressed throughout the life cycle of each information system; security awareness training for agency personnel, including contractors and other users of information systems th operations and assets of the agency; periodic testing and evaluation of the effectiveness of information security policies, procedures, and practices; ● a process for planning, implementing, evaluating, and documentingremedial action to address any deficiencies th and milestones; and procedures for detecting, reporting, and responding to security incidents. In particula agencies evaluate the associated risk according to three categories: (1) confidentiality, which is the risk associated with unauthorized disclosure of the information; (2) integrity, the risk of unauthorized modification or destruction of the information; and (3) availability, which is the risk of disruption of access to or use of information. Thus, each agency should assess the risk associated with personal data held by the agency and develop appropriate protections. r, FISMA requires that for any information they hold, The agency can use this risk assessment to determine the appropriate controls (operational, technical, and managerial) th will reduce the risk to an acceptably low level. For exampl agency assesses the confidentiality risk of the personal information as high, the agency could create control mechanisms to help prote ct the data from unauthorized disclosure. Besides appropriate policies,at e, if an these controls would include access controls and monitoring systems: Access con confidentiality of information. Organizations use these controls to grant employees the authority to read or modify only the information the employees need to perform their duties. In addition access controls can limit the activities that an employee c perform on data. For example, an employee may be given the right to read data, but not to modify or copy it. Assignment of right s and permissions must be carefully considered to avoid giving users unnecessary access to sensitive files and directories. trols are key technical controls to protect the To ensure that controls are, in fact, implemented and that no violations have occurred, agencies need to monitor compliance with security policies and investigate security violations. It is crucial to determine what, when, and by whom specific actions are taken on a system. Organizations accomplish this by implementing system or security software that provides an audit trail that they can use to determine the source of a transaction or attempted transaction and to monitor users’ activities. The way in which organizations configure system or security software determines the nature and extent of information that can be provided by the audit trail. To be effective, organizations should configure their software to collect and maintain audit trails that are sufficient to track security events. A comprehensive security program of the type described is a prerequisite for the protection of personally identifiable information held by agencies. In addition, agencies are subject to requirem specifically related to personal privacy protection, which come primarily from two laws, the Privacy Act of 1974 and the E- Government Act of 2002. The Privacy Act places lim disclosure, and use of personal information maintained in systems of records. The act describes a “record” as any item, collect ion, or grouping of information about an individual that is maintained by an agency and contains his or her name or another personal identifier.itations on agencies’ collection, It also defines “system of records” as a group of records under the control of any agency from which information is retrieved by the name of the individual or by an individual identifier. The Privacy Ac t requires that when agencies establish or make changes to a system of records, they must notify the public by a “system-of-records notice”: that is, a notice in the Federal Register identifying, among other things, the type of data collected, the types of individuals about whom information is collected, the intended “routine” uses o data, and procedures that individuals can use to review and corr personal information. Among other provisions, the act also requires agencies to define and limit themselves to specific predefined purposes. The provisions of the Privacy Act are consistent with and large based on a set of principles for protecting the privacy and security of personal information, known as the Fair Information Practices, which have been widely adopted as a standard benchmark for evaluating the adequacy of privacy protections; they include such principles as openness (keeping the public informed about privacy policies and practices) and accountability (those controlling the collection or use of personal information should be accountable for taking steps to ensure the implementation of these principles). The E-Government Act of 2002 strives to enhance protection for personal information in government information systems by requiring that agencies conduct privacy impact assessments (PIA PIA is an analysis of how personal information is collected, st shared, and managed in a federal system. More specifically, according to OMB guidance, a PIA is to (1) ensure that handling conforms to applicable legal, regulatory, and policy requirem ents regarding privacy; (2) determine the risks and effects of collecting maintaining, and disseminating information in identifiable form in , an electronic information system; and (3) examine and evaluate protections and alternative processes for handling information to mitigate potential privacy risks. To the extent that PIAs are made publicly available, they provide explanations to the public about such things as the information that will be collected, why it is bein collected, how it is to be used, and how the system and data will b maintained and protected. Federal laws to date have not required breaches to the public, although breac important role in the context of security breaches in the private sector. For example, requirements of California state law led ChoicePoint, a large information reseller, to notify its customer a security breach in February 2005. Since the ChoicePoint notification, bills were introduced in at least 44 states and enacted in at least 29 that require some form of notification upon a breach. agencies to report security h notification has played an A numbe in 2005 in the wake of the ChoicePoint security breach as well as incidents at other firms. In March 2005, the House Subcommittee on Commerce, Trade, and Consumer Protection of the House Energy r of congressional hearings were held and bills introduced and Commerce Committee held a hearing entitled “Protecting Consumers’ Data: Policy Issues Raised by ChoicePoint,” which focused on potential remedies for security and privacy concern s regarding information resellers. Similar hearings were held by th House Energy and Commerce Committee and by the U.S. Senate Committee on Commerce, Science, and Transportation in spring 2005. In carrying out its mission of providing health care and benefits to veterans, VA relies on a vast array of computer systems and telecommunications networks to support its operations and store sensitive information, including personal information on veterans. VA’s networks are highly interconnected, its systems support many users, and the department has increasingly moved to more interactive, Web-based services to better meet the needs of its customers. Effectively securing these computer systems and networks is critical to the department’s ability to safeguard its assets, maintain the confidentiality of sensitive veterans’ health and disability benefits information, and ensure the integrity of its financial data. In this complex IT environment, VA has faced long-standing challenges in achieving effective information security across the department. Our reviews identified wide-ranging, often recurring deficiencies in the department’s information security controls (attachment 2 provides further detail on our reports and the areas of weakness they discuss). Examples of areas of deficiency include the following. ● Access authority was not appropriately controlled. A basic management objective for any organization is to protect the resources that support its critical operations from unauthorized access. Electronic access controls are intended to prevent, limit, and detect unauthorized access to computing resources, programs, and information and include controls related to user accounts and passwords, user rights and file permissions, logging and monitoring of security-relevant events, and network management. Inadequate controls diminish the reliability of computerized information and increase the risk of unauthorized disclosure, modification, and destruction of sensitive information and disruption of service. However, VA had not established effective electronic access controls to prevent individuals from gaining unauthorized access to its systems and sensitive data, as the following examples illustrate: ● User accounts and passwords: In 1998, many user accounts at four VA medical centers and data centers had weaknesses including passwords that could be easily guessed, null passwords, and passwords that were set to never expire. We also found numerous instances where medical and data center staff members were sharing user IDs and passwords. ● User rights and permissions: We reported in 2000 that three VA health care systems were not ensuring that user accounts with broad access to financial and sensitive veteran information had proper authorization for such access, and were not reviewing these accounts to determine if their level of access remained appropriate. ● Logging and monitoring of security-related events: In 1998, VA did not have any departmentwide guidance for monitoring both successful and unsuccessful attempts to access system files containing key financial information or sensitive veteran data, and none of the medical and data centers we visited were actively monitoring network access activity. In 1999, we found that one data center was monitoring failed access attempts, but was not monitoring successful accesses to sensitive data and resources for unusual or suspicious activity. ● Network management: In 2000, we reported that one of the health care systems we visited had not configured a network parameter to effectively prevent unauthorized access to a network system; this same health care system had also failed to keep its network system software up to date. ● Physical security controls were inadequate. Physical security controls are important for protecting computer facilities and resources from espionage, sabotage, damage, and theft. These controls restrict physical access to computer resources, usually by limiting access to the buildings and rooms in which the resources are housed and by periodically reviewing the access granted, in order to ensure that access continues to be appropriate. VA had weaknesses in the physical security for its computer facilities. For example, in our 1998 and 2000 reports, we stated that none of the VA facilities we visited were adequately controlling access to their computer rooms. In addition, in 1998 we reported that sensitive equipment at two facilities was not adequately protected, increas the risk of disruption to computer operations or network communications. ● mployees were not prevented from performing incompatible E duties. Segregation of duties refers to the policies, procedures, a organizational structures that help ensure that one individual cannot independently control all key aspects of a process or computer- related operation. Dividing duties among two or more individual organizational groups diminishes the likelihood that errors and s or wrongful acts will go undetected, because the activities of one individual or group will serve as a check on the activities of the other. We determined that VA did not assign employee duties an d responsibilities in a manner that segregated incompatible function among individuals or groups of individuals. For example, in 1998 we s reported that some system programmers also had security administrator privileges, giving them the ability to eliminate any evidence of their activity in the system. In 2000, we reported thattwo VA health care systems allowed some employees to request, approve, and receive medical items without management approva l, violating both basic segregation of duties principles and VA policy; in addition, no mitigating controls were found to alert management of purchases made in this manner. S oftware change control procedures were not consistently implemented. It is important to ensure that only authorized tested systems are placed in operation. To ensure that changes to systems are necessary, work as intended, and do not result in the loss of data or program integrity, such changes should be documented, authorized, tested, and independently review ed. We found that VA did not adequately control changes to its operating systems. For example, in 1998 we reported that one VA data centerhad not established detailed written procedures or formal guidance for modifying operating system software, for approving and testing operating system software changes, or for implementing these changes. The data center had made more than 100 system softw changes during fiscal year 1997, but none of the changes included evidence of testing, independent review, or acceptance. We reporte d in 2000 that two VA health care systems had not established procedures for periodically reviewing changes to standard application programs to ensure that only authorized program was implemented. ervice continuity planning was not complete. In addition to S protecting data and programs from misuse, organizations must ensure that they are adequately prepared to cope with a loss of operational capability due to earthquakes, fires, accidents, sabo or any other disruption. An essential element in preparing for such catastrophes is an up-to-date, detailed, and fully tested service continuity plan. Such a plan is critical for helping to ensure thatinformation system operations and data can be promptly restore the event of a disaster. We reported that VA had not completed or tested service continuity plans for several systems. For example, in1998 we reported that one VA data center had 17 individual disaster recovery plans covering various segments of the organization, but it did not have an overall document that integrated the 17 separate plans and defined the roles and responsibilities for the disaster recovery teams. In 2000, we determined that the service continu plans for two of the three health care systems we visited did not include critical elements such as detailed recovery procedures, provisions for restoring mission-critical systems, and a list of key contacts; in addition, none of the health care systems we visited were fully testing their service continuity plans. hese deficiencies existed, in part, because VA had not implemented T key components of a comprehensive computer security program. Specifically, VA’s computer security efforts lacked ● clearly delineated security roles and responsibilities; ● regular, periodic assessments of risk; ● security policies and procedures that a VA’s interconnected environment; an ongoing security monitoring pro investigate unauthorized, unusual, or suspicious access a and ctivity; ● a process to measure, test, and report on the continued effectiveness of computer system, network, and process controls. As a result, we made a number of recommendations in 2002 that were aimed at improving VA’s security management. Among the primary elements of these recommendations were that (1) VA centralize its security management functions and (2) it perform other actions to establish an information security program, including actions related to risk assessments, security policies and procedures, security awareness, and monitoring and evaluating computer controls. The department has taken steps to address the weaknesses that we described, but these have not been sufficient to fully implement a comprehensive information security program. Examples of actions that VA has taken and still needs to take include the following: ● Central security management function: The department realigned its information technology resources to place administration and field office security functions more directly under the oversight of the department’s CIO, consolidating all administration-level cyber security functions under the department’s cyber security office. In addition, to provide greater management accountability for information security, the Secretary instituted information security standards for members of the department’s senior executive service. The cyber security officer organized his office to focus more directly on critical elements of information security control, and he updated the department’s security management plan and information security policies and procedures. However, the department still needed to develop policy and guidance to ensure (1) authority and independence for security officers and (2) departmentwide coordination of security functions. ● Periodic risk assessments: VA is implementing a commercial tool to identify the level of risk associated with system changes and also to conduct information security risk assessments. It also created a methodology that establishes minimum requirements for such risk assessments. However, it has not yet completed its risk assessment policy and guidance. VA reported that such guidance was forthcoming as part of an overarching information system security certification and accreditation policy that was to be developed during 2006. Without these elements, VA cannot be assured that it is appropriately performing risk assessments departmentwide. ● Security policies and procedures: VA’s cyber security officer reported that VA has action ongoing to develop a process for collecting and tracking performance data, ensuring management action when needed, and providing independent validation of reported issues. VA also has ongoing efforts in the area of detecting, reporting, and responding to security incidents. For example, it established network intrusion prevention capability at its four enterprise gateways. It is also developing strategic and tactical plans to complete a security incident response program to monitor suspicious activity and cyber alerts, events, and incidents. However, these plans are not complete. ● Security awareness: VA has taken steps to improve security awareness training. It holds an annual department information security conference, and it has developed a Web portal for security training, policy, and procedures, as well as a security awareness course that VA employees are required to review annually. However, VA has not demonstrated that it has a process to ensure compliance. ● Monitoring and evaluating computer controls: VA established a process to better monitor and evaluate computer controls by tracking the status of security weaknesses, corrective actions taken, and independent validations of corrective actions through a software data base. However, more remains to be done in t For example, although certain components of VA reported vulnerability and penetration testing to evaluate controls on intern and external access to VA systems, t o ngoing departmentwide program. his area. his testing was not part of an Since our last report in 2002, VA’s IG and independent auditors continued to report serious weaknesses with the department’s information security controls. The auditors’ report on interna l controls, prepared at the completion of VA’s 2005 financial statement audit, identified weaknesses related to access control, segregation of duties, change control, and service continuity—a of weaknesses that are virtually identical to those we identified years earlier. The department’s FY 2005 Annual Performance a Accountability Report states that the IG determined that many information system security vulnerabilities reported in nationa l audits from 2001 through 2004 remain unresolved, despite the department’s actions to implement IG recommendations in pre audits. The IG also reported specific security weaknesses and vulnerabilities at 45 of 60 VA health care facilities and 11 of 21 VA regional offices where security issues were reviewed, placing VA at ss and risk that sensitive data may be exposed to unauthorized acce improper disclosure, among other things. As a result, the IG determined that weaknesses in VA’s i controls were a material weakness. In response to the IG’s findings, the department indicates tha are being implemented to address the material weakness in information security. According to the department, it has maximi limited resources to make significant improvement in its overall security posture in the near term by prioritizing FISMA re activities, and work will continue in the next fiscal year. mediation Despite these actions, the department has not fully implemented the key elements of a comprehensive security management program, and its efforts have not been sufficient to effectively protect its information systems and information, including personally identifiable information, from unauthorized disclosure, misuse, or loss. In addition to establishing a robust information security program, agencies can take other actions to help guard against the possibility that personal information they maintain is inadvertently compromised. These include conducting privacy impact assessments and taking other practical measures. It is important that agencies identify the specific instances in which they collect and maintain personal information and proactively assess the means they intend to use to protect this information. This can be done most effectively through the development of privacy impact assessments (PIAs), which, as previously mentioned, are required by the E-Government Act of 2002 when agencies use information technology to process personal information. PIAs are important because they serve as a tool for agencies to fully consider the privacy implications of planned systems and data collections before those systems and collections have been fully implemented, when it may be relatively easy to make critical adjustments. In prior work we have found that agencies do not always conduct PIAs as they are required. For example, our review of selected data mining efforts at federal agencies determined that PIAs were not always being done in full compliance with OMB guidance. Similarly, as identified in our work on federal agency use of information resellers, few PIAs were being developed for systems or programs that made use of information reseller data, because officials did not believe they were required. Complete assessments are an important tool for agencies to identify areas of noncompliance with federal privacy laws, evaluate risks arising from electronic collection and maintenance of information about individuals, and evaluate protections or alternative processes needed to mitigate the risks identified. Agencies that do not take all the steps required to protect the privacy of personal information risk the improper exposure or alteration of such information. We recommended that the agencies responsible for the data mining efforts we reviewed complete or revise PIAs as needed and make them available to the public. We also recommended that OMB revise its guidance to clarify the applicability of the E-Gov Act’s PIA requirement to the use of personal information from resellers. OMB stated that it would discuss its guidance with agency senior officials for privacy to determine whether additional guidance concerning reseller data was needed. Besides strategic approaches such as establishing an information security program and conducting PIAs, agencies can consider a range of specific practical measures for protecting the privacy and security of personal information. Several that may be of particular value in preventing inadvertent data breaches include the following: Limit collection of personal information. One item to be analyzed as part of a PIA is the extent to which an agency needs to collect personal information in order to meet the requirements of a specific application. Limiting the collection of personal information, among other things, serves to limit the opportunity for that information to be compromised. For example, key identifying information—such as Social Security numbers—may not be needed for many agency applications that have databases of other personal information. Limiting the collection of personal information is also one of the fair information practices, which are fundamental to the Privacy Act and to good privacy practice in general. Limit data retention. Closely related to limiting data collection is limiting retention. Retaining personal data longer than needed by an agency or statutorily required adds to the risk that the data will be compromised. In discussing data retention, California’s Office of Privacy Protection recently reported an example in which a university experienced a security breach that exposed 15-year-old data, including Social Security numbers. The university subsequently reviewed its policies and decided to shorten the retention period for certain types of information. As part of their PIAs, federal agencies can make decisions up front about how long they plan to retain personal data, aiming to retain the data for as brief a period as necessary. Limit access to personal information and train personnel accordingly. Only individuals with a need to access agency databases of personal information should have such access, and controls should be in place to monitor that access. Further, agencies can implement technological controls to prevent personal data from being readily transferred to unauthorized systems or media, such as laptop computers, discs, or other electronic storage devices. Security training, which is required for all federal employees under FISMA, can include training on the risks of exposing personal data to potential identity theft, thus helping to reduce the likelihood of data being exposed inadvertently. Consider using technological controls such as encryption when data need to be stored on portable devices. In certain instances, agencies may find it necessary to enable employees to have access to personal data on portable devices such as laptop computers. As discussed, this should be minimized. However, when absolutely necessary, the risk that such data could be exposed to unauthorized individuals can be reduced by using technological controls such as encryption, which significantly limits the ability of such individuals to gain access to the data. Although encrypting data adds to the operational burden on authorized individuals, who must enter pass codes or use other authentication means to convert the data into readable text, it can provide reasonable assurance that stolen or lost computer equipment will not result in personal data being compromised, as occurred in the recent incident at VA. A decision about whether to use encryption would logically be made as an element of the PIA process and an agency’s broader information security program. While these suggestions do not amount to a complete prescription for protecting personal data, they are key elements of an agency’s strategy for reducing the risks that could lead to identity theft. In the event a data breach does occur, agencies must respond quickly in order to minimize the potential harm associated with identity theft. The chairman of the Federal Trade Commission has testified that the Commission believes that if a security breach creates a significant risk of identity theft or other related harm, affected consumers should be notified. The Federal Trade Commission has also reported that the overall cost of an incident of identity theft, as well as the harm to the victims, is significantly smaller if the misuse of the victim’s personal information is discovered quickly. Applicable laws such as the Privacy Act currently do not require agencies to notify individuals of security breaches involving their personal information; however, doing so allows those affected the opportunity to take steps to protect themselves against the dangers of identity theft. For example, California’s data breach notification law is credited with bringing to the public’s notice large data breaches within the private sector, such as those involving ChoicePoint and LexisNexis last year. Arguably, the California law may have mitigated the risk of identity theft to affected individuals by keeping them informed about data breaches and thus enabling them to take steps such as contacting credit bureaus to have fraud alerts placed on their credit files, obtaining copies of their credit reports, scrutinizing their monthly financial account statements, and taking other steps to protect themselves. Breach notification is also important in that it can help an organization address key privacy rights of individuals, in accordance with the fair information practices mentioned earlier. Breach notification is one way that organizations—either in the private sector or the government—can follow the openness principle and meet their responsibility for keeping the public informed of how their personal information is being used and who has access to it. Equally important, notification is consistent with the principle that those controlling the collection or use of personal information should be accountable for taking steps to ensure the implementation of the other principles, such as use limitation and security safeguards. Public disclosure of data breaches is a key step in ensuring that organizations are held accountable for the protection of personal information. Although the principle of notifying affected individuals (or the public) about data breaches has clear benefits, determining the specifics of when and how an agency should issue such notifications presents challenges, particularly in determining the specific criteria for incidents that merit notification. In congressional testimony, the Federal Trade Commission raised concerns about the threshold at which consumers should be notified of a breach, cautioning that too strict a standard could have several negative effects. First, notification of a breach when there is little or no risk of harm might create unnecessary concern and confusion. Second, a surfeit of notices, resulting from notification criteria that are too strict, could render all such notices less effective, because consumers could become numb to them and fail to act when risks are truly significant. Finally, the costs to both individuals and business are not insignificant and may be worth considering. FTC points out that, in response to a security breach notification, a consumer may cancel credit cards, contact credit bureaus to place fraud alerts on credit files, or obtain a new driver’s license number. These actions could be time-consuming for the individual and costly for the companies involved. Given these potential negative effects, care is clearly needed in defining appropriate criteria for required breach notifications. While care needs to be taken to avoid requiring agencies to notify the public of trivial security incidents, concerns have also been raised about setting criteria that are too open-ended or that rely too heavily on the discretion of the affected organization. Some public advocacy groups have cautioned that notification criteria that are too weak would give companies an incentive not to disclose potentially harmful breaches, and the same concern would apply to federal agencies. In congressional testimony last year, the executive director of the Center for Democracy and Technology argued that if an entity is not certain whether a breach warrants notification, it should be able to consult with the Federal Trade Commission. He went on to suggest that a two-tiered system may be desirable, with notice to the Federal Trade Commission of all breaches of personal data and notice to consumers where there is a potential risk of identity theft. The Center for Democracy and Technology’s comments regarding the Federal Trade Commission were aimed at commercial entities such as information resellers. A different entity—such as OMB, which is responsible for overseeing security and privacy within the federal government—might be more appropriate to take on a parallel role with respect to federal agencies. Once a determination has been made that a public notice is to be issued, care must be taken to ensure that it does its job effectively. Designing useful, easy-to-understand notices has been cited as a challenge in other areas where privacy notices are required by law, such as in the financial industry—where businesses are required by the Gramm-Leach-Bliley Act to send notices to consumers about their privacy practices—and in the federal government, which is required by the Privacy Act to issue public notices in the Federal Register about its systems of records containing personal information. For example, as noted during a public workshop hosted by the Department of Homeland Security’s Privacy Office, designing easy-to-understand consumer financial privacy notices to meet Gramm-Leach Bliley Act requirements has been challenging. Officials from the FTC and Office of the Comptroller of the Currency described widespread criticism of these notices—that they were unexpected, too long, filled with legalese, and not understandable. If an agency is to notify people of a data breach, it should do so in such a way that they understand the nature of the threat and what steps need to be taken to protect themselves against identity theft. In connection with its state law requiring security breach notifications, the California Office of Privacy Protection has published recommended practices for designing and issuing security breach notices. The office recommends that such notifications include, among other things, ● a general description of what happened; ● the type of personal information that was involved; ● what steps have been taken to prevent further unauthorized acquisition of personal information; ● the types of assistance to be provided to individuals, such as a toll- free contact telephone number for additional information and assistance; information on what individuals can do to protect themselves from identity theft, including contact information for the three credit reporting agencies; and information on where individuals can obtain additional information on protection against identity theft, such as the Federal Trade Commission’s Identity Theft Web site (www.consumer.gov/idtheft). The California Office of Privacy Protection also recommends making notices clear, conspicuous, and helpful by using clear, simple language and avoiding jargon, and it suggests avoiding using a standardized format to mitigate the risk that the public will become complacent about the process. The Federal Trade Commission has issued guidance to businesses on notifying individuals of data breaches that reiterates several key elements of effective notification—describing clearly what is known about the data compromise, explaining what responses may be appropriate for the type of information taken, and providing information and contacts regarding identity theft in general. The Commission also suggests providing contact information for the law enforcement officer working on the case, as well as encouraging individuals who discover that their information has been misused to file a complaint with the Commission. Both the state of California and the Federal Trade Commission recommend consulting with cognizant law-enforcement officers about an incident before issuing notices to the public. In some cases, early notification or disclosure of certain facts about an incident could hamper a law enforcement investigation. For example, an otherwise unknowing thief could learn of the potential value of data stored on a laptop computer that was originally stolen purely for the value of the hardware. Thus it is recommended that organizations consult with law enforcement regarding the timing and content of notifications. However, law enforcement investigations should not necessarily result in lengthy delays in notification. California’s guidance states that it should not be necessary for a law enforcement agency to complete an investigation before notification can be given. When providing notifications to the public, organizations should consider how to ensure that these are easily understood. Various techniques have been suggested to promote comprehension, including the concept of “layering.” Layering involves providing only the most important summary facts up front—often in a graphical format—followed by one or more lengthier, more narrative versions in order to ensure that all information is communicated that needs to be. Multilayering may be an option to achieving an easy-to-understand notice that is still complete. Similarly, providing context to the notice (explaining to consumers why they are receiving the notice and what to do with it) has been found to promote comprehension, as did visual design elements such as a tabular format, large and legible fonts, appropriate white space, and simple headings. Although these techniques were developed for other kinds of notices, they can be applied to those informing the public of data breaches. For example, a multilayered security breach notice could include a brief description of the nature of the security breach, the potential threat to victims of the incident, and measures to be taken to protect against identity theft. The notice could provide additional details about the incident as an attachment or by providing links to additional information. This would accomplish the purpose of communicating the key details in a brief format, while still providing complete information to those who require it. Given that people may be adversely affected by a compromise of their personal information, it is critical that they fully understand the nature of the threat and the options they have to address it. In summary, the recent security breach at VA has highlighted the importance of implementing effective information security practices. Long-standing information security control weaknesses at VA have placed its information systems and information, including personally identifiable information, at increased risk of misuse and unauthorized disclosure. Although VA has taken steps to mitigate previously reported weaknesses, it has not implemented a comprehensive, integrated information security program, which it needs in order to effectively manage risks on an ongoing basis. Much work remains to be done. Only through strong leadership, sustained management commitment and effort, disciplined processes, and consistent oversight can VA address its persistent, long-standing control weaknesses. To reduce the likelihood of experiencing such breaches, agencies can take a number of actions that can help guard against the possibility that databases of personally identifiable information are inadvertently compromised: strategically, they should ensure that a robust information security program is in place and that PIAs are developed. More specific practical measures aimed at preventing inadvertent data breaches include limiting the collection of personal information, limiting data retention, limiting access to personal information and training personnel accordingly, and considering using technological controls such as encryption when data need to be stored on mobile devices. Nevertheless, data breaches can still occur at any time, and when they do, notification to the individuals affected and/or the public has clear benefits, allowing people the opportunity to take steps to protect themselves against the dangers of identity theft. Care is needed in defining appropriate criteria if agencies are to be required to report security breaches to the public. Further, care is also needed to ensure that notices are useful and easy to understand, so that they are effective in alerting individuals to actions they may want to take to minimize the risk of identity theft. We have previously testified that as Congress considers legislation requiring agencies to notify individuals or the public about security breaches, it should ensure that specific criteria are defined for incidents that merit public notification. It may want to consider creating a two-tier reporting requirement, in which all security breaches are reported to OMB, and affected individuals are notified only of incidents involving significant risk. Further, Congress should consider requiring OMB to provide guidance to agencies on how to develop and issue security breach notices to the public. Messers. Chairmen, this concludes our testimony today. We would be happy to answer any questions you or other members of the committee may have. If you have any questions concerning this testimony, please contact Linda Koontz, Director, Information Management, at (202) 512-6240, koontzl@gao.gov, or Gregory Wilshusen, Director, Information Security, at (202) 512-6244, wilshuseng@gao.gov. Other individuals who made key contributions include Idris Adjerid, Barbara Collier, William Cook, John de Ferrari, Valerie Hopkins, Suzanne Lightman, Barbara Oliver, David Plocher, Jamie Pressman, J. Michael Resser, and Charles Vrabel. Information Systems: VA Computer Control Weaknesses Increase Risk of Fraud, Misuse, and Improper Disclosure. GAO/AIMD-98- 175. Washington, D.C.: September 23, 1998. VA Information Systems: The Austin Automation Center Has Made Progress in Improving Information System Controls. GAO/AIMD-99-161. Washington, D.C.: June 8, 1999. Information Systems: The Status of Computer Security at the Department of Veterans Affairs. GAO/AIMD-00-5. Washington, D.C.: October 4, 1999. VA Systems Security: Information System Controls at the North Texas Health Care System. GAO/AIMD-00-52R. Washington, D.C.: February 1, 2000. VA Systems Security: Information System Controls at the New Mexico VA Health Care System. GAO/AIMD-00-88R. Washington, D.C.: March 24, 2000. VA Systems Security: Information System Controls at the VA Maryland Health Care System. GAO/AIMD-117R. Washington, D.C.: April 19, 2000. Information Technology: Update on VA Actions to Implement Critical Reforms. GAO/T-AIMD-00-74. Washington, D.C.: May 11, 2000. VA Information Systems: Computer Security Weaknesses Persist at the Veterans Health Administration. GAO/AIMD-00-232. Washington, D.C.: September 8, 2000. Major Management Challenges and Program Risks: Department of Veterans Affairs. GAO-01-255. Washington, D.C.: January 2001. VA Information Technology: Important Initiatives Begun, Yet Serious Vulnerabilities Persist. GAO-01-550T. Washington, D.C.: April 4, 2001. VA Information Technology: Progress Made, but Continued Management Attention is Key to Achieving Results. GAO-02-369T. Washington, D.C.: March 13, 2002. Veterans Affairs: Subcommittee Post-Hearing Questions Concerning the Department’s Management of Information Technology. GAO-02-561R. Washington, D.C.: April 5, 2002. Veterans Affairs: Sustained Management Attention is Key to Achieving Information Technology Results. GAO-02-703. Washington, D.C.: June 12, 2002. VA Information Technology: Management Making Important Progress in Addressing Key Challenges. GAO-02-1054T. Washington, D.C.: September 26, 2002. Information Security: Weaknesses Persist at Federal Agencies Despite Progress Made in Implementing Related Statutory Requirements. GAO-05-552. Washington, D.C.: July 15, 2005. Privacy: Key Challenges Facing Federal Agencies. GAO-06-777T. Washington, D.C.: May 17, 2006. Personal Information: Agencies and Resellers Vary in Providing Privacy Protections. GAO-06-609T. Washington, D.C.: April 4, 2006. Personal Information: Agency and Reseller Adherence to Key Privacy Principles. GAO-06-421. Washington, D.C.: April 4, 2006. Data Mining: Agencies Have Taken Key Steps to Protect Privacy in Selected Efforts, but Significant Compliance Issues Remain. GAO- 05-866. Washington, D.C.: August 15, 2005. Aviation Security: Transportation Security Administration Did Not Fully Disclose Uses of Personal Information during Secure Flight Program Testing in Initial Privacy Notices, but Has Recently Taken Steps to More Fully Inform the Public. GAO-05- 864R. Washington, D.C.: July 22, 2005. Identity Theft: Some Outreach Efforts to Promote Awareness of New Consumer Rights are Under Way. GAO-05-710. Washington, D.C.: June 30, 2005. Electronic Government: Federal Agencies Have Made Progress Implementing the E-Government Act of 2002. GAO-05-12. Washington, D.C.: December 10, 2004. Social Security Numbers: Governments Could Do More to Reduce Display in Public Records and on Identity Cards. GAO-05-59. Washington, D.C.: November 9, 2004. Federal Chief Information Officers: Responsibilities, Reporting Relationships, Tenure, and Challenges, GAO-04-823. Washington, D.C.: July 21, 2004. Data Mining: Federal Efforts Cover a Wide Range of Uses, GAO-04- 548. Washington, D.C.: May 4, 2004. Privacy Act: OMB Leadership Needed to Improve Agency Compliance. GAO-03-304. Washington, D.C.: June 30, 2003. Data Mining: Results and Challenges for Government Programs, Audits, and Investigations. GAO-03-591T. Washington, D.C.: March 25, 2003. Technology Assessment: Using Biometrics for Border Security. GAO-03-174. Washington, D.C.: November 15, 2002. Information Management: Selected Agencies’ Handling of Personal Information. GAO-02-1058. Washington, D.C.: September 30, 2002. Identity Theft: Greater Awareness and Use of Existing Data Are Needed. GAO-02-766. Washington, D.C.: June 28, 2002. Social Security Numbers: Government Benefits from SSN Use but Could Provide Better Safeguards. GAO-02-352. Washington, D.C.: May 31, 2002. Full citations are provided in attachment 1. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The recent information security breach at the Department of Veterans Affairs (VA), in which personal data on millions of veterans were compromised, has highlighted the importance of the department's security weaknesses, as well as the ability of federal agencies to protect personal information. Robust federal security programs are critically important to properly protect this information and the privacy of individuals. GAO was asked to testify on VA's information security program, ways that agencies can prevent improper disclosures of personal information, and issues concerning notifications of privacy breaches. In preparing this testimony, GAO drew on its previous reports and testimonies, as well as on expert opinion provided in congressional testimony and other sources. For many years, significant concerns have been raised about VA's information security--particularly its lack of a robust information security program, which is vital to avoiding the compromise of government information, including sensitive personal information. GAO and the department's inspector general have reported recurring weaknesses throughout VA, including the Veterans Benefits Administration, in such areas as access controls, physical security, and segregation of incompatible duties. The department has taken steps to address these weaknesses, but these have not been sufficient to establish a comprehensive information security program. For example, it is still developing plans to complete a security incident response program to monitor suspicious activity and cyber alerts, events, and incidents. Without an established and implemented security program, the department will continue to have major challenges in protecting its information and information systems from security breaches such as the one it recently experienced. In addition to establishing robust security programs, agencies can take a number of actions to help guard against the possibility that databases of personally identifiable information are inadvertently compromised. A key step is to develop a privacy impact assessment--an analysis of how personal information is collected, stored, shared, and managed--whenever information technology is used to process personal information. In addition, agencies can take more specific practical measures aimed at preventing data breaches, including limiting the collection of personal information, limiting the time that such data are retained, limiting access to personal information and training personnel accordingly, and considering the use of technological controls such as encryption when data need to be stored on portable devices. When data breaches do occur, notification of those affected and/or the public has clear benefits, allowing people the opportunity to protect themselves from identity theft. Although existing laws do not require agencies to notify the public of data breaches, such notification is consistent with agencies' responsibility to inform individuals about how their information is being accessed and used, and it promotes accountability for privacy protection. That said, care is needed in defining appropriate criteria for triggering notification. Notices should be coordinated with law enforcement to avoid impeding ongoing investigations, and in order to be effective, notices should be easy to understand. Because of the possible adverse impact of a compromise of personal information, it is critical that people fully understand the threat and their options for addressing it. Strong leadership, sustained management commitment and effort, disciplined processes, and consistent oversight will be needed for VA to address its persistent, long-standing control weaknesses. |
We found that the 11 high-value, long-term SSACs we selected for review from three of the five VAMCs we visited took nearly 3 years (33.8 months) on average to develop and award. (See fig. 1.) The total time required for the development and award of these 11 high-value, long-term SSACs ranged from 18 to 46 months and the longest contracting phases were the solicitation and negotiation phases. According to leadership officials and contracting officers from all five of the network contracting offices we visited, establishing high-value, long- term SSACs in a timely manner has been challenging for several reasons, including (1) not always receiving a complete, actionable, and timely initial information package from the VAMC that contains information the contracting officer needs to begin acquisition planning; (2) lengthy review processes for high-value, long-term SSACs; (3) negotiation challenges with the affiliates on the price of high-value, long- term SSACs; and (4) VAMC resistance to developing and pursuing high- value, long-term SSACs. VAMC-based contracting officer’s representatives and medical directors from all five of the VAMCs we visited also explained that establishing high-value, long-term SSACs has presented challenges for them. Specifically, 9 of the 14 contracting officer’s representatives we spoke with noted that they are often asked to resubmit initial information packages to the contracting officer throughout the development of a SSAC due to form updates or policy changes that occurred since the time they created these documents. Moreover, VAMC officials from all five VAMCs we visited indicated that the length of time it takes to develop and award high-value, long-term SSACs presents many challenges for their VAMCs, including the potential for gaps in patient care and the need to repeatedly establish short-term solutions. We also found that VHA has not developed standards that can be used to measure the timeliness of developing high-value, long-term SSACs. However, during fiscal year 2016, VHA developed estimates for the maximum duration of each contracting phase, referred to as procurement action lead times (PALT). Currently, the PALT goal for the development and award of a high-value, long-term SSAC is between 20 and 21 months; however, we found that 10 of the 11 high-value, long-term SSACs we reviewed exceeded these PALT goals by as little as 1.4 months and as many as 25.8 months. According to VHA officials we interviewed, PALT goals are not used as performance standards for VAMC, network contracting office, and Medical Sharing Office staff responsible for the development of high-value, long-term SSACs. These officials told us that VHA is currently developing and conducting validity tests of revised PALT goals for several types of contracts, including SSACs, but there is no planned end date for these tests and they do not expect to implement revised PALT goals across VHA until at least fiscal year 2017. These officials explained that the revised PALT goals will be used for setting expectations with VAMC officials for the length of time it should take to develop and award several types of contracts, including SSACs. Federal internal control standards recommend establishing and reviewing performance standards at all levels of an agency. Absent such standards, VHA cannot ensure that its high-value, long-term SSACs are being developed in a timely manner. Additionally, we found that VHA does not collect data on the length of time each contracting phase took to complete for any SSACs, including the 11 high-value, long-term and 12 short-term SSACs we selected for review. Federal internal control standards state that information should be recorded and communicated to management and others within the agency that need it in a format and time frame that enables them to carry out their responsibilities. However, in contrast with these standards, VA is unable to analyze the time spent in each phase of SSAC development, and this inability has disadvantages in terms of management decisions and accountability for SSAC development. The absence of real-time data on the amount of time being spent within each contracting phase limits VA’s ability to make informed management decisions, including changes to the assignment of staff that are either overburdened by their workloads or in need of additional training to build their competency with a particular type of contract or contracting phase. This lack of information prevents VHA from effectively setting clear and consistent objectives for organizational performance and making improvements as needed. Our report concluded that a lack of attention to the time spent to develop and award high-value, long-term SSACs has resulted in VHA’s inability to ensure that contracts are being developed in a timely manner. To ensure the timely development of high-value, long-term SSACs, we recommended that VA (1) establish performance standards for appropriate development time frames for high-value, long-term SSACs and use these performance standards to routinely monitor VAMC, network contracting office, and Medical Sharing Office efforts to develop these contracts; and (2) collect performance data on the time spent in each phase of the development of high-value, long-term SSACs and periodically analyze these data to assess performance. VA concurred with these recommendations and said that it will take steps to address these weaknesses, including the creation of a workgroup that will establish performance standards for development time frames for high- value, long-term SSACs and the designation of an office within VHA to routinely monitor these performance standards. VA also said that it will assess its current data systems to determine whether a new or different system would be needed to capture all relevant data and that the Medical Sharing Office will collaborate with other stakeholders to determine the need for and the mechanism to collect additional data. We found that short-term SSACs are used to provide coverage to bridge the gap between an expired or expiring high-value, long-term SSAC and its replacement. Specifically, 6 of our 12 selected short-term SSACs were awarded as bridge contracts, which creates duplicative work for VAMC and contracting staff because they must simultaneously develop both the short-term SSAC bridge contract and the replacement high-value, long- term or low-value, long-term SSAC. Of the remaining 6 short-term SSACs we reviewed, 5 were awarded to allow affiliate services to begin while new high-value, long-term SSACs were being developed for the same services and 1 was awarded to fill a short-term staffing need at a VAMC. In addition, we found that the use of these 12 short-term SSACs was consistent with reasons reported by from the majority of the medical sharing team supervisors from the 21 network contracting offices. Specifically, 12 medical sharing team supervisors from the 21 network contracting offices (57 percent) reported that the most prevalent reason that they opt to award short-term SSACs is to avoid any gaps in services due to the length of time it takes to develop and award high-value, long- term SSACs. Federal internal control standards state an agency should provide for an assessment of the agency’s risk associated with achieving its objectives, including identifying risks through forecasting and strategic planning. However, in contrast with these standards, VHA does not have a policy that requires VAMCs and network contracting offices to engage in timely acquisition planning to ensure that expiring high-value, long-term SSACs are replaced without the need to use a short-term SSAC as a bridge contract. Moreover, VA’s governing directive for the development of SSACs does not specify when VAMC and network contracting office staff should begin acquisition planning activities to replace an existing high- value, long-term SSAC. As a result, VHA lacks assurance that its staff are performing and accountable for their roles in ensuring that replacement high-value, long-term SSACs are developed in time and that the agency is minimizing duplicative work when short-term SSACs are used as bridge contracts. We also found that VHA was further exposed to potential risks associated with using short-term SSACs because the Medical Sharing Office, the VHA Central Office entity with oversight authority of SSACs, does not consistently review available data on all SSACs awarded throughout VHA; in particular, it does not review the level of reliance on short-term SSACs. While this office creates monthly reports for all VISNs and network contracting offices that provide information on the status of their medical sharing contracts, including all SSACs, they rely on network contracting offices to determine if they are selecting the appropriate term for their contracts. This can potentially be problematic because 7 of the medical sharing supervisors from the 21 network contracting offices we contacted and leadership teams and contracting officers from 3 of the 5 network contracting offices we visited told us that at times they have purposefully developed short-term SSACs in lieu of high-value, long-term SSACs because the Medical Sharing Office does not review any short- term SSACs. In fact, we found 6 of the 12 short-term SSACs we selected for review were extended beyond their initial performance periods for up to 11 months resulting in total values for these 6 contracts that ranged from almost $686,000 to $1.4 million—well beyond the $500,000 Medical Sharing Office review threshold. Standards for internal control in the federal government state that control activities should occur at all levels of an agency to help ensure that management’s directive are carried out by staff and that top-level reviews of actual performance by agency management are needed to track major agency achievements and compare these to plans, goals, and objectives that were previously established. In addition, we found that 7 of the 12 short-term SSACs we selected for review from two network contracting offices did not follow VA and VHA policy for the development of SSACs. Specifically, we found 5 short-term SSACs we reviewed from one network contracting office where (1) a solicitation was not issued to the affiliate, (2) the affiliate did not provide VHA a formal proposal outlining its services and instead submitted a price quote, and (3) negotiations were not conducted to address potential pricing issues before awarding the final contract. The contracting officer responsible for these 5 short-term SSACs explained that he was often given as little as 10 business days to develop and award a short-term SSAC before the prior short-term SSAC expired and that he did not have the skills needed to conduct negotiations with the affiliate. We found that this contracting officer’s supervisor had reviewed all 5 of these contracts prior to their award; however, the review process did not identify the areas that did not adhere to VA and VHA policy requirements for the development of SSACs. Federal internal control standards recommend that agencies establish processes to ensure the proper execution of transactions, including the provision of the proper amount of supervision. However, without ensuring that contracting officers are adhering to VA and VHA policies and network contracting offices are effectively reviewing the development of short-term SSACs as required by VA and VHA policies, VHA may be at risk for overpaying for affiliate services provided through these contracts. Our report concluded that the lack of attention to this overreliance on short-term SSACs as bridge contracts exposes VHA to risks. To ensure the effective development and use of short-term SSACs, we recommended VA (1) develop requirements for VAMCs and network contracting offices to effectively engage in early acquisition planning for the replacement of expiring high-value, long-term SSACs, (2) prioritize the review of SSAC contract data to identify patterns of overreliance on short-term SSACs that avoid appropriate Medical Sharing Office oversight, and (3) develop standards for the minimum amount of time necessary to develop and award short-term SSACs to minimize cases of nonadherence to VA policy for these contracts. VA concurred with these recommendations, and laid out plans to develop new requirements and standards while also charging the Medical Sharing Office with conducting data reviews of short-term SSACs. We found a high level of turnover among medical sharing contracting officers in all 21 network contracting offices that was exacerbated by a high level of inexperience among contracting officers responsible for developing SSACs. Network contracting office medical sharing teams experienced significant turnover in recent years, with 23 percent (49 of 217) of medical sharing contracting officer full-time employee equivalents (FTEE) in fiscal year 2014 and 27 percent (65 of 239) of FTEEs in fiscal year 2015 either resigning or transferring to another VHA contracting team. Medical sharing supervisors offered several potential explanations for turnover on medical sharing teams, including job burnout, the complexity of medical sharing contracts, the workload associated with medical sharing teams, and frustration with the layers of review required for these contracts. Medical Sharing Office officials told us that this turnover hinders the SSAC development process because newer contracting officers have greater difficulty developing high-value, long-term SSACs due to a lack of experience and knowledge. They also told us that they believe it takes approximately 5 years for a contracting officer to become experienced in developing medical sharing contracts, including SSACs. We found, however, that more than half of medical sharing contracting officers had 2 years or less medical sharing contract experience and less than one-quarter had more than 4 years of experience developing medical sharing contracts. Federal internal control standards state that effective management of an organization’s workforce, such as having the right personnel on board, is essential to achieving results. However, in contrast to these standards, VHA does not have a plan to address medical sharing contracting officer turnover. As a result, VHA lacks assurance that network contracting offices can maintain and develop the contracting officers’ skillsets that are necessary for developing complex medical sharing contracts, such as SSACs. Moreover, we found that limited training opportunities for medical sharing contracting officers further erodes VA’s knowledge base for developing high-quality and cost-effective SSACs. The Medical Sharing Office has developed and offered three in-person training courses designed to progressively build a contracting officer’s competence in developing medical sharing contracts, including SSACs. Medical Sharing Office officials reported in February 2016 that over 90 percent of all participants for each of the training classes reported that the trainings increased their medical sharing competency and that the information presented would contribute to their job performance. Since fiscal year 2015, however, VHA has not consistently provided training for medical sharing teams in network contracting officers throughout VHA. VHA has canceled some of their course offerings due to budget constraints. In addition, VHA Central Office requested that the Medical Sharing Office cut the class size of each course offering by 25 percent. Federal internal control standards state that agencies should establish good human capital policies and practices, such as appropriate practices for training. In contrast to these standards, VHA has not determined how to either provide the existing training courses or develop alternatives that do not require travel in response to a changing budgetary environment. As a result, VHA cannot build the skills of its medical sharing contracting officers and overcome the challenges associated with their inexperience. Our report concluded that instability in the medical sharing workforce, due to high levels of turnover among medical sharing contracting officers, has limited VHA’s ability to develop high-quality SSACs throughout VHA. To develop and maintain medical sharing expertise within the network contracting offices, we recommended that VA (1) create a plan to increase retention of contracting officers that work in medical sharing teams, and (2) develop mechanisms to either provide existing training courses or create training courses that do not require travel for contracting officers working within network contracting offices. VA concurred with both of these recommendations and summarized planned steps to address these recommendations, including the development of a retention plan and soliciting agency leadership for assistance in resource prioritization to fund VHA health care contracting training courses. We found that representatives from the five affiliates that provide services through SSACs to our selected VAMCs noted challenges related to receiving information on changes to VA and VHA requirements for SSACs. These included communication from VHA about what services the VAMC needed from the affiliate, the documentation requirements affiliates needed to submit to support their physician salary pricing, and changes to VHA’s approach to negotiations. The affiliate representatives also noted coordination challenges related to responding to SSAC solicitations. For example, representatives reported that it was challenging for them to provide services to VAMCs under short-term SSACs because the length of these contracts does not provide a commitment from VHA for the physicians hired by the affiliate to fulfill the contract. These affiliate representatives explained that it can take a year or longer to recruit a well-qualified academic physician and short-term SSACs do not provide the funding commitment needed by the affiliate to recruit these physicians. Federal internal control standards state that information should be communicated both internally and externally to enable the agency to carry out its responsibilities; for external communications, these standards state that management should ensure that there are adequate means of communicating with, and obtaining information from, external stakeholders that may have a significant impact on the agency achieving its goals. In contrast to these standards, VHA’s efforts to cultivate better communication and coordination with affiliates at the national level have been limited, consisting of three regional forums with all its affiliates in fiscal year 2012. Since 2012, VA has relied primarily on local coordination with affiliates in lieu of regional forums, due to travel restrictions associated with VA’s recent budget shortfalls. As a result, VHA cannot ensure that it is effectively responding to the concerns of its affiliates. Our report concluded that concerns about VA’s communication and coordination with its affiliates, as voiced by representatives from the five affiliates we spoke with, demonstrate potentially ineffective communication streams with these critical partners. To ensure VHA effectively communicates with its affiliates regarding SSACs, we recommended that VA reach out to all its affiliates, identify any concerns, and determine the most effective method of communicating with affiliates regarding SSAC development. VA concurred with this recommendation and said that the VHA’s Office of Academic Affiliations and Medical Sharing Office will re-engage with the American Association of Medical Colleges to determine the best ways to gather input from affiliates on their concerns and determine the most effective method of communication with them regarding SSAC development. Furthermore, VA added that these offices will evaluate VA’s current partnerships with affiliates to identify both highly functional relationships that could be highlighted as best practices and partnerships that could benefit from targeted intervention. Chairman Coffman, Ranking Member Kuster, and Members of the Subcommittee, this concludes my prepared statement. I would be pleased to answer any questions that you may have at this time. If you or your staff members have any questions concerning this testimony, please contact me at (202) 512-7114 or williamsonr@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Other individuals who made key contributions to this testimony include Marcia A. Mann, Assistant Director; Cathleen Hamann; Katherine Nicole Laubacher; Dharani Ranganathan; and Said Sariolghalam. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | This testimony summarizes the information contained in GAO's May 2016 report, entitled VA Health Care: Improvements Needed for Management and Oversight of Sole- Source Affiliate Contract Development (GAO-16-426). GAO found it took nearly 3 years on average to develop and award 11 selected high- value, long-term sole-source affiliate contracts (SSAC) from three of the five Department of Veterans Affairs (VA) medical centers (VAMC) GAO visited. The two remaining VAMCs GAO visited did not use high-value, long-term SSACs. High-value, long-term SSACs generally require the most oversight of all SSACs by the Veterans Health Administration (VHA), have total initial values of $500,000 or more, and provide affiliate services for more than 1 year. Officials from all five VAMCs GAO visited said that the lengthy development time frames of these contracts can impact VAMCs in several ways—including creating potential gaps in patient care and the need to repeatedly establish short-term solutions. GAO found that 10 of these 11 selected high-value, long-term SSACs exceeded the informal estimates created by VHA as planning guides for the expected development time frames that high-value, long-term SSACs should take. According to VA officials, these informal estimates are not used to measure the performance of this process and VHA has not established standards for the timely development of high-value, long-term SSACs. Federal internal control standards recommend establishing and reviewing performance standards at all levels of an agency. Absent such standards, VHA cannot ensure that its high-value, long-term SSACs are being developed in a timely manner. VHA uses short-term SSACs to overcome lengthy high-value, long-term SSAC development time frames, but lacks effective oversight for the development and use of short-term SSACs. Short-term SSACs have total initial values of less than $500,000, provide affiliate services for up to 1 year, and are not reviewed by VHA Central Office. Instead they are developed and awarded independently by contracting officers within VHA's network contracting offices. Of the 12 short-term SSACs that GAO reviewed, 7 did not adhere to VA and VHA policy for the development of short-term SSACs—including 5 where (1) a solicitation was not issued to the affiliate (a required document detailing VA's performance requirements to enable a prospective contractor to prepare its proposal); (2) the affiliate did not provide VHA a formal proposal outlining the services to be provided and instead submitted a price quote; and (3) negotiations were not conducted between the contracting officer and affiliate to address potential pricing issues before awarding the final contract. The contracting officer responsible for these five contracts cited the lack of adequate time to develop and award the contracts and a lack of contract negotiating skills as the primary factors that impacted his ability to ensure that these short-term SSACs adhered to VA and VHA policy requirements. By not developing standards for short-term SSACs, VA has limited assurance that contracting officers have enough time to develop and award these contracts and also adhere to VA and VHA policy requirements. GAO found a high level of inexperience among contracting officers responsible for developing SSACs in all 21 of VHA's network contracting offices. Specifically, about one-third of medical sharing contracting officers had 1 year or less experience developing medical sharing contracts, including SSACs, and more than half of medical sharing contracting officers had 2 years or less medical sharing contract experience. The high level of inexperience can be attributed in part to high turnover in recent years. About one-quarter of medical sharing contracting officers working within network contracting offices either left VA or were reassigned to other contracting teams. Inconsistent with federal internal control standards, VA does not have a plan to address the retention of its contracting workforce, nor has it taken adequate steps to expand training opportunities to enhance the level of competence. |
The 1993 Midwest flood, termed “the Great Flood of 1993,” was unprecedented in the United States in terms of the amount of precipitation, the recorded river levels, the duration and extent of the flood, the damage to crops and property, and the economic impact. The intense rainfall that deluged the upper Mississippi River basin in the spring and summer of 1993 caused the largest flow ever measured at St. Louis. Affecting nine midwestern states, the rainfall generated record-high flood levels at 95 measuring stations on the region’s rivers. Because of the catastrophic flooding, 38 people died, millions of acres were inundated, property and agriculture sustained heavy damage, tens of thousands of people were evacuated from their homes, and transportation, business, and water and sewer services were disrupted. President Clinton declared 505 counties to be federal disaster areas, and estimates of the damage have ranged from $12 to $16 billion. The 1993 flood affected most of the upper Mississippi River basin. The basin drains all or part of 13 states and encompasses about 714,000 square miles, or 24 percent of the contiguous United States. The upper basin includes the Mississippi River from its source in Minnesota to its confluence with the Ohio River at Cairo, Illinois. Its principal tributary is the Missouri River, which joins the Mississippi at St. Louis, Missouri. Other major tributaries include the Minnesota, Wisconsin, Iowa, Des Moines, and Illinois rivers. Figure 1.1 shows the Mississippi River basin and the area of the 1993 flooding, and figure 1.2 compares two satellite images of the St. Louis, Missouri, area at the confluence of the Illinois, Mississippi and Missouri rivers during a severe drought and during the 1993 flood. The conditions that produced the flood began in the summer of 1992. According to the Department of Commerce’s National Weather Service (NWS), July, September, and November 1992 were much wetter than normal in the upper Mississippi River basin. Winter precipitation was near normal, but a wet spring followed. The period from April to June 1993 was the wettest observed in the upper basin in the last 99 years. As a result, soils were saturated, and many streams were flowing well above normal levels when summer rains began. A persistent atmospheric pattern during the summer of 1993 caused excessive rainfall across much of the upper Mississippi River basin. Major flooding resulted primarily from a series of heavy rainfalls from mid-June through late July. A change in the upper air’s circulation pattern created drier conditions in late July and early August, but heavy rainstorms brought more flooding to parts of the upper basin in mid-August. The rainfall over the upper Mississippi River basin from May to August 1993 is unmatched in the historical records of the central United States. Generally, rainfall from the Dakotas to Missouri and Illinois was well above normal. Figure 1.3 locates the heaviest concentrations of rainfall from January through July 1993 in the flood region. Rivers rose above flood levels at about 500 measurement points in the nine-state region, NWS reported. Record flooding occurred at 95 measurement points in the Upper Midwest—44 on the upper Mississippi River system, 49 on the Missouri River system, and 2 on the Red River of the North system. Water flow rates along major parts of the upper Mississippi and lower Missouri rivers equalled or exceeded floods with an annual probability of 1 percent—commonly called a 100-year flood. Figure 1.4 shows where the heaviest flooding occurred. Extreme flooding of major river systems like the Mississippi and Missouri rivers seldom occurs in the summer. During a typical midwestern summer, a few localized heavy rains are scattered throughout the region. In 1993, the rare combination of closely timed and record-level rainfall occurred on both the lower Missouri and upper Mississippi basins, causing a record flood at St. Louis. NWS reported that the extended duration of the flood was also extremely rare. Typically, periods of above-average rainfall during a midwestern summer last from 2 to 5 weeks, sometimes persisting up to 8 weeks. In 1993, major flooding continued throughout the summer along the Missouri and Mississippi rivers. For example, as of September 1, 1993, Hannibal, Missouri, had experienced 153 consecutive days, or about 22 weeks, of water above flood level. Flooding continued through the middle of September in many regions along the Mississippi River. The administration established the Interagency Floodplain Management Review Committee to evaluate the performance of existing floodplain management programs in light of the 1993 flood. Their review of existing damage estimates for the flood found that these estimates ranged from $12 billion to $16 billion. The available estimates are from such federal agencies as NWS and the Federal Emergency Management Agency (FEMA), which develop estimates for specific program purposes, such as disaster response and assistance. The Committee estimated that $4 billion to $5 billion in damage was to crops in upland areas outside the floodplain, which were destroyed by the heavy precipitation there. The Committee attributed about $2.5 billion in agricultural damage directly to the flooding. Other significant damage occurred to about 100,000 residences, more than 5,000 businesses, many bridges, hundreds of miles of roads and railroads, and 33 airports. The flood also closed the major rivers to navigation and affected about 200 municipal water systems, 388 wastewater facilities, and other public facilities, such as public buildings and parks. FEMA reported that about 6.6 million acres in the floodplain were flooded in 1993, of which 63.4 percent were agricultural lands and 2.5 percent were urban areas. The remaining acres in the floodplain were normally covered by water, were wetlands, and/or were used for other purposes. The primary federal agency involved in flood control is the U.S. Army Corps of Engineers (the Corps). The Department of Agriculture’s Natural Resources Conservation Service (NRCS) is indirectly involved in flood control when it addresses the effects of flooding in agricultural watersheds. After a series of disastrous floods affected wide areas, the Congress enacted the Flood Control Act of 1936. This act established a nationwide policy that (1) flood control was in the interest of the general public and (2) the federal government would cooperate with the states and local entities to carry out flood control activities. The Corps’ flood control programs are designed to reduce the susceptibility of property to flood damage and to relieve human and financial losses. The Corps has invested over $23 billion in flood control projects nationwide. It has constructed more than 600 projects, including reservoirs and about 10,500 miles of levees and floodwalls. Flood control reservoirs often provide the capacity to store water for multiple uses, including municipal and industrial water supplies, navigation, irrigation, production of hydroelectric power, conservation of fish and wildlife, maintenance of water quality, and recreation. Levees and floodwalls are usually turned over to local sponsors for operations and maintenance. The Corps is also authorized to perform emergency activities, such as fighting floods, repairing and restoring flood control works, and supplying emergency clean water to communities. It also performs emergency assistance work requested and funded by FEMA. Permanent repairs to levees and other flood control facilities are provided under a levee rehabilitation program. Five of the 37 Corps districts performing civil works activities were involved in the 1993 flood: St. Paul, Minnesota; Rock Island, Illinois; St. Louis and Kansas City, Missouri; and Omaha, Nebraska. Of the 251 Corps levees located in these districts, 193 were in the flooded area. The Corps operates 98 reservoirs in the upper Mississippi River basin to reduce flood damage. Of these, 22 were constructed by the Bureau of Reclamation. While not all of the reservoirs were in the flooded area, most had some impact on the flood because they stored water. For example, Corps headquarters officials said the reservoirs stored more than 20 million acre-feet of floodwater on August 1, 1993, reducing flood levels throughout much of the flood area—for example, lowering the crest of the Mississippi River at St. Louis on that day by 5 feet. In addition to the reservoirs, the Corps has built or improved more than 2,200 miles of levees for the protection of communities and agriculture in the basin. After the 1993 flood, the Congress funded a broad 18-month effort by the Corps to assess floodplain management in the upper Mississippi River and lower Missouri River basins. This effort, which was separate from the work of the interagency committee, describes the existing resources in the floodplain, identifies alternatives for the future use of the floodplain, and suggests policy changes and areas for further study. The assessment was conducted in collaboration with numerous federal, state, and local governments and interested parties. The Corps presented its findings and conclusions in a report published on June 30, 1995. NRCS’ programs are designed to protect and prevent flooding in small watersheds, repair or relocate agricultural levees that are damaged in flooding, and convert cropland to wetland reserves. The Small Watershed Program authorized by the Watershed Protection and Flood Prevention Act of 1954 (P.L. 83-566) provides for NRCS to install land conservation measures and flood damage reduction works nationally. NRCS traditionally works on smaller projects affecting watersheds of fewer than 400 square miles; the Corps addresses needs in larger watersheds. In addition, NRCS has authority under the Flood Control Act of 1944 (P.L. 78-534) for a flood prevention program for 11 watersheds. In the nine midwestern states affected by the 1993 flood, NRCS has performed soil and water conservation work on 3 million acres, installed 2,964 reservoirs, and worked on 818 miles of channel. NRCS estimated that its watershed projects prevented $400 million in damage from the 1993 flood. When a disaster strikes, NRCS implements the Emergency Watershed Protection Program under section 403 of the Agricultural Credit Act of 1978 (P.L. 95-334). This program provides assistance—including repairs to damaged levees—to reduce hazards to life and property. Levees are linear earthen embankments whose primary purpose is to prevent high water from reaching the floodplain. They normally extend from high ground along one side of a floodplain and around it to another area of high ground. Levees protect the area between the levee and the high ground. For stability, an earthen levee is normally constructed so that its bottom width is several times its height; hence, a levee requires considerable land area. In urban areas where space is limited, the Corps builds masonry floodwalls. A long levee system may include a combination of several segments of earthen levees and floodwalls. Figure 1.5 represents a cross-sectional view of a typical earthen levee. Levees reduce but do not eliminate flooding in the floodplain because levees may be overtopped by floods larger than those for which they are designed. Generally, the Corps analyzes the risks, costs, and benefits of constructing a levee to various heights; determines, with the participation of a local cost-sharing sponsor, how much protection the levee should provide; and proposes a plan to the Congress. After reviewing the Corps’ analyses, the Congress can authorize and fund the plan. The height of a levee is based on the maximum flow (discharge) of floodwater, measured in cubic feet per second (cfs), associated with flooding of a particular frequency, or average recurrence level, at the levee’s location. For example, the Kaskaskia Island levee in the Mississippi River in Illinois, which is designed to withstand a flood with an average frequency (or recurrence interval) of 50 years, was built to withstand a water level of 45.7 feet and a flow of 1,010,000 cfs. At other locations, floods of the same frequency will be associated with different heights and flows. For example, the Des Moines-Mississippi River levee in Missouri, which also protects against a flood with an average recurrence interval of 50 years, was built to withstand a water level of 24 feet and a flow of 371,000 cfs. Both levees provide the same degree, or level, of protection but have different performance criteria because the river’s channel and flows differ greatly at these levees. Engineers have accounted for uncertainties in the water level and for unknown factors—such as wave action, bridge openings, and the effects of urbanization—by adding 1 to 3 feet to the overall elevation of a levee’s design; this addition is known as freeboard. A levee should withstand floods up to and including the flood for which it was designed. Floods larger than the design flood may overtop or breach the levee—that is, cut a hole through it. The ways that floods can damage levees—by overtopping, piping, saturation, and underseepage—are depicted in figure 1.6. Hydrologists have several ways of describing the size of a flood at a specific location. They may refer to the water’s level and flow, or they may refer to the average interval between occurrences of that particular water level and flow. They also refer to the annual probability that the same water level and flow may occur. Because floods occur randomly, the interval between extreme water levels of the same height are far from uniform—a large flood in one year does not preclude the occurrence of an even larger flood the next year. For example, a flood with an average recurrence interval of 100 years has a 1-percent chance of being equaled or exceeded every year. This means that a “100-year” flood may occur several times within a 100-year period, or it may not occur at all. As communities and farms have grown on the floodplains of the upper Mississippi and Missouri rivers since the early to mid-1800s, levees have been constructed by various nonfederal entities, ranging from cities to individuals, to protect floodplains from seasonal flooding. Owners have wanted to protect the floodplain from flooding because it has often contained the most fertile land for farming. While no inventory of nonfederal levees exists, the Interagency Floodplain Management Review Committee estimated that such levees extend over 5,800 miles in the upper Mississippi River basin. The Corps estimated that about 1,100 of the 1,358 nonfederal levees in the area covered by the five Corps districts involved in the 1993 flood failed to keep the flood out of the areas they were designed to protect or were otherwise damaged. Corps officials told us, however, that this estimate was approximate and incomplete. Corps and nonfederal levees protect nearly all of the floodplain in the upper Mississippi River basin. Above Rock Island, Illinois, the Mississippi River floodplain is narrow and is filled largely with navigation pools. The remaining floodplain contains wildlife refuges, some farmland, and a few levees; scattered towns are protected by urban levees. Below Rock Island, the floodplain widens to as much as 6 miles, and because the extensive floodplain is used for crops, the river is almost continuously lined with Corps agricultural levees to Fort Madison, Iowa, and from Keokuk, Iowa, to Cairo, Illinois. In addition, many cities and towns, including St. Louis, are protected by levees and floodwalls in this section of the river. Missouri River floodplains, used predominantly for agriculture, are protected to varying degrees by levees. Between Omaha and Kansas City, Missouri, the river is heavily lined with Corps agricultural levees. Between Kansas City and St. Louis, the Missouri River has four Corps levees, but the river is heavily lined with nonfederal levees. Developed floodplains with larger urban areas—such as Omaha/Council Bluffs, Kansas City, and St. Louis—are largely protected by Corps urban levees. Near Kansas City and St. Louis, several residential, industrial, and commercial areas are built on floodplains behind levees. The Ranking Minority Member of the Subcommittee on Water Resources and Environment, House Committee on Transportation and Infrastructure, and Representative William L. Clay of Missouri asked GAO to review the extent to which (1) the Corps’ flood control levees prevented flooding and reduced damage during the event; (2) these federal levees increased the height of the flooding and added to the damage; and (3) federal, state, and local governments exercise control over the design, construction, placement, and maintenance of nonfederal levees. To address these objectives, we obtained information from the Corps and other federal agencies, state agencies, and other public and private organizations. We also interviewed officials and obtained documents from these agencies and organizations, as well as from individuals. Appendix I provides further details on our scope and methodology. We conducted our review between November 1993 and July 1995 in accordance with generally accepted government auditing standards. We discussed the facts in our report with responsible officials of the five agencies primarily involved: the Chiefs of the Readiness, Hydraulics and Hydrology, Central Planning Management, and Policy Development branches in the Corps’ Civil Works Directorate; the Director of FEMA’s Program Implementation Division; the Deputy Chief for Natural Resources Conservation Programs and the Acting Director of the Watershed Projects Division at NRCS headquarters; the Chief of the Science and Applications Branch and staff from the Office of Surface Water at USGS headquarters; and the Chief of NWS’ Hydrological Service Branch. Generally, these officials agreed with the basic information provided but offered comments, corrections and suggestions to improve the accuracy and clarity of the report. We made changes to the report where appropriate. According to the Corps, 157 (81 percent) of the 193 Corps levees located in the area affected by the 1993 flood prevented rivers from severely flooding about 1 million acres. However, some of these acres were flooded by smaller streams behind the levees and by seepage under the levees. Nevertheless, the Corps estimated that the 157 levees prevented about $7.4 billion in damage during the flood. Another 32 Corps levees withstood flood flows until the water exceeded their design capacity and overtopped the levees. Three other levees were breached without being overtopped by floodwaters, and an opening in an urban floodwall for railroad tracks was not closed in time to prevent flooding. The Corps estimated that flooding at the 36 levees caused about $450 million in damage. To assess the levees’ performance, we compared information on the levees’ design values for either flows or water levels (design capacity) with the flows and water levels recorded during the 1993 event. Data were not available on either the levees’ design capacity and/or the 1993 flows or water levels for 12 of the 193 levees. Data on the other 181 levees show that 177 withstood the flows and water levels at least as well as designed and 4 did not. The flood eventually overtopped some of the levees. Nevertheless, local flood-fighting efforts at some locations permitted levees to withstand flows and water levels that exceeded the levees’ design capacity. Most levees also withstood saturation far longer than they were designed to do. Corps officials told us that generally three basic design criteria apply to each levee. Two of these are flood level, expressed in feet, and flow, expressed in cfs. The five Corps districts involved in the 1993 flood used data on either flood level or flow or on both criteria to judge the performance of levees. The third design criterion is the extent to which a levee can be saturated and still withstand its design flood. The Corps designs levees to withstand saturation. However, according to Corps district officials, the time required to reach a levee’s maximum saturation point varies by flood. A Corps manual (No. 1110-2-1913, Mar. 31, 1978) provides that levees are expected to be exposed to flood flows for only a few days or weeks per year. Embankments that will be exposed to flows for longer periods must meet more stringent criteria for earthen dams. Water standing against a levee for an extended period may move through or under the levee, leading to problems such as sinkholes or sandboilson the landward side. The higher pressure of the floodwater will eventually overcome the materials within the levee and its foundation, and a breach may occur. To determine whether a levee performed to its design capacity, we attempted to compare its design flow capacity with actual or estimated flows during the flood. If the levee’s flow capacity was not available, we used the levee’s flood level capacity. Corps district officials agreed that these measures were acceptable bases for assessing a levee’s performance. As an additional criterion, we considered the length of time the levee withstood saturation from flooding. We asked the Corps to give us the design capacity of, and the flow rates or water levels at, the 193 levees involved in the flood. District personnel told us that all or some of the data were not readily available for 12 levees. They said that Corps field staff or local officials advised them that the land behind the 12 levees was not flooded. Therefore, they said they can reasonably assume that these 12 levees performed within their design capacity. None of the 12 levees were designated as overtopped on Corps levee repair schedules. Appendix II lists the levees for which insufficient data were available for our comparison. Of the 181 levees for which comparative data were available, 177 clearly performed up to their design capacity and sometimes exceeded it during the 1993 flood. Many levees withstood flows that, in some cases, were greater than those for which the levees had been designed because flood-fighting efforts extended their performance by raising their height. In addition, many levees experienced saturation far longer than they were designed to do. Of the 177 levees that clearly met performance criteria, 145 prevented the river from entering the protected floodplain. The flood eventually exceeded the design capacity of the remaining 32 levees and overtopped them. Only four levees allowed floodwater to enter the protected floodplain before the levees were overtopped. Appendix III lists the 145 Corps levees in the flood area that the Corps said prevented flooding. Figure 2.1 displays the location of the 193 levees in the flooded area and identifies the levees that met design criteria but were overtopped, as well as the levees that were breached or otherwise failed without first being overtopped. Corps officials said that, in some instances, flood-fighting at levees prevented water from entering protected areas when the water reached flow rates or elevations beyond the levees’ design capacity. They said that workers prevented overtopping by piling sandbags and building other makeshift barriers on the top and landward sides of the levees. In these cases, a levee exceeded its design capacity in three ways. First, the added height permitted the levee to continue holding back the flood even when the water rose above the top of the levee. Second, the base of the levee withstood the additional water pressure created by extending the height. Third, because flood-fighting prevented or delayed overtopping, the levee withstood saturation far longer than anticipated. Flood-fighting techniques effectively increased the design capacity of many levees. For other levees, such as those at St. Louis or North Kansas City, the success or failure of flood-fighting determined whether the levees were able to meet their original design capacity. Examples of flood-fighting efforts during the 1993 flood are described in appendix IV. According to the Corps, flooding caused 32 Corps levees to be overtopped. These levees were designed to protect against floods whose average recurrence intervals ranged from 20 to 500 years. Of the 32 levees, 26 were on the Mississippi and Missouri rivers where the flood was greatest. Five of the other six levees were located on the Illinois River, and one was located near the Missouri River. Table 2.1 lists, by Corps district and by river, the 32 levees that the flood overtopped, the design capacity of each levee, and the flood’s estimated flow or water level at each levee. We compared either the design flow capacity of the levee with the flood flow recorded at the gauge nearest the levee, or the design height of the levee with the flood level recorded at the gauge nearest the levee. For 29 of the 32 levees, either the flood flow exceeded the design flow capacity of the levee or the flood level exceeded the design height of the levee. For example, the peak flood flow for the South River levee along the Mississippi River just south of Hannibal, Missouri, was 524,000 cfs, far above the levee’s design flow capacity of 349,000 cfs. The flood overtopped three levees, identified in table 2.1, even when the data indicated that the flood flow did not exceed the design flow capacity of the levees. Corps officials said that these cases may be generally explained by (1) a decline from the levee’s design flow capacity, which they attribute to a change in the relationship between the flood level and the flow rate at the levee that resulted in higher flood levels for the same flow rate; (2) the distance between the levee and the gauge used to measure the flood flow, which resulted in an inaccurate flood flow estimate for the levee location; and (3) the location of the overtopping. For example, the main portion of Missouri River Levee Unit R-520 was not overtopped, just as the data indicate. However, the levee was overtopped at a point along a creek where the levee’s design flow capacity was much lower. The 32 levees whose design capacity was exceeded by the flood were concentrated in six general areas: above St. Louis, below St. Louis, the lower Illinois River, southeast Nebraska/northwest Missouri, north of Kansas City, and central Missouri. With one exception, the levees in these areas were built by the Corps to protect agricultural lands from floods. One levee was built to protect Elwood, Kansas, a community of about 1,000 residents, which lies across the river from St. Joseph, Missouri. Three Corps levees, two on the Mississippi River and one on the Missouri River, were breached without first being overtopped by floodwaters. In addition, a railroad opening in a floodwall along the Raccoon River was not closed to prevent flooding in the city of Des Moines. Table 2.2 lists the design criteria and flood flows for these four levees. Corps officials said that they would not classify the breaches of the Bois Brule and Kaskaskia Island levees as performance failures for two reasons. First, these breaches occurred when the water reached the freeboard area of the levee, which is the safety zone above the levee’s design height. They said that when water rises above a levee’s design height, the levee cannot be expected to continue holding against a flood. Second, because the duration of the 1993 flood far exceeded the design standard for levees, some Corps officials believe that the breaches should not be characterized as performance failures. According to Corps district staff, the breach in the third levee, Missouri River Levee Unit L-550, was caused by the use in the construction of the levee of a material that allowed underseepage. Corps staff told us that the fourth levee failed to prevent flooding because a railroad opening in the floodwall protecting the city of Des Moines was not closed in time. Each of these situations is described in appendix V. In its April 1994 report to the Congress on flood damage for fiscal year 1993, the Corps qualified the accuracy and completeness of estimates of the damage prevented and incurred because of the broad scope of the damage and the rapid compilation of preliminary estimates. Given their methodologies for estimating flood damage, the Corps and NWS said that their estimates of both the damage prevented and the damage incurred, presented in this report, are probably understated. The Corps estimated that its levees prevented the flooding of about 1 million of about 1.4 million acres protected. The overtopping and breaching of levees caused about 400,000 acres to be flooded. However, in areas behind levees that held, some flooding still occurred because of heavy seepage through and under the levees and heavy flows from streams draining areas behind the levees. Corps officials said that they have not estimated the number of acres flooded from these sources. According to the Corps, the levees prevented about $7.4 billion in flood damage during the 1993 flood. To calculate the damage prevented, the Corps uses damage curves that employ the principles of hydrology and economics to graphically depict the estimated costs of the damage that would occur if a Corps levee were not protecting the area. Economists compare curves depicting the estimated damage with and without a levee. The difference is the estimated value of the damage prevented by the project. Corps staff said that the severity and duration of the 1993 flood in the St. Louis district were so great that the existing damage curves could not be used to estimate the damage accurately. As a result, Corps personnel extrapolated damage estimates for the 1993 flood from damage estimates for the large 1973 flood and used their professional judgment. Corps staff pointed out that because the dollar values on the damage curves tend to be outdated, the estimates of flood damage derived from these curves are probably understated. They said that some of the dollar values they developed when they constructed the levee have not been updated to reflect the value of development that has since occurred behind the levee. They also mentioned that obtaining information about urban land values requires extensive fieldwork and research, which are labor-intensive, expensive, and time-consuming. Corps officials said that because their resources are limited, they give priority to updating information on the largest urban centers where the greatest threat of damage exists. The four Corps districts where levees were overtopped and breached estimated that about $450 million in damage was incurred behind the levees. According to the Corps, it did not have sufficient funding to complete on-site field surveys to estimate the damage incurred from the 1993 flood and, as an alternative, used a variety of techniques to estimate the damage. In the Rock Island District, for example, estimates were based on data from county-level sources that, according to Corps district officials, were rough estimates and were specific to only a few sites. The Omaha and St. Louis districts used computer programs and actual river elevations to compute the damage incurred, supplementing these sources with information from surveys of businesses, local and state government officials, and field personnel from the Department of Agriculture. The Kansas City District used water levels and damage curves to estimate the damage incurred. NWS makes the overall estimates of flood damage suffered each year that the Corps reports to the Congress. For fiscal year 1993, NWS estimated that all floods caused more than $16 billion in damage. According to staff from NWS’ Office of Hydrology, loss estimates can be considered only approximate because they are developed as an ancillary function to NWS’ primary mission of forecasting weather and floods. They said the quality of resulting estimates is uneven because of insufficient resources, inconsistent methods and sources, incomplete data collection by field offices, and early reporting deadlines. While data on the design flow capacity of the Corps levees and the actual flows during the 1993 flood are incomplete, the Corps levees in the area affected by the 1993 flood generally performed as designed. In fact, some levees withstood significantly greater flood flows and elevations than they were designed to withstand, especially when the duration of the flood is considered. Where the levees did not prevent water from entering protected areas, the Corps levees were overwhelmed by the size of the Great Flood of 1993. By confining floodwaters within a smaller portion of a floodplain than they would otherwise occupy, levees pressure the waters to rise higher and flow faster than they would do without restraint. Whether levees significantly increase flood levels varies by location. Computer simulations of the 1993 flood estimated that the nearby Corps levees added up to 2.7 feet to the flood crest at St. Louis and up to 7.3 feet to the flood crest at other locations. Corps officials acknowledge that levees increase flood levels and induce some flooding. However, they emphasized that the net effect of levees, reservoirs, and navigation structures in the upper Mississippi River basin is to reduce flood levels and damage. There is no consensus among researchers, however, about the long-term effects of these structures on flood levels. Many factors besides levees help determine the peak level of a flood. These include the amount of water entering a river from precipitation, the size and shape of the river’s channel and floodplain, and other natural factors. Human activities, such as clearing the floodplain for cultivation, have also affected flood levels. Studies show that cumulative changes within the basins have caused higher flood levels for Mississippi and Missouri river flows. In addition, available evidence suggests that precipitation in the upper Mississippi basin may be increasing. If these trends continue upward, rather than cycling downward, more frequent and more extensive flooding will occur and future damage from flooding may increase. Available studies and the experts we spoke with agreed that levees generally contribute to higher flood levels above and within the levied part of a river. They said that the construction of a levee in a floodplain forces the water into an artificial floodway, causing the water to back up, just as the loss of a traffic lane on a busy highway causes heavy traffic to become congested. This restriction raises the level of the water both upstream of the levee and at the levee itself. It also forces the water to flow faster than it would if it were permitted to spread out across the whole floodplain. Figure 3.1 depicts a cross-sectional view of a floodplain and the floodway created by the construction of a levee. According to the results of a 1994 study conducted by a federal interagency team, a levee can increase flood levels from a few inches to several feet. Factors such as the size of the flood, the height of the levee, and the dimensions of the floodway compared with those of the natural floodplain account for much of the variation. The U. S. Army Engineer Waterways Experiment Station, a Corps research facility, identified the impact of some of these factors in a 1991 study requested by FEMA. Using a model developed by NWS, the Corps simulated a range of hypothetical floods and levees for 150 miles along the Missouri River from Jefferson City to Waverly, Missouri. This portion of the river is lined by mostly small private and other nonfederal agricultural levees that do not allow for the floodway recommended by FEMA. These levees are built to varying heights and, as a result, do not provide a consistent level of protection. Overall, these levees prevent only small floods—those with average frequency intervals below 10 years—from entering the protected floodplain. The 1991 study found that the impact of levees on flood levels increases with the size of the flood until the flood is large enough to overtop the levees. It also estimated that raising the height of the existing levees to contain the 100-year flood would further increase flood levels by about 5 inches. Moving the existing levees back to FEMA’s recommended floodway boundary would reduce peak flood levels for the 100-year flood by an average of about 1.2 to 1.5 feet. The study concluded that flood levels are more sensitive to the size of the available floodway than to the height of the levees. The results of three studies show that Corps levees and nonfederal levees contributed to the flood levels experienced during the 1993 flood. These studies used computer modeling to simulate flows and water levels in part of a river during the 1993 flood, with and without levees. Observed differences, therefore, can be attributed directly to the presence of levees. However, the accuracy of a model’s results depends on the accuracy of the information describing the flood and the floodplain. It also depends on how well mathematical equations in the model represent actual hydrologic processes. These studies are the only modeling efforts on the 1993 flood to date and use models created by or accepted by the Corps. Because of the extremely complex nature of hydrologic computer models, we did not review the accuracy of the models or their results. In 1993, the St. Louis Post-Dispatch commissioned an associate professor of civil engineering at the University of Illinois to simulate the 1993 flood. According to the researcher, the simulation included the Mississippi River beginning just north of St. Louis and extending 50 miles downstream to Prairie Du Rocher, Illinois. He used a Corps model that, for simplicity, assumes that floodwaters flow at a steady rate rather than at varying actual rates. The simulation estimated that Corps agricultural levees added about 1.0 to 1.5 feet to the flood crest at St. Louis. The flood’s crest passed St. Louis just as two large downstream agricultural levees were being overtopped and the land behind them flooded. The Post-Dispatch simulation indicates that if these levees had not been overtopped or if they had been overtopped at another time, they would have added from 1.4 to 1.8 feet to the crest. This would have brought the crest to within a foot of the design capacity of the St. Louis floodwall. After the 1993 flood, the administration established an interagency Scientific Assessment and Strategy Team to provide data and an analysis of the 1993 flood. At the request of the Interagency Floodplain Management Review Committee, the team simulated the flood. The simulation included the Mississippi River and major tributaries from near Hannibal, Missouri, to Cairo, Illinois, and the Missouri River from Hermann, Missouri, to the mouth at St. Louis, Missouri. This area contains Corps urban and agricultural levees and smaller private and other nonfederal levees. The team used a recently developed model capable of simulating the varying rates of flow that occur during a flood. The simulation estimated that Corps agricultural levees added a few inches to 7.3 feet to the flood crests at 14 locations. For St. Louis, it estimated that the levees added a maximum of about 1.5 feet to the flood crest. These results were consistent with the results simulated for the Post-Dispatch. The Corps, as part of its 1995 Floodplain Management Assessment, also ran computer simulations of the 1993 flood. Its analysis, which included the lower Missouri River and the middle and upper Mississippi River basins, was performed on a systemwide basis using the model employed by the Scientific Assessment and Strategy Team. One simulation estimated that agricultural levees added up to 7.2 feet to the flood levels recorded at 11 locations in the Corps’ St. Louis District, including about 2.7 feet at St. Louis. Similarly, it estimated that agricultural levees added from a few inches to 4.7 feet to the flood levels along the Missouri River with one exception: Agricultural levees reduced the 1993 flood levels at Hermann, Missouri, by about 1 foot. No evidence is available to show the extent of the damage brought about by the addition to the flood’s height attributable to the levees, but Corps officials acknowledge that damage was caused by the levees. The Corps also points out that its levees provided substantial benefits in 1993 by preventing flooding on about 1 million acres in the developed floodplain. The Corps estimates that its levees in the Kansas City, Omaha, Rock Island, St. Louis, and St. Paul districts prevented about $7.4 billion in flood damage. Although unprotected areas within reach of levee backflows are subject to greater flooding than would occur if no levees existed, Corps officials believe that the damage prevented by levees greatly outweighs that induced by levees. In addition, they said that the capacity of reservoirs to store floodwater compensates for the increases in flood levels caused by levees. For example, the Scientific Assessment and Strategy Team found that storing water in flood control reservoirs reduced the peak flood level at St. Louis by 5 feet in 1993. Thus, levees and reservoirs at St. Louis achieved a net reduction in flood levels. Levees are one type of navigation or flood control structure that can affect flood levels. While the impact of levees on a particular event, such as the 1993 flood, has been estimated with sophisticated models, the long-term effects of such structures on flood levels in the Mississippi River basin is less certain. Over the years, researchers have used trend analysis to assert a relationship between long-term increases in flood levels and these structures. However, the value of this analysis is limited by a lack of accurate information about historic flood flow rates, as well as by the conflicting results these studies have yielded. Much of the research is based on flow rates estimated for extreme floods at St. Louis. Some of the flow rates for extreme floods in the historical record are estimates based on observations of water levels recorded at the time of the floods. Other flow rates in the historic record are based on methods or equipment now shown to be inaccurate. In any case, measuring flow rates during extreme floods is very difficult and sometimes impossible. Not only can a flood destroy recording equipment, but it can also prevent access to the best measurement sites. As a result, 7 of the 10 highest flow rates recorded for St. Louis before 1993 are estimates rather than actual measurements. A more accurate and now generally used device for measuring flow, the Price current meter, was not used exclusively at St. Louis until 1931.Before that time, various methods of measuring flow rates were used. Some researchers question the accuracy of these methods for very high flows because of the findings of two studies conducted at the University of Missouri at Rolla. The first study, conducted in 1976, found that estimates of very high flow rates are subject to large errors and that flows above the banks of the river could not be estimated satisfactorily. The second study, conducted in 1979, tested the accuracy of measurement devices used at St. Louis before 1931. Although the author concluded that most of the pre-1931 measurements were valid for use in analyses, he found that 57 percent of the test measurements taken for flows above the banks of the river using double floats exceeded the measurements taken with a Price current meter by more than 10 percent. Double floats were used at St. Louis from October 1881 through December 1930. On the basis of his findings, the author recommended that historical estimates of high flow rates be used only for the relative ranking of floods. In 1985-86, researchers at the Corps’ Waterways Experiment Station found a further indication of errors in pre-1933 flow rate information. Using a physical model of the Mississippi River, the researchers found that they could reproduce the high water marks at St. Louis for the 1844 and 1903 floods using flow rates 33 percent and 23 percent lower than historic flow estimates published by USGS. Three studies performed during the 1970s analyzed the trends in water levels for similar flows at specific locations and produced dissimilar results. In addition, two of the studies used pre-1939 historical data that are of questionable accuracy. Given the data and/or methodological problems associated with these studies, no conclusions can be drawn from them about the long-term impact of navigation structures (such as dikes) or flood control structures (such as levees) on flood levels. The following paragraphs outline the results of each study. In a 1974 study, researchers at Colorado State University addressed the impact of constructing levees and channelizing the river for navigation on water levels and flow rates at St. Louis before and after 1900. Their comparison found that water levels were higher for all flows above 300,000 cfs, but maximum annual water levels and average and maximum annual flow rates remained unchanged. They also found that lower water levels existed after 1900 for all flows below 300,000 cfs. The researchers asserted that the construction of navigation dikes and levees between 1900 and 1940 caused both decreases and increases in the water level’s relationship to the flow rates at St. Louis. Another study, which was published in 1975 by an associate professor of geology at St. Louis University and was widely cited during the 1993 flood, compared the relationship between maximum annual water levels and flow rates at St. Louis in 1973 with the same relationship during a base period from 1861 through 1927. This study found that the 1973 flow rate of about 851,100 cfs produced a flood elevation about 7.9 feet higher than it did during the base period. The researcher attributed the rise in water level per flow rate to a combination of navigation works, levees, and riverbed sedimentation. He has since refined his results and attributes about 4 to 5 feet of the total increase to levees, about 2 to 3 feet to navigation works, and about 1 foot to riverbed sedimentation. The Corps has questioned the results of both studies because the studies used the suspect historical data. In addition, a USGS headquarters hydrologist who specializes in the statistical analysis of hydrologic information told us that modeling is better than trend analysis for identifying the effects of navigation dikes and levees. The USGS hydrologist said that trend analysis can never prove what caused the changes identified. Thus, changes in water levels and flow rates that occur after the construction of navigation dikes and levees may suggest but do not prove that these structures are the source of the changes. In 1975-76, researchers from the University of Missouri at Rolla attempted to study the effects of levees on flood levels. Unlike the authors of the 1974 and 1975 studies, they used only post-1930 data to avoid questions about the accuracy of historical data. As a result, they concluded that the Mississippi River had not experienced enough floods of sufficient size from 1930 through 1976 to evaluate the effect of levees on floods. Their study also examined the effect of navigation dikes on water levels in the middle Mississippi River. They found that changes in water level per flow rate between 1934 and 1974 at three major gauging stations were dissimilar. The University of Missouri researchers concluded that changes in water level per flow rate at the study gauges showed no association with dike construction. The study stated that although the constriction of the channel caused by building an individual dike must have at least a temporary effect on the relationship between the water level and the flow rate, the dissimilar findings at the three gauging stations suggest that the effect may be restricted to the area immediately around the dike. Therefore, they said data on water levels and flow rates cannot be extrapolated from a single point of record to an entire reach of the river. Although floods result from heavy rainfall during a short time or above-normal rainfall over a long time—sometimes in combination with snowmelt—this precipitation interacts with the atmosphere, land topography, vegetation, soils, channel geometry, and human activities to determine the amount of runoff. The chief determinants of a flood’s peak level at a particular location in a river are the amount of water reaching the river as runoff, the size and shape of the river channel, and the size of the floodway. Assessing the impact of a single factor, such as levees, on water level is very difficult because hydrologic models can only approximate the complex processes that move and store water. Variations in flood levels under like conditions are not uncommon. According to the Corps, records of the relationship between high flows and water levels at St. Louis show about a 5-foot variation in water levels for like flows. For example, two floods passed St. Louis within a month in the spring of 1983 with similar flows but with crests whose height differed by 2.7 feet. According to the Corps, no accurate accounting for this variation exists. Over time, some of the factors that help determine a flood’s peak level also shift the range of water levels produced by like flows. Several studies have addressed the factors, both natural and man-made, that affect flood levels. Natural variables that help determine a flood’s peak level include (1) the flood’s duration and whether it is rising or receding, (2) the seasonal level of vegetation in the floodway, (3) the way the flood carries sediment, and (4) the water’s temperature. Long-term changes in the river’s channel from erosion, past floods, and earthquakes, as well as the growth of vegetation in the floodplain, particularly in the floodway, also affect peak flood levels. A 1994 study on the relationship between flow rate and flood level simulated hypothetical floods of 4.5 days, 9 days, and 13.5 days with the same peak flow rate through a channel approximating the Mississippi River at St. Louis. The study found that the speed with which a flood reaches its peak flow and the duration of that peak flow help determine the flood’s peak water level. For instance, in 1993, flow rates of 1,030,000 cfs at St. Louis on 2 consecutive days increased the water level by half a foot on the second day. According to the Scientific Assessment and Strategy Team’s study, the vegetation in the floodway affects flood levels because it obstructs and slows the flow of water, causing the water to rise. Consequently, the water level in an area covered with shrubs and trees would be higher than in an area covered with grass. Similarly, the same flood can be higher during the summer than during the late fall, winter, or early spring because of the summer foliage. Researchers have found that swiftly moving floodwater can cause intense erosion and sedimentation. The transport and deposition of sediment during a flood can increase or decrease water levels at various locations. In addition, changes in water temperature affect the amount and shape of sediment in the river. Cold water carries more sediment and enlarges the size of the sediment particles, increasing the friction that, over time, can scour the channel and increase its flood-carrying capacity, reducing all water levels. Floodplains reduce flood levels by providing space for the temporary storage of floodwaters until natural drainage can carry them away. They also reduce flood velocities. In addition to flood control, human activities in the last 175 years in agriculture, navigation, and urban development have altered the floodplains in the upper Mississippi River basin. These activities have altered water flow rates, the width and depth of the river channel, the size of the floodway, the pattern of erosion and sedimentation, the level of vegetation, and the speed with which precipitation flows into streams. Early development in the upper Mississippi River valley was closely tied to the rivers. By the late 1800s, settlers had cleared millions of acres in the floodplain for cultivation. Vegetation in the unaltered floodplain, especially in wetlands, created resistance to flow. Researchers have shown with modeling that removing resistance reduces flood levels and increases flow velocities and erosion. They speculate that clearing the floodplain for agriculture had the same effects. Between 1780 and 1980, an estimated 57 percent of the original wetlands in the nine midwestern states affected by the 1993 flood were converted to other uses, mainly agriculture. Wetlands temporarily store and hold some floodwater for later drainage. According to the 1994 report from the Interagency Floodplain Management Review Committee, the loss of wetlands contributes to higher flood levels for smaller, more frequent floods, like 25-year or smaller floods. Agricultural land management practices affect the processes of erosion, sedimentation, and runoff. For example, the Illinois Natural History Survey found that planting crops in rows and plowing with moldboards (which lift and turn the soil) increased the rate at which Illinois lakes were filling with sediment. Although researchers have observed the influence of agricultural land management practices on small watersheds, the influence of these practices on major rivers is still largely speculative. Travel to and commerce with early settlements along the rivers created a demand for improved navigation on the rivers. The Congress first approved a plan for improving the Mississippi River’s channel in 1881. However, most channel improvements for navigation on the Mississippi and Missouri rivers were made between 1927 and 1944. The improvements generally narrowed the natural channels and shortened the rivers. The Corps has constructed about 3,100 wing dikes to create and maintain navigation channels on the Mississippi and Missouri rivers. Wing dikes are embankments built in the river perpendicular to the shoreline to increase channel depths by reducing channel widths and increasing flow rates. The dikes help keep sediment from accumulating in the main channel and trap it along the shoreline. The previously cited 1974 Colorado State University study found that, between 1888 and 1968, wing dikes decreased the average width of the middle Mississippi River by 2,100 feet, or by about 40 percent. The report also stated that degradation of the riverbed has occurred along the middle Mississippi River whenever the channel has been narrowed. The Corps also stabilizes and dredges hundreds of miles of river bank and channel, respectively, to maintain open water navigation on the Missouri and Mississippi rivers. Stabilizing the banks reduces shoreline erosion, and dredging deepens the channel. In addition, the Corps has built and operated a system of locks and dams on the upper Mississippi River since the 1930s. This system converted the upper river into a series of pools to maintain channel depths at low and normal flows. A 1988 study of the long-term effects of the oldest lock and dam on the upper Mississippi River found that initially the river’s width and volume had increased behind the dam. However, the long-term impact has been to trap sediment. As a result the river has steadily lost both width and volume and returned to near pre-dam water levels and flow. Studies have shown that the growth of urban areas increases the speed of water running off the land into streams. Rain that falls on paved, tiled, or other impervious surfaces and runs into storm drainage systems is delivered to streams more quickly than it could run off porous surfaces. Hence, urban runoff produces higher, sharper flood peaks on small rivers and streams than rural runoff. However, as the water from each small stream joins the water in larger streams, the effects of urbanization on flood levels are diluted. One expert told us that researchers have not been able to measure the impact of urban development on the flooding of the Mississippi or Missouri rivers because the effects of urbanization are too small to isolate. Recent analyses show that water levels for high flows may be increasing for some locations in the upper Mississippi River basin. A continuing Corps study of Missouri River water levels shows that flow rates that once nearly filled the channel have been producing higher flood levels since the late 1920s. Similarly, a 1994 study of flow rates on the Mississippi and Missouri rivers found that flood levels for like flow rates have increased over time. Evidence also suggests that precipitation in the upper Mississippi basin may be increasing. These trends concern the Corps because they increase the frequency and extent of flooding, thereby increasing the damage from flooding. According to the Corps’ Missouri River Division, seven of nine gauges on the Missouri River have slowly and consistently produced higher water levels for the same high flow rates since about 1927. The rising trends are most noticeable at Nebraska City and Omaha, Nebraska, and at St. Joseph, Missouri. Table 3.1 shows the approximate increases in water levels for selected high flow rates. Statistical studies performed in 1994 at the Environmental Management Technical Center at Onalaska, Wisconsin, also examined water levels and flow rates for the period of record at six gauges: St. Louis, Chester, and Thebes on the Mississippi River; and St. Joseph, Waverly, and Hermann on the Missouri River. The studies found that water levels for like flood-level flow rates have been increasing at all six gauges at an average of about 1.2 inches annually. Over the periods of record, which range from 51 years at Chester to 132 years at St. Louis, increases in water levels at the six gauges ranged from about 3 to 9 feet. Table 3.2 shows the approximate increase at each gauge. The studies also found that, except at Thebes, the trend for maximum annual water levels is increasing. The studies found no changes in trends for flow rates. According to the 1994 Natural Disaster Survey Report issued by NWS, the duration and size of the 1993 Midwest flood and the wet conditions leading up to it suggest a significant variation in climate. An air circulation feature, called El Niño/Southern Oscillation, driven by abnormal sea surface temperatures occurred in both 1992 and 1993. Preliminary NWS modeling using the temperatures associated with the El Niño episode produced large-scale atmospheric results resembling the abnormal precipitation and temperature pattern experienced in 1993. However, NWS stated that it requires more in-depth and thorough analyses to understand the role played by El Niño in the extreme precipitation. According to a 1993 report on the Midwest flood by the Illinois State Water Survey, increases in the volume of water flowing down the Mississippi River are most closely related to the overall climate and precipitation. The report states that climate and resulting precipitation exert such a strong impact on streamflow that it masks changes from other sources, such as physical changes to the basin. The Illinois State Water Survey reports that, for most locations along the Mississippi River, average streamflows for the 28-year period since 1965 have been the highest on record. On the basis of an 11-year moving average, the Survey calculated that average flow rates at Clinton and Keokuk, Iowa, and St. Louis, Missouri, have increased by about 25 to 33 percent over the long-term average since the mid-1960s. An 11-year moving average is used to describe the trend because it smooths out natural fluctuations in the data. Also, average streamflows at Keokuk and St. Louis between 1965 and 1992 are about 17 and 13 percent, respectively, above the long-term average streamflows based on over 100 years of record. Average streamflows for the same period for the tributaries between Clinton and Keokuk, Iowa, for the Des Moines River, and for the Illinois River are even further above their long-term average streamflows. According to officials of the Survey, these deviations are significant because, historically, average streamflow has remained remarkably consistent with long-term average streamflow. USGS headquarters officials cautioned that because weather is cyclical and variable, any trends in climate are difficult to distinguish from normal weather cycles of higher precipitation and drought. For the 10-year period from 1983 through 1992, NWS estimates that damage to the United States from flooding totaled about $20.5 billion, or an average of about $2.1 billion annually, unadjusted for inflation. With the inclusion of data for 1993, the total estimated flood damage became $36.9 billion, or about $3.4 billion annually over the 11-year period. According to the Corps, upward trends in water levels are of concern, whether they are caused by increased streamflow or by higher water levels for the same flow rates, because they increase the frequency of flooding and the area subjected to flooding. For instance, between 1928 and 1959, flows of 100,000 cfs at St. Joseph, Missouri, never exceeded water levels of 17 feet, the official level at which flooding begins. Since 1959, flows of 100,000 cfs have exceeded flood level 16 times. If changes in the climate of the upper Mississippi River basin increase precipitation in the future and if water levels for like flow rates continue to rise, then the damage from flooding will rise unless the ability of flooding to cause damage is mitigated. The administration’s Interagency Floodplain Management Review Committee was formed to identify the major causes and consequences of the 1993 Midwest flood, evaluate the performance of existing floodplain management programs, and recommend changes that make the programs more effective. Among the Committee’s findings and recommendations are many related to flood control activities, particularly levees. See appendix VI for a summary. That levees increase flood levels is subject to little disagreement. Whether this increase is significant varies from location to location, but whether unprotected lands are more likely to be flooded than protected lands depends on the increase in flood levels after the construction of a levee. Proponents of levees point out that the impact of a levee should not be isolated because the net effect of all flood control projects has been to reduce flood levels and prevent billions of dollars in flood damage. Levees are only one of many natural and man-made factors that help determine the peak level of a flood. Cumulative changes within the upper Mississippi River basin have caused higher water levels for similar flows. These trends could mean that the damage from flooding may increase in the future because higher water levels are associated with more frequent and more extensive flooding. No federal program regulates the design, placement, construction, or maintenance of nonfederal levees. However, the federal government can exercise some control over nonfederal levees through programs that regulate navigable waters and wetlands and that provide flood insurance and disaster and emergency assistance. Overall, 17 of the 50 states have specific programs for regulating levees. Five of the nine states involved in the 1993 flood have regulatory programs that, to varying degrees, affect nonfederal levees. Most often, states are responsible for the overall coordination of floodplain management activities within the state and across state lines. In some cases, states may regulate local land use when localities are unable or unwilling to take the actions needed to reduce the risk of flooding. Local governments usually exercise control over nonfederal levees in response to requirements of the National Flood Insurance Program and state regulatory programs. However, in states without a regulatory program, local land use regulations generally affect the placement and construction of levees. Nonfederal levees are regulated to some extent under two federal programs that require permits to construct or modify levees affecting navigable waters and wetlands. In addition, the programs of three federal agencies that provide flood insurance and emergency and disaster assistance to repair flood-damaged levees affect nonfederal levees. Under the National Flood Insurance Program, FEMA exempts communities from certain requirements of the flood insurance program if they can show that the levees protecting them are designed, constructed, located, and maintained according to specified criteria. As part of its mission to provide disaster assistance, the Corps is authorized to repair levees through its levee rehabilitation program. The Department of Agriculture’s Natural Resources Conservation Service (NRCS) provides funds and technical assistance for the emergency repair of nonfederal levees that are damaged during a flood. In all of these programs, the federal government exercises some direct or indirect control over the nonfederal levees’ design, placement, construction, or maintenance. However, all of these elements are not always affected by each program. Table 4.1 lists the federal programs affecting levees, and table 4.2 shows whether these programs affect a nonfederal levee’s design, placement, construction, or maintenance. Section 10 of the Rivers and Harbors Appropriation Act of 1899 prohibits the obstruction or alteration of navigable U.S. waters without a permit from the Corps. Section 404 of the Clean Water Act requires a permit from the Corps when wetlands are to be altered. Under both of these authorities, the Corps may control the design, placement, and construction of a nonfederal levee when the levee affects areas regulated by these programs. However, the owner, sponsor, or builder is responsible for informing the Corps of the intent to construct a levee and for obtaining a permit. The Corps’ data do not indicate how many of the 15,000 permits requested for activities in wetlands or navigable waters in fiscal year 1994 were for constructing or modifying levees. For the 1993 flood, the Corps issued a nationwide permit for the construction of temporary levees and emergency repairs to levees. This permit encompassed U.S. waters—including rivers, streams, lakes, and wetland areas—in the counties that had been declared flood disaster areas. The permit was specific and identified conditions where it applied. For example, the permit required any levee that was reconstructed or repaired to be maintained after the flood. Since some levees were being constructed in new locations, the Corps could also require that the former location of the levee be restored to its previous condition. The permit also required that any temporary levees built must minimize damage to U.S. waters and that measures must be taken to maintain near-normal downstream flows. Furthermore, all levees were to be designed and constructed so as to prevent the channel from being constricted or redirected and erosion from occurring upstream or downstream. Section 10 prohibits the obstruction of the navigable capacity of waters of the United States without a permit from the Corps. To clarify the extent of the Corps’ authority to regulate levees under this section, we asked the Corps to provide its interpretation of this authority. Specifically, we asked whether the Corps has the authority to regulate upland nonfederal levees. Currently, these levees are not regulated by the Corps. In a November 1994 response, the Chief Counsel of the Corps indicated that the Corps might be able to assert jurisdiction over upland nonfederal levees on the basis of section 10. While this provision prohibits the obstruction of the navigable capacity of waters of the United States, it also prohibits the alteration or modification of the course, location, condition, or capacity of “any navigable water of the United States unless recommended by the Chief of Engineers and authorized by the Secretary of War” before work begins. From his review of court cases dealing with section 10, the Chief Counsel opined that a prerequisite to the Corps’ exercising jurisdiction would be circumstances demonstrating an alteration or modification of the course, location, condition, or capacity of navigable U.S. waters. However, he noted that a court might require that jurisdiction be founded on a negative effect and that, arguably, nonfederal levees have an essentially beneficial effect on navigable capacity. The Chief Counsel concluded that if the Corps exercised jurisdiction over upland federal levees under section 10 without clear authority and direction from the Congress, this action would likely be overturned upon appeal to the federal courts. FEMA administers a levee certification program under the auspices of the National Flood Insurance Program. To be certified and to enable communities to enact less stringent building codes, levees must meet FEMA regulations specifying design, placement, construction, and maintenance standards. In all cases, the levee must, at a minimum, protect against a 100-year flood and, in most cases, have at least 3 feet of height (or freeboard) above the 100-year level of protection. When FEMA maps a community, it designates the 100-year floodplain and floodway. When a community joins the insurance program, it must require that all buildings constructed or substantially improved be protected to the base flood elevation. This is usually accomplished by elevating the structure above the base flood elevation or, in some instances, by floodproofing. While increasing construction costs, these requirements make buildings more resistant to flood damage. FEMA officials said that experience confirms that elevated buildings are less susceptible to flood damage and, therefore, cost less to repair when a flood occurs. When FEMA certifies a levee as protecting against a 100-year flood, it exempts new construction and substantial improvements from the requirement to build above the base flood elevation. The effect of this exemption is to increase the costs of repairing buildings that are more susceptible to flood damage because they are not elevated. The Corps will repair, on a cost-shared basis, nonfederal levees that are damaged by floods if the levees meet the qualifying standards of the program and are in good standing with the program at the time of flooding. In order to be accepted into the Corps’ rehabilitation program for damaged nonfederal flood control levees, a sponsor’s levee must meet or be improved to meet the Corps’ minimum design standards before flood damage occurs. The owner may need to modify the levee in order for the levee to be admitted to the program. Once in the program, the levee must be maintained in accordance with the Corps’ criteria. For levees that have qualified, the Corps provides 80 percent of the repair costs and the levee’s sponsor pays the rest. To qualify, the levee must (1) be publicly sponsored to ensure that the 20-percent share can be paid, (2) be a primary levee providing a 10-year level of protection for urban areas and a 5-year level of protection for agricultural areas, (3) be properly maintained and regularly inspected, and (4) provide benefits that equal or exceed the cost of the levee’s repair. The rehabilitation program includes levees maintained by local sponsors, such as levee districts, individual municipalities, or Indian tribes. The Corps received 546 requests for assistance under the levee rehabilitation program after the 1993 flood. Of these requests, as many as 345 either were declared ineligible for assistance because the levees did not meet the basic requirements identified above or had been repaired under other disaster assistance programs, such as NRCS’ program. As of October 1994, the Corps estimated that the costs of repairing the 201 levees that met its established criteria would be $250 million. Also, during 1993, the Corps worked with the Department of Commerce’s Economic Development Administration (EDA) to identify and repair levees that were deemed to affect the economic development of the community where the levee was located. The Corps assisted EDA in assessing the levees that had not qualified for the Corps’ program in advance primarily because they lacked public sponsorship. Because one of EDA’s missions is to provide economic assistance to areas experiencing sudden and severe economic distress, the agency provided funding on a cost-shared basis to repair levees that protect critical public infrastructure. Working with the Corps, EDA reviewed all applications and funded those that identified local sponsors to share costs and maintain the levees. EDA has funded the repair of 13 levees at a cost of about $4.2 million. NRCS’ Emergency Watershed Protection Program, authorized by the Agricultural Credit Act of 1978 (P.L. 95-334), funds repairs to levees. To be eligible for these funds, the levee project must protect property threatened by a watershed emergency and the owner must either have exhausted, or lack, the funds needed to remedy the problem. To be eligible for assistance after the 1993 Midwest flood, potential applicants were also required by NRCS to show that the levee restoration project protected property or life, that benefits exceeded the costs of repair, and that regular maintenance would be performed. Since the program supports the repair of nonfederal levees to their preflood condition, it does not control the original design, placement, or construction of nonfederal levees. However, because regular maintenance is required for future assistance, the program does set maintenance standards. NRCS also received funding in a 1994 supplemental appropriation to make repairs to levees otherwise ineligible for assistance on condition that a sponsor share the cost and that the levee qualify for the Corps’ levee rehabilitation program. The levee also had to be environmentally sound and provide a 5-year level of protection. While NRCS’ Emergency Watershed Protection Program provides for a smaller proportion of the funding for repairs than the Corps’ levee rehabilitation program (75 percent versus 80 percent), NRCS would provide 100 percent of the funding in the case of extreme need or hardship on the part of the levee sponsor. As of February 1, 1995, NRCS had approved repairs for 375 levees at a cost of $10.2 million under the Emergency Watershed Protection Program. NRCS had also determined that 16 levees were eligible for repairs, costing a total of $650,000, under the supplemental appropriation for fiscal year 1994. NRCS officials reported the agency is evaluating repairs for 26 more levees under the supplemental appropriation. Some of these repairs were for levees whose drainage area exceeded NRCS’ threshold of 400 square miles and would, therefore, normally have fallen under the Corps’ levee rehabilitation program. However, NRCS was given the responsibility for repairing these levees under the supplemental appropriation. In 1992, the Association of State Floodplain Managers issued a reportdocumenting the scope of floodplain management programs in the states and found that 17 states regulate levees as part of their floodplain management programs. Of the nine states involved in the 1993 flood, five states (Iowa, Kansas, Minnesota, North Dakota, and Wisconsin) have programs to regulate levees. These programs have standards regulating the design, construction, placement, and maintenance of levees. We judgmentally selected Iowa as an example of a state that regulates its levees. Iowa’s program for regulating levees addresses the design, placement, construction, and maintenance of nonfederal levees. The program, which is described in Iowa’s statute on floodplain management, is administered by the state’s Environmental Protection Division within the Department of Natural Resources and has been in existence since at least 1967. In general, Iowa requires permits to construct, operate, and maintain levees in rural areas where the watershed drainage area exceeds 10 square miles and in urban areas where the watershed drainage area exceeds 2 square miles. Specific requirements must be met to comply with the statute’s criteria for level of protection, placement, interior drainage, freeboard, or design. For example, Iowa requires agricultural levees, at a minimum, to protect against a 10- to 25-year flood and urban levees to protect against a 100-year flood. Urban levees must provide at least 3 feet of freeboard above the design flood profile. Furthermore, any agricultural levee whose protection level is increased must comply with the state’s equal and opposite conveyance rule: An increase in the levee on one side of a river’s floodplain cannot cause a disproportionate increase in the flooding of the river’s opposite floodplain. Iowa officials told us that only one FEMA regulation more strictly controls levees. Under this rule, development within FEMA’s identified 100-year floodway cannot increase the area designated as the 100-year floodway. The effect of this rule is that agricultural levees may be placed farther away from the river than the equal and opposite conveyance rule may initially require. Of the nine states involved in the 1993 flood, four (Illinois, Missouri, Nebraska, and South Dakota) have yet to develop specific regulatory programs for levees. All of these states have a floodplain management program that may (1) provide technical expertise to localities and individuals on complying with FEMA’s or other federal agencies’ requirements, (2) regulate local land use when localities are unable or unwilling to take needed actions to reduce the risk of flooding, or (3) coordinate local and regional floodplain management programs. We judgmentally selected Missouri as an example of a state that does not regulate levees. Missouri does not currently regulate its levees but has identified areas of its floodplain management program that need to be improved. A task force established by Missouri’s governor to review and evaluate the state’s floodplain management program recommended the development of a levee oversight program to help decrease the risk to life and property from flooding. The task force recommended in a July 1994 report that Missouri (1) adopt a permit program for constructing and modifying levees; (2) identify all existing levees for the purpose of developing design criteria and policy guidelines; (3) determine the need for setbacks, relocations, and construction standards; and (4) enact legislation that would make it easier for levee districts to form and obtain public sponsorship for participation in the Corps’ levee rehabilitation program. As of June 1995, the Missouri General Assembly had not yet enacted these recommendations into law. Local governments generally exercise more control over local floodplains and levees than do the states or the federal government. This is often because FEMA requires the adoption of community floodplain regulations as a condition for joining its flood insurance program. The 1992 report by the Association of State Floodplain Managers noted that every state has granted its localities enough authority to meet the regulatory requirements of FEMA’s flood insurance program. As a result, localities may enact ordinances that allow them to (1) require certain building codes for development in the floodplain, (2) issue zoning regulations describing the types of land uses permitted in the floodplain, or (3) require that development incorporate improvements to alleviate potential flood hazards. Two exceptions to this local authority exist, however: First, most localities cannot regulate federal or state property or development by other localities, and, second, some states have exempted from local control certain activities that are important to the state’s economy (i.e., transportation, agriculture, mining). Under these exceptions, a local government could be prevented from exercising control over a nonfederal levee if, for example, the state determined that such control would negatively affect the economy of the river’s transportation industry. In a state that regulates levees, the locality is generally required to enforce and comply with the state’s standards for the design, placement, construction, and maintenance of levees. For example, in Iowa, local regulations must meet state requirements. In Minnesota, the state government may directly regulate levees and enforce compliance if the local government does not. In states that do not regulate levees, such as Missouri, local governments are bound by the standards of FEMA’s flood insurance program if the community participates in the program. According to the city engineer in St. Louis, the city complies with FEMA’s standards for floodplain management but has not established local levee regulations. As part of its review of floodplain management in the aftermath of the 1993 flood, the Interagency Floodplain Management Review Committee reported that few states control the design, placement, construction, and maintenance of nonfederal levees. As explained in detail in appendix VI, the Committee found that the states’ involvement in many floodplain management activities—including issuing permits for levees, flood-fighting, and repairing levees—was highly variable and in need of enhancement. The Committee recommended that the states assume responsibility for regulating the location, alignment, design, construction, upgrading, maintenance, and repair of levees and flood-fighting at levees. While no comprehensive federal program regulates the design, placement, construction, or maintenance of nonfederal levees, control over nonfederal levees is exercised through various federal programs for regulating navigable waters and wetlands and for providing flood insurance and disaster and emergency assistance. Not all aspects of levees are regulated in every program, and not all nonfederal levees are affected. Five of the nine states involved in the 1993 flood have regulatory programs that, to varying degrees, affect the design, placement, construction, or maintenance of nonfederal levees. Most often, states are responsible for the overall coordination of floodplain management activities within the state and across state lines, and local governments exercise direct control over levees through land-use regulations. Local governments often impose regulations to comply with the requirements of the National Flood Insurance Program and state regulatory programs. | Pursuant to a congressional request, GAO examined the operations of levees involved in the 1993 Midwest flood, focusing on: (1) whether the Army Corps of Engineers' flood control levees prevented flooding and reduced damage or increased the flooding and added to the damage; and (2) the governments exercise control over the design, construction, placement, and maintenance of nonfederal levees. GAO found that: (1) 157 of the 193 Army Corps of Engineers levees found in the areas affected by the 1993 flood prevented rivers from flooding and $7.4 billion in damages; (2) 32 of the Corps levees withstood floodwaters until the water rose above the levees and overtopped them; (3) 4 of the Corps levees allowed water into protected areas before their design capacity was exceeded; (4) the Corps estimated that the breaching of these levees caused about $450 million in damages; (5) although levees allow floodwater to rise higher than it normally would because they confine a flood to a portion of a floodplain, the Corps believe that its levees have the net effect of reducing flooding; (6) no federal programs specifically regulate the design, placement, construction, or maintenance of nonfederal levees, however, flood insurance and disaster assistance programs may exercise control over certain levees; and (7) 17 states and various local governments have programs to regulate levees, many of which are in response to the requirements of the National Flood Insurance Program. |
Title I grants are intended to help elementary and secondary schools establish and maintain programs that will improve the educational opportunities of low-income and disadvantaged children. Title I funds are intended to provide instruction and instructional support for these disadvantaged children so that they can master challenging curricula and meet state standards in core academic subjects. The law does not stipulate exactly how Title I funds are to be spent. Instead, Title I is an example of flexible funding that local and state educational agencies may use as they deem best. Title I funds are directed toward states and school districts with greater numbers and percentages of poor children regardless of the level of funding they receive from state and local sources. Although the amounts that states and localities spend on education vary due to differing resource bases and funding priorities, Title I funds are not intended to compensate for this variation. Federal Allocation Process
The U.S. Department of Education each year determines the distribution of Title I funds according to the mandates of the law. The authorizing legislation in effect through the 2001-02 school year provided for four different kinds of Title I grants: Basic Grants are the primary vehicle for Title I funding and are the easiest grants for which school districts can qualify. Districts are eligible for basic grants if they have at least 10 poor children and the number of poor children is more than 2 percent of the district’s school-age children.Nationally, about 92 percent of school districts (containing over 99 percent of poor children) receive basic grants, which accounted for about 85 percent of the Title I funds distributed in fiscal year 1999. Concentration Grants are somewhat more directed to poor districts than basic grants because district eligibility criteria for concentration grants are stricter than those for basic grants. Districts are eligible for concentration grants if they have more than 6,500 poor children or the number of poor children is more than 15 percent of the district’s school-age children. Nationally, about 60 percent of school districts (containing about 85 percent of poor children) receive concentration grants, which accounted for about 15 percent of the Title I funds distributed in fiscal year 1999. Targeted Grants were not funded until fiscal year 2003. Targeted grants would be directed more to high-poverty states and districts; as the number and percentage of poor children in the district increase, the targeted grant amount would increase both in absolute dollars and proportionally to other districts. A district would be eligible for targeted grants if it had at least 10 poor children and these children accounted for at least 5 percent of its school-age children. In the 2001-02 school year, about 86 percent of school districts, containing 99 percent of poor children would have been eligible to receive targeted grants had they been funded. Incentive Grants were not funded until fiscal year 2003. Incentive grants would not be distributed on the basis of poverty, but would provide additional funds to states that demonstrate high state spending relative to their tax base and states that have less disparity in funding among their districts. Under this formula, states would distribute funds to districts in proportion to the remainder of their Title I allocations. Title I funds are distributed from the federal government to the states, based on data that are measured at the school district and state levels. Since school year 1999-2000, for each school district meeting eligibility requirements based on numbers and/or percentages of poor children, Education has based its formula calculations on the number of poor children in the district. (Prior to that time, formula calculations were based on counties rather than school districts. The change occurred, in part, due to concern that poor school districts in otherwise wealthy counties were not receiving the Title I funds they needed.) The funding formula for basic and concentration grants principally involves multiplying the number of poor children in a school district area, as measured by Census and other data sources, by the state’s average per-pupil expenditure, although these amounts are subject to hold-harmless provisions and a “small state minimum” provision. These key elements of the funding formula for basic and concentration grants are described below: Poverty: the number of poor children in the school district area, as estimated by decennial Census data and updated by the Census using statistical modeling techniques. (The poverty threshold for the 2000-01 school year for a family of four was an annual income of $17,050.) Expenditures: forty percent of the state’s average per-pupil expenditure, limited to a minimum of 80 percent and a maximum of 120 percent of the national average expenditure. This measure is included as a proxy for education costs. Small state minimum: Each state is guaranteed a minimum level of funding, which is the smaller of either one-quarter of 1 percent of the overall appropriation, or the average of one-quarter of 1 percent of the overall appropriation and the state’s number of eligible students multiplied by 150 percent of the national average per-pupil payment. Hold-harmless provisions: guarantee each state and district a minimum level of funding based on past allocations. That is, the allotment cannot be less than a specified percentage of the preceding year’s allotment. Such provisions are intended to moderate the effects of any large year-to-year shifts in program funding, numbers of eligible children, or state education spending. The hold-harmless provisions protect states with declining numbers of poor children from reductions in their allocations, but in the absence of increased overall funding, this leaves fewer funds available for states and districts with growing numbers of poor children. Specific hold- harmless rules have changed over time. In the authorizing statute, the hold-harmless level for the basic and targeted grants is set at 85, 90, or 95 percent of the prior year’s allocation, depending on the percentage of children in the district who are eligible under the formula (formula children). (The concentration grant does not have a hold-harmless provision in the authorizing statute.) However, for the 1998-99, 1999-2000, and 2000-01 school years, provisions in the appropriations legislation set the hold-harmless level at 100 percent of the past year’s allocation. This 100-percent rule was modified for school year 2001-02. Under this new rule, rather than being guaranteed 100 percent of their prior year’s funding amount, districts were guaranteed the larger of either their previous year’s allotment or the amount they would have received had the authorizing statute’s hold-harmless rules been in effect. The Title I program was reauthorized in January 2002, with some significant changes. See appendix III for details. A state’s allocation is the sum of the district allocations determined by Education. These allocations, however, are not the final amounts that a district will receive. The state must adjust the allocations determined by Education to reserve funds for state administration (up to 1 percent of the amount allocated to the state) and for school improvement activities (no more than 0.5 percent of the amount allocated to the state but no less than $200,000) and account for changes in district boundaries, district consolidations, and the creation or existence of special districts, such as charter schools or regional vocational/technical schools, that are eligible for Title I funds but may not be reflected in Education’s allocations. In the case of special districts that meet the basic and concentration grant eligibility criteria, the state must ensure that those districts receive the Title I funds to which they are entitled. This may involve reducing the allocations of districts from which these special districts draw children. In addition, the statute gives states the flexibility to use alternative poverty data, which Education must approve, to redistribute Education- determined basic and concentration grant allocations and re-determine eligibility for “small” districts serving areas with fewer than 20,000 total residents. This provision arose out of concerns about the quality of Census poverty estimates for small districts. Roughly 79 percent of all school districts nationally have a total resident population of less than 20,000. Currently, nine states use alternative data to redistribute allocations determined by Education among their small districts. Most of these states use free and/or reduced-price lunch data either exclusively or in combination with Census poverty data. Once funds have been allocated to the district level, districts can in turn allocate funds to the schools. Districts have considerable discretion— more so than states—in how they allocate Title I funds. Districts may use Title I funds for district-level activities—including professional development, preschool programs, school improvement initiatives, program administration, and parental involvement efforts. Districts then generally allocate the remaining Title I dollars to the schools. In distributing these dollars, districts are subject to several key restrictions. For example, a district must serve eligible schools or attendance areas in rank order according to their poverty percentage. A district must serve those areas or schools above 75 percent poverty, including any middle or high schools, before it serves any with a poverty percentage below 75 percent. Once all of the schools and areas with a percentage above 75 percent have been served, the district may serve lower-poverty areas and schools either by continuing with the districtwide ranking or by ranking its schools below 75 percent poverty according to grade-span groupings (i.e., K-6, 7-9, 10-12). If a district ranks by grade-span, it can compare the school’s poverty percentage to either the districtwide poverty average or the poverty average for the respective grade-span grouping. Districts are not required to allocate the same per-pupil amount to each school, but if they choose not to, they must allocate higher per-pupil amounts to poorer schools than they allocate to schools with lower concentrations of poverty. In addition, districts may apply for and receive waivers of any of these allocation rules. The Title I statute also requires that school districts provide eligible private school children with Title I educational services or other benefits that are “equitable” to those provided to eligible public school children. The school district provides these services directly to the private school children rather than giving funds to the private school itself. Within the rules, districts may allocate funds to schools as they like. School districts differ in the size of their enrollment and their percentages of poor children, as shown in figure 1. For example, there are about 7,000 school districts with no more than 1,000 children enrolled and about 300 districts with more than 20,000 children enrolled. Among districts with less than 1,000 children, about half of those districts have more than 35 percent of their children eligible for free or reduced-price lunch, while the other half have fewer children eligible for free or reduced-price lunch. On the whole, total Title I grants were allocated to states and school districts on the basis of their numbers of children from low-income families in the 1999-2000 school year, but individual states and school districts received different grant amounts for each poor child. Generally, a state’s share of poor children was roughly proportional to its share of funds; however, even small differences between the two resulted in substantial differences in funding per poor child. School districts, like states, as a whole received Title I allocations that were in accordance with the numbers of poor children they had enrolled, but actual funding per poor child varied among individual school districts with similar numbers of poor children. The pattern of Title I grant distributions in terms of numbers of poor children among urban and rural school districts is to some extent dependent upon the measure of poverty that is used. When school districts allocated funds to individual schools, they used the flexibility of the Title I program to distribute funds where they believed the need was greatest. Even with this flexibility, most school districts allocated the majority of funds to schools in which at least half of the children were classified as poor. Generally, states with higher numbers of poor children received higher amounts of Title I basic grant funds, and states with lower numbers of poor children received lower amounts of Title I basic grant funds.However, even small differences between a state’s share of poor children and its share of funds resulted in substantial differences in funding per poor child. For example, in the 1999-2000 school year, Texas had about 10 percent of the nation’s Census poor children and received about 9 percent of the Title I basic grant dollars while Minnesota had about 1 percent of the poor children and 1.2 percent of the dollars. However, Texas’s share of the basic grants was about 14 percent less than its share of the poor children while Minnesota’s share of the basic grants was about 16 percent more than its share of the children. As a result, Texas’s basic grants amounted to $581 per poor child while Minnesota’s basic grants amounted to $793 per poor child. Table 1 displays for each state and the District of Columbia its share of Census poor children, share of basic grant dollars, the percentage difference between these shares, and the resulting basic grant amount per poor child. The shares of basic grant dollars that states received differed from their shares of poor children due to the factors in addition to the number of poor children that are included in the federal funding formula: state per- pupil expenditures, the small state minimum provision and the hold- harmless provision. For example, Vermont, Wyoming, New Hampshire, and Alaska, which each received much larger basic grants per poor child than the rest of the states, are each also among the small-population states that benefited from the small state minimum provision in the 1999-2000 school year. Another of the formula factors — the state per-pupil expenditures factor — also had an important impact on the distribution of Title I dollars to states. States with higher percentages of poor children would be expected, all else being equal, to have lower tax bases and tend to have lower state education expenditures. As a result, states with higher percentages of poor children tended to receive smaller basic grant amounts per poor child because Title I allocations are based, in part, on the amount a state spends on education. We found that of the 17 states with more than 20 percent of their children living in poverty in 1997, all but 4 had basic grant dollars per poor child below the national average in the 1999-2000 school year. An examination of the allocations that states reported making to their school districts in the 1999-2000 school year shows that school districts, like states, as a whole received Title I allocations that were in accordance with the numbers of poor children they had enrolled, but that actual funding per poor child varied considerably among individual school districts with similar numbers of poor children. For example, among districts with between 101 and 250 children eligible for free or reduced- price lunch, the median basic grant funding per poor child was $370, but ranged from $0 to $2,573. Concentration grants among school districts with similar numbers of poor children also varied. For example, among districts receiving concentration grants with between 101 and 250 children eligible for free or reduced-price lunch, the median concentration grant funding per poor child was $63, but ranged from $1 to $3,547. Figures 2 and 3 illustrate these findings. The variation illustrated for the national level in figures 2 and 3 is similar to the type of variation that occurs within states as well. In other words, within states, individual districts with similar numbers of children received different allocation amounts. Not only did districts with similar numbers of poor children receive different allocations per poor child, but also in some cases, school districts with larger numbers of poor children received less funding per poor child than districts with smaller numbers of poor children. For example, in California, one school district with 961 poor children received a basic grant of $363 per poor child, while a school district with 13 poor children received $1,656 per poor child.Concentration grants for these two districts varied similarly and similar examples of variation occurred in most states. Differences in Title I funding per poor child among school districts is due, in part, to hold-harmless provisions. The pattern of Title I grant distributions in terms of numbers of poor children among urban and rural school districts in the 1999-2000 school year is to some extent dependent upon the measure of poverty that is used, as shown in table 2. When Census data are used as the measure of poverty, both urban and rural school districts as a whole received shares of both basic and concentration grants roughly proportional to their share of poor children, with some small differences. However, when eligibility for free or reduced-price lunch is used as the measure of poverty, urban districts received shares of both basic and concentration grants that were greater than their share of poor children. Even small differences between shares of grants and shares of poor children resulted in per child dollar differences between urban and rural districts. For example, when Census data are used as the measure of poverty, rural districts overall had a higher median basic grant amount per poor child ($705) than did urban districts ($610). This resulted, in part, because the share of basic grants that rural districts received was 1 percentage point greater than their share of poor children while the share of basic grants that urban districts received was equivalent to their share of children. Because the pattern of Title I grant distributions among urban and rural school districts is dependent on the measure of poverty used, it is worth noting that currently all states base their allocations on Census data for areas with populations greater than 20,000 residents, and the overwhelming majority of states (41) base their allocations on Census data rather than subsidized lunch for areas with populations smaller than 20,000 residents. Title I allocations to charter schools that are independent school districts are lower than per-pupil allocations to other similarly sized school districts. Charter schools are not included in Education’s Title I formula calculations, but are guaranteed funding on an equal basis with other school districts. Consequently, those states with independent charter schools must reallocate resources from other school districts to these schools. In the 1999-2000 school year, 14 states and the District of Columbia had operating charter schools that were considered independent school districts; 533 of these 734 charter schools received basic grants. In these states combined, charter school districts that received Title I grants had almost 6 percent of the children receiving free or reduced-price lunch and received less than 5 percent of the basic grant funding that was distributed to districts with fewer than 2,500 students. Average basic grant funding per poor child in the charter school districts was $365 and was $481 in the other school districts. When school districts allocate funds to individual schools, they use the flexibility of the Title I program to distribute funds where they believe the need is greatest and in ways that they believe best provides needed activities efficiently and consistently. Unlike school district allocations, which are based on numbers of poor children, allocations to individual schools are required by law to be based on the percentages of poor children within grade span or within the district as a whole. From our national survey of school districts, we estimate that a majority of school districts (63 percent) prioritized Title I funding to eligible primary or elementary grades before funding other grade spans, while an estimated 23 percent of school districts prioritized funds to schools with higher percentages of poor children, regardless of grade span. (See fig. 4.) A similar pattern was found among rural districts, of which an estimated 67 percent targeted funds to needy primary schools. In contrast, equal percentages of urban school districts targeted funds to primary schools (an estimated 42 percent) and to schools with higher percentages of poor children (an estimated 42 percent). In addition to distributing funds to individual schools, we estimate that over half the districts (58 percent) reserve at least some of their Title I grant, typically no more than 10 percent, for administration or other district activities. An estimated 61 percent of districts that reserve funds for such purposes do so because they believe it is more efficient or promotes consistency across schools in the district. Districts use the reserved funds for such activities as professional development for teachers, parental involvement programs, preschool and summer school programs, and before and after school enrichment activities. In addition to the flexibility districts are allowed under the Title I regulations, a district may apply for a waiver from the regulations if a district finds that it cannot use Title I funds to best serve the needs of its schools within the existing regulations. Waivers are used most commonly to allow districts to (1) serve schools of the same grade span without regard for their rank order in terms of poverty or (2) allow lower-poverty schools to use Title I funds in schoolwide programs rather than for specific students, a practice otherwise allowed only in schools with greater than 50 percent of their children living in poverty. Because districts have flexibility in making Title I allocations to schools, and not all schools with poor children receive funds from their districts, the amount of Title I funds that individual schools receive per poor child differs among schools. Only three states, California, Mississippi, and Georgia, were able to provide us with electronic information on enrollment, free and reduced-price lunch eligibility, and Title I dollar amounts allocated to individual schools. Analyses of these three states’ data show that in the 1999-2000 school year, among the schools receiving Title I funds, schools with higher percentages of poor children received a larger percentage of the funds than schools with lower percentages of poor children, as shown in table 3. Although these findings are based on one year of data from only three states, they are consistent with Education’s findings that in the 1997-98 school year, nationally 66 percent of schools receiving Title I funds had more than 50 percent of their children eligible for free or reduced-price lunch, and 35 percent of schools receiving Title I funds had at least 75 percent of their children eligible for free or reduced-price lunch. When the number of poor children from low-income families shifts between states, Title I allocations do not fully adjust in response. More frequent updates of Census poverty data over the past decade have provided Education the data it needs to adjust Title I formula calculations to shifts in poverty more quickly than it could in the past. However, recent appropriations hold-harmless provisions and the small state minimum provision in the formula have limited the extent to which Title I allocations can shift at all, even when Education has data indicating that shifts in poverty have occurred. Even if these rules were changed, allowing formula calculations to be based more completely on poverty, the remaining lags in Census poverty data would continue to prevent Title I formula calculations from fully adjusting to shifts in poverty. To the extent that the number of poor children among the states has shifted, Title I funding has not completely adjusted in response. For example, in 1980, California had 9 percent of the nation’s poor children and received 10 percent of all Title I dollars. By 1997, California had 16 percent of the nation’s poor children, but received just 12 percent of all Title I dollars. This type of disparity has occurred in a number of states. In 1997, 33 states received Title I allocations that differed from their shares of poor children by more than 10 percent. A lack of updated poverty data contributed to the mismatch between poverty and funding over the last two decades. Education relied on 1980 decennial Census data to make allocations from 1984 to 1993, at which point Education began using 1990 decennial Census data to adjust allocations. As shown in table 4, since 1990, the lag in poverty data has decreased. This decrease occurred, in part, because in 1994, the Congress authorized the Census Bureau to update the data more frequently. Although the Census updates provided Education with more timely poverty estimates, the hold-harmless and small state minimum provisions limited the extent to which Title I formula allocations could change in response. These provisions limit the extent to which Title I grants are reduced for districts and small-population states with declining numbers of poor children. As a result, funds are not sufficient to provide “full” formula amounts to districts and states with increasing numbers of poor children. Because of the effects of the hold-harmless and small state minimum provisions, between school years 2000-01 and 2001-02, holding other things constant, most states with increased numbers of poor children would have received a decreased Title I allocation per poor child, while most states with decreased numbers of poor children would have received an increased Title I allocation per poor child. States with increased numbers of poor children would have lost an average of $25 per poor child under the basic grant and $10 per poor child under the concentration grant. States with decreasing numbers of poor children between these 2 years would have gained $23 per poor child under the basic grant and $4 per child under the concentration grant. While the hold-harmless and, to a lesser extent, small state minimum provisions create imbalances between poverty and funding, in the absence of these provisions, unavoidable lags in poverty data would prevent Education from fully adjusting Title I allocations to shifts in poverty. Title I allocations do not encourage states to target their own funds to children from low-income families. Our review of the Title I statute and regulations found no formal monetary, statutory, or regulatory incentives for states to target their funds in this way. In our interviews with Title I directors in each of the 50 states, only four reported that they have programs for disadvantaged children that model their eligibility criteria on the Title I program. Furthermore, our study based on 1991-92 school year data found large differences in the extent to which state funding was targeted to school districts on the basis of poverty criteria, indicating that states were not systematically following the allocation model of the Title I program. The incentive grant, if funded, could provide an incentive for states to spend more of their own dollars. However, the grant would not encourage states to target their own funds to children from low-income families. In addition, the amount of money that could be provided through an incentive grant is not likely to be sufficient to create changes in states’ behaviors. A number of policy options could affect the extent to which Title I funds are allocated to states and school districts with greater numbers and percentages of children from low-income families. First, using a less restrictive hold-harmless provision would reduce the variation in funding among school districts with similar numbers and similar percentages of poor children and allocate more funding to states with more rapidly growing numbers of poor children. Second, funding targeted grants and raising the basic grant eligibility thresholds would each shift funding toward districts with higher percentages of poor children and away from districts with lower percentages of poor children. Third, replacing the measure of state per-pupil expenditures with an alternative cost indicator would also have the effect of shifting funding to districts with higher percentages of poor children and reducing the variation in funding among districts with similar percentages of poor children. The effect of such changes would depend on the appropriations provisions. The effect of the changes would also depend on adjustments to the formula-calculated amounts made by state officials. Each policy option involves trade-offs between the competing goals of providing similar funding amounts to districts with similar numbers and percentages of poor children and providing stable funding to districts with rapidly declining numbers and percentages of poor children. Without increases in total funding, each change also would increase funds for some districts while decreasing funds for others. Different hold-harmless rules affect the extent to which Title I funding is allocated on the basis of numbers of poor children. The less restrictive the hold-harmless rules are, all else equal, the more Title I funding would be allocated on the basis of numbers of poor children. We considered four possible hold-harmless rules. In order from most to least restrictive, they were a 100-percent hold-harmless rule (districts receive 100 percent of their previous year’s allocation); the school year 2001-02 hold-harmless rule (districts guaranteed the larger of either their previous year’s allocation or the amount they would have received had the authorizing statute’s hold-harmless rule been in effect); the authorizing statute’s hold-harmless rule (districts receive 85, 90, or 95 percent of the previous year’s allocation, depending on the percentage of children in the district who are eligible under the Title I formula); and no hold-harmless rule (districts receive allocations based on current application of the Title I formula with no regard to the previous year’s allocation). Using a less restrictive hold-harmless rule would not noticeably redistribute funding between school districts with large and small numbers of poor (formula) children. However, less restrictive hold- harmless rules would substantially reduce the funding variation among the smallest districts with similar numbers of poor children. Disparities in funding are greatest under the 100-percent hold-harmless rule, less under the 2001-02 rule, further reduced under the authorizing statute and would be the least if there were no hold-harmless, as shown in table 5 for both small and large districts, in terms of their numbers of formula children. Using a less restrictive hold-harmless rule would also have the effect of reducing variation among districts with similar percentages of poor children. (See fig. 5.) Using a less restrictive hold-harmless provision distributes more funds to high-growth school districts and accordingly to high-growth states. The effect of using a less restrictive hold-harmless rule would be to increase the responsiveness of Title I funding to the growth in numbers of poor children in states. In figure 6, states are rank-ordered based on the growth in the number of formula-eligible children between school years 2000-01 and 2001-02. Figure 6 shows the percent change in funding that would have resulted had no hold-harmless rule been in effect. Under the no hold- harmless scenario, 12 of the 15 states with the highest growth in low- income children would have received more funding and 13 of the 15 slowest growth states would have received less.Appendix II shows the data on which figure 6 is based. Funding targeted grants instead of concentration grants would provide noticeably more Title I funds to districts with both higher numbers and percentages of poor children and reduce funding for districts with lower numbers and percentages of poor children. Districts are eligible for concentration grants if they have more than 6,500 poor children or the number of poor children is more than 15 percent of the district’s school- age children. Concentration grants are allocated to eligible districts based on their numbers of poor children. In contrast, districts would be eligible for targeted grants if they had at least 10 poor children and these children accounted for at least 5 percent of their school-age children. As the number and percentage of poor children in the district increase, the targeted grant amount would increase both in absolute dollars and proportionally to other districts In the 2001-02 school year, districts with the highest percentages of poor children received $864 per poor child compared with $758 per poor child in districts with the lowest percentages of poor children. If targeted grants had been funded instead of concentration grants, the funding for districts with the highest percentages of poor children would have increased by 5.5 percent, to $912 per poor child, while the funding for districts with the lowest percentages of poor children would have decreased by 5.7 percent, to $714, as shown in table 6. Table 6 also shows that funding targeted grants would have a similar effect on districts with larger and smaller numbers of poor children. The per-pupil expenditure factor was originally included in the Title I formulas to take into account cross-state differences in the cost of providing education services. While per-pupil expenditures reflect the cost of providing education services to some extent, expenditures are also explained by other factors not related to costs. For example, states with high-income taxpayers may spend more on education than those whose taxpayers have lower incomes. In addition, spending differences may result from differences in the “willingness” of a state’s taxpayers to fund public education. One alternative cost measure is a geographical cost-of-education index developed for the Department of Education’s Office of Educational Research and Improvement. The purpose of this experimental cost index is to make cost comparisons based on the cost of teachers, non-teaching school personnel, and other factors that may affect costs, but which are beyond the ability of local officials to control. The index includes cost factors for both states and school districts, unlike earlier experimental measures that had only cross-state cost factors. We use these cost factors for illustrative purposes only and do not necessarily endorse any particular measure. The intent of replacing state per-pupil expenditures with either a state or district-level cost-of-education factor is to more accurately reflect educational costs; however, as a by-product, doing so would shift funding somewhat toward districts with higher percentages of poor children. If the state per-pupil expenditure factor had been replaced with a cost factor in the 2001-02 school year, districts with the highest percentages of poor children would have seen an increase in funding of approximately 3 percent, while districts with the lowest percentages of poor children would have seen a decrease in funding of 2 to 3 percent (see table 7). If in addition, the authorizing statute’s hold-harmless rules had been adopted, funding to districts with the highest percentages of poor children would have increased by about 5 percent while funding in districts with the lowest percentages of poor children would have decreased by about 5 percent. The more substantial effect of replacing the state per-pupil expenditure factor, however, would be to reduce the variation in funding among districts with higher percentages of poor children (see figure 7). With the educational cost factor, the variation in funding among districts with higher percentages of poor children would be reduced from 19 percent to 13 percent, a reduction of 30 percent. In contrast, the variation in funding among districts with low percentages of poor children would be largely unaffected. However, if the current hold-harmless rules were also replaced with the authorizing statutes rules, very substantial reductions in funding disparities would result among all school districts. Funding variations would be cut by more than half between the highest and lowest poverty districts. Title I grants have sometimes been criticized because the poverty threshold for basic grant eligibility is so low and that nearly all districts can participate in the program. It is often noted that by funding nearly all districts, less funding is available for high-poverty districts. One policy option is to raise the basic grant eligibility threshold, making fewer districts eligible. With fewer districts eligible, the remaining districts would receive more funds per poor child, if total funding were to remain constant. Table 8 shows how increasing the current 2 percent poverty threshold to poverty thresholds of 5 and 10 percent, respectively, would have this effect. For example, a 10 percent threshold would result in 26 percent of all districts, which contain 7.7 percent of all formula children, becoming ineligible. This would allow funding per child in the remaining districts to increase by 8.3 percent or $57. The effects of these thresholds are shown on a state-by-state basis in table 9. Changes to the Title I allocation formulas will change the amount of funds states receive and also would be expected to result in changes in the amounts districts receive; however, there are limits on how precisely changes in the formula can be expected to affect school districts because states alter the formula-calculated amounts. In the aggregate, relatively few poor children and Title I funds were associated with districts whose allocations differed widely from their formula-calculated amounts in the 1999-2000 school year. As a result, state adjustments did not appear to alter the overall extent to which available funding was allocated on the basis of the number of poor children. However, for some individual states and school districts, state adjustments were substantial. When allocating the funds they receive from the federal government, states adjust for changes in school district boundaries and the creation of charter schools. In our work, we found that among the school districts operating in the 1999-2000 school year, there were more than 900 school districts, containing about 126,000 children eligible for free or reduced- price lunch and receiving about $79 million in Title I funds, that were not included in Education’s formula calculations. In addition, states alter the formula-calculated amounts to adjust for the poverty measure used for school districts in small areas, as well as to fund statewide activities and program administration. States are allowed to withhold up to 1.5 percent of their Title I funds for statewide activities and program administration, so this much variation from the formula-calculated allocations is expected. Considering only the districts that were included in Education’s calculations, we found that during the 1999-2000 school year 49 percent of the districts received total Title I grants that differed by more than 1.5 percent from the formula-calculated allocation; 16 percent of districts’ allocations differed by more than 10 percent. Among the districts included in Education’s calculations, over half of the Title I funds were allocated to districts whose allocation amounts differed from their formula calculations by no more than 1.5 percent. These districts also contained just over half of the poor children. Only about 5 percent of the funds were allocated to districts whose actual allocations differed from their formula calculations by more than 10 percent. These districts also contained about 5 percent of the poor children. (See table 10.) The variation between actual allocations and the formula calculations is greater in some states than in others. There are some states, for example, Georgia, Louisiana, and South Dakota, where actual allocations to school districts are very close to the formula calculations. (See table 11.) However, these states are the exception, as most states, including Delaware, North Dakota, and Maine, had many districts receiving very different allocations than the formula calculations. (See table 12.) However, states do not appear to alter the overall extent to which available funding is allocated on the basis of the number of poor children. Table 13 shows for each state the percentage of its districts whose total Title I grants differed from their formula calculations by less than 1.5 percent, between 1.5 and 10 percent, and by more than 10 percent. Although Title I funding generally reflects the distribution of poor children, there are many instances of states, districts, and schools with either similar numbers or similar percentages of poor children receiving widely differing amounts of funding per poor child. These differences result, in part, from formula provisions that attempt to balance several, sometimes competing, goals. These goals include allocating funds based on the distribution of poor children, ensuring that states and districts are provided funding stability even in light of declining numbers of poor children, and addressing differences across school districts and states in the costs of providing educational services. Choosing among the policy options discussed in this report will entail, in part, weighing the goal of increased targeting with other goals. Enacting any of the policy options— using less restrictive hold-harmless provisions, funding targeted grants, using an alternative cost factor, or raising the eligibility threshold—would result in changes for many states and school districts in terms of their formula calculations. In addition, under any of these policy options, states and school districts would still have flexibility in making allocation decisions—flexibility that allows states and school districts to use these funds in a manner that they believe best meets the needs of disadvantaged children. In written comments on our draft report, the Department of Education generally agreed with the findings presented in the report. Education suggested that our report be updated to reflect the passage of the “No Child Left Behind Act of 2001” and the fiscal 2002 appropriations act. Appendix III describes the impact of this legislation on the aspects of Title I discussed in the report. Education’s written comments are printed in appendix IV. In written comments on our draft report, the U.S. Department of Agriculture (USDA) said that the Food and Nutrition Service (FNS) has become increasingly aware of the limitations of free and reduced-price lunch data as a measure of low-income status, which could have implications for the targeting of Title I funds. As described in the report, we recognize that there are limitations of these data as a measure of poverty. Despite these limitations, however, we chose to use subsidized lunch data as one of our measures of poverty for several reasons. We used these data as a poverty measure at the school level because the Department of Education has found these data to be the best available source of poverty data at the school level. We used these data as a poverty measure at the school district level because subsidized lunch data are available at the school, district, and state levels, and thus provide a consistent measure across all three levels. Also, subsidized lunch data are available for nearly all school districts, including charter schools, whereas Census poverty estimates are available only for the somewhat limited number of school districts included in Education’s database. While recognizing the limitations of subsidized lunch data, we believe the use of it, along with Census poverty estimates, strengthens our report findings. USDA’s written comments are printed in appendix V. We are sending copies of this report to the Secretaries of Education and Agriculture and interested congressional committees. We will also make copies available to others upon request. If you have any questions concerning this report, please contact me on (202) 512-7215. Other GAO contacts and staff acknowledgments are listed in appendix VI. As mandated by the Congress (Public Law 106-554 Sec. 305), we designed our study to provide information on (1) the extent to which Title I funds are allocated to states, school districts and schools with the greatest numbers and percentages of school-age children from low-income families; (2) the extent to which allocations of such funds adjust to shifts in numbers of children from low-income families; (3) the extent to which the allocation of Title I funds encourages the targeting of state funds to school-age children from low-income families; and (4) what options might improve targeting of funds, especially to states and school districts with higher numbers and percentages of poor children, to more effectively serve those children. To determine the extent to which Title I funds are targeted to poor children, we used two measures of poverty and two types of allocation data. We used the Census Bureau’s updated decennial poverty data for one measure. Census poverty data are used by the U.S. Department of Education to calculate Title I formula allocation amounts. For the second measure of poverty, we used eligibility for free or reduced- price lunches through the National School Lunch Program, a federal food assistance program administered by the U.S. Department of Agriculture for children from low-income families. The subsidized lunch program provides the best source of data on low-income students at the school- level, according to the Department of Education, and these data are also available for the district and state levels. We determined how states actually allocated Title I funds to each of their school districts by collecting 1999-2000 allocation information directly from state Title I officials in every state and the District of Columbia. In this report, we refer to these data as “actual allocations” or simply “allocations.” We compared the actual allocations with the amounts generated by the Title I formula calculations for the 1999-2000 school year, which we obtained from the Department of Education. In this report, we refer to these data as “formula calculations.” Formula calculations are the data typically used in analyses of the Title I program. The formula calculations accurately reflect the amount of funds allocated to each state, but these data do not reflect changes that states subsequently make to the formula-calculated amounts when allocating the funds to their school districts. We also interviewed state Title I directors in each of the 50 states and the District of Columbia about their experiences and perceptions of the Title I program. We examined school district policies for allocating Title I funds to schools by surveying a nationally representative, stratified sample of school districts. In addition, we reviewed school-level allocation data from the few states that were able to provide it. We examined the responsiveness of state-level Title I allocations to shifts in poverty by analyzing Census data and Title I allocation data from the Department of Education for the period of 1980 to the present. We analyzed the relevant statutory provisions and reviewed our previous reports to identify incentives for states to target their own funds. Finally, we determined the consequences of various policy options by examining the Title I formulas and running simulations of Education’s formula calculation process for states and school districts for the 2001-02 school year. We conducted our work from December 2000 to December 2001 in accordance with generally accepted government auditing standards. Federal Funding Formulas From the Title I statute, we obtained the formulas that Education is required to use to calculate Title I grant amounts. We met with Education officials to discuss their procedures for using the formulas and data to calculate grant amounts. We used these procedures to replicate Education’s formula calculations for the 1999-2000 and 2000-01 school years and as the basis for computer simulations of various changes that could be made to the formulas. From the Department of Education, we obtained the grant amounts calculated for school districts for the 1999-2000, 2000-01, and 2001-02 school years, which Education generated using the federal funding formulas included in the Title I statute. The school districts for which Education calculated grant amounts in the 1999-2000, 2000-01 school years were those known to Education in the 1995-96 school year. The school districts for which Education calculated grant amounts in the 2001-02 school year were those known to Education in the 1997-98 school year. At no time have the calculations included charter schools. The data set from Education also included 1995 Census data on the characteristics of these school districts, such as numbers of school-age poor children, total numbers of school-age children, and total resident populations, which Education used in calculating grant amounts. The formula calculations were used to examine their relationship to poverty and other characteristics of school districts and to compare the formula calculations to the actual allocations that school districts received. We collected data from state Title I program directors on the dollar amounts of Title I funds, if any, that they disbursed to each of their school districts in the 1999-2000 school year. We collected allocation data on basic and concentration grants to 14,682 school districts in all 50 states and the District of Columbia, including data on charter schools that are independent school districts, as they existed in the 1999-2000 school year. Where possible, we matched the school district data provided by the states with the school district data provided by Education. Where the lists of school districts differed, we called state officials to verify the accuracy of their data. In most cases, state officials clarified that districts had been created, consolidated, eliminated, or had changed names since Education’s data were updated in 1995. For those school districts that were identified both by state officials and Education, we compared the actual allocations with Education’s formula calculations and examined the relationships between the actual allocations and Census poverty and other school district characteristics. We also used the actual allocation data to examine the relationship between the actual allocations and poverty, as measured by eligibility for free or reduced-price lunches. From state Title I directors, we requested data on the dollar amount of Title I funds that each of their states’ schools received in the 1999-2000 school year, if these data were available in an electronic format. From state food services officials, we also requested for each school, electronic data on enrollment and the number of students receiving free or reduced- price lunches. Only three states (California, Georgia, and Mississippi) could provide us with school-level allocations, enrollment, and school lunch data in an electronic format for each of their schools. For each school in these three states, we matched the allocation data to the enrollment and school lunch data and calculated both the percentages of children eligible for free or reduced-price lunch and the amount of funds received per poor child. Because we obtained data on every school within these three states, there is no estimation or sampling error associated with our results. However, our findings based on these data are not generalizable beyond the state or school year for which the data were collected. To estimate the numbers and percentages of children in poverty in every school district, we obtained data from state school food service officials on both the numbers of children receiving free or reduced-price lunches through the National School Lunch Program (NLSP) and the total number of students enrolled in each school district in the 1999-2000 school year. We also obtained these data at the school level, where available. Children from families with incomes at or below 130 percent of the poverty level are eligible for free meals through NSLP; those with incomes between 130 and 185 percent of the poverty level are eligible for reduced-price meals. We requested NSLP participation data for every school district in every state, including charter schools that are independent school districts, where available. These data were combined with school district-level data on Title I allocations in order to calculate the amount of Title I allocations per poor child received in each district. We chose participation in the NSLP as a measure of school district poverty because it is the measure used most commonly by school districts to determine allocations to schools and is the best source of poverty data that is available at the state, district, and school levels. Because participation in the NSLP is voluntary, there is some concern that participation rates may reflect, in part, the effort schools make to encourage participation, and may not consistently reflect actual program eligibility rates across schools and school districts. There is also concern that high school students are less likely to participate in the program than younger students due to the associated stigma. Nonetheless, a National Research Council panel concluded that NSLP participation is an indicator of low family income and that the quality of NSLP data are neither appreciably better nor worse than Census data for measuring poverty, especially for areas as small as school districts. To obtain information on the number and types of waivers granted to districts under the Elementary and Secondary Education Act, we reviewed Education’s annual reports to the Congress for 1998-2001 and met with program officials. To obtain information on the number and types of waivers granted by “Ed-Flex” states under the 1994 and 1996 Ed-Flex Demonstration Project and the Ed-Flex Partnership Act of 1999, we met with program officials and reviewed states’ Ed-Flex applications on file with Education. We used this information to determine the most common types of waivers that were granted overall. To obtain information on both states’ roles in the Title I allocation process and the opinions of state Title I directors, we conducted telephone interviews with the directors of the Title I program in every state and the District of Columbia between December 2000 and May 2001, using a semi- structured interview protocol. We asked the directors to explain exactly how they generate dollar allocations to school districts once they receive the information from the Department of Education, including how they apportion funds to small size districts and districts whose boundaries have changed, how charter schools are handled in the allocation process, what data they use, and how recent those data are. Finally, we asked whether their states had compensatory education programs that target funding to high-poverty schools and districts and, if so, how allocations for that program were related to the allocation of Title I dollars. Survey of School Districts
We surveyed a stratified nationally representative sample of school district administrators drawn from the approximately 13,000 school districts nationwide for which Education had calculated an initial Title I allocation amount for the 1999-2000 school year. In addition to providing information on their school districts’ schools and communities, survey respondents provided information on how they measure poverty in their schools, their priorities and rationales in distributing funds, and their use of funds for district-level activities. The sample was stratified into four categories according to the number of school-age children living in the school district boundaries, as follows: 2- 500 children; 501-2,500 children; 2501-50,000 children; over 50,000 children. A random sample was drawn from each of the first three strata; all of the 96 school districts with greater than 50,000 children were included in the survey. Table 14 provides information on the total numbers of students and districts, the number of districts sampled, and the response rate for each of the strata. This sample design allows us to generalize our results to all school districts of similar sizes, including the very smallest school districts. The survey was conducted between July and October 2001 and reflects school district decisions in the 2000-01 school year. Because our estimates are based on samples, they are subject to sampling error. Table 15 shows each of our estimates and indicates the extent of each estimate’s sampling error by showing the 95-percent confidence interval around that estimate. There is a 95-percent chance that the actual total falls within the interval. During our work, we consulted with representatives from the following agencies and organizations who have knowledge of the Title I program and related issues: U.S. Department of Education, U.S. Department of Agriculture, National Research Council, U.S. Census Bureau’s Small-Area Income and Poverty Estimates Panel, Congressional Research Service, Council of Chief State School Officers, Center on Education Policy, American Association of School Administrators, Council of the Great City Schools, and the National Association of State Title I Directors. Local school districts were put into one of five groups with each group containing an equal number of poor children. The groups ranged from the lowest percentages of poverty to the highest percentages of poverty. Each group represents approximately 20 percent of all formula eligible children. For this analysis, we expressed the number of formula-eligible children as a percentage of the number of children ages 5-17. Table 16 shows the average funding per child allotted by formula, formula allotments under each of the simulations, and percent differences in funding per child compared to 2001 formula allotments. Table 17 shows the coefficients of variation for figure 5. Table 18 shows the data used to construct figure 6. We analyzed three formula scenarios that replaced the state per-pupil spending factor with an alternative cost factor developed by the Department of Education: a state-level cost factor, a district-level cost factor, and a district-level cost factor combined with the hold-harmless rules described in the authorizing statute. Table 19 reports the average funding per child in each poverty group and table 20 reports the coefficients of variation in funding per child within each group that were reported in figure 7 of the report. On January 8, 2002, President Bush signed into law the “No Child Left Behind Act of 2001,” reauthorizing Title I and other Elementary and Secondary Education Act programs, with some significant changes. Two days later, he signed the related appropriations law. The changes to Title I relevant to this report are outlined below: Funding for Targeted Grant Formula: For the first time, the Congress appropriated funds for targeted grants. The 2001 Act requires that the amounts allocated through basic and concentration grants are to be the same as they were in fiscal year 2001 and that any additional funds remaining (i.e., any new funds) are to be allocated through the targeted grant formula. As under prior law, a tiered weighting system would provide proportionately greater funding per poor child to districts with higher numbers and percentages of poor children. The new law changed the cut-points between the tiers slightly, based on updated Census poverty estimates, so that each tier would continue to contain roughly equal numbers of poor children. Funding for Revised Finance Incentive Grant Formula: Not only was the finance incentive grant funded for the first time, but the grant formula and other provisions also were significantly revised. In prior law, the incentive grant formula was designed to provide additional funds to states that demonstrated high state education spending relative to their tax base and states that had less disparity in funding among districts. The new law maintains these provisions and adds several more that give proportionately more funds to states and districts with higher numbers and percentages of poor children, as follows: Allocations will be based on each state’s number of poor children, rather than its total school-age population. Districts are required to have at least 10 poor children, making up at least 5 percent of enrollment to qualify for finance incentive funds, whereas there had been no such enrollment requirement in prior law. Allocations will be made to school districts on the basis of a tiered weighting system, like that in the targeted grant formula. The incentive grant weighting system provides proportionally more funds not only to districts with greater numbers and percentages of poor children but also to districts in states with less funding disparity among districts. Districts are newly required to allocate finance incentive funds to schools in the same way that they allocate the other Title I funds (e.g., in rank order of poverty) and to use finance incentive funds only for Title I purposes. In addition, the new formula includes a per-pupil expenditure factor, like that for the other grants, but more narrowly limited to a minimum of 34 percent and a maximum of 46 percent of the national average per pupil expenditure, rather than the 32 percent minimum and 48 percent maximum in the other grant formulas. Increased Overall Funding for Fiscal Year 2002: The education appropriations legislation includes a combined increase of nearly $1.8 billion in funding for Title I basic, concentration, targeted, and finance incentive grants. (See table 21.) Hold-Harmless: The prior authorizing legislation included a hold- harmless provision only for basic grants (districts were guaranteed 85, 90, or 95 percent of the previous year’s funding, depending on percentage of poor children in the district). However, as described in the body of the report, appropriations language in recent years created more restrictive hold-harmless provisions, including (1) a 100-percent hold-harmless provision for basic grants and (2) a hold-harmless provision for concentration grants that allowed even districts no longer meeting the concentration grant eligibility criteria to continue receiving concentration grants. In contrast, the fiscal year 2002 appropriations language does not include any provisions that override the authorized hold-harmless provisions. However, under the new authorizing legislation, hold-harmless provisions will apply not only to basic grants but also to concentration grants and targeted grants. In addition, under the new authorizing legislation, districts that become ineligible for concentration grants will continue to receive concentration grant allocations for up to 4 consecutive years. As a result of these changes, operative hold-harmless provisions for basic grants are somewhat less restrictive than under prior law and, therefore, will allow basic grant allocations to be more reflective of the number of poor children in a school district. Likewise, the newly authorized hold- harmless provisions for concentration grants are somewhat less restrictive than the hold-harmless provisions previously included in appropriations law. In the end, allocations under these grants will be more reflective of numbers of poor children than in the past, but not as reflective of them as would have been the case under the prior authorizing legislation alone. Small State Minimum: The minimum level of funding guaranteed to each state is increased over prior law. Previously, each state was guaranteed the smaller of 0.25 percent of total appropriations for that year, or the average of that amount and the state’s number of eligible students multiplied by 150 percent of the national average per-pupil payment. The new law uses essentially the same calculation but instead of 0.25 percent of the total appropriations, the new calculation will use 0.25 percent of the amount appropriated in 2001, plus 0.35 percent of any subsequent increases in appropriations over the 2001 level. States are required to reserve 2 percent of their Title I funds for school improvement, increasing to 4 percent in fiscal year 2004. Previously, states were permitted, but not required, to reserve up to 0.5 percent of their funds for school improvement. States must pass at least 95 percent of these funds directly to school districts. Districts must continue to reserve at least 1 percent of their Title I allocations for parental involvement activities, as was required under prior law, but the new law also requires that they pass 95 percent of these reserved funds to Title I schools. Districts may use Title I funds for schoolwide programs, rather than targeting funds to specific students, in schools where at least 40 percent of the children in the school or school attendance area are from low-income families. Previously, schoolwide programs were allowed only in schools in which at least 50 percent of the children in the school or school attendance area were from low-income families. In addition to those named above, the following people made significant contributions to this report: Natalie Britton, Karen Brown, Patrick DiBattista, Robert Dinkelmeyer, Jerry Fastrup, Sarah Glavin, Sonya Harmeyer, Peter Minarik, and Michael Williams. Jon Barker, Richard Burkard, and Robert Parker also provided key technical assistance. | The Title I program spends $8 billion each year on elementary and secondary education. Although state and local funds account for more than 90 percent of national education expenditures, Title I has been an important source of funding for many poor school districts and schools since 1965. In the 1999-2000 school year, Title I funds were targeted on the basis of numbers and percentages of poor children, but the complex allocation process resulted in differences in actual funding per poor child. When the numbers of children from low-income families shift among states, Title I allocations adjust, but not completely, and a state whose share of the nation's poor children changed from year to year would not necessarily see a corresponding change in its Title I allocation amount. The following two factors account for this: lack of current poverty data and various hold-harmless provisions. GAO found no monetary, statutory, or regulatory incentives for states to target their own funds to children from low-income families. Several policy options could increase Title I funds allocated to states and school districts with high numbers and percentages of poor children. These options include changing the appropriations hold-harmless provisions, funding the targeted grant, using an alternative cost factor, and raising the basic grant eligibility threshold. |
In the late 1980s, the Marine Corps began efforts to improve its chemical and biological protective suits because the ones used by U.S. armed forces at the time were bulky and heavy and could not be reused after laundering. Starting in 1989, the Marine Corps’ exploratory development efforts and subsequent MCLIST Advanced Technology Transition Demonstration program identified, evaluated, and field tested a number of chemical defense materials. During the same time period, the Army began an exploratory development effort to evaluate materials that offered less weight and bulk and enhanced protection, and the Air Force and the Navy began pursuing improved materials for their chemical protective uniforms. Separate from the MCLIST program, the Marine Corps fielded the Saratoga chemical and biological protective overgarment in 1991. The Saratoga garment, made from materials manufactured by Blucher, a German company, had an inner liner made with activated carbon spheres rather than the carbon-impregnated foam used by the military services at the time. All of the military services eventually combined their chemical defense material improvement efforts. Marine Corps, Army, Air Force, and Navy officials signed a memorandum of agreement, effective in November 1993, that defined the new JSLIST program and set forth the services’ respective responsibilities for the development, production, and deployment of the next-generation chemical and biological protective suits. The Marine Corps Nuclear, Biological, and Chemical Defense Office was designated as the lead agency for the JSLIST program. The MCLIST and Army programs, which ultimately became JSLIST, were research and development activities. Research and development activities are generally conducted in a more flexible environment than acquisition programs and often involve informal communications with prospective participants. DOD used market research to seek out existing foreign and domestic technologies for testing. The minimum requirements that applied to DOD’s efforts to identify improved chemical defense materials were published in a series of annual Broad Agency Announcements. These announcements described various areas of scientific or technical interest, provided specific points of contact, and included a statement that proposals would be evaluated based on merit and responsiveness to a government requirement. According to Marine Corps officials, the JSLIST program evaluated 57 material combinations (including liners, outershells, and combinations of both). The program evaluation was conducted in two phases. At the completion of phase I, 5 of the 57 material combinations were qualified to enter phase II for developmental and operational testing. At the completion of phase II in April 1997, only one of the five material combinations tested was determined to be acceptable for use in protective suits—Blucher’s Saratoga liner with a nylon and cotton outershell. (See app. I for further information on the JSLIST test program phases.) DOD is contracting for the suits through the National Industries for the Severely Handicapped and a company under section 8(a) of the Small Business Act. The contracts require that the suits be made from the Blucher material combination. Appendix II contains additional information on the JSLIST contracts. The JSLIST acquisition strategy includes a pre-planned product improvement program. DOD announced this new competition in the Commerce Business Daily in June 1997. DOD is seeking additional materials to address JSLIST objectives that were not met in the initial evaluation, such as a chemical protective garment that lasts for 60 days and improved chemical protective gloves and socks. Production is scheduled to begin in fiscal year 2000. Appendix III shows a timeline of key events in the MCLIST and JSLIST programs. Congress has raised concerns about the fact that the JSLIST material is produced by a sole-source company. The National Defense Authorization Act for Fiscal Year 1998 conference report noted that DOD should consider taking actions necessary to qualify additional sources of supply for chemical protective garment materials. Congress directed the Secretary of Defense to address this issue in DOD’s next Nuclear, Biological, and Chemical Defense Annual Report to Congress. Because the MCLIST and Army demonstrations were research and development activities, they were not subject to the formal notification requirements that apply to acquisitions. However, DOD provided adequate notice of its interest in identifying improved chemical protective materials through three different mechanisms. First, DOD conducted informal market research efforts and contacted industry directly to identify materials to include in the research and development programs. These efforts involved foreign as well as domestic suppliers. Officials from the Army’s Natick Research, Development, and Engineering Center, which managed MCLIST for the Marine Corps, told us that they traveled worldwide to locate promising technologies. Second, ongoing exploratory development projects at Natick identified technologies that were subsequently included in the demonstration programs. Marine Corps officials told us that they directed certain new and promising materials from these projects into the MCLIST demonstration program. Last, notice of three Broad Agency Announcements, published in the Commerce Business Daily, alerted industry of the government’s interest in developing new technologies for chemical defense materials. The dates of the announcements are shown in table 1. Although these announcements did not specifically cite the MCLIST program or the Army’s exploratory development effort, points of contact were named so that industry representatives could obtain further information concerning DOD’s research and development efforts. Participation in the JSLIST program was limited only to those materials submitted under the MCLIST or Army’s demonstrations; new companies or materials were excluded from entering JSLIST. DOD tested 57 material combinations, submitted by 13 companies, in the JSLIST program. However, reliance on informal communications—common in a research and development environment—resulted in some companies getting different information about deadlines for material submissions. Outside of the research and development programs, a company submitted an unsolicited proposal to the Marine Corps for an improved chemical defense material. The Marine Corps’ subsequent rejection of the proposal was consistent with the Federal Acquisition Regulation. However, the Marine Corps could have referred the company to the Army’s exploratory development program so that the material could have been considered for inclusion in JSLIST. Industry learned of the opportunity to participate in the MCLIST and Army research and development programs through informal discussions with DOD officials or by calling the points of contact listed in the Broad Agency Announcements. Participation in these programs resulted in 57 material combinations, submitted by 13 companies, being tested in JSLIST. However, because the MCLIST and Army programs operated separately, each had different timeframes and cutoff dates for submitting materials. For example, materials for the MCLIST program had to be submitted to Natick in time to make garments before field testing, which took place from April to June 1992. According to Marine Corps officials, industry was required to submit garments by September 30, 1992, for participation in the Army’s demonstration, which was held in March 1993. If industry was submitting materials rather than fully constructed garments, the deadline was June 1992. The Army’s program was industry’s final opportunity to have materials entered in JSLIST. MCLIST participants were advised informally of the final date to submit materials to the Marine Corps program, but not all companies learned of the opportunity to submit materials to the Army program. For example, officials from the complainant company said that they were not told of the Army program’s deadline for submission of materials until September 30, 1992, the last day submissions were allowed. Therefore, the company did not submit an improved material to Natick that could have been included in the Army’s demonstration and JSLIST program. Officials from this company did not call the points of contact listed in the Broad Agency Announcement. Rather, they contacted a Natick official who had been involved with MCLIST but was not directly involved with JSLIST. In contrast, a company that fared poorly in MCLIST was advised by a Natick official who was also not named in the Broad Agency Announcements of its opportunity to submit an improved material for evaluation in the upcoming Army program. Because contacts between DOD officials and industry were informal and largely undocumented, we could not determine whether all companies that contacted the Natick officials named in the Broad Agency Announcements were told of the opportunity to participate in the Army’s demonstration program. According to Marine Corps officials, the closing date of each Broad Agency Announcement represented the cutoff dates for material submissions to the demonstration programs. For example, the officials said that the closing date of the final announcement—September 30, 1992—was the deadline for entering materials in the Army’s demonstration program. However, Natick officials told us that the Broad Agency Announcements were not connected to cutoff dates for material submissions to the demonstration programs. For example, the officials noted that the cutoff date for the MCLIST demonstration was simply tied to their need to receive materials and make garments in time for field testing. The three Broad Agency Announcements did not cite particular programs—such as MCLIST or JSLIST—or cutoff dates for material submissions and thus were not explicitly tied to the deadline for material submissions. Nothing in the third announcement sets it apart from the two previous announcements to indicate that September 30, 1992, was industry’s last opportunity to submit a material to JSLIST via the Army’s demonstration program. Furthermore, a fiscal year 1993 Broad Agency Announcement, issued a year after entry to JSLIST was closed, is nearly identical to the three previous announcements. In June 1992, 3 months before the closing date of the fiscal year 1992 Broad Agency Announcement, the complainant company submitted an unsolicited proposal to the Marine Corps for an improved chemical defense material. The company had developed this improved material after its initial submission to the MCLIST program. The proposal—submitted outside of the Advanced Technology Transition Demonstration process—was for a competitive alternative to the Saratoga suits that the Marine Corps had begun to field in 1991. The company offered to absorb a substantial share of the testing costs if its product did not meet the Marine Corps’ requirements. In April 1993, more than 6 months after the deadline for submissions to the Army’s demonstration program, the Marine Corps rejected the unsolicited proposal, stating that it was duplicative in nature to an existing Marine Corps effort. The rejection letter noted that the Marine Corps had already initiated a research and development effort based on the carbon sphere technology identified in the proposal. The Federal Acquisition Regulation provided that an unsolicited proposal be rejected when it resembled an ongoing acquisition or did not demonstrate an innovative and unique concept. The Marine Corps’ actions in rejecting the proposal were therefore consistent with the Federal Acquisition Regulation. However, we believe the Marine Corps missed an opportunity to refer the firm to the Army’s demonstration program, which ultimately became part of JSLIST, before it was too late for entry. Because a joint program was already planned at that time, the Marine Corps could have informed the company that the Broad Agency Announcement was still open, thus allowing the firm’s potential entry into the Army’s demonstration program. We could not determine whether the outcome of the JSLIST program would have differed if the firm’s material had been assessed. The complainant company asserted that the transition to JSLIST had caused a significant change in requirements. The company was primarily concerned that the requirement for a single-use garment in MCLIST had changed to one that was launderable in JSLIST. Officials from the company stated that the changed requirements placed it at a disadvantage relative to the competition because it was not allowed to submit a different material under JSLIST. However, the basic requirements for a lightweight, launderable, chemical protective overgarment did not change during the transition. The requirements for MCLIST and JSLIST were based on assessments of the services’ missions and the threat. DOD’s basic requirement for both programs was to develop a lightweight, launderable, durable material that protected against chemical agent penetration and reduced heat stress. These fundamental requirements were reflected in the three Broad Agency Announcements issued from fiscal years 1990 to 1992. The latter two announcements added a requirement for protection from toxic aerosols. MCLIST and JSLIST both sought improved materials that could be laundered multiple times and reused, unlike the single-use garment that the Army, Air Force, and Navy were using at the time, which could not be laundered. Although the fundamental requirements did not change, each service added various service-unique requirements to the JSLIST program. For example, the Army required, in addition to the standard 30-day garment, a single-use, 7-day garment that would weigh less than the 30-day garment and provide the same level of chemical protection. In addition, the Air Force required chemical protection for liquid dispersed by air burst munitions or spray tanks. Although materials were tested that would accommodate these variations, the basic requirements for the overgarment did not change. The National Defense Authorization Act for Fiscal Year 1998 conference report urged DOD to consider taking actions necessary to qualify additional sources of supply for its chemical protective garment materials. The conferees directed the Secretary of the Army, as executive agent for the chemical-biological defense program, to report to the congressional defense committees on any plans to qualify such sources. The conferees also directed the Secretary of Defense to address the issue as a special area of interest in DOD’s next annual report to Congress on the Nuclear, Biological, and Chemical Defense program. However, the subsequent annual report to Congress, issued in February 1998, did not address the issue of qualifying additional sources of supply. A DOD official told us that the reporting requirement had been overlooked. In May 1998, DOD issued an addendum to its February 1998 report to address the congressional reporting requirement. The addendum cited DOD’s ongoing pre-planned product improvement program as a potential mechanism for identifying additional sources of supply for the requirements that were not achieved in the JSLIST program. The report emphasizes, however, that DOD’s primary goal remains to provide the U.S. armed forces with the best chemical protective ensemble available. Because the MCLIST program and the Army’s exploratory development efforts—which ultimately became the JSLIST program—were research and development activities, they were not subject to the same procedural requirements that apply to acquisition programs. In this context, we believe that the MCLIST and JSLIST programs were conducted fairly. DOD provided industry adequate notice of the government’s interest in improved chemical defense garment materials and sufficient opportunity to participate in the programs. The basic requirement for a lightweight, less bulky overgarment that could be reused after laundering did not change in the transition from MCLIST to JSLIST, although each service added certain unique requirements. Although the informal nature of communications that characterizes the research and development environment may have contributed to a missed opportunity for DOD to evaluate the complainant’s improved chemical defense garment material, the Marine Corps’ rejection of the unsolicited proposal was consistent with the Federal Acquisition Regulation governing such proposals. DOD reviewed a draft of this report and fully concurred with the information as presented. To determine whether DOD provided sufficient notice to industry of its interest in new chemical defense materials and industry had sufficient opportunity to participate in the MCLIST demonstration, the Army’s exploratory development effort, and the JSLIST program, we obtained and analyzed Broad Agency Announcements and information about DOD’s formal and informal communications with industry regarding the programs. We gathered and analyzed information pertaining to the intent, timeframes, and participants in the MCLIST and Army’s demonstration programs, which afforded industry entry to JSLIST. We discussed these programs with officials from the Office of the Secretary of Defense, Director of Defense Procurement; the Marine Corps Systems Command; the Army’s Natick Research, Development, and Engineering Center; and selected contractor locations. We also reviewed the unsolicited proposal a company sent to the Marine Corps in June 1992 and the Marine Corps’ response to the company. We determined whether the Marine Corps’ rejection of the proposal was done in accordance with the Federal Acquisition Regulation. To determine whether basic requirements that materially affected the competition changed in the transition from MCLIST to JSLIST, we obtained and analyzed requirements documents, such as the Marine Corps’ 1986 Required Operational Capability statement and the 1995 Joint Operational Requirements Document for JSLIST, as well as test plans for JSLIST. We discussed MCLIST and JSLIST requirements with officials from the Marine Corps Systems Command and industry officials. We also analyzed the requirements as stated in the Broad Agency Announcements. To report on DOD’s response to the congressional reporting requirement in the Fiscal Year 1998 National Defense Authorization Act conference report, we interviewed officials at the Office of the Assistant to the Secretary of Defense for Nuclear, Chemical, and Biological Defense Programs and the Marine Corps Systems Command. We also reviewed DOD’s February 1998 Nuclear, Chemical, and Biological Defense Annual Report to Congress and the May 1998 addendum to that report. We did not attempt to assess the accuracy of the JSLIST test results or determine whether the outcome of the JSLIST program would have differed if additional chemical defense materials had been evaluated. We conducted our review from February to June 1998 in accordance with generally accepted government auditing standards. We are sending copies of this report to appropriate congressional committees, the Secretary of Defense, and the Commandant of the Marine Corps. We will also provide copies to other interested parties on request. Please contact me at (202) 512-4841 if you or your staff have any questions concerning this report. Major contributors to this report are listed in appendix IV. Because the Joint Service Lightweight Integrated Suit Technology (JSLIST) program was a joint effort, standardized test methods and procedures had to be developed that would be acceptable to all the services and recognize service-unique requirements. Testing occurred during two phases. Phase I began in October 1993; phase II began in August 1995 and resulted in the selection of the only material combination that was determined to be acceptable for use in the Department of Defense’s chemical defense garments. During phase I, 600 suits made from 57 candidate material combinations (outershells, liners, or both) were tested. Phase I testing involved chemical agent penetration, aerosol penetration, heat stress, laundry tests, wear tests, suit design evaluation, physical properties and flame resistance tests, and user size and fit tests. Wear tests were conducted over 14 days at Camp Pendleton, California, and Camp Lejeune, North Carolina. Suit design evaluations took place at Camp Pendleton, California; Camp Lejeune, North Carolina; Brooks Air Force Base, Texas; U.S. Army Natick Research, Development, and Engineering Center, Massachusetts; Fort Rucker, Alabama; Norfolk Naval Station, Virginia; and Fort Bragg, North Carolina. To pass the tests, candidate materials had to equal or exceed the performance of the Battledress Overgarment, which was the suit used by the Army at that time, or the Saratoga, which was replacing the Battledress Overgarment in the Marine Corps. Those materials that did not pass chemical agent penetration and laundry tests were disqualified. Test garments and individual swatch samples were coded to ensure unbiased testing. Phase I culminated in a decision by the Marine Corps Systems Command to proceed to the engineering and manufacturing development phase, or phase II. Of the 57 material candidates, 5 were selected for phase II testing. During phase II, more than 3,000 suits (330 suits per candidate material plus the Battledress Overgarment and Saratoga control garments) were tested at 10 worldwide field locations using mission-oriented scenarios and individual user tasks. The purpose of these tests was to collect operational and technical data to assess the performance of each material configuration during mission-oriented activities and provide worn suits for follow-on laboratory tests. The coding procedure used in phase I was also used in phase II to continue unbiased testing. Arctic, tropic, and desert environmental conditions were represented during the tests. The wear periods were 7, 15, 30, and 45 days, and suits were laundered up to 6 times. Laboratory tests included chemical agent, flame resistance, aerosol penetration, and heat stress. As with phase I, the Army’s Battledress Overgarment and the Marine Corps’ Saratoga garment were the established baselines. Only one material combination was found to meet all program requirements—the nylon/cotton outershell over the Blucher Saratoga liner. Because Blucher’s Saratoga liner with a nylon/cotton outer shell was the only material combination to pass the JSLIST phase II tests, the JSLIST production material is sole source. The Department of Defense is contracting for the JSLIST suits through the National Industries for the Severely Handicapped and another contractor under section 8(a) of the Small Business Act. Tex-Shield, Inc., Blucher’s American distributor, is a directed sole source for the material from which the suits are to be fabricated. The contracts are not awarded competitively. Suit fabrication services are on the procurement list developed under the provisions of the Javits-Wagner-O’Day Act. Under the act, once a commodity or service has been added to the list by the Committee for Purchase from the Blind and Other Severely Handicapped, contracting agencies are required to procure the commodity or service directly from the workshops for blind or other severely handicapped individuals affiliated with the National Industries for the Severely Handicapped. For the JSLIST suits, the Marine Corps offered the requirement to the National Industries for the Severely Handicapped, which accepted as much of the work as it could, and the Marine Corps contracted for the balance of work with another contractor under section 8(a) of the Small Business Act. Neither contracting through the National Industries for the Severely Handicapped nor the contract with the section 8(a) contractor required publication of a notice in the Commerce Business Daily or the use of competitive procedures. MCLIST program began. First Broad Agency Announcement was released in Commerce Business Daily. Marine Corps fielded Saratoga suits. Second Broad Agency Announcement was released in Commerce Business Daily. Third Broad Agency Announcement was released in Commerce Business Daily. First MCLIST wear test was conducted at Camp Lejeune (hot/humid climate testing). A company submitted an unsolicited proposal to the Marine Corps. Second MCLIST wear test was conducted at Camp Pendleton (hot/dry climate testing). Deadline for submissions to Army demonstration program. According to Marine Corps, no additional materials were accepted after this time. Army wear test was conducted at Camp Pendleton. Marine Corps rejected the unsolicited proposal. JSLIST phase I testing began with 57 material combinations. JSLIST memorandum of agreement became effective. JSLIST phase II testing began. Phase II testing was completed, and Saratoga material was selected. JSLIST contracts were awarded. Preplanned Product Improvement Program was announced in Commerce Business Daily. John A. Carter William T. Woods The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO examined the Department of Defense's (DOD) research and development programs for improved chemical defense garment materials for U.S. armed forces, focusing on whether DOD: (1) provided sufficient notice to industry of the government's interest in identifying improved chemical defense materials; (2) provided industry sufficient opportunity to participate in the programs; (3) changed requirements in the transition of the Marine Corps Lightweight Integrated Suit Technology (MCLIST) program to the Joint Service Lightweight Integrated Suit Technology (JSLIST) program; and (4) responded to a congressional reporting requirement concerning sources of supply for the military's chemical defense materials. GAO noted that: (1) because the MCLIST program and the Army's exploratory development efforts--which ultimately became the JSLIST program--were research and development activities, they were not subject to the same procedural requirements that apply to acquisition programs; (2) in this research and development context, DOD provided sufficient notice to industry of its interest in improved chemical defense materials through market research, direct industry contacts, ongoing exploratory development projects, and Broad Agency Announcement notices published in the Commerce Business Daily; (3) DOD provided industry an adequate opportunity to participate in the research and development programs; (4) on the basis of the notice DOD provided to industry, a total of 57 material combinations, submitted by 13 companies, were evaluated in JSLIST after initial testing in MCLIST and the Army demonstration; (5) however, reliance on informal communications resulted in companies receiving different information about submission deadlines to MCLIST and the Army programs; (6) due in part to the lack of formal communications regarding deadlines of material submissions, the complainant company did not get an improved material into JSLIST for evaluation; (7) the complainant company, which already had one material in the MCLIST demonstration, submitted an unsolicited proposal to the Marine Corps for an improved chemical defense material after entry into MCLIST was closed; (8) the Marine Corps rejected the proposal in accordance with the Federal Acquisition Regulation because the material duplicated an existing Marine Corps effort; (9) GAO could not determine whether the outcome of JSLIST would have differed if this material had been assessed; (10) the basic requirements for a lightweight, launderable, chemical protective garment did not change in the transition from MCLIST to JSLIST; (11) however, the individual military services added certain mission-specific requirements under JSLIST; (12) DOD overlooked the congressional reporting requirement in the National Defense Authorization Act for Fiscal Year 1998 conference report; and (13) a May 1998 addendum to DOD's February 1998 Nuclear, Biological, and Chemical Defense Annual report to Congress stated that the preplanned product improvement program might identify additional sources of supply to meet those requirements that were not achieved in the JSLIST program. |
The Army and Marine Corps use established mechanisms that allow them to categorize and track personnel with specific skill sets. Specifically, each service uses Military Occupational Specialties (MOS) and additional skill identifiers to categorize the specific jobs and skill sets of officers and enlisted servicemembers. Department of the Army Pamphlet 600-3, Department of the Army Pamphlet 600-25, Department of the Army Pamphlet 611-21, and the Marine Corps’ Military Occupational Specialties Manual outline the MOS and skill identifiers for officers and enlisted personnel.extend over one or more grades. For instance, infantrymen in the Army have an MOS of 11B, while those in the Marine Corps have an MOS of 03. Each MOS outlines a path for career development and management of each MOS by grade. This path helps direct the training, education, and experience that will guide a servicemember in career development, but it does not guarantee that individual’s success. Instead, it describes the full spectrum of development opportunities that an officer or enlisted servicemember is expected to pursue throughout his or her career. For example, majors seeking promotion to lieutenant colonel in the field artillery branch of the Army are expected to continue their professional development by completing certain required military education and serving in key assignments such as fires brigade S3, as well as The MOS describes a set of related duties and tasks that developmental assignments such as training developer. Further, these majors are expected to continue self-development efforts to build organizational leadership and strategic perspective, and to hone operational skills. In addition to the MOS, a skill identifier is a career-management tool that identifies specific skills that are required to perform the duties of a particular position and are not related to any one branch, functional area, or career field. For example, all Army infantry officers who have met ranger training requirements will have a skill identifier of 5R, and those who have met parachutist training requirements will have a skill identifier of 5P. As servicemembers obtain additional skills not related to their primary skill set, they can be awarded an additional skill identifier associated with that skill set, based on their meeting criteria laid out in training standards or manuals. MOS and additional skill identifiers are maintained in the personnel data systems for the Army and Marine Corps and can be searched to identify personnel with certain experience. DOD has established a requirement for the services to identify and track personnel who have completed SFA-related training, education, or experience in the Defense Readiness Reporting System with a relevant skill-designator indicating their SFA qualifications. According to Department of the Army headquarters officials, the Army has developed a mechanism to identify and track personnel who have completed certain SFA-related training and education and is considering steps to take toward identifying and tracking personnel with SFA-related experience, but it has not made a decision on what approach it will take. Further, the Army has not developed a plan with goals and milestones to identify and track personnel with SFA-related experience. The Marine Corps is in the process of implementing a means to identify and track individuals with SFA-related training, education, and experience. In October 2010, DOD issued an instruction—DOD Instruction (DODI) 5000.68— that broadly defines SFA and outlines responsibilities for key stakeholders to plan, prepare for, and execute SFA activities. The instruction states that DOD shall develop and maintain the capability within DOD general purpose forces, special operations forces, and the civilian expeditionary workforce to conduct SFA activities in support of U.S. policy and in coordination with the relevant U.S. government departments or agencies. The instruction further states that DOD shall conduct SFA activities with the appropriate combination of general purpose forces, special operations forces, and civilian expeditionary workforce under a variety of conditions that include: (1) politically sensitive environments where an overt U.S. presence is unacceptable to the host-country government; (2) environments where a limited, overt U.S. presence is acceptable to the host-country government; and (3) environments where a large-scale U.S. presence is considered necessary and acceptable to the host-country government. The instruction further requires that the secretaries of the military departments identify and track individuals who have completed SFA-related training, education, or experience in the Defense Readiness Reporting Systemskill-designator indicating their SFA qualifications. According to DODI 5000.68, the secretaries shall develop, maintain, and institutionalize capabilities of servicemembers to support DOD efforts to organize, train, equip, and advise foreign military forces. Having the ability to identify and track individuals with SFA-related training, education, and experience will assist DOD in institutionalizing SFA capabilities. Officials at the Office of the Secretary of Defense, Department of the Army, and Headquarters Marine Corps stated that it was important to be able to identify and track servicemembers with SFA-related training, education, and experience so that they could be utilized for future SFA missions. Having this tracking capability will assist the Army and Marine Corps in identifying the right people to carry out SFA activities. with a relevant In April 2011, the Joint Requirements Oversight Council issued a memorandum endorsing certain recommended SFA actions. As we reported in 2012, the memorandum identified 25 tasks related to the implementation of SFA across DOD in the areas of doctrine, organization, training, materiel, leadership and education, personnel, and facilities. According to SFA Working Group officials, this memorandum does not constitute policy. However, like DODI 5000.68, it emphasizes the need to identify and track personnel with SFA-related training, education, and experience. For example, one of the tasks that was emphasized for the services was the need to establish skill identifiers for military personnel with SFA-related training and experience, and the need to track these personnel. According to the memorandum, this task was to be completed by the Under Secretary of Defense for Personnel and Readiness, the services, the Defense Security Cooperation Agency, and United States Special Operations Command by April 2013. According to officials, an SFA Steering Committee and Working Group were established in January 2011 to coordinate the department’s efforts to develop SFA policy and capabilities and manage the implementation of DODI 5000.68. According to officials with the Office of the Under Secretary of Defense for Personnel and Readiness, SFA continues to be a priority for DOD. As a result, officials stated that an instruction on irregular warfare and SFA is being drafted in order to emphasize how DOD will conduct SFA activities. Among other things, according to officials, the instruction will reiterate the importance of identifying and tracking personnel with SFA-related training, education, and experience so that they can be utilized for future missions. Irregular warfare is defined as a violent struggle among state and nonstate actors for legitimacy and influence over the relevant populations. It favors indirect and asymmetric approaches, though it may employ the full range of military and other capacities, in order to erode an adversary’s power, influence, and will. The Army has taken steps toward identifying and tracking personnel with certain SFA-related training and education, but efforts to identify and track SFA-related experience are incomplete. Officials told us that since 2008, they have been identifying and tracking soldiers who attend certain SFA training with a Personnel Development Skill Identifier code. The Army uses Personnel Development Skill Identifier codes in combination with an area of concentration or MOS to identify unique skills, training, and experience personnel may obtain during their careers. Normally, the principal organizations or agency may request establishment of a Personnel Development Skill Identifier code for tracking unique skills, training, or experience where identification of qualified personnel would be beneficial to the Army. Personnel Development Skill Identifier codes identify skills, training, and experience that do not meet the minimum standards for establishment of an additional skill identifier or have too few individuals meeting the qualifications to warrant the creation of an additional skill identifier. For example, the Army has created Personnel Development Skill Identifier codes to identify and track soldiers who have completed congressional fellowship training, completed training and 6- month deployment in support of United Nations peacekeeping operations, and completed training for the maintenance of various weapons systems. With regard to certain SFA-related training and education, the Army has established Personnel Development Skill Identifier codes for the following training activities: successfully completing a resident advisor training course at Fort Riley and Fort Polk; successfully completing an advisor training course at the Phoenix Academy in Iraq; and successfully completing advisor training offered by the mobile training team from Fort Polk. Army officials acknowledged that while these Personnel Development Skill Identifier codes can be used to identify and track soldiers who attend these SFA-related training courses, the codes do not identify and track the actual experience that a soldier obtains from serving in an SFA- related role or any other SFA-related training and education that soldiers might receive. For example, according to an Army official, one brigade that is regionally aligned to Africa receives SFA-related training through a course called Dagger University at Fort Riley that is not captured by any of the existing Personnel Development Skill Identifier codes. Further, Army officials noted that some personnel trained in these activities might not have actually served in these roles. Army officials stated that the Army has been considering steps to take toward identifying and tracking personnel with SFA experience, but has not yet developed a mechanism with which to do so. Officials from Department of the Army Headquarters Operations and Plans office stated that they have ongoing efforts to identify the best approach, which may include developing or expanding on existing Personnel Development Skill Identifier codes, or creating an additional skill identifier or MOS to meet the requirement in DODI 5000.68 to identify and track personnel with SFA-related experience. Army officials told us that they could create new SFA-related Personnel Development Skill Identifier codes or expand existing ones to identify and track personnel with SFA-related experience. The officials stated that they use a similar process to capture not only the education and training, but also the experience of individuals who have served in the Afghanistan-Pakistan Hands Program. For example, Personnel Development Skill Identifier code T2A identifies soldiers who completed language, counterinsurgency, and culture training as part of that program. Personnel Development Skill Identifier code T2B identifies the soldiers who successfully completed a 1-year deployment to Afghanistan or Pakistan and utilized skills and training from the Afghanistan-Pakistan Hands Program. Officials indicated to us that they could potentially enhance existing SFA-related Personnel Developer Skill Identifier codes to capture the SFA-related experience of soldiers. However, as of the time of our review, Army headquarters officials stated that the Army had not decided which approach it would use. As a result, while they can identify personnel with certain SFA-related training and education, they have not developed a mechanism to identify and track personnel with SFA-related experience. Army officials knowledgeable about these efforts told us that they have not yet determined what mechanism they might use to identify and track personnel with SFA-related experience for several reasons, including a lack of clarity regarding what activities and experience constitute SFA. For example: Determining activities that constitute SFA. According to officials from the Office of the Under Secretary of Defense Personnel and Readiness, SFA continues to evolve to meet emerging threats. Army officials stated that they are still trying to determine what activities constitute SFA. An Army headquarters official stated that identifying an activity as SFA is subjective and the Army is still trying to figure out how SFA fits into irregular warfare. Determining SFA experience. Army officials with whom we spoke stated that they are trying to determine the level of SFA responsibility that a soldier must assume in order for it to be recognized as sufficient for SFA experience and thereby tracked. For example, some individuals serve as junior members on SFA teams and may not continually engage in an advising role, whereas the leader of the SFA team is continuously advising foreign military officials. Officials from the Under Secretary of Defense for Personnel and Readiness office told us they do not intend to issue any more prescriptive guidance to the services about SFA. Similarly, members of the SFA Working Group stated that the information available on SFA is adequate and they believe it is unnecessary to provide the services with any additional SFA guidance. We found several issued documents that define SFA or identify the characteristics that individuals conducting SFA activities should possess. These documents include the following: The 2013 Army Field Manual on Army Support to Security Cooperation, which provides details on the advisor roles, considerations for working effectively with foreign security forces, and personality traits of advisors. According to this manual, not every soldier is well suited to be an advisor. It identifies tolerance for ambiguity, open-mindedness, and empathy, among others, as traits that greatly enhance the advisor’s ability to adapt and thrive in a foreign culture. The Security Force Assistance Lexicon and Framework, issued in November 2011, was intended to promote a common understanding of SFA and related terms by providing greater clarity on the definition of SFA and how it relates to other existing terms, such as security cooperation and security assistance. The International Security Assistance Force Guide 2.0, which was updated in January 2014, defines SFA and provides information about advisor characteristics for the advising mission in Afghanistan. According to this guide, not everyone is qualified to perform advisory functions. Similar to the Army Field Manual on Army Support to Security Cooperation, it identifies a tolerance for ambiguity, empathy, and open-mindedness, among others, as characteristics that will enhance an advisor’s ability to adapt and thrive in a foreign culture. In January 2014, the Under Secretary of Defense for Personnel and Readiness issued a memorandum with guidance that identifies the common training standards for the SFA mission. The memorandum lays out a list of skills—such as patience, adaptability, judgment, and initiative—related to SFA that are common across the forces, as well as standards that can be used as guidelines for measuring the qualifications of individuals and collective forces. It identifies a set of benchmarks for the services to use to identify, train, and track individuals and collective forces conducting SFA activities. For example, according to the memorandum, a servicemember should be proficient in cross-cultural communications, which include the ability to identify and discern cultural differences, and should demonstrate adaptability to engage in unfamiliar situations. In 2012, we reported that combatant commanders lacked a clear understanding of SFA and therefore DOD made efforts to clarify the meaning of SFA. In spite of DOD’s efforts, Army officials with whom we spoke stated that confusion still remains regarding what constitutes SFA. According to SFA Working Group members, there may be a need for additional education regarding available resources that define and list activities that constitute SFA. Although Army officials stated that the Army has considered steps to identify and track personnel with SFA-related experience, the Army has not established a plan with goals and milestones for how or when these steps will be completed. Since the Army does not have a mechanism to identify and track personnel with SFA-related experience, the Army would have to review duty descriptions in personnel files individually in order to determine which servicemembers have SFA-related experience. Manually reviewing files can be a laborious, time-consuming effort. Some DOD officials told us that it is not an efficient way to conduct business. Further, reviewing personnel files individually to identify servicemembers with SFA-related experience does not meet the requirement to identify and track individuals who have completed SFA-related training, education, or experience in the Defense Readiness Reporting System with a relevant skill-designator indicating their SFA qualifications. According to DODI 5000.68, individuals with SFA-related training, education, or experience were to be identified and tracked in the Defense Readiness Reporting System. However, since the Army has not established a method to identify and track personnel with SFA-related experience, this information is not being integrated into that system. Instead, personnel at Human Resources Command said they are able to manually review personnel files in order to identify servicemembers with SFA-related experience. As noted above, the Army has not decided what approach it will take to identify and track personnel with SFA-related experience. Further, according to a headquarters Army official, no time frame has been set for meeting this requirement. Standard practices for project management call for agencies to conceptualize, define, and document specific goals and objectives in the planning process, along with appropriate steps, milestones, time frames, and resources to achieve those results. Without goals and milestones, it is unclear how long the Army’s implementation of the DODI 5000.68 requirement to identify and track personnel with SFA-related experience might take. As a result, the Army is at risk for not being able to readily identify the right personnel with the right SFA-related skills and experience to serve in an SFA mission. This could potentially limit the effectiveness of the advisor teams and the Army’s ability to develop, maintain, and institutionalize the capabilities of servicemembers to conduct SFA activities to build the capacity and capability of foreign military forces. The Marine Corps has approved and is in the process of implementing an officer and enlisted Free MOS to identify and track personnel with SFA- related training, education, and experience. A Free MOS is an auxiliary MOS that can be assigned to any Marine to reflect the acquisition of certain skill sets beyond those related to the primary MOS. Initially, in 2011, the Marine Corps developed an online irregular-warfare skills tracker to provide this capability, and the information in it was not readily accessible, could not be grouped by unit, and was not integrated with any other system in the Marine Corps. While the tracker met the intent of DODI 5000.68 to identify and track personnel with SFA-related experience and will be developed to meet emerging irregular warfare requirements, it was not sufficient in helping the Marine Corps assign servicemembers to SFA positions. As a result, in April 2013, the Marine Corps decided to develop a Free MOS that could be used to identify and track personnel with SFA-related training, education, and experience. The Marine Corps currently uses Free MOS to capture personnel with other skill sets and experience, such as personnel who have served as foreign area officers and regional area officers. The Marine Corps tasked its human resources command—Manpower and Reserve Affairs— as the lead to integrate the Free MOS into existing personnel systems in accordance with standard procedures for the creation of a new Free MOS. In December 2013, the Commanding General of the United States Marine Corps Training and Education Command approved the Foreign Security Force Advisor Free MOS for officers and enlisted personnel. The Free MOS is currently being finalized and is expected to be included in the Marine Corps fiscal year 2015 MOS Manual. The Marine Corps is planning for the Free MOS to be available by October 1, 2014, to assign to those Marines who meet specified qualifications. Those qualifications are identified as follows in the memorandum that approved the language for the Free MOS that will be added to the MOS manual: Prerequisites. The Marine must complete the appropriate Regional, Culture, and Language training for that Marine’s grade. In addition, the Marine must have an endorsement from a battalion- or squadron- level commanding officer certifying that the Marine has met the prerequisites, is mature, and is capable of independent operations. Requirements. The Marine must complete the Marine Corps Advisor Course. Alternatively, if the Marine does not complete the Marine Corps Advisor Course, the Marine can be certified by a commanding officer after demonstrating sufficient expertise in advising foreign security forces in an on-the-job training environment spanning a cumulative period of at least 6 months while still serving in the position. According to the Marine Administrative Message that the Marine Corps has developed to announce approval of the Free MOS, Marines who have already graduated from an advisor course or who have 6 months or more of cumulative on-the-job experience advising foreign security forces may apply to be grandfathered in to receive the Free MOS. The Marine Corps Security Cooperation Group will convene a panel in order to review these applicants’ packages, which should include a letter from the applicant to the commanding officer of the Marine Corps Security Cooperation Group, endorsement from battalion- or squadron-level or higher commanding officer, and documentation demonstrating that the Marine graduated from an advisor training course, among other things. The Marine Corps Security Cooperation Group is responsible for managing the Foreign Security Force Advisor MOS and the Marine Advisor Course. In the event that the Marine Corps Security Cooperation Group is disbanded, the Training and Education Command commanding officer would be responsible for determining who would provide the training going forward. The Army and Marine Corps have established processes to guide and evaluate career progression. Both services have taken a number of actions to give consideration to those serving in SFA-related positions during the promotion process. However, DOD cannot evaluate the effect that serving in an SFA-related role could have on career progression because the Army and Marine Corps have not established a mechanism to identify and track servicemembers with SFA-related experience. The Army and Marine Corps convene promotion boards to review personnel for career advancement. These promotion boards consist of 5 to 21 servicemembers whose rank is above that of the promotion candidates. The promotion cycles for officers and enlisted personnel occur throughout the year. The promotion process is confidential, and each board member is briefed on and reviews the same information for each candidate, to include performance evaluations, assignment history, and professional development and education. According to DOD officials, board members review the entirety of servicemembers’ careers when making decisions about promotions. Both the Army and the Marine Corps provide guidance to promotion boards to assist them in their deliberations. Specifically, the Army provides every promotion board member with a Memorandum of Instruction signed by the Secretary of the Army for officers and the Army Chief of Staff for enlisted personnel. Similar to the Army, the Marine Corps provides every promotion board member with instructions, also referred to as precepts, that are signed by the Secretary of the Navy for officers and the Commandant of the Marine Corps for the enlisted personnel. According to the Memoranda of Instruction and precepts, promotion board members should not consider demographic information such as gender and ethnic background when making promotion decisions, as they have no bearing on servicemembers’ merit and abilities. However, they should consider education, experience, and physical fitness in their deliberations. More specifically, for example, one of the Army’s Memoranda of Instruction that we reviewed states that it is important that officers selected by board members for promotion possess the right mix of field and headquarters experience, as well as the training and education to meet current and future leadership requirements. Further, it states that all assignments are important to sustain a trained and ready Army. The Army and the Marine Corps have taken a number of steps to ensure that servicemembers with SFA-related experience receive consideration for serving in these positions during the promotion process. Both services have incorporated language into the guidance given to each promotion board member every time a promotion board is held to ensure that appropriate consideration is given to individuals who have served on advisor teams. For example, both the Army’s Memoranda of Instruction and the Marine Corps’ precepts for promotion boards state that, given the current operational environment, promotion board members should pay particular attention to servicemembers with unique key and developmental positions, to include serving on an advisor team. Specifically, as it relates to SFA, one of the Army’s Memoranda of Instruction states that the absence of command, combat experience, or support of deployed forces should not be a basis for nonselection for promotion. With regard to service as an advisor, the memorandum cites the important role that advisor teams play in enabling the United States to hand over security responsibilities to host-nation security forces and notes that advisor teams operate under very austere conditions and coach, teach, and mentor host-nation security forces while simultaneously conducting combat operations as they are embedded with host-nation security forces. The memorandum instructs the promotion board to appreciate the challenging nature of demands of advisor team jobs and to provide appropriate consideration in the overall evaluation of the record of each officer and enlisted servicemember who has served on these teams. Similarly, the Marine Corps’ precepts direct each promotion board to be especially diligent in weighing the qualifications of officers and enlisted personnel serving in advisor teams. They further state that, in board deliberations, service in these critical positions should be weighted equally with traditional Marine Corps officer and enlisted positions in the operational forces supporting overseas contingency operations because assignments are made in the best interest of the Marine Corps. To emphasize the importance of giving proper consideration to servicemembers who have served in SFA positions, according to a Marine Corps official, the Marine Corps is planning to draft additional language that will be briefed to members of each promotion board prior to their ranking of servicemembers for promotion. In addition, according to an Army headquarters official, in an effort to ensure that servicemembers who serve on advisor teams are given credit for SFA-related experience, the Army made changes to its MOS publications. In 2008, the Army Chief of Staff stated that soldiers who served on advisor teams were developing the skills and experience that are vital for the Army to accomplish tasks from direct combat to stability operations, recognizing that these skills will be a major part of military operations in the future. To ensure that majors serving on advisor teams receive recognition and credit for their service, the Army Chief of Staff directed the Army to designate service on an advisor team as a key assignment for infantry majors in Department of the Army Pamphlet 600- 3. For majors, serving on a brigade or battalion advisor team is therefore comparable to being assigned as, among other things, a battalion executive officer or brigade operations officer. According to an Army personnel official, the Army Chief of Staff’s statement also served as a driver for the Army not only to change Department of the Army Pamphlet 600-3, but also to change Department of the Army Pamphlet 600-25, as well as the Memorandum of Instruction provided to promotion board members. The pamphlets have been updated to indicate that service in an advisor role is a key developmental opportunity. For example, the MOS of Army infantry majors and lieutenant colonels now reflect that service on advisor teams can be considered part of their key developmental assignments. We found differing opinions regarding the effect that Army and Marine Corps guidance have on the manner in which individuals with SFA-related training, education, and experience are considered during the promotion process. Several officials from the SFA brigades as well as officials at the Office of the Secretary of Defense, Army and Marine Corps headquarters, Human Resources Command, and Manpower and Reserve Affairs stated that serving as an SFA advisor could negatively affect career progression. Specifically, in 14 of the 24 interviews we conducted at these organizations, officials stated that serving in an SFA-related role could potentially negatively affect career progression, or that their own careers had been negatively affected as a result of serving in an SFA-related role. According to officials, serving in an SFA-related position could cause a servicemember to miss or be delayed in completing a key developmental step outlined in the MOS publication, which could negatively affect the member’s career progression. For example: One servicemember stated that he was taken out of company command to serve on an advisor team. This cut his company commander time down, which he believed made him less competitive compared to other servicemembers in his year group who had completed company command in accordance with the professional development outlined for his MOS. One soldier said that he wanted to apply for an opportunity that would allow him to transfer from a combat arms position into the Foreign Area Officer branch. According to the soldier, the time spent serving on an advisor team shortened the amount of time he could spend completing key developmental steps that were a prerequisite for him to transfer. The soldier stated that as a result of not completing the key developmental steps in his MOS, he did not qualify to apply for the transfer. Conversely, some servicemembers with whom we spoke stated that they felt their careers had not been negatively affected. For example, one servicemember told us that he served on an advisor team and is now a colonel. He stated that serving in an SFA-related role is viewed in a positive manner and that soldiers seek these positions. Similarly, officials from the Army’s Operations and Plans office stated that some soldiers’ careers could potentially be enhanced as a result of serving in an SFA- related position. In addition, officials from Army Human Resources Command and Marine Corps Manpower and Reserve Affairs, as well as officials from the Office of the Under Secretary of Defense for Personnel and Readiness, Department of the Army Headquarters, and the Marine Corps Security Cooperation Group, stated that other factors likely play into decisions not to promote an individual. They noted that the promotion process considers the entirety of an individual’s career and not just the time spent as an advisor. According to the Army’s MOS manual, success does not depend on the number or type of positions held, but rather on the quality of duty performance in every assignment. As a result, officials stated that serving as an advisor was unlikely to be the sole explanation for a failure to be promoted. According to officials, if a soldier performs well, that soldier will be promoted regardless of the role he or she was assigned. According to DOD officials, it is difficult if not impossible to determine why a servicemember did not get promoted, because the promotion boards do not maintain a record of or discuss the reasons why a person did not get promoted. Officials from the Army Human Resources Command and the Marine Corps Manpower and Reserve Affairs Department informed us that they monitor where servicemembers are in their career to try to ensure that servicemembers are able to obtain a key developmental step that may be missed. Although some servicemembers stated that their careers could potentially be negatively affected as a result of serving in an SFA-related role, the Army and Marine Corps are currently unable to assess this potential negative effect or to take any needed corrective actions. Specifically, information about who has SFA-related experience is necessary in order to isolate this population from those who have not served in an SFA-related role. As noted above, while the Army has developed a Personnel Development Skill Identifier to identify and track soldiers with SFA-related training and education, the Army has not developed a mechanism with which to identify and track personnel with SFA-related experience, and the Marine Corps’ mechanism will not be available until October 2014. As a result, officials are not able to conduct an assessment of personnel to determine whether there are different promotion rates for people who served on advisor teams versus those who did not, which could be a potential indicator that servicemembers with SFA-related experience were negatively affected during the promotion process. We recently reported that the Army’s use of SFA brigades to form advisor teams has enabled it to meet requirements, but this practice results in those brigades leaving behind large numbers of personnel at the brigades’ home stations. Officials from each of the four SFA brigades with whom we met confirmed that this continues to be the case and said that they left behind a large rear detachment when they deployed. For example, according to officials from one of the brigades we visited, the brigade deployed approximately 500 people to create advisor teams, leaving approximately 3,000 personnel behind at their home station. Officials from another brigade that deployed as an SFA brigade stated that approximately 1,900 personnel were deployed, leaving behind about the same number in the rear detachment. In 2013, we reported that because the advisor team requirement calls for high numbers of company- and field-grade officers and senior noncommissioned officers, staffing the teams requires the brigades to deploy a significant portion of their leadership and expertise, including the brigade commanders and many battalion, company, and platoon commanders, for the advisor mission. Army Forces Command officials confirmed that this continues to be the Army’s process and that as a result, in providing these senior servicemembers to the advisor teams, the brigades are forced to rely on more junior officers and enlisted personnel at their home stations to manage the rear detachment. For example, according to SFA brigade officials we interviewed, in some cases they relied on captains, who would typically be responsible for leading a company, to serve as rear- detachment battalion commanders. The rear-detachment commanders and other SFA brigade officials with whom we spoke stated they faced numerous challenges with managing personnel who were left at home stations. Officials from two of the SFA brigades we spoke with stated that their biggest challenge stemmed from effectively managing medical, legal, and other administrative issues while concurrently maintaining readiness and responding to installation tasks. These officials also stated that another challenge they faced was ensuring that equipment was accounted for properly. During this review, officials from the SFA brigades we met with stated that the brigades’ predeployment planning efforts ensured that the brigades left behind sufficient leadership to manage the rear detachments and develop training plans to maintain readiness. A number of other actions were also taken to maintain readiness that included the following: Consolidating Units: According to officials from two of the brigades we spoke with, to ensure that leadership is available to conduct training, some brigades consolidated units. For example, one rear-detachment battalion commander told us that his battalion teamed up with another battalion to ensure they could conduct training to maintain their readiness. Leaving Key Leadership Personnel: Officials told us that leaving the right personnel behind to manage the rear detachment is just as important as deploying the right personnel. Officials from the four SFA brigades we met with stated that brigade leadership had to undertake significant planning to ensure that the right leadership remained at the home station. One SFA brigade official with whom we spoke said that rear-detachment leadership candidates were interviewed prior to being assigned as rear-detachment commanders to ensure that they had the personal and professional attributes needed to effectively manage the rear detachment. Officials from two of the SFA brigades we met with stated that the right leadership was needed to manage the training, legal, medical, and other administrative issues of the soldiers who remained at the home station. Conducting Training and Maintaining Readiness: Army Forces Command issued guidance for the training and employment of rear detachments during advisor team deployments, including missions the force may be assigned to, training expectations, and equipment maintenance responsibilities. According to Army Forces Command officials, Army Forces Command allowed the brigade leadership to determine the type and amount of training that could be conducted based on the number of personnel remaining at the home station. For example, officials from each of the four SFA brigades with whom we met stated that they developed training plans for what the rear detachment would train on while the rest of the brigade was deployed. Reporting Readiness: Prior to October 2013, brigades were required to report only on the status of deployed forces and equipment. The Department of the Army issued guidance in October 2013 to gain better visibility over all available personnel and equipment. It required units to report on personnel and equipment that were deployed and at the rear detachment. However, it is still up to the rear-detachment brigade leadership to determine the level of readiness that they can achieve with the personnel and equipment remaining at the home station. SFA has been and continues to be an enduring part of the U.S. military effort in Afghanistan and other locations around the world. Accordingly, DOD wants to ensure that it has the ability to develop, maintain, and institutionalize the capabilities of servicemembers to provide SFA to foreign military forces. To ensure that the Army and Marine Corps have the right people with the right skill sets to serve in an SFA mission, DOD must be able to identify and track personnel with SFA-related training, education, and experience. Without a systematic means to identify and track personnel with SFA-related training, education, and experience, the Army and Marine Corps (1) cannot monitor the career progression of servicemembers with SFA-related experience; and (2) are limited in their ability to evaluate the effectiveness of their guidance. Currently DOD lacks this capability, and the Army has not developed a plan—including goals and milestones—to complete the implementation and institutionalization of its ability to identify and track personnel with SFA- related experience. Until the Army develops such a plan, it is at risk of not being able to readily identify the right personnel with the right skills and experience to serve in an SFA mission. Further, the absence of such a plan could limit the effectiveness of the advisor teams and DOD’s ability to develop, maintain, and institutionalize the capabilities of servicemembers to provide SFA to foreign military forces. To enable the Army to address the requirement to identify and track personnel with SFA-related experience, we recommend that the Secretary of Defense direct the Secretary of the Army to develop and implement a plan with goals and milestones for how it will develop the means for systematically identifying and tracking personnel with SFA- related experience. We provided a draft of this report to DOD for review and comment. In written comments on a draft of this report, DOD partially concurred with the report’s recommendation. DOD’s comments are summarized below and reprinted in appendix II. DOD partially concurred with the recommendation that the Secretary of Defense direct the Secretary of the Army to develop and implement a plan with goals and milestones for how it will develop the means for systematically identifying and tracking personnel with SFA-related experience. In its letter, DOD stated that it concurred with our underlying assessment that the Army has had difficulty in systematically identifying and tracking personnel with SFA-related experience. However, DOD stated that sufficient guidance and direction exists for the Army to continue refining its processes and procedures. Specifically, DOD stated that the Army, through the application of its Personnel Development Skill Identifier (PDSI), is capable of identifying and designating all soldiers who have satisfactorily completed SFA training. DOD stated that it will continue to work with the Army in this important area. We agree that the Army may have sufficient guidance to continue refining its processes and procedures. As noted in the report, the Army uses PDSI codes to capture certain SFA-related training activities; however, some SFA-related training and education may not be captured. Moreover, as noted in the report, the PDSI codes do not capture SFA-related experience. DODI 5000.68, which was issued in October 2010, requires the Army to, among other things, identify and track individuals with SFA- related training, education, and experience. As noted in the report, Army officials told us that they were discussing the best method to identify and track personnel with SFA-related experience, but the Army has not established a plan with goals and milestones to develop this capability. Such a plan will provide the necessary direction to enable the Army to develop the means for systematically identifying and tracking personnel with SFA-related training, education, and experience, as required by DODI 5000.68. DOD also provided technical comments, which we incorporated as appropriate. We are sending copies of this report to the appropriate congressional committees; the Secretary of Defense; the Secretary of the Army; and the Commandant of the Marine Corps. The report is also available at no charge on GAO’s website at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-5431 or russellc@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix III. The objectives of this engagement were to assess the extent to which (1) the Army and Marine Corps identify and track personnel with SFA-related training, education, and experience; and (2) the Army and Marine Corps consider SFA-related training, education, and experience in the promotion process; and to describe the process the Army uses to prepare units to perform their core missions while some unit members are deployed to support SFA activities. To determine the extent to which the Army and Marine Corps identify and track personnel with SFA-related training, education, and experience, we reviewed relevant policies, directives, and other documents. The documents that we reviewed included the Department of Defense Instruction 5000.68, the 2013 Army Field Manual on Army Support to Security Cooperation, International Security Assistance Force SFA Guide 2.0, and Marine Corps memorandum approving a Free MOS for SFA, as well as other memorandums as appropriate. We also interviewed officials from the Office of the Secretary of Defense, Department of the Army, and Headquarters Marine Corps, and personnel from Army Human Resources Command and Marine Corps Manpower and Reserve Affairs. To determine the extent to which the Army and Marine Corps consider SFA-related training, education, and experience in the promotion process, we analyzed guidance and other documents that identify critical elements of a career path and reviewed guidance provided to promotion boards to instruct them on how to incorporate advisor experience in their evaluation of individuals for promotion. In addition, we interviewed Department of the Army, Headquarters Marine Corps, and service-level officials about the promotion process. We discussed their concerns about the potential negative effect of serving in SFA-related positions on their careers and the careers of others. Information that we received is not generalizable. To describe how the Army prepares units to perform their core missions when some unit members are deployed to support SFA activities, we reviewed documentation and interviewed knowledgeable officials from four SFA brigades, and Army headquarters and service-level officials to ascertain any potential challenges to units having the ability to train and manage their readiness for core missions and mitigation strategies, if any, they developed. As part of this review, we selected an illustrative, nongeneralizable sample of brigades that sent SFA advisor teams to Afghanistan. We worked with Army Forces Command officials to identify brigades that recently deployed as SFA Brigades in Afghanistan and left a large rear detachment. Army Forces Command provided us with a list of five brigades that had recently served as SFA brigades in Afghanistan and had left a large rear detachment. From the list that it provided, we contacted the five brigades, but only four of them were available for us to interview. We could not conduct interviews with the fifth one due to conflicts with its deployment. From the four brigades that left a large rear detachment when they deployed and that were available to meet with us, we interviewed officials from the division, brigade, and battalion levels that were responsible for managing the rear detachment at each of the brigades we visited. We obtained testimonial evidence from the four selected brigades to ascertain any potential readiness challenges presented by the need to meet SFA requirements. Officials from these brigades were also able to provide information about their perceptions of the effect that serving in an SFA-related role has on promotion opportunities. We did not report on the readiness of the rear detachment for the Marine Corps because the Marine Corps’ sourcing approach does not create the same type of large rear detachments that result from the Army’s use of SFA brigades. We visited or contacted officials from the following organizations during our review: Office of the Secretary of Defense, Arlington, Virginia Personnel & Readiness Military Personnel and Policy Security Force Assistance Steering Committee and Working Department of the Army Headquarters, Office of the Deputy Chief of Staff for Personnel (G-1), Arlington, Virginia Department of the Army Headquarters, Operations and Plans (G- 3/5/7), Arlington, Virginia Human Resources Command, Fort Knox, Kentucky 101st Airborne Division, Fort Campbell, Kentucky A. B. C. D. 1/101st Brigade Combat Team 2/101st Brigade Combat Team 3/101st Brigade Combat Team 1/101st Combat Aviation Brigade 3rd Infantry Division, Fort Stewart, Georgia 1st Infantry Division, Regionally Aligned Forces, Fort Riley, Kansas United States Marine Corps Marine Corps Security Cooperation Group, Arlington, Virginia Manpower and Reserve Affairs Department, Quantico, Virginia We conducted this performance audit from August 2013 to July 2014 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. In addition to the contact named above, James A. Reynolds, Assistant Director; Emily Biskup; Richard Burkard; Michael Silver; Sonja Ware; and Cheryl Weissman made key contributions to this report. Security Force Assistance: More Detailed Planning and Improved Access to Information Needed to Guide Efforts of Advisor Teams in Afghanistan. GAO-13-381. Washington, D.C.: April 30, 2013. Building Partner Capacity: Key Practices to Effectively Manage Department of Defense Efforts to Promote Security Cooperation. GAO-13-335T. Washington, D.C.: February 14, 2013. Afghanistan: Key Oversight Issues. GAO-13-218SP. Washington, D.C.: February 11, 2013. Afghanistan Security: Long-standing Challenges May Affect Progress and Sustainment of Afghan National Security Forces. GAO-12-951T. Washington, D.C.: July 24, 2012. Security Force Assistance: Additional Actions Needed to Guide Geographic Combatant Command and Service Efforts. GAO-12-556. Washington, D.C.: May 10, 2012. Iraq and Afghanistan: Actions Needed to Enhance the Ability of Army Brigades to Support the Advising Mission. GAO-11-760. Washington, D.C.: August 2, 2011. | SFA—DOD activities that contribute to supporting the development of foreign security forces and their supporting institutions—is a key component of U.S. efforts to create sustainable security around the world. These activities are carried out by DOD personnel serving as advisors who may have SFA-related training, education, and prior experience to conduct the advising mission. GAO was mandated to review the Army's and Marine Corps' approaches to the SFA mission. GAO assessed the extent to which the Army and Marine Corps (1) identify and track personnel with SFA-related training, education, and experience and (2) consider SFA-related training, education, and experience in the promotion process. This report also describes the Army's process for preparing units to perform their core mission while some members are deployed to support SFA activities. GAO reviewed DOD policies, directives, and other documents and interviewed cognizant DOD and service officials. The Army and Marine Corps have taken steps to identify and track personnel with Security Force Assistance (SFA)-related training, education, and experience, but the Army does not have a plan with goals and milestones to fully capture advising experience. A key element in conducting SFA missions is being able to identify the right personnel with the right training, education, and experience to execute SFA activities. The Department of Defense (DOD) established a requirement for the services to identify and track personnel with SFA-related training, education, and experience. Although the Army is able to identify and track soldiers with certain SFA-related training and education, it does not have a mechanism to identify and track SFA-related experience. Moreover, the Army has not developed a plan with goals and milestones on how it will capture this information. As a result, it is unclear how long it will take the Army to implement DOD's requirement and be able to readily identify the right personnel to serve in the SFA mission. The Marine Corps is implementing a mechanism to identify and track personnel with SFA-related training, education, and experience that is planned to be available in October 2014. The Army and the Marine Corps have taken steps to ensure the consideration of SFA-related training, education, and experience in the promotion process. Both services have incorporated language into the guidance given to promotion boards to ensure that appropriate consideration is given to individuals who have served on advisor teams. However, opinions differ regarding the effect the guidance has had on the manner in which individuals with SFA-related training, education, and experience are considered during the promotion process. Some officials said that serving in an SFA role could potentially negatively affect career progression. Others noted that the promotion process considers the entirety of an individual's career and not just the time spent as an SFA advisor. However, because the Army and Marine Corps do not yet have comprehensive information on SFA advisors, officials are unable to determine whether promotion rates differ for people who served on SFA advisor teams versus those who did not. Such differences could be a potential indicator that servicemembers with SFA-related experience were negatively affected during the promotion process. According to Army officials, the Army has taken a number of actions to ensure that units are prepared to perform their core mission when part of the unit is deployed in support of SFA activities. Officials stated that some actions taken to manage units that remain at the home station, also known as the rear detachment, when part of the unit deploys include consolidating units so that they can conduct training, ensuring the right personnel are left behind to lead the rear detachment, developing training plans for what the rear detachment would train to, and reporting on the availability of all personnel and equipment. GAO recommends that the Army develop and implement a plan, with goals and milestones, for how it will develop the means for systematically identifying and tracking personnel with SFA-related experience. DOD partially concurred with the recommendation, stating that sufficient guidance and direction exists for the Army to continue refining its processes and procedures. GAO continues to believe the recommendation is valid, as discussed in the report. |
Title XIX of the Social Security Act is a federal and state entitlement program that pays for medical assistance for certain categories of low- income adults and children. This program, known as Medicaid, became law in 1965 and is jointly funded by the federal and state governments (including the District of Columbia and the Territories). Medicaid is the largest source of funding for medical and health-related services for America’s poorest people. More than 50 million persons enrolled in the Medicaid program in fiscal year 2006. In fiscal year 2006, according to CMS, total outlays for Medicaid (federal and state) were approximately $324 billion, of which about $185 billion was paid by the federal government. Medicaid is jointly funded by the federal and state governments. The federal government shares in a state’s Medicaid service costs through a matching formula. The federal matching rate for the cost of services provided to Medicaid beneficiaries is related to a state’s per capita income and in federal fiscal year 2006 ranged from 50 percent to 76 percent. Although the federal government establishes general guidelines for the Medicaid program, requirements are established by each state. CMS, within the Department of Health and Human Services (HHS), is responsible for administering federal matching funds to the states and for legislation and regulations affecting the Medicaid program. CMS also provides guidelines, technical assistance, and periodic assessments of state Medicaid programs. Title XIX of the Social Security Act allows considerable flexibility within the states’ Medicaid plans. Within broad national guidelines established by federal statutes, regulations, and policies, each state (1) establishes its own eligibility standards; (2) determines the type, amount, duration, and scope of services; (3) sets the rate of payment for services; and (4) administers its own program—including enrollment of providers. Medicaid policies for eligibility, services, and payment are complex and vary considerably, even among states of similar size or geographic proximity. Thus, a person who is eligible for Medicaid in one state may not be eligible in another state, and the services provided by one state may differ considerably in amount, duration, or scope from services provided in a similar or neighboring state. In addition, state legislatures may change Medicaid eligibility, services, or reimbursement during the year. To receive payment for services or goods provided to beneficiaries from Medicaid, providers must first enroll in the Medicaid program. To enroll, providers must submit a Medicaid enrollment application to the state or their fiscal agents who are responsible for determining whether the providers meet federal and state requirements for enrollment. The state or its fiscal agents are responsible for screening the applications based on CMS and state policies. Once an applicant is deemed eligible by the state or its fiscal agents, Medicaid providers can submit their claims to the state for payment. The state is responsible for claims processing and verifying the claim is accurate, complete, medically necessary, and covered under the state’s Medicaid plan. After the claim is approved, the state pays the claim. The typical Medicaid payment process is illustrated in figure 1. When a Medicaid beneficiary receives care from a health care provider such as a hospital, physician, or nursing home, the provider bills the state Medicaid program for its services. The state in turn pays the provider from a combination of state funds and federal funds, which have been advanced by CMS each quarter. The state then files an expenditure report, in which it claims the federal share of the Medicaid expenditure as reimbursement for its payment to providers and reconciles its total expenditures with the federal advance. In addition to reimbursement for medical services, the state may claim federal reimbursement for functions it performs to administer its Medicaid program, such as enrolling new beneficiaries; reviewing the appropriateness of providers’ claims; and collecting payments from third parties, which are payers other than Medicaid, such as Medicare, that may be liable for some or all of a particular health claim. Transaction 1: A Medicaid beneficiary receives care from a health care provider, such as a physician; the provider bills the state Medicaid program; and the state pays the provider by drawing on a pool of state funds combined with a quarterly advance on federal matching funds. Transaction 2: The state files an expenditure report, in which it claims the federal Medicaid matching share as reimbursement for its payments to providers and reconciles total quarterly expenditures with the federal advance. States may file claims for medical services and for administrative functions. Our analysis found that over 30,000 Medicaid providers at the selected states had over $1 billion in unpaid federal taxes as of September 30, 2006. This represents over 5 percent of the approximately 560,000 Medicaid providers paid by the selected states during federal fiscal year 2006. The amount of unpaid federal taxes we identified among Medicaid providers is likely understated because (1) we intentionally limited our scope to providers with agreed-to federal tax debt for tax periods prior to 2006, and (2) the IRS taxpayer data reflect only the amount of unpaid taxes either reported by the taxpayer on a tax return or assessed by IRS through its various enforcement programs and thus the unpaid tax debt amount does not include entities for which IRS had not identified that they did not file tax returns or underreported their income. As shown in figure 2, 87 percent of the approximately $1 billion in unpaid taxes was comprised of individual income and payroll taxes. The other 13 percent of taxes included corporate income, excise, unemployment, and other types of taxes. As shown in figure 2, over half of the unpaid taxes owed by Medicaid providers were payroll taxes. Employers are subject to civil and criminal penalties if they do not remit payroll taxes to the federal government. When an employer withholds taxes from an employee’s wages, the employer is deemed to have a responsibility to hold these amounts “in trust” for the federal government until the employer makes a federal tax deposit in that amount. To the extent these withheld amounts are not forwarded to the federal government, the employer is liable for these amounts, as well as the employer’s matching Federal Insurance Contribution Act contributions for Social Security and Medicare. Individuals within the business (e.g., corporate officers) may be held personally liable for the withheld amounts not forwarded and they maybe assessed a civil monetary penalty known as a trust fund recovery penalty (TFRP). Willful failure to remit payroll taxes can also be a criminal felony offense punishable by imprisonment up to 5 years, while the failure to properly segregate payroll taxes can be a criminal misdemeanor offense punishable by imprisonment of up to 1 year. The law imposes no penalties upon an employee for the employer’s failure to remit payroll taxes since the employer is responsible for submitting the amounts withheld. The Social Security and Medicare trust funds are subsidized or made whole for unpaid payroll taxes by the general fund. Thus, personal income taxes, corporate income taxes, and other government revenues are used to pay for these shortfalls to the Social Security and Medicare trust funds. A substantial amount of the unpaid federal taxes shown in IRS records as owed by Medicaid providers had been outstanding for several years. As reflected in figure 3, about 56 percent of the $1 billion in unpaid taxes was for tax periods from calendar year 2000 through calendar year 2004, and approximately 29 percent of the unpaid taxes was for tax periods prior to calendar year 2000. Our previous work has shown that as unpaid taxes age, the likelihood of collecting all or a portion of the amount owed decreases. This is due, in part, to the continued accrual of interest and penalties on the outstanding tax debt, which, over time, can dwarf the original tax obligation. The amount of unpaid federal taxes reported above does not include all tax debts owed by Medicaid providers due to statutory provisions that give IRS a finite period under which it can seek to collect on unpaid taxes. There is a 10-year statute of limitations beyond which IRS is prohibited from attempting to collect tax debt. Consequently, if the Medicaid providers owe federal taxes beyond the 10-year statutory collection period, the older tax debt may have been removed from IRS’s records. We were unable to determine the amount of tax debt that had been removed. Although over $1 billion in unpaid federal taxes owed by Medicaid providers as of September 30, 2006, is a significant amount, it likely understates the full extent of unpaid taxes owed by these or other businesses and individuals. The IRS tax database reflects only the amount of unpaid federal taxes either reported by the individual or business on a tax return or assessed by IRS through its various enforcement programs. The IRS database does not reflect amounts owed by businesses and individuals that have not filed tax returns and for which IRS has not assessed tax amounts due. For example, during our audit, we identified instances from our case studies in which Medicaid providers failed to file tax returns for a particular tax period and IRS had not assessed taxes for these tax periods. Consequently, while these providers had unpaid federal taxes, they were listed in IRS records as having no unpaid taxes for that period. Further, our analysis did not attempt to account for businesses or individuals that purposely underreported income and were not specifically identified by IRS as owing the additional federal taxes. According to IRS, underreporting of income accounted for more than 80 percent of the estimated $345 billion annual gross tax gap. Finally, our analysis did not attempt to identify Medicaid providers who owed taxes under a separate TIN from the TIN that received the Medicaid payments. For example, sole proprietors and certain limited liability companies may file Medicaid claims under their Social Security numbers (SSN). If these Medicaid providers had employees, they would typically report the payroll taxes under an employer identification number (EIN) and not their SSNs. Consequently, the full extent of unpaid federal taxes for Medicaid providers is not known. In addition to the IRS tax database not reflecting all assessed tax amounts due, our past audits have also found that the IRS tax database contains coding errors that adversely affect IRS’s collection activities. IRS’s collection process is heavily dependent upon its automated computer system and the information that resides within this system. In particular, the codes in each taxpayer’s account in IRS’s tax database are critical to IRS in tracking the collection actions it has taken against a tax debtor and in determining what, if any, additional collection actions should be pursued. For example, IRS uses these codes to identify cases it should exclude from the continuous levy program, which is an automated method of collecting tax debt by offsetting certain federal payments made to individuals and businesses, as well as cases it should exclude from other collection actions. For all 25 cases that we audited and investigated, we confirmed that their activities were abusive and in many instances found criminal activity related to the federal tax system. Of these cases, 17 involved businesses with employees who had unpaid payroll taxes, most dating as far back as the late 1990s. However, rather than fulfill their role as “trustees” of this money and forward it to IRS, these Medicaid providers diverted the money for other purposes, including their own salaries. As stated earlier, willful failure to remit payroll taxes can be a criminal felony offense punishable by imprisonment up to 5 years, while the failure to properly segregate payroll taxes can be a criminal misdemeanor offense punishable by imprisonment of up to 1 year. Table 1 highlights 10 cases of businesses and individuals with unpaid taxes. Our investigations revealed that, despite their businesses owing substantial amounts of taxes to IRS, some owners had substantial personal assets—including expensive homes and luxury cars. We are referring the 25 cases detailed in our report to IRS for appropriate collection action and criminal investigation. The following provides illustrative detailed information on four cases we audited and investigated. Case 1: During the time the owners of the hospital owed over $5 million in payroll taxes, the owners purchased a vacation home worth about $1 million. IRS assessed a trust fund recovery penalty (TFRP) of nearly $2 million against the owners, filed federal tax liens totaling nearly $8 million against the owners and hospital, attempted to levy the owners’ bank accounts, and proposed an injunction to close the hospital because the business continued to accumulate tax debt. The hospital received over $9 million in Medicaid payments during fiscal year 2006. Case 2: While owing over $2 million in unpaid payroll taxes, the nursing home owner and business were fined for jeopardizing the health and safety of their patients. The nursing home owner attempted to sell the business and other real estate property and promised to pay tax debts in full. However, the owner did not sell the business or real estate and took other actions to avoid federal tax liens. IRS fined the owner over $400,000 in a recent year for intentionally disregarding IRS’s tax reporting and filing requirements. The owner also has a related business that owes over $1 million in unpaid taxes. The nursing home received over $6 million in Medicaid payments during fiscal year 2006. Case 4: The managing officer of a pharmacy sold off business assets without notifying IRS while knowing that the business owed over $800,000 in unpaid payroll taxes over 7 years. In an attempt to collect unpaid debts from the officer, IRS assessed a TFRP of nearly $3 million and filed federal tax liens against the officer and the business. The officer owns a related entity that also owes a large amount of taxes, and recently started up a new corporation using the same address as the pharmacy. The pharmacy received nearly $100,000 in Medicaid payments during fiscal year 2006. Case 8: A medical clinic owner owns a house worth nearly $4 million, several luxury vehicles, and a pleasure boat while owing taxes. The owner also borrowed over $2 million from the business and sold properties for about $1 million at the same time the business owed over $1 million in unpaid payroll taxes. In addition, IRS generated a tax return for the business in a recent year because the business owner did not file it. The medical clinic received over $2 million in Medicaid payments during fiscal year 2006. In addition to the 25 cases that we identified through IRS tax records, we separately also found a Medicaid provider that was recently convicted for failure to pay employment taxes owed by several nursing homes. The nursing home businesses received over $25 million in Medicaid payments during fiscal year 2006. According to court documents, the nursing homes owed over $14 million in unpaid taxes. At the same time the businesses owed taxes, the owner bought a 10,000 square foot house with a current estimated value of over $2 million. The court records indicate that the owner spent tens of thousands of dollars furnishing the house including crystal chandeliers, a 132-piece set of Haviland Bavarian porcelain china, and oriental rugs. The owner used company funds to pay personal expenses such as a housekeeper, children’s nanny, monthly pension for a parent who never worked at the company, a sailboat, and jet-skis. While owing taxes, the owner also went on vacations to Hawaii and gambling trips to Las Vegas and Reno, Nevada. Court records also indicate that while in Hawaii, the owner bought a $16,000 Rolex watch, the day before one of the required federal tax deposits was due. CMS and the selected states do not prevent health care providers who have tax debts from enrolling in or receiving payments from Medicaid. CMS has not developed regulations to require states to (1) screen health care providers for unpaid taxes and (2) obtain consent for IRS disclosure of federal tax debts. CMS officials stated that the primary focus of the Medicaid program is to provide health care services for low income people and not the administration of taxes. Further, federal law generally prohibits the disclosure of taxpayer data to CMS and states and thus, CMS and states do not have access to tax data directly from IRS unless the taxpayer provides consent. Further, none of the seven states we contacted have ever implemented a continuous federal tax levy for Medicaid payments. Thus, Medicaid payments to providers that owe federal taxes are not being continuously levied. Federal law does not prohibit providers with unpaid federal taxes from enrolling in and billing Medicaid. Federal regulations and policies require the states, as part of their responsibilities for determining whether the providers meet Medicaid requirements for enrollment, to verify basic information on potential providers, including whether the providers meet state licensure requirements and whether the providers are prohibited from participating in federal health care programs. However, federal regulations and policies do not require the states to screen these providers for federal tax delinquency nor do they explicitly authorize the states to reject the providers that have delinquent tax debt from participation in Medicaid. CMS officials stated that the primary focus of the Medicaid program is to provide health care services for low income people and not the administration of taxes. CMS officials stated that such a requirement could be a burden to the states in their enrollment of providers and could adversely impact states’ ability to provide health care to the poor. Consequently, the selected states’ processes generally do not consider federal tax debts of prospective providers in the Medicaid enrollment process. Further, due to a statutory restriction on disclosure of taxpayer information, even if tax debts specifically were to be considered in enrollment in Medicaid, no coordinated or independent mechanism exists for the states to obtain complete information on providers that have unpaid tax debt. Federal law does not permit IRS to disclose taxpayer information, including tax debts, to CMS or Medicaid state officials unless the taxpayer consents, which neither CMS nor the states currently seek. Thus, certain tax debt information can only be discovered from public records if IRS files a federal tax lien against the property of a tax debtor or if a record of conviction for tax offense is publicly available. Consequently, CMS and state officials do not have ready access to information on unpaid tax debts to consider in making decisions on Medicaid providers. Although a provision of the Taxpayer Relief Act of 1997 authorizes IRS to continuously levy certain federal payments made to delinquent taxpayers, no tax debt owed by Medicaid providers has ever been collected using this provision of the law. In the 10 years since its passage, IRS had not determined whether Medicaid payments are federal payments and thus subject to the continuous levy program or determined the feasibility of incorporating these payments into the program. If there had been an effective levy program in place, we estimate that the selected states could have levied payments for the federal government and collected between $70 million to about $160 million of unpaid federal taxes during fiscal year 2006. This estimate was based on those debts that IRS reported to the Treasury Offset Program (TOP) as of September 30, 2006. Officials from all these selected states stated that they have a continuous levy program to offset Medicaid payments against their state debts. Available data indicate that the vast majority of Medicaid providers appear to pay their federal taxes. However, our work has shown that over 30,000 Medicaid providers have taken advantage of the opportunity to avoid paying their federal taxes. While Medicaid providers are relied on to deliver significant medical services to those most in need, they must also pay their fair share of federal taxes. Many of the individuals involved in our cases have consistently not paid their taxes yet have received millions of dollars in Medicaid payments and have faced no criminal consequences. At the same time, some of these individuals are living lives of luxury, financed in part by Medicare and Medicaid payments. Also, IRS has taken little action to explore the continuous levy of Medicaid payments, which over time potentially could have resulted in millions of dollars of collections or to aggressively pursue collection and criminal investigation of the individuals involved in our 25 case studies. We recommend that the Commissioner of the Internal Revenue Service take the following two actions: Conduct a study to determine whether Medicaid payments can be incorporated in the continuous levy program. Evaluate the 25 referred cases detailed in this report for appropriate additional aggressive collection action and criminal investigation as warranted. We received written comments on a draft of this report from the Acting Commissioner of Internal Revenue (see app. III) and Acting Administrator of CMS (see app. IV). We also received an e-mail response from FMS. IRS concurred with our recommendations. In response to our recommendation that IRS conduct a study to determine whether Medicaid payments can be incorporated in the continuous levy program, IRS stated that both a subgroup of the Federal Contractors Tax Compliance (FCTC) task force and IRS General Counsel have completed an independent study on whether the Medicaid payments can be incorporated into the continuous levy program. Both the FCTC task force and IRS General Counsel concluded that Medicaid disbursements do not qualify as federal payments and therefore cannot be incorporated in the continuous levy program. In response to a draft of our report, CMS expressed concern about the tone and language we used to discuss our findings. Specifically, CMS interpreted our finding that over 30,000 Medicaid providers had over $1 billion of unpaid federal taxes as implying that "there is some direct correlation between owing taxes and being a Medicaid provider." Our report clearly states that the vast majority of Medicaid providers are paying their taxes. For the 5 percent of Medicaid providers with tax debt, we simply reported on the facts of what we found, which do not require additional evaluation to satisfactorily address our objective. Furthermore, regarding our third objective, CMS interpreted our report as implying that there is an underlying connection between the activity (preventing providers with tax problems from participating in the Medicaid program) and the authority and responsibility to perform such activity. Again, it appears that CMS misinterpreted our findings. We specifically stated that federal law does not prohibit providers with unpaid taxes from enrolling in and billing Medicaid. Although CMS is not required to screen potential providers for tax debts, we are concerned that CMS stated it would be inappropriate to prevent medical providers that owe federal taxes participating in the Medicaid program—which would presumably include those egregious cases we identified in this report. We believe that any CMS action to prevent medical providers who refuse to pay their taxes from participating in the Medicaid program would help ensure the integrity of the Medicaid program and does not necessarily conflict with CMS's role in providing health care to low-income individuals. Both CMS and FMS expressed concern with their agencies involvement in the continuous levy program. CMS stated that we implied that CMS and the Medicaid agencies should be conducting the continuous levy on these payments. FMS stated that our report indicated that, because Medicaid payments include funds the states receive from the federal government, the Medicaid payment is a federal payment. Our report did not state that CMS and the Medicaid agencies should be conducting the continuous levy on Medicaid payments nor did we state that Medicaid payments are federal payments. However, we did report that IRS had not determined whether Medicaid payments are federal payments and recommended that IRS conduct a study to determine whether Medicaid payments can be incorporated in the continuous levy program. CMS and FMS also provided us technical corrections to the report which we incorporated, as appropriate. As agreed with your office, unless you publicly release its contents earlier we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copies of this report to the Secretary of the Treasury, the Commissioner of the Financial Management Service (FMS), the Acting Commissioner of Internal Revenue, the Acting Administrator of Centers for Medicare & Medicaid Services (CMS) and interested congressional committees. The report is also available at no charge on the GAO Web site at http://www.gao.gov. If you have any questions concerning this report, please contact Gregory D. Kutz at (202) 512-6722 or kutzg@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. To identify the magnitude of unpaid federal taxes owed by Medicaid providers, we used a nonrepresentative selection of states. We selected the states of California, Colorado, Florida, Maryland, New York, Pennsylvania, and Texas based on the magnitude of payments made to Medicaid providers and the geographical location of those states. We obtained and analyzed Internal Revenue Service (IRS) tax debt data as of September 30, 2006. We also obtained and analyzed the selected states’ federal fiscal year 2006 approved Medicaid payments to providers. We matched the Medicaid payment data to the IRS unpaid assessment data using the taxpayer identification number (TIN) field. To avoid overestimating the amount owed by Medicaid providers with unpaid tax debts and to capture only significant tax debts, we excluded from our analysis tax debts and paid claims meeting specific criteria to establish a minimum threshold for the amount of tax debt and for the amount of paid claims to be considered when determining whether a tax debt was significant. The criteria we used to exclude tax debts are as follows: tax debts that IRS classified as compliance assessments or memo accounts for financial reporting, tax debts from calendar year 2006 tax periods, and Medicaid providers with total unpaid taxes and Medicaid paid claims of less than $100. These criteria were used to exclude tax debts that might be under dispute or generally duplicative or invalid, and tax debts that were recently incurred. Specifically, compliance assessments or memo accounts were excluded because these taxes have neither been agreed to by the taxpayers nor affirmed by the court, or these taxes could be invalid or duplicative of other taxes already reported. We excluded tax debts from calendar year 2006 tax periods to eliminate tax debt that may involve matters that are routinely resolved between the taxpayer and IRS, with the taxes paid or abated within a short period. We further excluded tax debts and Medicaid-paid claims of less than $100 because they are insignificant for the purpose of determining the extent of taxes owed by Medicaid providers. Our analysis also did not attempt to identify Medicaid providers who owed taxes under a separate TIN from the TIN under which the Medicaid payments were received. As a result, the full extent of unpaid federal taxes for Medicaid providers is understated. To identify indications of abuse or potentially criminal activity, we selected 25 Medicaid providers for a detailed audit and investigation. The 25 providers were chosen using a nonrepresentative selection approach based on our judgment, data mining, and a number of other criteria. Specifically, we narrowed down providers to 25 with unpaid taxes based on the amount of unpaid taxes, number of unpaid tax periods, amount of payments reported by Medicaid, and indications that owner(s) might be involved in multiple companies with tax debts. For these 25 cases, we obtained copies of automated tax transcripts and other tax records (for example, revenue officer’s notes) from IRS and performed additional searches of criminal, financial, and public records. In cases where record searches and IRS tax transcripts indicated that the owners or officers of a business were involved in other related entities that have unpaid federal taxes, we also reviewed the related entities and the owner(s) or officer(s), in addition to the original business we identified. In instances where we identified related parties that had both Medicaid payments and tax debts, our case studies included those related entities, combining unpaid taxes and combined Medicaid payments for the original individual/business as well as all related entities. Because our investigations were generally limited to publicly available information, our audit of the 25 cases may not have identified all related parties or all significant assets (i.e., personal bank data, companies established to hide assets) that the Medicaid providers own. To determine the extent to which Centers for Medicare & Medicaid Services (CMS) officials and the states are required to consider tax debts or other criminal activities in the enrollment of providers into Medicaid, we examined CMS policies and procedures, Medicaid regulations, and the selected policies for enrollment. We also discussed policies and procedures used to enroll providers into Medicaid with officials from the selected states. As part of these discussions, we inquired whether the selected states specifically consider tax debts or perform background investigations to determine whether a prospective provider is qualified before the enrollment to Medicaid is granted. To determine the extent to which Medicaid payments to providers are continuously levied to pay tax debts, we examined the statutory and regulatory authorities that govern the continuous levy program and interviewed officials from CMS, IRS, and Department of the Treasury’s Financial Management Service (FMS) to determine whether any legal barriers existed. To determine the potential levy collections on Medicaid payments during fiscal year 2006, we used 15 percent and 100 percent of the total paid claim or total tax debt amount reported to the Treasury Offset Program (TOP), whichever was less. A gap will exist between what could be collected and the maximum levy amount calculated because (1) tax debts in TOP may not be eligible for immediate levy because IRS has not completed due process notifications and (2) tax debts may become ineligible for levy because of a change in collection status (e.g., tax debtor filed for bankruptcy). We conducted our audit work from July 2006 through August 2007 in accordance with U.S. generally accepted government auditing standards, and we performed our investigative work in accordance with standards prescribed by the President’s Council on Integrity and Efficiency. To determine the reliability of the IRS unpaid assessments data, we relied on the work we performed during our annual audits of IRS’s financial statements. While our financial statement audits have identified some data reliability problems associated with the coding of some of the fields in IRS’s tax records, including errors and delays in recording taxpayer information and payments, we determined that the data were sufficiently reliable to address this report’s objectives. Our financial audit procedures, including the reconciliation of the value of unpaid taxes recorded in IRS’s masterfile to IRS’s general ledger, identified no material differences. For the selected states’ Medicaid payment databases and FMS’s TOP databases, we interviewed the selected states’ and FMS officials responsible for their respective databases. In addition, we performed electronic testing of specific data elements in the databases that we used to perform our work. On the basis of our discussions with agency officials, review of agency documents, and our own testing, we concluded that the data elements used for this testimony were sufficiently reliable for our purposes. This appendix presents summary information on the abusive or potentially criminal activity associated with 15 of our 25 case studies. Table 2 shows the remaining case studies that we audited and investigated. As with the 10 cases discussed in the body of this report, we also found substantial abuse and potentially criminal activity related to the federal tax system during our review of these 15 Medicaid providers that also received Medicaid payments in federal fiscal year 2006. The case studies involving businesses primarily involved unpaid payroll taxes. In addition to the contact named above, the following individuals made major contributions to this report: Matthew Valenta, Assistant Director; Erika Axelson; Ray Bush; Jeremiah Cockrum; Bill Cordrey; Kenneth Hill; John Kelly; Tram Le; Barbara Lewis; Andrew McIntosh; John Ryan; Steve Sebastian; Robert Sharpe; Barry Shillito; Pat Tobo; and Jenniffer Wilson made key contributions to this report. | In fiscal year 2006, outlays for Medicaid were about $324 billion; about $185 billion was paid by the federal government. Because GAO previously identified abusive and criminal activity associated with government contractors owing billions of dollars in federal taxes, the subcommittee requested GAO expand our work to Medicaid providers. GAO was asked to (1) determine if Medicaid providers have unpaid federal taxes, and if so, the magnitude of such debts; (2) identify examples of Medicaid providers that have engaged in abusive or criminal activities; and (3) determine whether the Centers for Medicare & Medicaid Services (CMS) and the states prevent health care providers with tax problems from enrolling in Medicaid or participating in the continuous levy program to pay federal tax debts. To perform this work, GAO analyzed tax data from the Internal Revenue Service (IRS) and Medicaid data from seven selected states based on magnitude of Medicaid payments and geography. GAO also performed additional investigative activities. Over 30,000 Medicaid providers, about 5 percent of those paid in fiscal year 2006, had over $1 billion of unpaid federal taxes. These 30,000 providers were identified from a nonrepresentative selection of providers from seven states: California, Colorado, Florida, Maryland, New York, Pennsylvania, and Texas. This $1 billion estimate is likely understated because some Medicaid providers have understated their income or not filed their tax returns. We selected 25 Medicaid providers with high federal tax debt as case studies for more in-depth investigation of the extent and nature of abuse and criminal activity. For all 25 cases we found abusive and related criminal activity, including failure to remit individual income taxes or payroll taxes to IRS. Rather than fulfill their role as "trustees" of federal payroll tax funds and forward them to IRS, these providers diverted the money for other purposes. Willful failure to remit payroll taxes is a felony under U.S. law. Individuals associated with some of these providers diverted the payroll tax money for their own benefit or to help fund their businesses. Many of these individuals accumulated substantial assets, including million-dollar houses and luxury vehicles, while failing to pay their federal taxes. In addition, some case studies involved businesses that were sanctioned for substandard care of their patients. Despite their abusive and criminal activity, these 25 providers received Medicaid payments ranging from about $100,000 to about $39 million in fiscal year 2006. CMS and our selected states do not prevent health care providers who have federal tax debts from enrolling in Medicaid. CMS officials stated that such a requirement for screening potential providers for unpaid taxes could adversely impact states' ability to provide health care to low income people. Further, federal law generally prohibits the disclosure of taxpayer data to CMS and states. No tax debt owed by Medicaid providers has ever been collected through the continuous levy program. During our audit, IRS had not made a determination on whether Medicaid payments are considered "federal payments" and thus eligible for its continuous levy program. For fiscal year 2006, if an effective levy was in place for the seven selected states, GAO estimates that the federal government could have collected between $70 million and $160 million. |
In 1980, Congress enacted the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). This act, which created the Superfund program, was intended primarily to clean up those sites considered to be the most serious of the hazardous waste sites identified in the United States. As of March 7, 1995, EPA reported 15,723 sites in its inventory, of which 1,363 are considered the most hazardous. EPA is authorized to compel parties responsible for causing the hazardous waste pollution, such as waste generators or haulers and site owners or operators, to clean up the sites. If these parties, known as potentially responsible parties (PRPs), cannot be found, or if a settlement cannot be reached, EPA can conduct the cleanup. EPA uses funds from a trust fund established by CERCLA when it performs such cleanups. This trust fund, currently authorized at $15.2 billion, is financed primarily by a tax on crude oil and certain chemicals and by an environmental tax on corporations. After completing a cleanup, EPA can take action against the responsible parties to recover costs and replenish the fund. These costs can cover such items as EPA cleanup studies, removal actions, and program administration, as well as costs incurred by other agencies, such as the Department of Justice, in helping to administer the Superfund program. The process of recovering costs includes (1) conducting searches to identify the PRP(s) and assessing their liability and financial viability, (2) issuing both notice and demand letters to the PRP(s) for the recovery of costs, and (3) if warranted, initiating judicial action with the assistance of the Department of Justice, if the PRP(s) decide not to participate in negotiations to settle the case or if negotiations are unsuccessful. These steps must be completed within specified time periods that are cited in CERCLA. EPA has reported expenditures of over $10.1 billion for cleaning up nonfederal Superfund sites through fiscal year 1994. Barring major changes to the program, we estimate that such sites may cost the federal government about $37 billion more between 1995 and 2019 (in 1993 discounted present-value dollars). EPA’s cost recovery workload has grown substantially over the years as cleanups have been completed and recoveries of costs have been sought from responsible parties. As of January 1995, EPA reported it had pursued actions to recover costs for 1,625 sites. Through the end of fiscal year 1994, EPA reported that the Superfund program had about $1.4 billion in binding agreements from responsible parties to reimburse the federal government. About $934 million of this amount had actually been collected, including about $9 million in fines and penalties. The remaining $8.7 billion of Superfund past costs include costs such as those not pursued, unrecoverable costs, and costs currently being sought through litigation. Although Superfund was enacted 15 years ago, the bulk of EPA’s cost recovery settlements has occurred in the last 7 years. For example, during the first 8 years of the program, cost recovery activities resulted in binding cost recovery agreements totaling about $104 million. In contrast, such binding agreements in fiscal year 1994 alone totaled about $207 million. EPA’s cost recovery workload to recover cleanup costs is likely to increase because the number of Superfund sites is expected to grow. In November 1994, we reported that between 2,500 and 2,800 nonfederal sites could be added to the then inventory of about 1,200 sites that were considered to be the most serious. After EPA has identified PRPs that are liable and able to pay, the success of EPA’s cost recovery efforts depends in large part on the ability of staff to access accurate and complete cost data and related supporting documentation. For a typical cost recovery case, EPA may amass thousands of pages of (1) documents identifying work that was authorized and performed, referred to as work-performed documents and (2) financial documents, including travel vouchers and contract-related documents, showing site costs that were invoiced, approved, and paid. EPA has a number of financial and records management information systems to help support its cost recovery efforts. For instance, EPA operates two financial information systems to maintain Superfund cost data and two more to generate reports: the Integrated Financial Management System (IFMS), the agency’s official financial information system, which contains all of the agency’s core financial data since March 1989; the Financial Management System (FMS), the predecessor system to IFMS, which contains financial data both before and after the implementation of IFMS in March 1989. the Management and Accounting Reporting System, which is used to produce reports from IFMS data; and the Software Program for Unique Reports, the reporting system for FMS, which generates reports containing both IFMS and FMS data. According to EPA officials, the functionality of the Financial Management System and the Software Program for Unique Reports will be completely replaced by IFMS and the Management and Accounting Reporting System as of October 1, 1995. EPA also has two information management systems developed specifically to support Superfund cost recovery: the Superfund Cost Recovery Image Processing System (SCRIPS), which allows cost recovery staff to electronically capture, store, display, and print images of original Superfund financial documents, such as contract invoices, travel vouchers, and payroll records; and the Superfund Cost Organization and Recovery Enhancement System, which is designed to organize and edit financial information into easy-to-read cost summaries. EPA’s Office of Solid Waste and Emergency Response has overall responsibility for the Superfund program. Other key EPA organizations with Superfund responsibilities include (1) the Office of Enforcement and Compliance Assurance, which is responsible for enforcement actions, and (2) the Office of Administration and Resources Management, which is responsible for financial management activities and the development of supporting information systems. EPA also has ten regional offices that have lead responsibility for carrying out the program within their geographical jurisdiction. These responsibilities include conducting or overseeing cleanup activities and pursuing cost recovery, including assembly of supporting documentation; negotiating settlements with PRPs; and collecting amounts owed the government. In December 1992, and again in February 1995, we reported that EPA’s management of the Superfund program was a high-risk area and noted that EPA had recovered only a fraction of the cleanup costs from responsible parties. We have also previously reported that the low priority EPA has given to the cost recovery program had resulted in a backlog of cost recovery cases. EPA also recognizes its problems with Superfund cost recovery, having reported it as a material weakness in its fiscal year 1994 Federal Managers’ Financial Integrity Act Report to the President and Congress. Concerning IFMS, EPA’s Office of Inspector General (OIG) reported in 1991 and 1994 deficiencies with the agency’s development and implementation of the system, such as problems with the integrity of payroll data and inadequate system development and user documentation. Also, IFMS has been on the Office of Management and Budget’s (OMB) high-risk list since 1990. Our work was performed at several offices at EPA headquarters including the Office of Solid Waste and Emergency Response; Office of Enforcement and Compliance Assurance; Office of Inspector General; and the Financial Management Division in the Office of Administration and Resources Management. These offices are located in Washington, D.C., and Arlington, Virginia. We also performed work at (1) EPA regional offices in New York, New York; Philadelphia, Pennsylvania; Chicago, Illinois; and San Francisco, California; (2) the Department of Justice in Washington, D.C.; and (3) the office of Leonard G. Birnbaum and Company in Springfield, Virginia. We conducted our review from January 1994 to July 1995, in accordance with generally accepted government auditing standards. We requested comments on a draft of this report from the Administrator, Environmental Protection Agency. In August 1995, we received the agency’s response from the Comptroller, the Director of the Financial Management Division, and the Director of the Policy and Program Evaluation Division. We have incorporated their comments where appropriate. Additional details on our scope and methodology are provided in appendix I. The automated information systems that EPA has in place fall short of providing the information and support that staff need to efficiently perform Superfund cost recovery work. Data in the central financial systems are insufficiently detailed, and are sometimes inaccurate or incomplete. Further, the records management systems do not provide for the efficient retrieval of supporting cost and work-performed documentation, which, if not located, can result in unrecovered costs. In addition, efforts to collect costs from responsible parties is more difficult, in part because the agency’s financial system, IFMS, is not sufficiently sophisticated to address the complexity of the repayment agreements. As a result of these limitations, the cost recovery process is often longer and more tedious than necessary and must be supported by manual searches and ad hoc information systems. Having sufficiently detailed financial information is essential for preparing and supporting cost recovery actions. The Chief Financial Officers Act of 1990 requires that an agency’s Chief Financial Officer develop and maintain an integrated agency accounting and financial management system that provides for (1) complete, reliable, consistent, and timely information that is responsive to the financial information needs of agency management and (2) the development and reporting of cost information. Further, the Joint Financial Management Improvement Program states that financial data reporting should be of proper scope, level of detail, timing, content, and presentation format to provide information of real value to users. EPA currently operates two financial management systems for maintaining Superfund cost data, IFMS and FMS. However, neither system currently records cost information at a level of detail that is often needed by EPA staff to prepare cost recovery packages. Specifically, EPA regions divide large or complex cleanup sites into smaller components called operable units. Cost recovery staff said that in order to properly assign the correct amount of costs to the appropriate PRP they need to be able to tracedetailed costs to these operable units. Because EPA systems do not currently record costs at the operable unit level, identifying which costs were incurred at different operable units becomes a time-consuming and tedious task. During the course of a cleanup, which often lasts for years, thousands of individual transactions are processed and stored, including payroll and travel costs for EPA employees, as well as contractor cleanup costs and costs incurred by other agencies, such as the Department of Justice and the U.S. Army Corps of Engineers. To trace these costs to individual operable units, EPA staff must identify all costs that have been recorded and accumulated by site, and then manually segregate the costs by operable unit. Staff in EPA’s regions told us that this data limitation has resulted in wasted staff resources. For example, one region we contacted was managing a site with 18 operable units, involving $2.8 million in cost recovery. In order to identify costs at the operable unit level, three staff had to work full time for over 4 months to manually allocate the costs. This required them to go through numerous records, including individual time sheets and travel records. Similarly, a staff person in another region estimated that about 10 percent of his time was spent manually allocating costs, which he believed could be avoided if costs were recorded in greater detail. The independent public accounting firm’s report on EPA’s fiscal year 1993 financial statements for the Superfund Trust Fund stated that the system limitation may adversely impact EPA’s ability to account for costs at Superfund sites and projects. The report noted that this could result in the failure to identify and recover these costs in cost recovery actions. EPA staff need accurate and complete financial data to efficiently and effectively pursue cost recovery actions. OMB Circular A-127 specifies that federal agencies should have financial management systems in place to process and record financial events effectively and efficiently and to provide complete, timely, and consistent information. It also states that these systems should have consistent internal controls over data entry, transaction processing, and reporting to ensure the validity of information and protection of federal government resources. Concerns exist about the integrity of data in IFMS. For example, in its 1994 report on IFMS, the OIG raised concerns about data integrity, including inaccuracies and omissions in the data. In our discussions with cost recovery staff, they too stated that they had encountered instances of inaccurate and incomplete data, including critical cost and site identification information, in the agency’s financial information systems. Several of the examples cited by these staff are described below. Staff in three regions stated that they had identified instances of duplicative data. For example, during initial negotiations, one region initially overstated costs for a PRP by about $822,000. While staff identified and corrected this overstatement prior to final negotiations, they determined that the error was largely due to a cost figure that had been duplicated in the financial system. EPA staff were unsure whether this was a random problem or a systemic one. One region discovered, while attempting to support a cost summary it had provided to a PRP, that approximately $23,000 had been erroneously charged to a site. The overcharge occurred because contract lab costs that should have been charged to a site in another EPA region had instead been charged to this site, possibly due to a data entry error. Five regions expressed concerns that certain costs associated with work performed at individual sites, under national contracts, were not being recorded by site in the agency’s financial management systems. For example, one EPA region reported in 1994 that about $90 million in technical assistance team contract charges associated with one of two national contracts could not be traced to specific sites through the agency’s financial systems. According to EPA, most of these costs were incurred for non site-specific activities and are recovered from responsible parties as indirect costs through the annual allocation process. However, the regional analysis concluded that some of the costs that were site-specific in nature were not reflected in individual site accounts in IFMS. Two regions provided examples of missing or invalid data in the site/project identification field. This was corroborated by a report generated by EPA’s Financial Management Division showing about 10,500 transactions, totaling about $129 million in expenditures, for which, according to EPA officials, the site/project identification field was missing. These examples are not intended to be representative of the overall integrity of data in the financial systems. However, EPA staff told us that as a result of these types of problems, they have to spend excessive time and effort in researching, reconciling, and correcting financial data needed to support cost recovery actions. EPA has no assurance that its application controls are sufficient to prevent these data quality problems. Such controls are critical for ensuring accurate data input, processing, and output. The independent public accounting firm that reviewed EPA’s financial statements for the Superfund Trust Fund for fiscal year 1993 noted that weaknesses with the internal controls governing data entry made it possible for inaccurate or incomplete account numbers to be entered into IFMS. For example, they found there was no error check control of the site/project code portion of IFMS’ account code. EPA officials believe IFMS contains adequate application controls. However, because these controls are not documented in accordance with federal policies, such as OMB Circular No. A-127 and the Joint Financial Management Improvement Program, we could not assess these controls to determine if they are sufficient to prevent data integrity problems. The lack of documentation for application controls was identified in the OIG’s February 1995 report, in which the OIG stated that it could not assess application processing controls due to a lack of technical system documentation. The OIG reported that such an internal control weakness could adversely affect EPA’s ability to ensure that (1) obligations and costs were in compliance with applicable laws, (2) funds, property, and other assets were safeguarded against unauthorized use or disposition, and (3) transactions were properly recorded to permit the preparation of reliable financial statements. The previously mentioned example of missing site identification data for technical assistance team costs could have been prevented had additional controls been in place. Such controls would have alerted senior financial managers that these costs had been approved and paid, but were at risk of being excluded from cost recovery actions because they had not been allocated, where possible, to a specific Superfund site. Until EPA addresses the need for documented controls, data integrity problems could continue to adversely affect the efficiency of performing cost recovery. In addition, when site/project codes are missing, EPA may lose the opportunity to recover related costs in specific cost recovery actions. To successfully defend its claims for cost recovery, EPA must be able to substantiate each cost item. To do this, the agency locates and provides a wide-range of supporting financial documents, such as invoices and travel vouchers, and supporting work-performed documents, such as contracts, contractor work assignments, and progress reports pertaining to a site. Such documents are needed to provide proof to PRPs and the courts that Superfund-led work to clean up hazardous waste sites was authorized, performed, invoiced, and paid. Despite the importance of these documents, EPA staff in regional offices believe that the difficulty in locating and retrieving supporting documents was a major contributor to the amount of time and effort required to assemble the packages detailing costs to be recovered. According to these staff, almost all financial documents generated since 1991 are available through the SCRIPS imaging system. However, most of the contract-related financial documents created prior to this time are not available from SCRIPS because the system was not operational until 1991. Pre-1991 contract-related financial documents are stored in EPA’s financial management center in Research Triangle Park, North Carolina, and must be manually retrieved for inclusion in the cost recovery packages. Cost recovery staff said that it usually takes about 20 working days to retrieve these documents once identified, and that the time required to assemble the requisite financial documents could be substantially decreased if these documents could also be retrieved using SCRIPS. Staff also noted that the situation is worse for work-performed documents. There are estimated to be over 11 million pages of work-performed documents occupying about 6,000 linear feet of shelf space in EPA’s ten regional offices. The regional offices maintain these work-performed documents as hard copy in various locations—some in off-site storage, some in records management centers, and some in working files maintained by EPA staff responsible for managing or overseeing the cleanup process. In many cases, cost recovery staff have to rely on their memories to identify which contractors were used at a site and where relevant documents might be located. Cost recovery staff also noted that if the documents cannot be found in EPA’s offices, they must then try to obtain replacements from the contractors’ files. Staff in several regions said that assembling work-performed documents from various locations inside and outside of the agency is a time-consuming or labor-intensive process. For example, in one region it typically takes 2 months to assemble such documents. Another region said it takes about 4 months to identify, retrieve, and review work-performed documents. Although locating supporting documentation can be labor-intensive, the effect of not locating needed documentation can be worse. According to cost recovery staff, if supporting documents cannot be located or otherwise supported, the corresponding cost items are removed from the cost recovery summary, even though these costs may be recoverable. We could not determine the amount that EPA has lost because of such missing documentation because EPA does not track this information. While EPA maintains a record showing the reasons why costs are excluded from final settlements with PRPs, costs excluded from initial negotiations due to missing documentation are not a part of this record. EPA regional offices are primarily responsible for managing accounts receivable after the government reaches cost recovery settlements with responsible parties. This requires EPA to establish accounts receivable in a timely manner, collect interest, accurately record collections, and identify and take action on delinquencies. OMB Circular A-127 requires that an agency’s financial management systems provide reliable and timely information on amounts owed the government. It also requires that agency financial systems satisfy the core financial system requirements developed by the Joint Financial Management Improvement Program, including a variety of functions to support the establishment, management, and collection of accounts receivable. These functions include calculating and generating customer bills, tracking receivables to be paid for under an installment plan, and accurately identifying receivables that are past due. IFMS does not meet these requirements. Although IFMS includes an accounts receivable module, which EPA began using in 1989, the module does not meet the special requirements needed to manage the settlement agreements reached with PRPs. It lacks the capabilities to compute compound interest and manage installment payments. This module also lacks the ability to produce accurate aging reports for Treasury and EPA management. EPA has recognized that it has a receivables problem. It has reported this problem as a material weakness in its fiscal year 1994 Federal Managers’ Financial Integrity Act Report. This weakness is very significant, especially given that EPA data show that uncollected Superfund cost recovery receivables totalled about $498 million at the end of fiscal year 1994. Because EPA has not yet resolved its problem with receivables, some regional offices have developed their own automated systems or manual procedures to overcome these limitations. For example, four regional offices have developed local PC-based systems to provide some of these accounts receivable capabilities, while another region uses a combination of manual procedures and IFMS capabilities. Staff in these regions pointed out that the locally developed systems or procedures give them the capability to perform basic debt-servicing functions that IFMS does not support. EPA has initiated efforts to address its information system limitations. These efforts include (1) reporting cost data in greater detail, (2) using a statistical tool to test the integrity of financial data, and developing a capability to require that the site/project field is complete and valid, (3) implementing and testing an imaging system to improve the agency’s identification and retrieval of Superfund work-performed documentation, and (4) developing a PC-based information system to better manage accounts receivable. However, additional actions are needed to fully address the limitations and ensure that the agency obtains the best possible systems support for its cost recovery efforts. To address the need for more detailed cost data, in October 1995, EPA plans to begin using an expanded 41-digit account code structure in IFMS. This expanded structure should provide the capability to record costs in greater detail, such as by site operable unit, and thus better support EPA’s cost recovery efforts. To assess financial data reliability, EPA’s Financial Management Division has recently developed an automated statistical sampling tool. The Division instructed the regions and finance centers in March 1995 to begin using this statistical tool as part of the agency’s internal control evaluations. In August 1995, EPA officials stated that the results of the initial testing are currently being reviewed. In response to our concerns, EPA officials told us they intend to issue guidance for automated statistical testing of the integrity of financial data needed for cost recovery. Regarding application controls, EPA officials acknowledged that the capability to require that the site/project field be completed when financial transactions are entered into IFMS would be beneficial. They said that a new project cost accounting subsystem of IFMS, scheduled for implementation by October 1995, should provide this capability. With respect to the requirement that financial systems be documented in accordance with federal policies, EPA officials also reported that they intend to work with the OIG in improving the documentation of application controls in IFMS. As noted earlier, difficulties in locating and retrieving financial and work-performed documentation has been a major contributor to the amount of time and effort required to assemble cost recovery packages. Although EPA has two efforts underway that may improve certain aspects of its records management capabilities, neither project, as currently planned, will address the agency’s difficulties in locating pre-1991 financial documents, or millions of work-performed documents that occupy growing amounts of space in EPA locations nationwide. One project involves the development of an imaging system, called the Superfund Document Management System (SDMS). SDMS is intended to provide a number of advanced capabilities, such as full-text indexing, electronic redaction, and security controls. The system is being tested in EPA’s regional office in San Francisco, California, using documents related to its largest Superfund site. This site accounts for about 25 percent of the region’s Superfund documents. Although SDMS may provide an effective means for locating Superfund-related program documentation, EPA has not assessed the use of SDMS for cost recovery in other regions. A second project, initiated in 1993, involves microfilming over a million pages of documentation pertaining to 60 expired nationally-managed contracts and creating an automated index of these documents. The project, which is being funded by EPA and implemented by the Department of Justice, is intended to overcome difficulties that EPA regions and Justice have experienced in obtaining copies of this documentation. This effort may substantially improve the accessibility and retrievability of work-performed documents related to the expired national contracts. However, EPA has no plans to assess whether this effort should be expanded to include other region-specific work-performed documents that are used extensively in cost recovery, such as documents pertaining to contracts managed by EPA regions. Although SCRIPS provides electronic access to financial documents generated since 1991, an EPA official in the Financial Management Division told us that the agency had not evaluated the costs or benefits of expanding this system to include pre-1991 financial documents, or included such a project in the agency’s Five-Year Plan. Agency officials explained that this has not been a high priority. Recognizing that IFMS’ accounts receivable management capabilities needed improvement, EPA has initiated plans to strengthen these capabilities beginning in early fiscal year 1996. The agency plans to implement a Cost Recovery Collection Tracking System (CTS), which is being developed in EPA’s Chicago, Illinois, regional office. CTS will run on personal computers that are connected to a local area network in the region. The system is intended to provide (1) a demand letter billing capability for actions initiated subsequent to an administrative or judicial order, (2) timely collection information to EPA managers, (3) tracking reports concerning cost recovery collections, and (4) direct uploading of collections data to IFMS. EPA’s Financial Management Division plans to have CTS designed, developed, and tested in the Chicago regional office by September 30, 1995, and plans to distribute CTS to all of its regional offices by December 31, 1995. Given that the development of receivables management capabilities could affect the collection of and accounting for billions of dollars, it is critical that EPA implement a system that effectively safeguards these public assets. As outlined in OMB Circulars A-123, A-127, and A-130, agencies are required to (1) perform an assessment of the potential risks associated with the operation of a system and (2) provide some assurance that appropriate controls are in place to reduce risks such as data entry errors and fraudulent manipulation of accounts receivable data. Although EPA officials told us that risk assessment was an inherent part of the development of CTS, they could not provide us with documentation demonstrating that the agency had performed a risk assessment or ensured that necessary controls will be in place. EPA’s financial and records management systems do not efficiently support cost recovery, a critical business process that is vital to the continued existence of the Superfund program. Because of limitations in these systems, cost recovery staff cannot fully rely on them to provide the information needed for cost recovery. Instead, they laboriously search and reconcile paper records to ensure that the information supporting cost recovery cases is accurate, reliable, and complete. Aware of these limitations, EPA is taking steps to improve support for cost recovery. However, the agency could further ensure that it is obtaining the best possible support for cost recovery by (1) implementing its planned automated statistical testing of the integrity of financial data needed for cost recovery and developing a baseline on the extent of any integrity problems; (2) improving the documentation of its financial systems’ application controls; (3) assessing how best to use records management systems to meet cost recovery users’ needs; and (4) ensuring that all risks associated with the collection and management of receivables have been addressed. These additional actions could further improve EPA’s efforts to recover billions of Superfund dollars through cost recovery actions, make cost recovery more efficient, and lower the risks of losing recoverable dollars. To improve EPA’s ability to recover costs associated with cleaning up hazardous waste sites, we recommend that the Administrator of the Environmental Protection Agency take steps to ensure that cost recovery data and supporting documentation are complete and accurate by implementing planned automated statistical testing of the integrity of financial data needed for cost recovery and developing a baseline on the extent of any integrity problems identified, improving the documentation of financial systems’ application controls to help ensure accurate data input, processing, and output, assessing whether efforts to improve records management systems for cost recovery should be expanded, including evaluating how best to improve the retrieval of pre-1991 financial documents, and performing a risk assessment and determining if additional controls are needed for accounts receivable. EPA officials, including the Comptroller, the Director of the Financial Management Division, and the Director of the Policy and Program Evaluation Division, provided comments on a draft of this report. Overall, the officials agreed with our recommendations and with our conclusions that the agency’s systems supporting cost recovery needed improvement. The agency provided additional information on the status of its improvement activities, which we have incorporated where appropriate. As arranged with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the date of this letter. At that time, we will send copies to the Administrator of the Environmental Protection Agency, Director, Office of Management and Budget, and interested congressional committees. Copies will also be made available to others upon request. Please call me at (202) 512-6253 if you or your staff have any questions concerning this report. Other major contributors are listed in appendix II. To evaluate how well EPA’s information systems support the Superfund cost recovery process, we used a structured interview document to discuss cost recovery efforts with staff from five of EPA’s ten regional offices: Region 2 (New York), Region 3 (Philadelphia), Region 5 (Chicago), Region 7 (Kansas City); and Region 10 (Seattle). We chose regions 2, 3, and 5 because they had the highest levels of direct expenditures on cleanups. We chose regions 7 and 10 because they provided geographical diversity. We analyzed numerous documents related to cost recovery from each of these regions. Because integrity of data in EPA’s financial systems has a direct impact on how well these systems support cost recovery, we sought information from cost recovery staff on the extent of problems with the financial data. However, because these staff were unable to provide quantified information on the extent of such problems, we relied on their oral responses and some documented instances in reaching our conclusions. We also contacted by phone records management officials in all ten EPA regions concerning the volume of documentation maintained and researched for supporting cost recovery. We met with representatives and analyzed workpapers and documents from the three firms involved in the audit of EPA’s fiscal year 1993 financial statements for the Superfund Trust Fund. These firms were Leonard G. Birnbaum and Company; KPMG Peat Marwick; and American Power Jet Company. We met with officials from EPA’s OIG and reviewed its past and current reports related to Superfund and cost recovery. We also met with officials in the Department of Justice’s Environment and Natural Resources Division concerning the quality of the cost recovery documentation that it receives from EPA and uses to pursue cost recovery actions. To evaluate the extent to which EPA’s planned information systems modifications could improve the efficiency of cost recovery efforts, we (1) applied relevant segments of the information systems audit methodology published by the EDP Auditors Foundation, (2) interviewed officials from several EPA headquarters offices in Washington, D.C., and from EPA regional offices involved in developing new information systems or modifications to existing systems, and (3) reviewed and analyzed documents on EPA’s actions, including documentation on users’ requirements, feasibility, costs, benefits, and detailed specifications pertaining to the agency’s efforts to enhance and develop system capabilities to support cost recovery. We also reviewed EPA planning documents, including the agency’s Five-Year Plan, and Strategy and Master Work Plan for IFMS. Our work was performed at several offices at EPA headquarters including the Office of Solid Waste and Emergency Response, Office of Enforcement and Compliance Assurance, Office of Inspector General, and the Financial Management Division in the Office of Administration and Resources Management. These offices were located in Washington, D.C., and Arlington, Virginia. We also worked at (1) EPA regional offices in New York, New York; Philadelphia, Pennsylvania; Chicago, Illinois; and San Francisco, California; (2) the Department of Justice in Washington, D.C.; and (3) the office of Leonard G. Birnbaum and Company in Springfield, Virginia. We conducted our review from January 1994 to July 1995, in accordance with generally accepted government auditing standards. We requested comments on a draft of this report from the Administrator, Environmental Protection Agency. In August 1995, we received the agency’s response from the Comptroller, the Director for the Financial Management Division, and the Director for the Policy and Program Evaluation Division. We have incorporated these comments where appropriate. Ronald W. Beers, Assistant Director William G. Barrick, Project Manager Robert C. Reining, Deputy Project Manager James V. Rinaldi, Senior Evaluator The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 6015 Gaithersburg, MD 20884-6015 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (301) 258-4066, or TDD (301) 413-0006. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (301) 258-4097 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO reviewed the adequacy of the Environmental Protection Agency's (EPA) information systems that support the Superfund cost recovery process, focusing on whether planned modifications to the information systems will improve the efficiency of EPA cost recovery efforts. GAO found that: (1) EPA financial and records management systems do not provide all the detailed cost information EPA staff need for the Superfund cost recovery process; (2) EPA staff have to conduct excessive manual searches and reconciliations to gather the needed data, which prolongs the cost recovery process; (3) EPA financial management systems are not sophisticated enough to cope with the complexity of certain transactions and the data contained in the systems are not always accurate; (4) EPA internal controls to prevent inaccurate data entry are inadequate and undocumented; and (5) although planned information systems modifications will improve cost information collection and retrieval, EPA needs to do more to enhance its records management capabilities. |
The President’s budget request is proposing to maintain taxpayer service levels with fewer staff by realizing efficiency gains; it also proposes to increase enforcement by adding staff. The President’s FY 2009 budget request of $11.4 billion for IRS is 4.3 percent more than the FY 2008 enacted budget and represents an increase of less than 1 percent for taxpayer service and 7 percent for enforcement, as shown in table 1. The budget request increases IRS-wide staff levels, measured in full-time equivalents (FTEs), by 2 percent, with a 1.4 percent decrease in taxpayer service FTEs and a 5.2 percent increase in enforcement FTEs, as shown in table 2. The President’s budget proposal is consistent with longer-term trends for IRS. Compared to actual spending in FY 2006, the proposed FY 2009 budget increases taxpayer service funding by 3.7 percent, a real decrease after inflation, while increasing IRS’s enforcement funding by 10 percent. The budget request proposes to maintain taxpayer service at recent levels. As an example, the key taxpayer service measures shown in table 3 are projected to remain relatively stable through FY 2009. In order to maintain taxpayer service at recent levels despite a decrease in real spending and staffing, IRS expects to realize efficiency gains. For instance, IRS expects to devote 207 fewer FTEs to the labor-intensive processing of paper returns because of expected increases in electronic filing. These expected efficiency gains are consistent with past trends— between 1999 and 2007, IRS reduced staff devoted to processing paper returns by about 1,800 FTEs. IRS’s ability to maintain or improve taxpayer service beyond 2009 will likely depend on its ability to continue to improve efficiency. To this end, in recent reports, we made recommendations to further increase electronic filing. We recommended that IRS determine the actions needed to require software vendors to include bar codes on printed returns, and we suggested that the Congress mandate electronic filing by certain paid tax preparers. IRS agreed with our bar code recommendation and outlined the actions it would take. Some of the real spending decrease proposed for FY 2009 is because of one-time investments made in FY 2008 or carryovers in funds from FY 2008. For instance, the budget request proposes a $31 million reduction in funding for taxpayer assistance centers and outreach. However, IRS officials told us that this reduction includes funding used for long-term investments in FY 2008 that would not need to be duplicated in FY 2009. IRS officials also told us that a $7.7 million decrease in funding for the Taxpayer Advocate offsets a funding increase in FY 2008 that is being used to lower the Advocate’s outstanding caseload. Finally, an $8 million reduction in the Volunteer Income Tax Assistance (VITA) program reflects FY 2008 funding that was not spent and carried over into FY 2009. The budget request for IRS’s enforcement programs includes nonlegislative and legislative initiatives. According to the proposal, the five nonlegislative enforcement initiatives would cost about $338 million in FY 2009 and are expected to raise about $2 billion of direct revenue annually starting in FY 2011. In addition, the budget request estimates that the enforcement initiatives would generate at least another $6 billion annually in indirect revenue. The indirect revenue results from improved voluntary compliance induced by taxpayers’ awareness of expanded IRS enforcement. The budget request also proposes increases in examination coverage for corporations with assets of $10 million or more from a planned 6.6 percent for FY 2008 to 6.8 percent for FY 2009. The coverage rate would increase to 7.6 percent in FY 2010 as new enforcement staff hired in FY 2009 complete training and can audit more returns. The budget request includes 16 legislative initiatives budgeted at $23 million for FY 2009 that it says would raise about $36 billion in revenue over 10 years; if none were enacted, IRS would not need the $23 million. We have reported on three of the proposals. In 2006, we suggested that the Congress consider an idea for reducing securities capital gains noncompliance. In 1991, we supported the notion that payments to corporations be reported on information returns. Finally, in 2007, we described ways to mitigate the compliance costs related to these information returns and to other information returns associated with credit and debit card payments. The revenue expected from IRS’s enforcement initiatives is modest compared to the net tax gap, which was last estimated at $290 billion for tax year 2001. As we noted in our statement to this Committee last year, no single approach, such as IRS enforcement, is likely to fully and effectively address noncompliance. Multiple approaches are needed because noncompliance has multiple causes and spans different types of taxes and taxpayers. Hiring needed staff for the nonlegislative initiatives will be challenging for IRS’s Large and Mid-Size Business (LMSB) and Small Business/Self- Employed (SB/SE) divisions. For instance, the initiatives call for adding 1,431 revenue agents in addition to those who must be replaced from attrition, a high number relative to past years. IRS divisions have previously hired large numbers of staff in a short time because of specific budget initiatives, but officials reported that hiring gradually over time would reduce challenges. If IRS were to fall behind in its hiring efforts, it would not need all $226 million of the funding for staff for FY 2009 initiatives. Responding to our recommendations from last year, IRS included more information on initiatives in the FY 2009 proposed budget, including ROI information for all nonlegislative initiatives. Last year, we recommended that IRS have available basic descriptive, cost, and expected performance information on all new initiatives and include such information in future budget submissions. This year, the budget request has sections explicitly entitled, for instance, “Initiative Summary,” “Implementation Plan,” “Expected Benefits,” and “ROI.” Four of the five nonlegislative enforcement initiatives for FY 2009 were revisions of FY 2008 initiatives, but with more total funds requested and generally more informative justifications than for FY 2008. However, IRS’s ROI calculations have limitations that reflect the challenges of estimating ROIs. For example, the calculations do not account for benefits that are harder to measure, such as improved voluntary compliance. Another example showing ROI limitations is the $51 million National Research Project (NRP) initiative for which IRS estimates the ROI to be $0.40 per $1.00 invested. NRP funds research audits in order to develop more effective enforcement programs. The ROI calculation only includes direct revenue resulting from the research audits, not the potential for increased revenue from improved enforcement programs; nor does the calculation include the benefits of the Department of the Treasury’s use of NRP data to provide the basis for legislative recommendations. Although the budget request for IRS provides performance measure data, it does not provide ROI analyses for programs or activities other than the new initiatives. As we noted in our recent report, analytic data such as ROI can be helpful to managers and the Congress when making resource allocation decisions. ROI analyses, even with their limitations, can help answer questions such as the following: What are the implications for IRS’s resource allocation of the lower costs per taxpayer contact for some services compared to others as shown in table 4? Are there extra benefits that offset the higher costs of some services, or could costs be reduced by promoting increased reliance on the lower- cost options? Similar questions can be asked about enforcement based on table 5: Is IRS appropriately allocating resources between field audits, often conducted at a taxpayer’s business, and correspondence audits, which are simpler and conducted by mail? For the rows in table 5 with average recommended additional tax per return greater for correspondence audits than for field audits, could resources be reallocated from field audits to correspondence audits in order to help close the tax gap? Are there other benefits to field audits, such as a greater impact on voluntary compliance, that are not captured in IRS’s data? We recognize that developing ROI estimates for IRS’s ongoing programs such as examinations and taxpayer service will be a challenge. However, because of the potential benefits of ROI analyses, we recommended in our previous report on the FY 2009 budget request that the Commissioner of Internal Revenue extend the use of ROI in future budget proposals to cover major enforcement programs. At that time, IRS officials said that because of the short time frame for our report, they did not have time to fully analyze its recommendations, and, therefore, were unable to respond. We have agreed to meet with IRS to further discuss the ROI recommendation. IRS’s BSM program, initiated in 1999, involves the development and delivery of a number of modernized tax administration, internal management, and core infrastructure projects that are intended to provide improved and expanded service to taxpayers as well as IRS internal business efficiencies. Key tax administration projects include CADE, which is intended to provide the modernized database foundation to replace the existing Individual Master File processing system that contains the repository of individual taxpayer information; AMS, which is intended to enhance CADE by providing applications for IRS employees and taxpayers to access, validate, and update accounts on demand; and MeF, which is to provide a single standard for filing electronic tax returns. We recently reported that while IRS has continued to make progress in implementing BSM projects and improving modernization management controls and capabilities, challenges and risks remain, and further improvements are needed. As shown in table 6, the FY 2009 budget request for the BSM program is less than the enacted FY 2008 budget by over $44 million and about $185 million less than the amount the IRS Oversight Board is proposing. When we asked about the impact of this reduction on its operations, IRS told us that the proposed funding level will allow it to continue developing and delivering its primary modernization projects but did not provide details on how plans to deliver specific projects or benefits to taxpayers would be affected. MeF is the project with the largest difference between the requested budget and the FY 2008 enacted amount. IRS has made progress in implementing BSM projects and meeting cost and schedule commitments for most deliverables, but three project milestones experienced significant cost or schedule delays. During 2007, IRS completed milestones of the Filing and Payment Compliance (F&PC), a tax collection case analysis support system; MeF; CADE; and AMS. Our analysis of reported project costs and completion dates showed that 13 of the 14 associated project milestones that were scheduled for completion during this time were completed within 10 percent of cost estimates, and 11 of the 14 milestones were completed within 10 percent of schedule estimates. However, a milestone for CADE exceeded its planned schedule by 66 percent and experienced a 15 percent cost increase; another milestone for the same project incurred a 153 percent schedule delay, and a milestone for MeF experienced a 41 percent schedule delay (see fig. 1). IRS has taken steps to address our prior recommendations to improve its modernization management controls and capabilities. However, work remains to fully implement them. For example, in July 2005, we recommended that IRS fully revisit the vision and strategy for the BSM program and develop a new set of long-term goals, strategies, and plans consistent with the budgetary outlook and IRS’s management capabilities. We also noted that the vision and strategy should include time frames for consolidating and retiring legacy systems. In response, IRS has developed a Modernization Vision and Strategy framework and supporting 5-year Enterprise Transition Plan. However, the agency has yet to develop long-term plans for completing BSM and consolidating and retiring legacy systems. We also recommended in February 2007 that IRS ensure that future BSM expenditure plans include a quantitative measure of progress in meeting scope expectations. We further recommended that, in developing this measure, IRS consider using earned value management since this is a proven technique required by the Office of Management and Budget for measuring cost, schedule, and functional performance against plans. While IRS has developed an approach to address our recommendation, it has not yet fully implemented it. Future BSM project releases continue to face significant risks and issues, which IRS is addressing. Specifically, the agency recently identified significant risks and issues with planned system deliveries of CADE and AMS and reported that maintaining alignment between the two systems will be a significant challenge and source of risk for the BSM program. IRS recognizes the potential impact of identified risks and issues on its ability to deliver projects within cost and schedule estimates and has developed mitigation strategies to address them. While mitigation strategies have been developed, the risks and challenges confronting future releases of CADE and AMS are nevertheless significant, and we will continue to monitor them and actions to address them. IRS also made further progress in addressing high-priority BSM program improvement initiatives during the past year. In September 2007, IRS completed another cycle of initiatives and initiated a new cycle, which was scheduled to be completed at the end of March 2008. Initiatives that were addressed in the 6-month cycle ending in September 2007 included IT human capital, information security, and process improvements (e.g., developing and implementing standardized earned value management practices for major projects). IRS’s program improvement process continues to be an effective means of regularly assessing, prioritizing, and incrementally addressing BSM issues and challenges. However, more work remains for the agency to fully address these issues and challenges. Finally, we recently reported that efforts to address human capital challenges continue, but more work remains. IRS developed an IT human capital strategy that addresses hiring critical personnel, employee training, leadership development, and workforce retention, and agency officials stated that they plan to undertake a number of human capital initiatives to support their human capital strategy, including conducting analyses of turnover rates and continuing efforts to replace key leaders lost to retirement. However, a specific plan with time frames for implementing these initiatives has not been developed. We recommended that IRS complete such a plan to help guide the agency’s efforts in addressing its IT human capital gaps and measure progress in implementing them. IRS agreed with our recommendation and stated that it intends to develop a plan to implement its IT human capital strategy. The Economic Stimulus Act of 2008 is resulting in a significant workload increase not anticipated in the FY 2008 budget. As part of the legislation, IRS received $202 million in a supplemental appropriation. However, because IRS could not find an alternative according to responsible officials, it has reallocated resources from enforcement to taxpayer service and is allowing some deterioration in telephone service. IRS will begin sending economic stimulus payments to more than 130 million households in early May, after the current tax filing season, and is scheduled to be done by mid-July. These include an estimated 20 million retirees and disabled veterans, and low-wage workers who usually are exempt from filing a tax return but will be eligible for stimulus payments. Taxpayers required to file a tax return must do so by April 15 in order to receive a stimulus payment by mid-July. People who are not required to file a tax return, but are doing so to receive a stimulus payment, are required to file an IRS Form 1040A by October 15, 2008. As part of the legislation, IRS received a supplemental appropriation of $202 million to help fund its costs for implementing the stimulus package. This funding will remain available until September 30, 2009. As shown in table 7, IRS plans to spend the bulk of the funding—$151.4 million—for Operations Support, most of it on postage for two mass mailings and on IT support. IRS also expects to spend $50.7 million for Taxpayer Services, including $26.2 million for staffing and overtime for telephone assistors. IRS is expecting 2.4 million additional telephone calls in March and April with questions for IRS assistors about the economic stimulus legislation. These calls are in addition to the more than 14 million calls typically answered by IRS assistors between January and mid-April. To help meet the increased telephone demand, IRS is shifting about half of its over 2,000 Automated Collection System (ACS) telephone staff from collecting delinquent taxes to answering economic stimulus telephone calls from March through May. To accommodate this shift, IRS stopped sending out some ACS-generated notices, such as notices of levy, several weeks ago. According to IRS officials, it takes about 3 to 4 weeks before this adjustment in ACS-generated notices affects the ACS workload. IRS originally estimated that the revenue foregone by shifting ACS staff to be up to $681 million. However, according to IRS officials, in early April, IRS revised its foregone revenue estimate down to $565 million, shown in table 7, largely because of lower-than-expected demand for telephone assistance in March. According to IRS officials, IRS’s priority is to respond to taxpayers’ questions about the stimulus program; therefore, the officials are monitoring call volume and adjusting the number of ACS staff answering telephones accordingly. When call volume is low, ACS staff work on outstanding ACS collection cases. However, IRS officials stated that this work does not produce the same revenue as the ACS-generated notices, particularly revenue generated from notices of levy. When IRS adjusts the volume of ACS-generated notices, it takes several weeks before that adjustment affects ACS workload. IRS officials do not want to resume sending ACS-generated notices until they are sure ACS staffers are available to handle the resulting workload. Should the lower-than-expected call volume continue, IRS may have an opportunity to shift the ACS staff back to their most productive collection work. This could further reduce the revenue foregone from using ACS staff to answer stimulus-related telephone calls. To date, IRS has not reduced its projections for future stimulus-related call volume. If the projections are reduced, IRS may be able to resume sending out at least some ACS-generated notices. According to IRS officials, IRS considered alternatives to shifting ACS staff, including contracting out, using other IRS staff, or using Social Security Administration or other federal staff, but decided the alternatives were not feasible. For example, contracting out was not deemed feasible because of insufficient time to negotiate the contract and conduct background checks and training. Another cost—although not measured in dollars—is the decline in telephone service shown in table 7. Because of the increased call volume, IRS expects its assistor level of service to drop from 82 percent (the 2008 goal) to as low as 74 percent—the lowest level since 2002. IRS is already experiencing some declines in telephone service. As of March 29, the level of service had dropped to 80 percent, taxpayers were waiting a minute and a half longer than last year, and they were hanging up 43 percent more often while waiting to speak to an assistor. Between March 3 and March 29, IRS assistors answered over 572,000 stimulus-related calls. IRS expects call volume to increase rapidly in upcoming weeks as taxpayers receive their stimulus notices in the mail. Because IRS is in the early stages of implementing the stimulus legislation, IRS officials do not have much information about the actual costs. Through March, IRS estimates that it has spent almost $103 million, mostly for postage. In commenting on a draft of our earlier report on the FY 2009 budget request and 2008 tax filing season, IRS officials said that, because of the short time frame for our report, they did not have time to fully analyze our recommendation and, therefore, were unable to respond at the time. They provided technical comments at that time and again for this statement, and we made those changes where appropriate. We have agreed to meet with IRS to further discuss the ROI recommendation. Mr. Chairman, this concludes my prepared statement. Mr. Powner and I would be happy to respond to questions that you or other Members of the Subcommittee may have at this time. For further information regarding this testimony, please contact James R. White, Director, Strategic Issues, on (202) 512-9110 or whitej@gao.gov or David A. Powner, Director, Information Technology Management Issues, on (202) 512-9296 or pownerd@gao.gov. Contact points for our offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Individuals making key contributions to this testimony include Joanna Stamatiades, Assistant Director; Carlos E. Diz; Sarah A. Farkas; Charles R. Fox; Leon H. Green; Carol M. Henn; Lawrence M. Korb; Paul B. Middleton; Karen V. O’Conor; Sabine R. Paul; Cheryl M. Peterson; and Neil A. Pinney. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The fiscal year 2009 budget request for the Internal Revenue Service (IRS) is a road map for how IRS plans to allocate resources and achieve ambitious goals for improving enforcement, improving taxpayer service, increasing research, and continuing to invest in modernized information systems. One complicating factor in implementing IRS's plans in the immediate future is the recent passage of the Economic Stimulus Act of 2008, which creates additional, unanticipated workload for IRS. GAO was asked to (1) assess how the President's budget request for IRS allocates resources and justifies proposed initiatives; (2) determine the status of IRS's efforts to develop and implement its Business Systems Modernization (BSM) program; and (3) determine the total costs of administering the economic stimulus legislation. To meet these objectives, GAO drew upon and updated recently issued reports. The President's fiscal year 2009 budget request for IRS is $11.4 billion, 4.3 percent more than last year's enacted amount. The request proposes to maintain taxpayer service at recent levels, in part by realizing efficiency gains from electronic filing, despite a decrease in staffing. It also proposes a 7 percent increase in enforcement spending, including spending for 21 legislative and nonlegislative initiatives. The legislative proposals are projected to cost $23 million in fiscal year 2009, funding that IRS would not need if the proposals are not enacted. Similarly, if IRS were to fall behind in its proposed enforcement hiring efforts, it would not need all $226 million of the associated funding. IRS justified its nonlegislative enforcement initiatives with return on investment (ROI) analyses, which are useful, despite limitations, for making resource allocation decisions. The budget request does not provide ROI information for activities that constitute a large part of the budget request--activities other than the proposed initiatives. The request for BSM is over $44 million lower than the fiscal year 2008 enacted amount. IRS said this funding level will allow it to continue its primary modernization projects, but it did not describe how specific projects or benefits to taxpayers would be affected. IRS has continued to make progress in implementing BSM projects and improving modernization management controls and capabilities. However, further improvements are needed. For example, the agency has yet to develop long-term plans for completing BSM and consolidating and retiring legacy systems. IRS estimated that the costs of implementing the economic stimulus legislation may be up to a total of $767 million--including a $202 million supplemental appropriation. In addition to the supplemental appropriation, IRS is reallocating hundreds of collections staff to answering taxpayer telephone calls, resulting in up to $565 million in foregone enforcement revenue. In addition, IRS expects some deterioration in telephone service because of the increased call volume. For example, IRS is expecting its assistor level of service to drop to as low as 74 percent compared to its goal of 82 percent. |
IDEA authorizes federal funding for the needs of school-age children with a range of disabilities, such as specific learning disabilities, speech and language impairments, or mental retardation, who need special education services (see fig. 1). In 2007-2008, students covered under IDEA made up about 13 percent of the nation’s public prekindergarten through 12th grade school enrollment and, in fiscal year 2007, states received $10.8 billion in federal funds to provide services to school-age students. To receive federal funds, states and local educational agencies must identify and evaluate children who have a disability and provide special education and related services, as well as supplementary aids and services when necessary, to those who are eligible. Such services and supports are formulated in an individualized education program (IEP), which is developed, discussed, and documented by a student’s IEP team. An IEP team must generally include, among others, the parents of the child, a general education teacher, a special education teacher, and the child when appropriate. IDEA also requires the placement of students in the least restrictive environment—in which students with disabilities are educated with students without disabilities—to the maximum extent appropriate. Regarding PE, IDEA regulations specify that schools must generally provide opportunities for students to participate in regular or general PE classes or, in some cases, specially designed PE if determined by the IEP team. Regarding extracurricular athletics, districts and schools must take steps to provide services to give students with disabilities an opportunity to participate in extracurricular activities, which may include athletics, equal to that of other students. Section 504 prohibits entities that receive federal financial assistance, including elementary and secondary (K-12) schools, from discriminating against otherwise qualified individuals with disabilities. The Americans with Disabilities Act of 1990, as amended (ADA) is also a broad antidiscriminatory law protecting individuals with disabilities. Education interprets the ADA and Section 504 in a similar manner. While IDEA students are covered under Section 504, other students who are not covered under IDEA may still have a disability as defined under Section 504. Several of the most common disabilities of students included under Section 504, but not always covered under IDEA, are attention deficit hyperactivity disorder, diabetes, and asthma. In 2006, about 444,000 students with disabilities were covered under 504, and not IDEA, based on the most recent estimates from Education. These students may have an education plan developed under Section 504 that sets forth their needed regular or special education and related aids and services, but the federal requirements for such a plan are much less detailed than for an IEP. Similar to IDEA, Education’s Section 504 regulations require that students with disabilities be provided a free appropriate public education and learn alongside students without disabilities to the maximum extent appropriate. These Section 504 regulations also require that students with disabilities must be provided equal opportunities to participate in PE courses and extracurricular athletics. Unlike IDEA, however, Section 504 does not authorize any federal funding to schools to provide services to students. (See table 1 for more information on IDEA and Section 504.) At the federal level, Education is responsible for administering IDEA, as well as Section 504, as it applies to many educational institutions. The Office of Special Education Programs (OSEP) allocates federal funds to states for their implementation of IDEA programs and monitors states’ ability to provide a free, appropriate public education. For example, states must report to OSEP the amount of time that students who are covered under IDEA spend in their state’s general education classes, including PE. The Office for Civil Rights (OCR), through its headquarters and 12 regional offices, carries out enforcement activities related to Section 504, the ADA, and other antidiscrimination laws. These activities include investigating complaints of discrimination on the basis of race, color, national origin, sex, disability, or age. OSEP and OCR also help states and schools meet their federal requirements and improve the quality of education by issuing policy guidance, disseminating information, and providing technical assistance to individual institutions. Education funds and oversees numerous national and regional research and technical assistance providers, such as regional educational laboratories and national clearinghouses that disseminate information, research, and other support to policymakers and educators at the state and local levels. For example, the Office of Special Education and Rehabilitative Services, which includes OSEP, supports a network of more than 40 national technical assistance centers, each with a focus on some aspect of special education. At state and district levels, various parties may be involved in providing students with opportunities in PE and extracurricular athletics. For instance, many states have requirements regarding their districts’ provision of PE, according to data from the Centers for Disease Control and Prevention and the National Association of State Boards of Education. However, policies may vary by state or district, such as the required grades in which PE should be offered, the number of minutes students should be in class, or the specific content and curriculum areas that should be taught. For extracurricular athletics, the National Federation of State High School Associations (NFHS) is the national body that writes the rules of competition for most high school interscholastic sports—the main form of extracurricular athletic competition in schools, which includes varsity and junior-varsity level teams. In turn, each state has its own state high school athletic association which is a voluntary member of the NFHS. According to an NFHS official, NFHS does not establish standards or policies specifically regarding students with disabilities and each state athletic association develops its own standards and policies. In addition, most state high school athletic associations are not part of state departments of education, but are private organizations, according to NFHS officials. Students with disabilities generally attend PE class about the same amount of time as students without disabilities, according to national data and our site visits. Estimates from YRBS 2005 on high school students indicate that students with physical disabilities or long-term health problems attend PE classes for approximately the same amount of time as those without disabilities. For example, 29 percent of students with physical disabilities or long-term health problems attend PE class 5 days a week compared to 34 percent of students without disabilities (see fig. 2). In 12 schools and seven districts we visited, officials also told us that most students with disabilities are in PE class the same amount of time as students without disabilities. Officials from a few schools said that they actually provide a greater amount of instruction time in PE for some students with disabilities who have specific needs and take both general PE classes and specialized PE classes. National data show that among districts that require PE, many have policies allowing student to be exempt from PE due to long-term physical, medical, or cognitive disabilities. However, exemptions for students with disabilities were very rare according to officials in the 20 schools we visited. Rather than exempting students with disabilities from PE class entirely, a few schools we visited excuse some students from certain activities. For example, officials from one high school in New Jersey we visited told us that a few students with medical conditions, such as asthma, may be exempt from particular activities but that they are given alternative activities. Schools generally provided students with and without disabilities comparable opportunities to participate in PE, according to national data and our site visits. According to 2005 YRBS estimates, students with and without physical disabilities or long-term health problems spend similar amounts of time actually engaged in physical activity per class. For instance, 8 percent of students with physical disabilities or long-term health problems spend 41 to 50 minutes in a typical PE class exercisin playing sports compared to 9 percent of students without these disabilities. students we spoke with, said that students with disabilities often participate in PE to the same extent as their peers without disabilities. In addition, almost all state, district, and school officials we spoke with reported they rarely received complaints about PE from parents of students with disabilities. Several state and district officials said that while they hear complaints from parents in other areas, PE is generally an area in which parents se satisfied. Many district and schools officials, as well as some Various factors may affect students’ experience in PE, such as their school level (e.g., elementary or middle) or their type of disability. For instance, some parents and school officials said that PE teachers in elementary school may be able to more easily integrate students with disabilities in y their classes than those in secondary schools because peers in elementar school are more accepting, the equipment is more varied, and there is less focus on competitive games than in secondary school, which may be harder for students with disabilities to participate. Some district and school officials also said that middle school can be particularly difficu some students with disabilities who may have more difficulty changing into PE uniforms or opening combination locks on their PE lockers. However, some schools overcame these challenges by waiving their PE uniform policy or providing key locks to accommodate students with disabilities. In addition, some parents and school officials noted that ir whether a student has a physical or cognitive disability may affect the experience. For example, officials in one school said that while students with physical impairments required more accommodations, they also may have an easier time expressing their needs than students with cognitive disabilities. On the other hand, many district and school officials told us that students with emotional or behavioral disabilities usually participate in the same activities as other students and often excelled physically. Some noted that PE class can provide them a physical outlet that helps them focus their energy. These estimates have 95 percent confidence intervals within plus or minus 3 percentage points. National data indicate that most students with disabilities who took PE did so in general education classes with students who do not have disabilities. Specifically, 92 percent of students with disabilities in grades 1 through 7 took PE in general education classes, according to SEELS estimates. In grades 7 through 12, 88 percent of students with disabilities took PE in general education classes, according to NLTS2 estimates. Similarly, in most of the schools we visited, the majority of IDEA students received PE in a general classroom without any accommodations or modifications. Officials from a number of schools we visited told us that for some students with disabilities who are in special education classes, PE is the one general education class that they attend on a regular basis. School and district officials also noted that the emphasis on educating students with disabilities in an inclusive environment with their peers without disabilities in all subjects also contributes to the high numbers of students in general PE classes. To facilitate participation for students with disabilities in general PE classes, teachers may make accommodations or modifications for students, which are often informal. According to 2006 SHPPS data and many teachers we interviewed, common informal adjustments that teachers use for students with disabilities in PE class include simplifying the instructional content, providing additional skill modeling or repetition, and having peers without disabilities assist in teaching or coaching (see fig. 3). School officials said that these types of informal accommodations or modifications are generally not determined through the IEP process. In contrast, schools we visited said that more formal accommodations or modifications are usually determined through the IEP process and noted on a student’s IEP. These may include modifying assessments (e.g., allowing students to have more time to take tests); using specialized equipment; or having individual aides or assistants work with students during class. For example, officials from a middle school in California told us that some students with disabilities receive accommodations when taking tests for all their classes, including PE, and this is written on students’ IEPs. Similarly, officials from schools we visited who had individual aides help some students with disabilities in PE class said the use of these aides was also determined on a case-by-case basis through the IEP process. Some schools accommodate students with more severe motor development delays by providing specialized or “adapted” PE instruction, using various approaches. Nationally, less than 11 percent of students with disabilities under IDEA received adapted PE in 2000 and 2001, according to SEELS and NLTS2 estimates. Among the schools and districts we visited, the provision of adapted PE varied. Officials from 8 of the 20 schools we visited reported that no students received adapted PE services at the time we visited. The remaining schools reported that between 2 and 29 percent of IDEA students—who represent most students with disabilities—received adapted PE. Many districts we visited told us that the provision of adapted PE for individual students was always determined through the IEP process. Districts in Minnesota also relied on standard statewide criteria to provide adapted PE, such as formal tests evaluating motor skills or fitness levels. PE muse taught PE techer with dditionpted PE licenbased on or review of te docment nd interview with te offici. According to te offici, techer trining progr encoge propective PE techer to oin thi licene to e more mrketable, nd thus, there re mny more certified pted PE techer in the te thctual poition. Schools we visited provide adapted PE using different approaches. Students may receive direct instruction from an adapted PE teacher, the adapted PE teacher may consult with the general PE teacher who actually provides the specific modifications or accommodations to the student, or schools may use a combination of both approaches. Some states require that adapted PE be taught by a PE teacher with a supplemental license, including California and Minnesota, based on state documents we reviewed and state officials we interviewed. Nationwide, 12 states offer or require a supplemental adapted PE license to teach adapted PE, according to the Center on Disability Health and Adapted Physical Activity. Some adapted PE teachers travel to several schools to provide services in the districts we visited. In one district we visited in Florida, the sole adapted PE teacher overcame challenges in scheduling her time by having students who took adapted PE bussed to a central school so she could provide them with services. Some parents we spoke with were pleased with their children’s adapted PE experiences, while others said that their experiences varied and depended on factors such as individual teaching styles and the extent to which students who received adapted PE were integrated into the general PE class. Cn Do It” progrm, ponored y the Ntionl Intitte of Helth in the Deprtment of Helth nd HService. Sdent with phyicl or cognitive disabilitiern Preidentil Active Lifetyle Awrd if they prticipte in 0 minte of ctivity 5 d week for 6 week. The ditrict o has hyrid PE nd port progrm for dent with disabilitie. Dring the chool dy, dentrticipte in ctivitie such as wheelchir basketball or pted sailing. The ditrict hasined pecilized equipment nd fcility ce throgh commnity prtnerhip, such as with the locsailing club nd conty golf coe. Fewer students receive PE only in separate or self-contained classes as opposed to receiving PE in general classes with other students, according to national data and our site visits. Our estimates from SEELS indicate that 7 percent were in a self-contained special education classroom for PE in 2000. Among eight districts we visited that could provide estimates, the percentage of students with disabilities who received all of their PE in a self-contained, adapted PE class ranged from 0 percent to 6 percent. However, some self-contained PE classes may follow the same curriculum as general PE classes. Officials in two schools we visited told us that students with severe disabilities were in self-contained special education classes and took general PE together as a class. A notable challenge to serving students with disabilities in general PE classes is the lack of sufficient training or experience among PE or classroom teachers, according to our interviews and other research. Many state, district, and school officials we interviewed said that PE teachers typically take one course on working with students with disabilities in their undergraduate training. This coursework may not always include practical experience working with students with disabilities, and several studies found that PE teachers reported feeling insufficiently prepared to teach students with disabilities. According to SHPPS, only 32 percent of teachers surveyed who taught required PE classes reported having received training on how to work with students with long-term physical, medical, or cognitive disabilities in the 2 years preceding the 2006 study. Officials from several disability, health, or PE associations said that general PE or classroom teachers who lack training or experience teaching students with disabilities may not know what to do or how to provide the appropriate level of accommodations or effectively adapt their instruction. Their lack of training and experience may result in their excusing some students with disabilities from portions of the class, according to two associations we interviewed. Some parents we spoke with described instances in which their children sat on the side during certain activities or were not encouraged to be as active as their classmates. Several parents noted that while PE teachers may not intend to be exclusionary, they were not sufficiently trained on how to include students with disabilities in their classes. Some state, district, and school officials we interviewed said teachers who teach general PE need more training opportunities regarding students with disabilities, yet resources for training are not always available. Some officials from states and districts said they provide professional development opportunities to teachers through online training, workshops, or conferences. However, budget shortfalls limited their ability to offer additional training opportunities for PE teachers. Some district officials told us that they would like to receive more information on meeting the needs of students with disabilities in general PE settings. For instance, officials from a rural district in Georgia said that they did not have any teachers who specifically taught adapted PE and all students were taught by a general PE or special education teacher. Officials from this district acknowledged that their teachers would benefit from additional training in adapted PE, but told us that they could only provide training if it were available at a low cost or if they had additional resources. Officials in many of the districts and schools we visited reported that budget constraints created a challenge to providing PE to students with disabilities, particularly in the general PE classroom. For example, many district and school officials, including PE teachers, told us that budget constraints in their schools have resulted in large general PE classes and have hindered teachers’ ability to provide students with disabilities the individual attention they may need. In both of the districts in California that we visited, teachers and school officials said that PE classes can include over 45 students, often with only one teacher. Several teachers noted that they would like to provide more individual attention to students with disabilities in their classes but cannot do so while managing the rest of the class. Another PE teacher in Minnesota said that large classes in a spacious and noisy environment can be particularly hard for students with emotional or behavioral disabilities who are very sensitive to such stimuli. District and school officials said that additional staff, such as paraprofessional aides who could work one-on-one with students with disabilities, would help address the challenge of large classes. However, officials said that budget constraints limit the use of aides. Finally, teachers or district officials in several states told us that limited resources for equipment, including special adapted equipment or facility renovations for outdated gymnasiums, hindered their ability to effectively provide opportunities to students with disabilities. A few state, district, school, and disability association officials we spoke with cited a lack of importance placed on PE as a school subject compared with academic subjects. Some state, district, and school officials noted that the greater emphasis on assessments for reading and math as required by federal law has led them to reduce the number of PE classes that their students are required to take or PE elective courses that they can offer. However, other district officials said their PE offerings have not changed because state laws dictate PE requirements for students, among other reasons. Officials from some districts, schools, disability associations, and parent groups also said that parents of children with disabilities may view PE as a relatively low priority and, subsequently, have low expectations with respect to their children’s participation or performance. National data show that students with disabilities participate in extracurricular athletics but do not fully distinguish whether these athletic opportunities are offered through schools or out-of-school (i.e., through community-based) programs. According to SEELS estimates, 41 percent of students with disabilities in grades 1 through 7 participated in school or community-based extracurricular athletics in 2000. Additionally, according to NLTS2 estimates, 33 percent of students with disabilities in grades 7 through 12 participated in school or community-based extracurricular athletics in 2001. In these studies, more students with disabilities reported participating in school or community-based sports teams than in other extracurricular activities surveyed, such as performance groups (e.g., band or choir), scouting, or special interest groups (e.g., chess or other hobby clubs). However, sports participation was consistently higher for boys with disabilities than for girls with disabilities across age groups. In addition, according to our interviews with Education officials and researchers, we found no national data that was reliable for our purpose of comparing the participation of students with disabilities to students without disabilities in extracurricular athletics. Among the schools we visited, we found that IDEA students participated in school-based extracurricular athletics at varying rates, but at a lower rate than their peers without disabilities. The percentage of IDEA students who participated in “traditional” school-based extracurricular athletics (that is, athletics not specially designed for students with disabilities) ranged from 6 to 25 percent among the 11 schools we visited that could provide estimates. In contrast, the percent of non-IDEA students who participated in traditional extracurricular athletics ranged from 18 to 73 percent in these schools (see fig. 4). In all cases, the participation rates for IDEA students were lower than for non-IDEA students, ranging from 10 to 56 percentage points lower. A number of factors may contribute to these differences among schools, such as outreach to students with disabilities, priorities of school officials, and the level of competitiveness among athletic teams. Some schools we visited in Florida and California actively reached out to some students with disabilities in an effort to recruit them for extracurricular athletics. Several of the special education teachers in a California school we visited also coached different sports teams; this dual role enabled them to encourage students with disabilities to participate, according to one official. Additionally, the leadership of school officials may affect students’ participation in extracurricular athletics, according to some parents, school officials, and officials from disability groups. One parent of a child with a learning disability in California told us that administrators at her daughter’s school recognized that extracurricular athletics was an area where she excelled and wrote into her IEP that she could participate even if her grades did not qualify her. In contrast, the competitiveness of some athletic teams can negatively affect participation among students with disabilities. A district official we spoke with in Texas told us that extracurricular athletics in his district were very competitive and that it was unlikely that many students with disabilities would make these teams. Additionally, one parent told us that she took her daughter who has a neurological disorder out of the school volleyball team because of competitive pressures from other parents and students. Perceptions among students, parents, and schools can also affect student participation in extracurricular athletics, according to a number of officials from schools and disability organizations we interviewed. Specifically, students with disabilities may not perceive themselves as athletic and therefore are reluctant to try out for teams, while parents or schools may resist the idea of students participating due to safety concerns. Student participation on sports teams varies by disability. In most of the schools we visited, officials told us that students with disabilities who participated in extracurricular athletics often had mild cognitive disabilities or learning disabilities. Additionally, national data from NLTS2 indicates that students with hearing impairments, speech impairments, learning disabilities, or other health impairments reported participating on sports teams at a higher rate compared to students with orthopedic impairments, mental retardation, visual impairments, autism, or multiple disabilities. Many officials from schools and disability associations we interviewed said that students with physical disabilities have fewer opportunities in extracurricular athletics compared to students with cognitive disabilities because fewer programs were designed for them. For example, officials said that while students with learning or cognitive disabilities could participate in traditional sports teams with little to no modifications, students with physical disabilities may not be able to do so. Officials from several schools and disability groups also noted that Special Olympics has had a strong influence nationally in providing opportunities for students with intellectual disabilities but that there is not a similar organization for students with physical disabilities. Officials from all 15 districts we spoke with told us that all students in their districts, including those with disabilities, are allowed to try out for their school’s interscholastic athletics teams. In addition, many schools we visited said they have at least some teams that accept all students who try out. These teams were usually those based on individual performance, such as track or swimming, according to school officials. In contrast, other teams, such as soccer or basketball, were very competitive and had a limited number of positions available. Many school officials told us that if a student did not get selected to play on a competitive team, they were encouraged by coaches to play on a team that accepted all students. School officials reported that they provide varying levels of accommodations for students with disabilities on traditional teams, usually depending on the student’s disability. For example, a high school we visited in New Jersey provided an interpreter for a deaf student on the baseball team, and a middle school we visited in California modified rules for deaf and blind students on their wrestling team. State athletic associations in Texas and Minnesota sometimes granted exceptions to their policy on student age limits for students with disabilities. Other methods to aid students with disabilities in extracurricular athletics were less formal. Coaches told us that they may vary their style to meet individual student needs as a practice of good coaching but did not provide any specific accommodations for students with disabilities. For example, coaches in schools that we visited in New Jersey and Florida told us that they repeat rules or give instructions by modeling an activity for students with cognitive or learning disabilities in order to help them better understand game rules. A few schools provided extracurricular athletic opportunities to students with disabilities through formal partnerships with community programs, most commonly Special Olympics. While many schools provided referrals to Special Olympics but did not otherwise facilitate participation, a few schools we visited provided transportation, coaching, or funds to enable students with intellectual disabilities to participate in Special Olympics. Some of these schools also incorporated Special Olympics coaching and activities into class time for students. In addition, a few schools and districts we visited partnered with other community organizations to provide extracurricular athletic opportunities. In Maryland, one high school we visited collaborated with a local golf club to provide extracurricular golf to students with disabilities and an elementary school partnered with the city’s parks and recreation program to provide a sports program for these students. More recently, it introdced pted owling to ddress chllenge tht ditrictced regrding the lck of sufficient ner of dent to form tem or gret dince etween ditrict tht mde regr competition difficlt. Sdent go to their locowling lley to compete “virtually” indent from other chool. Across ll port, dent with phyicl impirment nd cognitive impirment ply on epte te. Prticipting denty ern vity letter. A quarter of the ditrict in the te offer t least one pted port progrm, ccording to offici. allow students with disabilities an opportunity to participate to the fullest extent possible in mainstream PE and athletic programs; and (3) ensure that alternative PE and athletic programs are available, such as adapted sports (only students with disabilities) or unified sports (includes students with and without disabilities). County school systems must develop policies and procedures to promote and protect the inclusion of students with disabilities into mainstream PE and athletic programs and report annually to the State Department of Education. The State Board of Education must adopt a model policy to assist with implementation, while the State Department of Education must monitor compliance and provide technical assistance to county school systems. New Jersey passed a much less detailed law in August 2009 that directs the New Jersey State Interscholastic Athletic Association to establish interscholastic adapted athletics for students with visual impairments or physical disabilities who are participating in an adapted athletic program developed by a school district. The New Jersey State Interscholastic Athletic Association must require coaches of adapted athletics to receive training specific to that program but the law does not outline other requirements. Officials from several districts and state athletic associations said they generally lacked information that would help them provide extracurricular athletic opportunities for students with disabilities. For instance, a number of district officials said coaches do not receive training specifically on how to work with students with disabilities. Some parents we spoke with said that coaches who do not have such training can be overly focused on winning and fail to fully include students with disabilities on their teams. Other parents noted that coaches are rarely at IEP meetings, and one parent in Minnesota told us that a coach improperly removed her child from a team for being academically ineligible without considering the student’s IEP. Additionally, some states and districts lacked information on how to implement new adapted athletic programs for students with disabilities. For example, while the Florida High School Athletic Association planned to initiate a wheelchair track event in the coming year, state and district officials we spoke with said they had questions about how eligibility would be determined and how to provide the equipment needed for these students to participate. Other state athletic associations we interviewed were hesitant to consider offering adapted athletic programs because they did not know how competitive events would be held or how teams would be formed. Further, some districts lacked information and clarity regarding their responsibilities to provide opportunities under Section 504 for students with disabilities who want to participate in extracurricular athletics equal to those provided to other students. Officials in two districts and several disability associations told us that Education’s Section 504 regulations regarding schools’ responsibilities to provide extracurricular opportunities are ambiguous. For example, a few disability associations noted that there is lack of clarity regarding how “equal opportunity” should be defined. Officials from another district questioned whether their responsibilities included providing specifically designed programs for students with disabilities, such as separate adapted athletics, particularly within a school environment focused on greater inclusion for students with disabilities. State, district, and school officials often cited budget constraints as another key challenge to expanding athletic opportunities to all students, including those with disabilities. Officials from seven of the 10 districts we visited said that overall budget constraints have affected funding for their entire athletics program, including funds for personnel and equipment. For example, a rural district we visited in Florida told us that they do not have the resources to fund many personnel positions or buy equipment and that they had to eliminate their athletic director position this year. Another district in California told us that the state’s budget situation has negatively impacted all extracurricular athletics and the district is considering dropping their sports teams in the near future. In addition, district officials cited as a concern the high cost of transporting students with disabilities. Officials in one district in Florida noted that transporting students with disabilities was the largest cost in the district’s extracurricular athletic program. Some school and district officials we spoke with had concerns about the facilities used for competitions. A district official in Florida noted that many facilities are old and the costs associated with making these fields accessible for students with disabilities would be challenging. Finally, some school officials said that budget limitations for their athletic programs limit their ability to create new adapted teams, due to costs associated with training and maintaining coaches and aides to work with students. While OSEP monitors states’ implementation of IDEA and provides information, resources, and technical assistance to states and schools on teaching students with disabilities, very little of it is related to PE or extracurricular athletics, according to OSEP officials and our review of Education Web sites. Beyond technical assistance to current or prospective grantees for special education grants, OSEP has not provided any other relevant information or support. OSEP officials said that OSEP has not used its limited resources to provide such information or support because IDEA’s monitoring priorities are focused on other areas and the office has not received relevant concerns or complaints from states, schools, or parents. Similarly, OCR has not widely disseminated any detailed guidance or information on schools’ responsibilities to provide opportunities in PE or extracurricular athletics for students with disabilities under Section 504, according to agency officials and our review of OCR’s Web site. Based on data from OCR officials, 12,543 complaints were made to its office, from fiscal year 2005 to fiscal year 2009, regarding elementary and secondary school students with disabilities. Of these, 108 of the complaints filed between fiscal year 2005 and fiscal year 2009 (less than 1 percent) pertained to discrimination regarding student participation in PE or extracurricular athletics. From these 108 complaints, we reviewed a subset of 20 and found that 16 alleged that a student was denied equal opportunity to participate in extracurricular athletics. When resolving a Section 504 complaint investigation, OCR may provide detailed guidance or training to help individual districts or schools meet their responsibilities regarding PE or athletics. However, according to OCR officials, such guidance is tailored to the needs of the district or school based on facts obtained through an OCR investigation, and OCR has not yet issued guidance specifically on Section 504’s requirements concerning PE or athletics. Our review of OCR guidance found several documents that state that students with disabilities may not be excluded on the basis of a disability from an extracurricular activity, which may include athletics, and that students must be provided opportunities to participate in these activities equal to those of other students. However, with regard to PE and extracurricular athletics, these documents do not provide information beyond what is stated in the Section 504 regulations. In contrast, OCR has provided more detailed guidance in other civil rights areas, such as letters, pamphlets, and question and answer sheets. OCR officials said they have not yet issued additional guidance on extracurricular athletic opportunities for students with disabilities in part due to resource constraints and their focus on other areas of civil rights, such as gender discrimination in athletics or harassment based on race, sex, and disability. They also noted that these areas have received more official complaints and persistent concerns from constituents such as parents and community groups. Through several grant programs, Education has administered a relatively small amount of funds specifically to support PE or extracurricular athletic opportunities for students with disabilities. For example, OSEP administers a grant program that provides funding to institutes of higher education for the development of special education personnel. In fiscal year 2009, of the $91 million administered for these grants, about 2 percent was used to support programs intended to provide undergraduate or graduate training in the area of adapted PE. In the same fiscal year, OSEP also administered about $8 million, as specifically directed by Congress, to support Special Olympics programs. These federal funds are administered under the Special Olympics Sport and Empowerment Act of 2004 and are intended to increase the participation of individuals with intellectual disabilities in sports and competitive activities. They are also used to design and implement Special Olympics education programs that can be integrated into classroom instruction. In addition, the Office of Safe and Drug-Free Schools (OSDFS) administers the Carol M. White Physical Education Program—the only dedicated PE program sponsored by Education. However, according to OSDFS officials, less than 1 percent of the program’s overall funding of $77 million in fiscal year 2009 was used to serve the specific needs of students with disabilities. To serve these needs, grantees used funds to purchase specialized equipment or provide training to teachers and staff on how to adapt their general PE programming and activities for students with disabilities. Officials from several states and many districts we interviewed said that they could benefit from Education helping states and schools share relevant information, such as practices or resources regarding PE and extracurricular athletics for students with disabilities. For example, officials from one state told us that when they recently developed new guidance for schools in PE and extracurricular athletics for students with disabilities, they could have benefited from information on what other states were doing in these areas. Officials from another state noted that Education could collect information on promising practices and share these with institutes of higher education that offer PE teacher training programs. Additionally, officials from 14 out of 15 districts we interviewed indicated that additional information or resources from Education on how to offer PE or extracurricular athletic opportunities to students with disabilities would be useful. For example, an adapted PE coordinator in a California district said that while he draws on existing informational resources, such as private sector Web sites and state or local conferences, he is constantly looking for additional information from others in his field on good practices and policies. Similarly, a PE and health director in a Florida district noted that as more states implement new wheelchair track and field activities in the general high school athletic program, supporting information and procedural practices would be helpful. Further, a number of officials from various associations we interviewed, including those representing special education, PE, and adapted sports, said that Education should share information on successful practices or provide informational resources such as factsheets or toolkits. Some association officials also said that Education could give additional attention to PE or extracurricular athletics, by providing information online or promoting greater awareness through media campaigns. This attention would increase awareness among schools about the types of opportunities that could be provided, promote interest among students with disabilities to participate in such activities, and address concerns and questions among parents. Officials from many districts and several disability groups also said that additional guidance from Education, specifically on extracurricular athletics, would be useful. Officials in 12 of 15 districts we spoke with told us that additional clarification and communication from Education regarding schools’ responsibilities under Section 504 in providing extracurricular athletics to students with disabilities would be useful. Specifically, one district official noted that the lack of clarity in federal requirements regarding how schools should define and provide equal opportunities serves as a disincentive for schools to take proactive steps to provide additional opportunities for students with disabilities. Another said that while most districts give a “good faith effort” to provide equal opportunities in extracurricular activities, including athletics, district officials probably do not fully understand what it means to provide equal opportunities until a problem arises, such as parent concerns or complaints. According to this official, districts would benefit from additional information on the legal issues or processes involved in allowing students with disabilities to participate in extracurricular athletics, as well as how to document instances in which a student is or is not allowed to participate. Officials from another district noted that school responsibilities related to special education services or supports for activities that occurred after school, such as extracurricular athletics, as opposed to those during the school day, were not entirely clear. Similarly, officials from several disability groups noted that OCR could do more to ensure that schools provide equal opportunities in extracurricular athletics, such as by reminding schools of their responsibilities and what they entail (e.g., through informational factsheets). Officials from one disability association also said that they do not believe parents and students with disabilities know where to turn for help with respect to problems regarding extracurricular athletics and that OCR could help inform parents and schools by issuing a letter reminding them that OCR is a resource. All students, including those with disabilities, benefit from the positive effects that physical activity and school athletics have on an individual’s health, social well-being, and self-esteem. While available data show that students with disabilities generally are participating in physical education classes to a similar extent as students without disabilities and that some also are involved in school sports, we found that many districts and schools we visited are interested in improving how they provide physical education and athletic opportunities to these students. Garnering additional resources to provide such opportunities is difficult at a time when states and districts are operating under severe fiscal constraints. However, certain districts and schools may have developed approaches or activities to engage students with disabilities more fully in the PE classroom or in extracurricular athletics, and leveraging their expertise could be helpful to other schools that face challenges in these areas. Some schools may also be uncertain about their exact responsibilities to provide opportunities to students with disabilities equal to those of other students, particularly regarding extracurricular athletics. Without additional clarification from Education, schools may not be providing some students full access to these opportunities. While Education has not addressed these issues because it has targeted its limited resources to other areas, focusing some of its existing resources on helping schools provide opportunities in PE and extracurricular athletics could yield important benefits and enable students with disabilities to more fully experience the rewards of physical activity. To help states and schools access existing knowledge and resources, we recommend that the Secretary of Education facilitate information sharing among states and schools on ways to provide opportunities in PE and extracurricular athletics to students with disabilities. For instance, Education could provide Web site links to resources or practices used by states, districts, schools, or organizations in PE or extracurricular athletics for students with disabilities. Such information could be posted on its Web site or discussed at conferences or Webinars. To ensure that schools are aware of their responsibilities and that students with disabilities consistently have opportunities to participate in extracurricular athletics equal to those of other students, we recommend that Education clarify and communicate schools’ responsibilities under Section 504 of the Rehabilitation Act regarding the provision of extracurricular athletics. For example, Education could provide clarifying guidance to schools through its Web site or at conferences. We provided a draft of the report to the Department of Education for review and comment. Education’s written comments are presented in appendix IV. Education agreed with our two recommendations to provide more information and guidance. Specifically, Education concurred that further information sharing on providing opportunities in PE and extracurricular athletics to students with disabilities would be helpful to states, schools, and students. It plans to identify useful information on this subject and share such information by posting it on its Web site, www.ed.gov. Education also agreed that it is important for schools to be aware of their responsibilities under Section 504 to provide opportunities in extracurricular athletics equal to those of other students and plans to issue additional guidance addressing this in fiscal year 2011. It also provided technical comments, which we incorporated throughout the draft as appropriate. We are sending copies to appropriate congressional committees, the Secretary of Education, and other interested parties. The report also will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-7215 or ashbyc@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix V. To address the objectives of this study, we used a variety of methods. Specifically, we examined pertinent data from four nationally representative studies; conducted site visits in five states and phone interviews with two conducted 21 interviews with representatives from national associations, community organizations, parent advocacy centers, or researchers in the areas of disability, health, physical education (PE), special education, and athletics; and interviewed officials from the Department of Education (Education) about Education’s oversight and technical assistance efforts, and reviewed relevant federal laws, regulations, guidance, and other agency documentation and data pertaining to PE and extracurricular athletics for students with disabilities. To obtain national information on the extent to which schools provide students with disabilities opportunities in PE and extracurricular athletics, we identified four nationally representative studies with the most recent data available for our purposes. Specifically, we examined pertinent data from the 2005 Youth Risk Behavior Survey (YRBS) and cited published data from the 2006 School Health Policies and Programs Study (SHPPS). Both studies are administered by the Department of Health and Human Services’ Centers for Disease Control and Prevention. YRBS surveys students in grades 9 through 12 on health-risk behaviors, and SHPPS collects survey information on school health programs and practices from state, district, and school officials, as well as teachers and parents. We also examined 2001 data from Education’s National Longitudinal Transition Study-2 (NLTS2) and 2000 data from Education’s Special Education Elementary Longitudinal Study (SEELS). To calculate the percentage of students with disabilities in these studies who did not participate in sports, we included students who reported not participating in any extracurricular activities, as well as students who reported participating in some extracurricular activities but not sports. SEELS and NLTS2 consist of nationally representative samples of students who received special education services. Students in SEELS were 6 to 12 years old and in at least the first grade in 1999, and students in NLTS2 were 13 to 16 years old and in at least the seventh grade in 2000 (see table 2). We selected these sources, in consultation with GAO methodologists, because they contained nationally representative information about the participation of students with disabilities in PE and extracurricular athletics and were determined to be sufficiently reliable for the purposes of our report. We examined the studies’ technical documentation to determine their purpose, the population described, the sample surveyed, data collection methods, and the focus of questions and variables related to students with disabilities and PE and extracurricular athletics. Using this documentation, and through interviews with knowledgeable agency officials, we evaluated the reliability and quality of these data, as well as the extent to which they could be used to answer our questions. Additionally, while more recent data are available from NLTS2 and SEELS, we determined, through interviews with knowledgeable agency officials and our own analyses, that the data collected during the first wave of these two longitudinal studies were preferable, for substantive and technical reasons, for the purposes of our report. Such substantive issues included relevant student grade levels, and technical reasons included high attrition rates in later waves. We used the same methodology as Education to produce estimates on the participation in extracurricular athletics among students with disabilities. We determined that the data we analyzed were sufficiently reliable for our purposes. We assessed the quality, reliability, and usability of the data for reporting descriptive statistics on the demographic characteristics of students with disabilities (e.g., grade level, gender, and type of disability), and their participation in PE and extracurricular athletics. Because each of the three surveys we provided estimates for (SEELS, NLTS2, and YRBS) followed a selection procedure based on random selections, each sample is only one of a number of samples that might have been drawn. To assess the precision of estimates from these three surveys, we calculated confidence intervals at the 95 percent level for each measure. A confidence interval gives an estimated range of values, calculated from sample data, which is likely to include the true measure of the population. All percentage estimates from SEELS and NLTS2 used in this report have 95 percent confidence intervals within plus or minus 4 percentage points, unless otherwise noted. All percentage estimates from YRBS used in this report have 95 percent confidence intervals within plus or minus 7 percentage points, unless otherwise noted. All SHPPS estimates have 95 percent confidence intervals within plus or minus 5.5 percentage points of the estimates, unless otherwise noted. All comparisons made between groups are significant, unless otherwise noted. We analyzed data from YRBS, SEELS, and NLTS2 using SUDAAN (Survey Data Analysis) ®, a statistical analysis software package with capabilities to analyze surveys with a stratified, multistaged cluster sampling design, such as that found in YRBS, SEELS, and NLTS2. To address all three of our objectives, we conducted site visits in five states (California, Florida, Maryland, Minnesota, and New Jersey) and interviewed officials from state departments of education, state athletic associations, school districts, and schools. We also interviewed teachers, coaches, parents, and students. In each of these states we visited two school districts and four schools. In addition, we conducted phone interviews with officials from state agencies and two to three school district offices in Georgia and Texas. We selected these seven states because they were in the top half of states in numbers of both IDEA students and Section 504 students and varied in whether they had policies requiring elementary, middle, and secondary schools to teach PE, and geographic location. We chose school districts that provided variation in their percentages of students with disabilities, socio-economic status, and classification as urban, rural, or suburban. In deciding between multiple states or districts that met our selection criteria, we purposely chose some that external stakeholders recommended as having innovative or promising practices. We spoke with parents by contacting one to two Parent Training and Information Centers in each state we visited. These parent centers are funded by Education and provide information and support to families of children with disabilities of all ages. For this study, staff from these centers asked parent members to voluntarily meet with GAO to share their children’s experiences in PE and extracurricular athletics. At some meetings, children also attended with their parents and shared their views. We cannot generalize to all states, school districts, or schools based on these site visits. Through our site visits and phone interviews, we gathered information on relevant state and district policies, available data or estimates on students’ participation in PE and extracurricular athletics, and factors that may affect participation. We also inquired about the approaches districts and schools use to provide opportunities in PE and extracurricular athletics, key challenges to providing these opportunities, and areas in which federal assistance has been or would be useful. When possible, we reviewed written documentation and policies from state agencies, districts, and schools. We also observed some classrooms and athletic practices and spoke with groups of students at schools we visited. To enhance the consistency of the information we obtained from interviews, we used structured sets of questions for officials from state agencies, district offices, and schools, as well as parents. We analyzed interview information by developing and testing a coding structure and coding the interviews using NVivo, a qualitative analysis software. One team member coded all the interviews, and another reviewed 100 percent of the coded output for accuracy. Using a short questionnaire, we obtained estimates from officials in the districts and schools we visited of the number of students with disabilities in PE and extracurricular athletics. We inquired about students covered under IDEA, as well as those covered under Section 504, but not IDEA. However, we did not include numbers of students covered only under Section 504 in these tables due to the number of missing responses and the few students in each district or school in this group. We visited two schools in each district we visited. Unless otherwise noted, differences made between groups in these tables are not statistically significant at p-value less than 0.05. In addition to the contact above, Elizabeth Morrison (Assistant Director), Theresa Lo (analyst-in-charge), Hedieh Rahmanou Fusfield, and Michael Pahr made significant contributions to this report. Steven Putansu, Dae Park, and Shana Wallace assisted with the data analysis and methodology. Alexander Galuten provided legal support. James Bennett assisted with graphics. Susannah Compton and Jessica Orr assisted in report development. | Research has established that physical activity and participation in athletics provides important health and social benefits for children. Certain federal laws help ensure that kindergarten-12th grade schools provide students with disabilities opportunities to participate in physical education (PE) and extracurricular athletics equal to those of their peers. However, national associations have questioned whether students with disabilities receive opportunities similar to their peers. Regarding students with disabilities, GAO was asked to examine (1) what is known about the PE opportunities that schools provide, and how do schools provide these; (2) what is known about the extracurricular athletic opportunities that schools provide, and how do schools provide these; and (3) how the Department of Education (Education) assists states and schools in these areas. GAO analyzed federal survey data; reviewed relevant federal laws and regulations; and interviewed state, district, and school officials in selected states, as well as parents and disability association officials. Schools provide students with and without disabilities similar opportunities to participate in PE but face challenges when serving students with disabilities. Students with disabilities spend similar amounts of time in PE class and exercising in class as students without disabilities, according to national data and GAO site visits. Most students with disabilities take PE with other students in general PE classes. To facilitate their participation, teachers may make accommodations for some students, such as providing additional modeling or repetition. Many state, district, and school officials GAO interviewed cited teacher preparation and budget constraints as key challenges to serving students with disabilities in general PE classes. For example, they said general PE teachers need more training opportunities on working specifically with students with disabilities, yet resources for training are not always available. Limited national data suggest that students with disabilities participate in extracurricular athletics, but do not distinguish whether these opportunities are offered through schools or community programs. Among the schools GAO visited, students with disabilities participated in athletics at varying rates, but at consistently lower rates than students without disabilities. Several factors, such as a student's disability type or outreach to students, may affect participation. Some schools or districts GAO interviewed provided opportunities by partnering with community programs or offering athletics designed specifically for students with disabilities, such as wheelchair basketball. District and school officials GAO interviewed cited a lack of information on ways to expand athletic opportunities, lack of clarity regarding schools' responsibilities, and budget constraints as key challenges. Education has provided little information or guidance on PE or extracurricular athletics for students with disabilities, and some states and districts GAO interviewed said more would be useful. According to agency officials, Education has not provided much information or guidance because it has targeted its limited resources on other areas, such as monitoring priorities specified in federal law. Officials from several states and many districts said they could benefit from Education helping states and schools to share information on practices or resources regarding PE and athletics for students with disabilities. Officials from districts and disability groups also said more clarification from Education on schools' responsibilities under federal law on extracurricular athletics for students with disabilities would be useful. |
MDA is the agency responsible for designing, developing, producing, and procuring targets for testing the nation’s ballistic missile defense system. In 2001, MDA initiated a Targets and Countermeasures Program Office. The MDA’s Targets and Countermeasures program oversees the design, development, fabrication, and production of ballistic missile targets and countermeasures to be used in the MDA flight test program. These targets must represent the full spectrum of threat missile capabilities (separating and nonseparating reentry vehicles, varying radar cross sections, countermeasures, etc.) and ranges. The appropriate targets are engaged by both strategic and tactical missile defense systems developed by MDA, the Army, and the Navy. MDA originally used numerous contractors to design and build the various targets; however in December 2003 Lockheed Martin was awarded a contract to be the lead systems integrator for most of the existing targets. The lead systems integrator was chosen to provide system-level management, implement a single organization and management structure, and to manage all processes. Additionally, MDA contracted with Lockheed Martin to plan and develop a new family of targets called the Flexible Target Family (FTF) because existing targets were deemed no longer able to meet test needs due to aging components and an evolving threat. MDA’s original goals for FTF included commonality of components across the family of short, medium and long-range targets, the purchase of inventory, reduced cycle time in the building of individual targets, increased complexity, cost savings, and ground, air, and sea launch capabilities. Although the development of FTF began in 2005, the first FTF target has yet to be delivered. While MDA manages the targets program, the Missile Defense Executive Board (MDEB), which was created in March 2007, provides advice and recommendations on the missile defense program. Specifically, the MDEB reviews the implementation of strategic policies and plans, program priorities, and investment options for protecting the United States and its allies from missile attacks. The MDEB is chaired by the Under Secretary of Defense for Acquisition, Technology and Logistics, who is authorized to make recommendations to the Deputy Secretary of Defense on missile defense issues. According to its charter, the board will provide to the Under Secretary a recommended strategic program plan and a feasible funding strategy for approval. Recommendations will be based on a business case analysis that considers the best approach to field integrated defense capabilities in support of joint objectives. A consistent business case analysis approach will facilitate the delivery of capabilities to the war fighter in a timely, efficient, cost-effective manner. The board will review the missile defense program and recommend changes in the missile defense strategic plan. In addition, Congress has mandated that DOT&E annually assess the adequacy and sufficiency of the BMDS test program. Congress has also mandated that DOT&E perform other BMDS program related test and evaluation responsibilities. The number of target failures and anomalies during flight tests has increased over time. From 2002 through 2005, 7 percent (3 of 42 launches) of the targets launched had in-flight target failures and anomalies, while from 2006 through 2007, 16 percent (6 of 38 launches) of the targets launched had in-flight failures or anomalies. Though problems providing reliable and timely targets have negatively affected many of the missile defense elements, THAAD and GMD missile defense elements have been affected the most. GMD experienced a target failure in 2007 in which the primary objectives failed to be achieved. THAAD experienced one failure and two anomalies between 2006 and 2007. The immediate effect of a target delay or problem is on an individual flight test. For example, unavailable targets have delayed tests such as FTT-08 and FTM-15. There are longer term effects of these delays on subsequent tests and demonstration of capability in the flight test program. Specifically, MDA’s flight test program has been restructured to reflect target deficiencies and availability. For example, validating GMD capability has been delayed due to target availability and a failure. The THAAD flight test program has decreased the amount of flight tests due to reduced availability of targets. Additionally, target failures and anomalies caused THAAD to replan key objectives in their flight test plan due to the inability to collect necessary information during flight tests. Since we briefed your staff in July 2008, MDA has experienced two additional target anomalies. The target flown in a recent sensor test (FTX- 03) under-flew the planned trajectory. Test reports indicate that target trajectory fell short of what was planned and thus did not reach the anticipated intercept area. While the anomaly did not prevent MDA from achieving the sensor test objectives, had this been an intercept attempt, the target anomaly would have precluded the launch of the interceptor due to safety restrictions on the test range. In September 2008, the THAAD element was unable to complete a planned flight test because the target missile malfunctioned and did not have enough momentum to reach the intercept area. Therefore, the two THAAD interceptors taking part in the flight test were not launched. At the same time, the unit cost of targets has grown exponentially, from $4.5 to $8.5 million from 2002 to 2006 to current estimates of $32 to $65 million for the targets planned from 2008 to 2010. Many factors contribute to this cost growth, including increased complexity of targets to better reflect an evolving threat and late changes to target requirements on contract. In addition, the shift to a lead systems integrator resulted in an increase in composite hourly labor rates. MDA has also encountered development problems with the FTF program and the total cost of FTF remains unclear. According to information provided by MDA in July 2008, nonrecurring engineering costs plus the cost to purchase some test assets is at least $1 billion, of which about $553 million has already been spent. Difficulties with target availability are traceable to (1) supplier base and problems incorporating requirements into the contracts and establishing program baselines for existing targets and (2) the lack of a sound business case for FTF development. Regarding existing targets, inventory is aging for key components. For example, motors used in some targets are now over 40 years old. In addition, vendors who previously supplied targets to MDA are leaving the market due to lack of business. MDA has also experienced program and requirements issues with the existing targets on contract. According to MDA officials, several target contracts did not capture all of the needed test requirements. Furthermore, target technical, cost, and schedule baselines were not always established in a timely manner. MDA officials acknowledged that they have only recently established these baselines for all missions planned to take place in the next 2 years. MDA began the FTF without establishing a sound business case. We have previously reported on the importance of using a solid, executable business case before committing resources to a new product development effort. A sound business case demonstrates that (1) the identified needs are real and necessary and are best met with the chosen concept and (2) the chosen concept can be developed and produced with existing resources—such as technical knowledge, funding, time and management capacity. According to a July 2006 MDA briefing, the decision to commit to a lead systems integrator and FTF was not based on a formal cost or business case analysis. The briefing further indicated that the FTF cost estimates considered at the time only included integration costs and did not address nonrecurring engineering costs, unit costs for FTF configured targets, or the costs to purchase existing target systems through the prime contract. Nor was there any indication that FTF costs had been reviewed by independent cost organizations within the Department of Defense. It is also unclear whether MDA evaluated all alternatives before committing to a lead systems integrator and the FTF. MDA did perform a study to identify requirements design attributes and candidate configurations for a family of targets by reviewing existing systems. However, in a July 2006 assessment, MDA stated that other alternatives for improving legacy targets were not considered at that time. We also could not identify an original approved FTF acquisition strategy. Originally, the lead systems integrator proposed using existing launch vehicles for short- and medium-range missions, and developing a new long-range launch vehicle family while maximizing the reuse of existing target system designs. However, MDA committed to a riskier strategy that required developing a single common architecture across three target system families. Early in the development of FTF, MDA underestimated the technical and design challenges involved in the development of a new target family. By May of 2006, MDA recognized that the funding set aside for FTF development was no longer adequate and made a decision to focus developmental efforts on the ground-launched 72-inch long-range target needed for near-term missions and defer other key capabilities. Despite the focus on the 72-inch ground-launched target, development problems have delayed the first launch from July 2008 to April 2009. The 72-inch target qualification process has been more difficult and costly than expected and development is still not complete. Of the 69 components in development, 24 had not passed the qualification process as of August 2008. In addition, according to the targets program director, the qualification process completion date has moved from early October to November 2008. Delays have also occurred because development of the 72-inch target began without a stable design. Eleven of the remaining unqualified components are considered critical technologies and are currently being redesigned. MDA acknowledged both underestimating the difficulty in using components never qualified for this shock and vibration environment and underestimating the complexity of combining multiple launch vehicle and mission requirements. Similarly, the 52-inch target’s first flight will occur no earlier than fiscal year 2012, representing a 3 year slip from the original expectations of first flight in fiscal year 2009. This slip has resulted in the need to initiate an interim target development effort to satisfy near-term target requirements of the Aegis flight test program. MDA recently began implementing a series of management initiatives to improve the targets program, such as establishing technical, cost, and schedule baselines for all missions in the 2-year integrated master test plan. The new program director and staff have also begun drafting long- term target capability requirements, and developing and implementing a new cost model for targets. In addition, plans have been made to improve risk management by (1) considering program-wide issues, and (2) including programmatic risks, cost and schedule in the risk assessments. Overall, the success of the flight test program depends on the success of FTF. However, the current FTF acquisition strategy will yield fewer targets at higher costs and the entire FTF might not be fully pursued. For example, key capabilities such as the air-launched targets have been deferred and MDA is considering whether to continue with the 52-inch target development. Further, the initial plan for 10 to 12 72-inch target flights per year has been reduced to 4 to 5 flights per year. In addition, the development and production of the first four 72-inch targets cost has grown 63 percent, from $248 million to at least $405 million, and the cost to complete development of the entire FTF is not known. Given that the GMD and THAAD flight test programs have already been negatively affected by target difficulties, further changes to the flight test program could become necessary if the FTF program does not deliver its capabilities as scheduled. Targets are sophisticated and expensive missiles that are essential to flight testing the ballistic missile defense system. As such, management of the targets program must be both effective in supplying enough reliable targets and efficient as to their cost. Increasing problems with target failures and anomalies have caused degradations in the missile defense flight test program. The strategy of supplanting the use of existing targets, which contain aging components, with a new family of targets, has not gone as planned. The new FTF has been delayed, costs have increased, and the program’s scope is being reduced. The difficulties with executing FTF are due, at least in part, to the fact that a sound business case was not established before proceeding with the program and the attendant contracting strategy. MDA has recently initiated a series of actions to improve the targets program in general, and the FTF program specifically. Even so, the availability of targets for flight tests continues to be problematic, and as a result the scope of the flight test program has been reduced to better match available targets. The availability of targets is also dependent on the ultimate success of the FTF program. We recommend that the Secretary of Defense take the following two actions: Require the Director, MDA to establish a revised business case for providing reliable and timely targets for a robust flight test program. Such a business case should, among other things, determine the best approach for providing an ample supply of quality targets to support the missile defense flight test program, including consideration of approaches other than the FTF. demonstrate that the chosen approach—including the attendant acquisition and contracting strategy–can be executed with available technologies, funding, time, and management capacity. establish separate cost, schedule, and performance baselines for each class of target under development against which progress can be measured. reflect input from key organizations, such as the Missile Defense Executive Board, independent cost estimators, and the Director, Operational Test and Evaluation. Align MDA’s plans and resources with the targets program approach resulting from the above business case. The Congress has repeatedly stressed the need for robust testing of the Ballistic Missile Defense System and has become concerned with the health of the MDA targets program. The Congress has also expressed concern that the FTF program is proceeding at a slower pace and greater cost than expected. Therefore, Congress may wish to consider whether to require the Secretary of Defense to report the departmentally approved missile defense target development and procurement strategy, business case, and baselines to the Congressional Defense Committees. DOD provided written comments on a draft of this report. These comments are reprinted in appendix II. DOD also provided technical comments, which we incorporated as appropriate. In our first recommendation, we outlined four components that the Secretary of Defense should require in a revised business case for the missile defense targets program. DOD partially concurred with this recommendation. In responding to the first component of the business case, which is to determine the best approach for providing an ample supply of quality targets, DOD stated that MDA is reviewing the acquisition strategy for targets. It is important that the review be a fresh look at the business case for targets, which is why we recommended that approaches other than the FTF be considered as well. DOD’s response failed to make it clear whether such approaches will be included in the MDA review. In responding to the second component of the business case, to demonstrate that the chosen approach could be executed with available technologies, funding, time, and management capacity, DOD explained that the review underway will result in an approach that will be technologically achievable, executable within budget, and will support the flight test schedule. DOD noted that the delivery of the first four 72-inch FTF targets will inform the review in the area of cost. While we agree that such cost information will be important, we also note that the 72-inch targets have already experienced cost overruns, qualification problems, and delivery delays. As such, they remain unproven, which underscores the need for the business case review to remain open to other approaches to supplying targets. DOD also partially concurred with the third component of the business case, to establish cost, schedule, and performance baselines for each class of target. DOD indicated that baselines are established for each individual target procured and that a baseline for each class of target would be redundant. While individual baselines are appropriate for legacy systems in production, the key issue is establishing baselines for new classes of targets under development, like the FTF. In the course of our review, MDA could not provide a full cost estimate or a baseline for the FTF developmental effort. We have modified the recommendation to deal specifically with establishing baselines for new classes of targets under development, like the FTF. DOD partially concurred with the last component of the first recommendation, which recommends that the business case reflect input from key organizations, such as the MDEB, independent cost estimators, and DOT&E. DOD acknowledged that the MDEB and DOT&E inputs have been extremely valuable but that the decision regarding the targets business case should be reserved for the Director of MDA. Our recommendation calls for the input of key organizations but does not suggest that their approval is required. Thus, it does not pose a challenge to the MDA Director’s decision making authority. MDA has not always obtained such input on key decisions. For example, DOT&E was not consulted for the recent decision to cancel an important GMD flight test. Also, estimates of MDA costs have typically not been obtained from independent cost estimators. Regarding our second recommendation to align MDA’s plans and resources with the targets program, DOD partially concurred, stating that revisions to the acquisition strategy will be incorporated into the Program Budget development. Finally, DOD did not concur with a third recommendation, subsequently removed from the final report, to report the departmentally approved missile defense target development and procurement strategy, business case, and baselines to the Congressional Defense Committees. DOD preferred to rely on briefings to Congressional staff to convey this information. MDA began developing the new target family without a complete acquisition strategy, baselines, or a business case and the program has already experienced qualification problems, schedule delays and cost overruns. Because of the importance of an ample supply of reliable targets to the success of the Ballistic Missile Defense System, we continue to believe that the Secretary of Defense needs to ensure that MDA takes a fresh look at alternatives when it establishes a revised acquisition strategy and business case. The departmentally approved approach should then be reported to the Congressional Defense Committees because MDA’s targets program and strategy are of interest to the Congress, as evidenced by the language that requested our assessment. Accordingly, we have elevated the recommendation for executive action to a matter for congressional consideration. We will send copies of this report to the Secretary of Defense, the Director, MDA, appropriate congressional committees, and other interested parties. This report is also available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staff have any questions concerning this report and its appendixes, please contact me at (202) 512- 4841 or FrancisP@gao.gov. Contact points for our offices of Congressional Relations and Public Affairs may be found on the last page of this report. The major contributors are listed in appendix III. GAO undertook this work in response to a mandate contained in the Conference Report accompanying the National Defense Authorization Act for Fiscal Year 2008 (H.R. Rep. No. 110-477). We focused our review on the following three objectives: (1) Is the Missile Defense Agency (MDA) meeting the need for reliable and timely targets for the flight test program? (2) What are the causes of any deficiencies or delays? (3) What are the prospects for resolution of any problems identified? We will follow up on this briefing by September 30, 2008, with a report. Technical comments from the agency were incorporated as appropriate. Our scope & methodology is described on slide 29. The Ballistic Missile Defense System (BMDS) and its elements verify design performance and assess capabilities in a variety of ways, including flight testing BMDS missiles to track or intercept live targets that represent real-world missile threats. MDA’s Targets and Countermeasures program oversees the design, development, fabrication, and production of ballistic missile targets and countermeasures to be used in the BMDS test program. Target systems are ballistic missiles and are available in numerous configurations designed to test the performance of the system’s sensors and weapons. The target systems include launch vehicles, payloads, instrumentation, and flight controls. In December 2003, MDA awarded Lockheed Martin a contract to be the Lead Systems Integrator (LSI) for most of the existing targets and FTF target development. Lockheed Martin now subcontracts with vendors who previously supplied targets directly to MDA. In-flight target failures and anomalies All targets used to date were predecessors to FTF. FTF has not delivered a new development target yet. Existing targets include off-the-shelf components with some modifications, new productions of existing designs, and modifications. Effect of target problems on elements is greater than in-flight anomalies and failures. THAAD and GMD affected most by target problems. First FTF launch was scheduled for July 2008, now April 2009 Deferrals of subsequent developmental work LV-2 air launch, all sea launch, LV-3, Common-liquid Rocket Boosters, and the 52” configured FTF have been deferred. 52” configured FTF was assumed to be available for first use in FY 2009. increasing complexity of targets (including existing targets) late changes to target requirements on contract schedule delays shift to Lead Systems Integrator – higher labor rates, additional oversight of subcontractors Cost increases associated with recent target failures and anomalies are only partially estimated.~$87 M cost increase associated with FTG-03 failure ~$80 M associated with target availability and affordability Unknown cost implications for Radar Data Collection “anomalies” MDA has estimated some cost for FTF, but total nonrecurring development remains unclear because some cost figures include the purchase of test assets. According to information MDA provided July 15, 2008, nonrecurring cost plus the cost to purchase some test assets are estimated at about $1.2 billion. Sunk costs are about $553 million. Supplier base problems with existing targets: Aging inventory/diminishing sources Motors used in some targets are over 40 years old. Vendors are leaving the market due to lack of business. Target contracts have not captured all element testing requirements. Target technical, cost, and schedule baselines were not always established before targets were placed on contract. MDA acknowledges they have recently established these baselines for all missions in the next 2 years. Flat charges assessed to each element do not represent actual use. Flight test costs. According to MDA, contracting with an LSI increased costs; the magnitude of the increase exceeded expectations. For example, composite hourly labor rates were 32% higher in 2004 than prior to the contract with the LSI. Effective cost of subcontractor labor is higher on prime contract due to management oversight of subcontractors. Heritage components put in unverified environment, failures due to high vibration and shock levels.11 of 69 components are considered critical. As of April 2008, all 11 were in redesign or failure analysis. 5 of these critical components were not projected to complete qualification testing until September at the earliest. As of April 2008, 35 components had not completed qualification testing. By June 2008, 28 components remained unqualified. MDA acknowledged (1) underestimating the difficulty in using components never qualified for this shock and vibration environment and (2) underestimating the complexity of combining multiple launch vehicle and mission requirements. identify requirements, design attributes, and candidate configurations for a family of targets by reviewing existing systems. In a July 2006 assessment, MDA stated that other alternatives for improving legacy targets were not considered at that time. According to a July 2006 MDA briefing, the decision to adopt the FTF and a LSI was not based on a formal cost or business case analysis. FTF original cost estimates only included the integration costs and did not address: nonrecurring engineering costs, unit costs for FTF configured targets, or the costs to purchase existing target systems through the prime contract. A formal analysis of the relative merits and costs of each alternative could not be performed when the commitments to the LSI and the FTF were made because cost estimates were not available. We could not identify an original approved FTF acquisition strategy. MDA did, however, provide a May 2006 FTF briefing that stated: At that time, the majority of the investment required for the FTF LV (LV-2 and LV-3) family design effort had been expended. LV-2 development cost estimated at $248 million (FY06 dollars). Completion of entire FTF LV family is required to realize substantial yearly recurring cost reductions. Assumed 10 to 12 LV missions per year. Completion of the entire FTF family is required to support the increased pace and complexity of BMDS testing 2-4 years in the future. Design and cost implications of the FTF requirements were underestimated. In May 2006, MDA acknowledged that significant technology and design challenges were inherent in FTF. Initial FTF requirements “levied functional capabilities and operational environments far above existing targets’ current qualifications.” Choice of nonproprietary, single common architecture required new “clean-sheet-of-paper” designs. Only capable of marginal reuse of existing designs. Funding set aside was no longer adequate to resource the development. Decision made to focus on ground-launched LV-2 for near- term missions, and delay other key capabilities. Full FTF LV family may not be fully pursued. The LV-3 and the LV-2 air-launched capabilities are deferred; LV-2 air launch considered a key capability which provides greater test flexibility. Assumption of 10–12 LV-2 flights per year, including 4 GMD tests, is no longer valid, expect only 4–5 per year. The development and production costs of the first 4 LV-2s have grown from $248 M to $405 M (a 63 percent increase). MDA is considering whether to continue with 52” development First 52” flight will occur no earlier than FY12 (delayed from FY09). The cost to complete development of the FTF is not known. MDA recently began implementing a series of management initiatives for BMDS target acquisition. Establishing technical, cost, schedule baselines for all missions in the 2-year integrated master test plan. Implementing a requirements stabilization process requiring higher-level approval for target changes. Drafting long-term target capability requirements Drafting initial comprehensive plan to validate system-level models and simulations. Improving risk management by (1) considering program-wide issues, (2) including programmatic risks, cost & schedule. Developing and implementing a new cost model for targets. New program director and personnel. BMDS flight test program has changed to reflect target availability and deficiencies. THAAD flight test program extended. GMD flight test objectives delayed. Success of flight test program assumes success of FTF. If everything goes as planned, the targets program may be able to meet the future needs of the BMDS test plan. MDA’s Integrated Master Test Plan is modified as often as every quarter. Key deferred work not in funding plan–including air- launched LV-2 capability. without demonstrating a sound business case. Specifically, MDA did not show that the FTF could be developed and produced within existing resources. Assumptions underlying the FTF development have changed, including whether the concept of a “family” of targets can be accomplished. MDA has taken actions to improve target acquisition deficiencies, but these actions may not be enough to provide an adequate supply of targets. To assess the ability of MDA’s targets program to provide support for the BMDS test efforts, we analyzed test plans and schedules, flight test reports, budget documents, and program execution reviews. We also analyzed program directives and acquisition policies and procedures. We interviewed MDA officials from GMD, THAAD, Aegis BMD, and the Targets and Countermeasures program offices as well as officials from MDA’s Business, Test, and Engineering directorates. We also interviewed officials from DOD’s Office of the Director, Test and Evaluation, and the Space and Missile Defense Command. Our analysis covered data ranging from December 2003 through July 2008. We conducted this performance audit from February 2008 through July 2008, in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Major contributors to this report were David B. Best, Ivy G. Hübler, Steven B. Stern, Robert S. Swierczek, Letisha T. Watson, and Alyssa B. Weir. | The Missile Defense Agency (MDA) is likely to spend $460 million annually on missiles used as targets for flight tests. Executing these tests depends on the quality and availability of targets. Congress asked GAO to assess (1) if MDA is providing reliable targets; (2) the causes of any deficiencies; and (3) if resolutions exist for any problems identified. To do this, GAO analyzed acquisition policies and procedures; flight test data; and budget, program execution, and acquisition materials; and interviewed MDA and DOD officials. MDA has difficulty in both supplying targets for missile defense testing as well as in developing a new family of short, medium and long-range targets. The number of target failures and anomalies (failing to achieve one or more non-critical mission objective or partially achieving a critical mission objective) during flight tests has increased since 2006, contributing to delays in flight tests and modification of flight test objectives. In addition, the average unit cost of targets has grown significantly, from $4.5 million to $8.5 million from 2002 to 2006 to current estimates of $32 million to $65 million for the targets planned from 2008 to 2010. Many factors contribute to this cost growth, including increased complexity of targets to better reflect an evolving threat and late changes to target requirements on contract. MDA's difficulty in supplying existing targets is driven by diminishing sources for components, unanticipated costs, and problems incorporating requirements into contracts and establishing program baselines. MDA has also encountered problems developing the new family of targets, an effort currently estimated to cost at least $1 billion. The problems are due, at least in part, to the fact that a sound business case was not established before proceeding with the program and the attendant contracting strategy. The decision to pursue this new family of targets was made without a formal cost analysis and it is unclear whether MDA evaluated all alternatives before making this commitment. GAO also could not identify an original approved acquisition strategy for the new family of targets. Consequently, developmental problems have arisen in the new family of targets, leading to cost growth, delayed flight tests, and deferral of several key capabilities. MDA is taking a series of actions to address these problems, such as: (1) establishing technical, cost, and schedule baselines for all missions in the 2-year integrated master test plan; (2) drafting long-term target capability requirements; (3) developing a new cost model for targets; (4) making plans to improve risk management by considering program-wide issues, and including programmatic risks, cost and schedule in risk assessments. However, the prospects for resolution of the target acquisition problems are unclear and the overall success of the Ballistic Missile Defense System flight test program depends on the success of a new family of targets under development. |
The federal government supports multiple freshwater programs in the United States and abroad. Although agencies vary in how they describe their freshwater programs (activities, projects, or initiatives), the following terms generally describe the freshwater efforts covered in this review: Desalination—Activities and/or infrastructure related to the process of removing salts from saline water to provide freshwater. Drinking water supply—Activities and/or infrastructure designed to improve access to and quality and availability of clean drinking water. Flood control—Activities related to dredging, hydrologic forecasting, and the construction, operation, and maintenance of infrastructure designed to reduce flood damage. Irrigation—Activities and/or infrastructure related to the diversion, distribution, delivery, and drainage of water for agricultural purposes. Navigation—Activities and/or infrastructure related to dredging and the construction, operation, and maintenance of infrastructure (e.g., locks, channels, and dams), primarily for river-based transportation. Wastewater treatment—Activities and/or infrastructure designed to manage and treat storm water and domestic and/or industrial wastewater. Water dispute management—Adjudication, litigation, and negotiation activities to prevent or resolve water-related disputes, including water settlement payments. Water conservation—Activities and/or infrastructure designed to reclaim, recycle, and/or reuse potable or nonpotable water. Watershed protection, restoration, and management (watershed management)—Activities and/or infrastructure related to nonpoint source pollution prevention, wetlands restoration, and land-based activities related to source water protection and coastal zone management. Federal agencies provide financial support for freshwater programs through direct federal spending and grant, loan, and loan guarantee programs. For the purposes of this report, direct federal spending is a general term used to describe, among other things, work performed by federal employees or through contracts with private and nongovernmental entities on the federal government’s behalf. For example, direct federal spending includes the financial support spent by federal agencies for, among other things, the construction of reservoirs for flood control and irrigation. Grants, loans, and loan guarantees are different types of financial assistance programs provided by federal agencies to help communities with projects, such as construction of water and wastewater treatment facilities. In addition to providing financial support on the direct costs of freshwater programs, such as capital construction and operations and maintenance costs, agencies also provide funds for technical assistance and research and development efforts. Agencies rely on several sources of funding—including annual appropriations from the general fund and from dedicated funding sources, such as trust funds—to provide financial support for these programs. Agencies obligate these funds for multiple purposes as they implement their programs. Obligations represent amounts for orders placed, contracts awarded, services received, and similar transactions during a given period that will require payments during the same or a future period. Obligations differ from expenditures in that an expenditure is the issuance of a check, disbursement of cash, or electronic transfer of funds made to liquidate an obligation. In addition, in some circumstances expenditures fulfilling an obligation may occur during subsequent years. Consequently, obligations provide the best estimate of what an agency plans to spend during a fiscal year. The extent to which agencies track their freshwater efforts vary. Agencies tend to track the financial support they provide to large freshwater programs, such as EPA’s Drinking Water State Revolving Fund. Some programs, such as Agriculture’s Conservation Reserve Program, serve multiple purposes and specific program components, such as irrigation activities, are not specifically tracked. Similarly, freshwater projects (e.g., a dam) can serve multiple purposes, providing benefits such as water storage, flood control, and generation of hydroelectric power. The financial support provided for the overall project cost is generally available, but the costs for specific components of the project are not readily available. In addition, definitions of freshwater programs can vary depending on the agency’s role. An agency that is responsible for maintaining portions of the nation’s waterways may include lock infrastructure and waterway operations and maintenance activities in its navigation program; however, other agencies, which are responsible for research and development of freshwater resources, may include stream flow and sedimentation research within their navigation programs. Because of these differences, agencies may not be tracking their freshwater expenditures consistently across agencies. As such, financial information reported by the agencies on the federal government’s financial support of freshwater programs in the United States and abroad is an estimate of the minimum amount of funds provided for these efforts. Of the 27 agencies that provided about $49 billion in federal financial support specifically for freshwater programs in the United States during fiscal years 2000 through 2004, 3 agencies accounted for over 70 percent of the total. EPA provided about 31 percent of the total support, the Corps accounted for about 26 percent of the total, and Agriculture’s Rural Utilities Service accounted for about 16 percent of the total. See table 1 for more information on the federal financial support provided for domestic freshwater programs. Each of the agency’s financial support of freshwater programs generally reflects the nature of its mission. EPA—as the agency responsible for protecting the nation’s waters through enforcing clean water and safe drinking water laws, providing support for municipal wastewater treatment plants, and protecting watersheds and sources of drinking water—provides substantial financial support for freshwater programs. On the other hand, the Department of Health and Human Services’ (Health) Administration of Children and Families—an agency responsible for federal programs that seek to promote economic and social well-being—administers a grant program dedicated to helping rural, low-income communities with their water and wastewater systems and provides a relatively small amount of financial support. In addition, agencies, such as Agriculture’s Rural Utilities Service, the Department of Commerce’s (Commerce) Economic Development Administration, Health’s Indian Health Service, the Department of Housing and Urban Development’s (Housing) Office of Community Planning and Development, and EPA, that assist communities with the development of drinking water supply and wastewater treatment facilities provide far more financial support than agencies, such as Agriculture’s Cooperative State Research, Education, and Extension Service and Defense’s Army Material Command and Office of Naval Research, that primarily support research and development efforts. Agencies generally receive annual appropriations from the general fund to support their domestic freshwater programs. In addition to annual appropriations, 4 agencies—Agriculture’s Rural Utilities Service, the Corps, Defense’s Office of Naval Research, and the Appalachian Regional Commission—received funds from supplemental appropriations. The Department of Transportation’s (Transportation) Federal Highway Administration and Saint Lawrence Seaway Development Corporation received the majority of their annual budgets from dedicated funding sources available subject to appropriations (the Highway Trust Fund and the Harbor Maintenance Trust Fund, respectively). In addition, 4 other agencies—Agriculture’s Farm Service Agency and Natural Resources Conservation Service, the Corps, and Interior’s Reclamation—supplemented their annual budgets with dedicated funding sources available subject to appropriation. The Agriculture agencies received funds from the Commodity Credit Corporation, while the Corps and Interior’s Reclamation received funding from trust funds. Finally, certain agencies, including Interior’s Reclamation and Fish and Wildlife Service and Agriculture’s Rural Utilities Service, received funds from dedicated funding sources available without further appropriation. For example, Interior’s Reclamation received funds from the Lower Colorado River Basin Development Fund to finance the operation and maintenance of freshwater infrastructure in the Colorado River Basin for, among other things, drinking water supply, flood control, and irrigation efforts. Domestic drinking water supply and wastewater treatment programs were supported by 18 and 16 agencies, respectively. Agriculture’s Rural Utilities Service, Commerce’s Economic Development Administration, Health’s Indian Health Service, Housing’s Office of Community Planning and Development, and EPA primarily supported activities related to the development of drinking water supply and wastewater treatment infrastructure, and some of the other agencies mostly provided technical assistance and/or research and development assistance, such as water quality and water availability research. Sixteen agencies supported a variety of watershed management programs. While 7 agencies provided financial support for navigation programs, the Corps and Transportation’s Saint Lawrence Seaway Development Corporation provided the majority of the support for, among other things, operations and maintenance of the nation’s waterways. See table 2 for more information on the domestic freshwater programs these agencies supported. Agencies used several different funding mechanisms to provide financial support for domestic freshwater programs. Twenty of the 27 agencies reported that they used direct federal spending to provide financial support for freshwater programs. For the purposes of our review, we define direct federal spending to include (1) work carried out by federal employees, contractors, and private and nongovernmental organization sectors for the federal government; (2) the federal government’s portion of federal cost-share programs; and (3) funds provided from one agency to another agency to conduct work. Thirteen agencies used grant programs, and 2 agencies each used loan and loan guarantee programs. Many agencies used a combination of funding mechanisms to provide financial support for freshwater programs. See table 3 for more information on funding mechanisms used by agencies to provide financial support for freshwater programs in the United States. Grant programs and direct federal spending provided over $22 billion and about $22 billion, respectively, for domestic freshwater programs. Loan programs provided over $4 billion, and loan guarantee programs provided over $90 million in initial obligations. Initial obligations are the amount agencies obligated for the subsidy cost when a loan guarantee was made. They do not include subsequent reestimates. See figure 1 for the financial support provided by each funding mechanism. Direct federal spending ($21.8 billion) Grant programs ($22.5 billion) The $49 billion for domestic freshwater programs includes funding provided from one agency to another to conduct freshwater activities in the United States. For example, the Corps received financial support from EPA to conduct watershed management activities. We included funding for this effort in the financial information reported by EPA. See table 4 for examples of agencies that performed work on freshwater activities in the United States using financial support provided to them by other agencies. We also identified domestic programs that may provide financial support for freshwater activities, but are not included in the $49 billion because supporting freshwater activities is not the programs’ primary purpose and activity-level data is not readily available. For example, Housing’s Office of Community Planning and Development administers a loan guarantee program that may provide financial support for water infrastructure projects, but aggregate information on the use of loan guarantee authority for particular categories of activities is not readily available. Consequently, financial support provided by these types of programs is not included in table 1. Other agencies also have these types of programs that may support freshwater-related activities: Agriculture’s Farm Service Agency, Forest Service, and Rural Utilities Service; Commerce’s Economic Development Administration; Interior’s Fish and Wildlife Service; the Appalachian Regional Commission; and EPA. Furthermore, while these 27 agencies provided the majority of the federal financial support for freshwater programs in the United States, other agencies may also provide financial support for these types of programs. Appendix II provides information on some other agencies and programs that can provide financial support for freshwater-related activities. These agencies provide financial assistance primarily for specific regions of the United States and/or support a variety of programs, including freshwater-related activities. In the domestic support for freshwater programs, we also identified the United States’ financial contributions to three binational commissions—Border Environment Cooperation Commission, International Boundary and Water Commission, and International Joint Commission. These commissions support a variety of projects on both sides of the U.S. borders with Canada and Mexico. Typically, these commissions coordinate their efforts with EPA and/or State and are able to track how U.S. contributions are used to support their freshwater activities. Together, the three commissions used about $175 million in U.S. contributions to support a number of freshwater projects during fiscal years 2000 through 2004. For the purposes of our review, we include information on U.S. contributions to these commissions in the same section as the information for domestic freshwater programs because these projects are joint efforts among the United States, Canada, and Mexico along the shared borders; however, information on funding spent solely in the United States by the commissions is not readily available. Table 5 presents information on total U.S. contributions to these commissions for freshwater projects. See appendix II for more information on these commissions. Of the about $3 billion of U.S. financial support provided internationally for freshwater programs during fiscal years 2000 through 2004, an estimated $2 billion was spent throughout most of the world, and more recently another $1 billion supported freshwater projects in Afghanistan and Iraq. In addition to the financial assistance provided directly by federal agencies for freshwater programs abroad, the United States also indirectly supports these programs through its contributions to numerous international organizations, such as the World Bank and the United Nations. Eight federal agencies obligated an estimated $2 billion during fiscal years 2000 through 2004 for freshwater activities abroad, excluding Afghanistan and Iraq. USAID accounted for over 90 percent of the $2 billion. Interior’s Fish and Wildlife Service provided about 6 percent; Agriculture’s Foreign Agricultural Service and the U.S. Trade and Development Agency (an independent federal agency) each accounted for around 1 percent. The remaining 4 agencies (State, the Corps, the African Development Foundation, and the National Science Foundation) together provided about 1 percent of the support. See table 6 for information on the federal financial support provided for freshwater programs abroad, excluding aid provided to Afghanistan and Iraq. For international support on freshwater programs abroad, agencies generally receive annual appropriations from the general fund to support their freshwater programs. Some freshwater programs, such as those at Interior’s Fish and Wildlife Service, are also supported by permanent or dedicated funding sources, which remain available without further appropriation. Agriculture’s Foreign Agricultural Service receives funds for its freshwater programs from the Commodity Credit Corporation (a dedicated funding source at Agriculture subject to congressional appropriation). Seven of the 8 agencies reported that they provided financial support for wastewater treatment and watershed management programs abroad. In addition, 6 agencies each provided financial support for drinking water supply and irrigation programs. See table 7 for information on the freshwater programs supported by federal agencies abroad. About 99 percent of the estimated $2 billion in federal financial support for freshwater programs abroad was delivered through grant programs administered by 7 agencies. The Corps was the only agency that did not have a grant program to support freshwater programs abroad. Agriculture’s Foreign Agricultural Service, the Corps, State, and USAID provided a relatively small amount of financial support (about $15 million total) through direct federal spending. Additionally, USAID obligated $4 million in loan guarantees for water supply and wastewater treatment projects through its Development Credit Authority program to cover up to 50 percent of the risk in lending. Some of the $2 billion for freshwater programs abroad includes funding provided from one agency to another to conduct freshwater projects in foreign countries. For example, Interior’s Reclamation received funds from State to conduct desalination activities. We included funding for this effort in the financial information reported by State. See table 8 for examples of agencies that performed work on freshwater activities abroad using financial support provided to them by other agencies. Appendix III provides information on these and other agencies and programs that can also provide financial support for freshwater activities abroad. For Afghanistan and Iraq, USAID and Defense provided about $1 billion during fiscal years 2002 through 2004 to support the reconstruction and rehabilitation of freshwater infrastructure in these countries. For the purposes of our review, we report financial support for freshwater projects in Afghanistan and Iraq separately from the $2 billion total in foreign aid because these funds were primarily made available during fiscal year 2004. USAID provided financial support through grants and contracts and by transferring funds to the Corps to carry out work on water supply, wastewater treatment, irrigation, and watershed projects in Afghanistan and Iraq. USAID received funds from the Emergency Supplemental Appropriation for Defense and for the Reconstruction of Iraq and Afghanistan for Fiscal Year 2004 to support, among other activities, freshwater projects in those two countries. Within the Office of the Secretary of Defense, (1) the Defense Security Cooperation Agency’s Humanitarian Assistance Program supports, among other things, water infrastructure projects overseas and (2) the Army’s Commander’s Emergency Response Program is designed to respond to urgent humanitarian relief and reconstruction requirements at the local level in Afghanistan and Iraq. Also within Defense, the Project and Contracting Office awarded and continues to manage various contracts to support, among other things, the construction of rural water systems and the rehabilitation of drinking water supply facilities in major cities in Iraq. Table 9 presents information on financial support provided by USAID and Defense for freshwater projects in Afghanistan and Iraq. In addition to providing financial support directly through federal agencies for freshwater programs abroad, the United States also indirectly supports these programs through its contributions to numerous international organizations (e.g., the United Nations and the World Bank). In most cases, the United States makes contributions to the general budgets of these organizations and not to a specific project or program. In addition, these organizations usually combine U.S. contributions with other sources of funds, including contributions from other countries, to fund their freshwater programs. As a result, it is difficult to determine what portion of U.S. contributions to international organizations is used to support freshwater programs. The United States contributed to the general budgets of a number of multilateral development banks and financial institutions that support freshwater projects around the world, and some portion of these contributions was used to support freshwater efforts worldwide. Using funds contributed by the United States and other countries or borrowed from world capital markets, multilateral development banks finance economic and social development programs around the world. Together, these autonomous institutions are the largest single source of developmental assistance for developing countries. The United States is a member of, and has made financial contributions to, five multilateral development banks that support freshwater projects around the world. These multilateral development banks include the African Development Bank Group, the Asian Development Bank, the European Bank for Reconstruction and Development, the Inter-American Development Bank, and the World Bank Group. The United States also contributed to other international financial institutions—including the Global Environment Facility, the International Fund for Agricultural Development, and the North American Development Bank—that also support freshwater projects abroad. See table 10 for figures on total contributions the United States made to selected multilateral development banks and financial institutions. Some portion of these contributions supported freshwater projects abroad. The United States also contributed to the general budgets of a number of international organizations that support freshwater projects around the world, and some portion of these contributions was used to support freshwater efforts. Of these organizations, the United States contributed the most financial support to the United Nations. In addition to assisting with peacekeeping efforts, the United Nations provides funds for humanitarian, environmental, and development programs that support, among other things, water resources management efforts around the world. Table 11 presents figures on contributions the United States made to selected international organizations. Some portion of these contributions supported freshwater projects abroad. We provided the Departments of Agriculture, Commerce, Defense, Energy, Health and Human Services, Housing and Urban Development, the Interior, Transportation, and State and independent agencies, including the African Development Foundation, the Appalachian Regional Commission, EPA, the National Science Foundation, the Small Business Administration, USAID, and the U.S. Trade and Development Agency, with a draft of this report for review and comment. Three of these agencies—Interior, Health and Human Services, and USAID—provided us with written comments that are included in appendixes IV through VI. The 3 agencies agreed with the report and provided us with technical comments, which we have included as appropriate. The other 13 agencies provided us with technical comments orally or did not provide us with any comments. We have made changes in response to the technical comments throughout the report, as appropriate. We will send copies of this report to interested congressional committees; the Secretaries of Agriculture, Commerce, Defense, Energy, Health and Human Services, Housing and Urban Development, the Interior, Transportation, and State; the Administrators of EPA, Small Business Administration, and USAID; the Directors of the National Science Foundation, Office of Management and Budget, and U.S. Trade and Development Agency; the President of the African Development Foundation; the Federal Co-Chair of the Appalachian Regional Commission; and other interested parties. We will also make copies available to others upon request. In addition, the report will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff have any questions, please call me or Edward Zadjura at (202) 512-3841. Key contributors to this report are listed in appendix VII. The objectives of our review were to determine for fiscal years 2000 through 2004 how much financial support federal agencies provided for freshwater programs in the United States and abroad. To identify the agencies that support freshwater programs in the United States and abroad, we reviewed the Catalog of Federal Domestic Assistance and reports published by GAO, the Congressional Research Service, the United Nations, and the National Research Council of the National Academies. We also interviewed water and natural resources experts at the Congressional Research Service. For the purpose of our review, we examined freshwater programs that support desalination; drinking water supply; flood control; irrigation; navigation (primarily for river-based transportation); wastewater treatment; water conservation; water dispute management; and watershed protection, restoration, and management activities. We identified numerous agencies that support at least one of these freshwater programs. After conducting additional background research and interviews with officials at these agencies, we narrowed our focus to 32 agencies. These agencies received congressional appropriations during each of the fiscal years from 2000 through 2004 and, to some extent, tracked the amount of financial support they provided for freshwater programs. Together, these agencies accounted for the majority of the federal financial support for freshwater programs in the United States and abroad during fiscal years 2000 through 2004. These agencies include the following: Department of Agriculture (Agriculture): Agricultural Research Service; Cooperative State Research, Education, and Extension Service; Economic Research Service; Farm Service Agency; Foreign Agricultural Service; Forest Service; Natural Resources Conservation Service; Rural Utilities Service; Department of Commerce (Commerce): Economic Development Administration, National Oceanic and Atmospheric Administration; Department of Defense (Defense): Army Corps of Engineers (Corps), Army Material Command, Office of Naval Research; Department of Energy (Energy); Department of Health and Human Services (Health): Administration for Children and Families, Indian Health Service; Department of Housing and Urban Development (Housing): Office of Community Planning and Development; Department of the Interior (Interior): Bureau of Indian Affairs, Bureau of Land Management, Bureau of Reclamation (Reclamation), Fish and Wildlife Service, National Park Service, U.S. Geological Survey; Department of State (State); Department of Transportation (Transportation): Federal Highway Administration, Saint Lawrence Seaway Development Corporation; and Independent agencies: African Development Foundation, Appalachian Regional Commission, Environmental Protection Agency (EPA), National Science Foundation, U.S. Agency for International Development (USAID), U.S. Trade and Development Agency. To determine the amount of financial support these agencies provided for freshwater programs, we used a questionnaire to gather information on agencies’ freshwater programs, including funds provided on an annual basis to support these programs. Before distributing the questionnaire, we had staff from the Resources, Science, and Industry and American Law Divisions of the Congressional Research Service review the draft questionnaire, and we included their comments, accordingly. In addition, we pretested the questionnaire with an official from the Corps and incorporated her comments, as appropriate. We sent the questionnaires, along with a cover letter, to respondents and requested that agencies return the completed questionnaire within 4 weeks. All agencies returned their questionnaires. We conducted follow-up interviews with respondents to confirm the information and to clarify the information, if necessary. In our questionnaire, we requested financial information on the direct costs of agencies’ freshwater programs, such as capital construction and operations and maintenance costs of freshwater infrastructure. We also requested information on the amount of financial support provided for technical assistance and research and development efforts related to freshwater programs. We requested information on freshwater programs that agencies typically fund during a year and excluding atypical funding, such as financial support for freshwater projects in response to natural disasters. We obtained and reported on financial support for freshwater projects in Afghanistan and Iraq separately from the total foreign assistance because these funds were provided in recent years. We requested this information in the form of obligations because obligations provide a good estimate of what an agency plans to spend during a fiscal year. Additionally, financial information in the form of obligations are the best measure for comparing the amount of financial support provided from year to year over a period of time. Because we reviewed a fairly recent period of time during which inflation was minimal, we reported the figures in current dollars. For loan guarantee programs, we requested that the agency provide the amount they initially obligated the year the loan was guaranteed to cover potential defaults, rather than annual reestimates of these amounts. In cases where agencies do not track financial information in the form of obligations, we requested that senior agency officials provide the best available proxy for obligations. Some officials said that obligations are comparable to their appropriation or expenditure figures. Other officials stated that their agency collects information on actual disbursements or actual grant approval amounts. Agencies for which we do not report obligations include Agriculture’s Farm Service Agency and Forest Service, Housing’s Office of Community Planning and Development, Transportation’s Saint Lawrence Seaway Development Corporation, and the Appalachian Regional Commission—see notes to table 1 for more information on these agencies’ financial information. We did not independently assess the reliability of the financial information provided by agency officials because obligations are agency expectations for expenditures and there are no associated transactions to track. Although we requested program-level financial information, we opted to present this information at the agency-level because agencies’ definitions of freshwater programs vary. However, we collected the program-level financial information to ensure that agency officials reported financial support for freshwater programs consistently across agencies. As part of this effort, we confirmed with each agency that for each of the programs listed in the questionnaire, the agency only provided financial information on the freshwater portion of the program. In addition to programs that specifically supported freshwater activities, we also requested information on other programs that may provide funding for these activities. We did not include financial support from these programs in the agency totals because supporting freshwater activities is not their primary purpose and activity level data is not readily available. Consequently, the financial information we reported is an estimate of the minimum amount of financial support provided by the agencies. In addition to the agencies that specifically supported freshwater programs, we identified and obtained information on several binational commissions, international organizations, and multilateral development banks to which the United States made financial contributions and which support freshwater programs along U.S. borders or abroad. To identify how U.S. contributions to binational commissions were used to support freshwater programs along U.S. borders, we obtained financial information from officials at State and EPA and representatives from the commissions. To identify the annual amount of the United States’ financial contributions to multilateral development banks and other international organizations, we obtained financial information from officials at State and the Department of the Treasury. When necessary, we corroborated this information with support from other sources, including annual reports to Congress on U.S. contributions to international organizations. We conducted our review from March 2004 through January 2005 in accordance with generally accepted government auditing standards. This appendix discusses federal financial support of freshwater programs in the United States and along U.S. borders in three parts. First, we provide a general overview on the agencies responsible for the majority of the federal financial support for freshwater programs in the United States, along with the total amount of financial support provided for freshwater programs during fiscal years 2000 through 2004. Second, we briefly summarize the information on some other agencies that can also provide financial support to domestic freshwater programs. Third, we provide information on (1) binational commissions that used U.S. financial contributions for freshwater projects along the U.S. borders with Canada and Mexico and (2) total U.S. financial contributions to these commissions for their freshwater programs during fiscal years 2000 through 2004. The following agencies provide the majority of the federal government’s financial support specifically for freshwater programs in the United States. These programs, as described by agency officials, documents, and reports, are discussed below. The mission of the Department of Agriculture is to provide leadership on food, agriculture, and natural resources issues on the basis of sound public policy, the best available science, and efficient management. The department conducts work under a variety of mission areas, including farm services, natural resources and the environment, research and education, and rural development. The agency provides financial support for freshwater programs primarily for the construction of drinking water and wastewater facilities, watershed and wetland management, and freshwater- related research. The Agricultural Research Service is the department’s primary in-house scientific research agency. The agency conducts research to develop and transfer solutions to address agricultural problems to enhance natural resources, such as protecting and sustaining freshwater resources. The agency’s authority to conduct these efforts primarily falls under the Department of Agriculture Organic Act of 1862 and the Agricultural Research Act of 1935. Freshwater programs include technological improvements in irrigation, rural and urban water recycling and reuse, nonpoint source pollution prevention, stream restoration, and flood control structures. The agency also receives financial support from other federal agencies—such as Commerce’s National Oceanic and Atmospheric Administration and the Corps—to support domestic freshwater programs, including watershed management, irrigation, and water conservation. Additionally, the agency receives funds from State to conduct freshwater activities abroad, such as irrigation systems in Pakistan and Mexico. The Cooperative State Research, Education, and Extension Service was created in 1994 through the USDA Reorganization Act. The agency provides financial support—primarily through grants to universities, nonprofit associations, private industry, and other groups—for state and local research, education, and outreach activities. The agency conducts these activities primarily through the Hatch Act of 1887, as amended, the National Agricultural Research, Extension, and Teaching Policy Act of 1977, and section 406 of the Agricultural Research, Extension, and Education Reform Act of 1998. For example, through the Hatch Act, the agency provides block grants for agricultural research on an annual basis primarily to state land grant institutions. These funds are distributed according to a statutory formula. Although the scope of the agricultural research conducted under the Hatch Act is broad, portions of the financial support are directed toward research projects on freshwater resources. The Economic Research Service is Agriculture’s main source of economic information and research. Regarding freshwater resources, the agency primarily provides financial support for research and development programs. For example, the agency is currently conducting research on the impact of agriculture on water quality by examining (1) the influence of economic, environmental, and institutional factors affecting adoption of water conservation management practices and irrigation technologies; (2) the economics of alternative public policy mechanisms to encourage agricultural water conservation and improved water quality; and (3) the availability of water infrastructure and policy mechanisms to facilitate water reallocations and the implications for irrigated agriculture and resource costs. According to a senior agency official, these research activities are generally performed under the Agricultural Marketing Act of 1946. In addition to conducting its own freshwater programs, the agency receives funds from Agriculture’s Foreign Agricultural Service to support freshwater research and development efforts to foreign countries, such as a project to support hydrological modeling of river systems in North China. The Farm Service Agency was formed after a departmental reorganization in 1994 and incorporated programs from several agencies. One of the agency’s primary missions is to help farmers conserve both land and water resources. The agency supports several multipurpose programs—such as the Debt for Nature Program and the Conservation Reserve Program—that may also benefit freshwater resources, but financial information specifically for freshwater efforts is not readily available. For example, through the Conservation Reserve Program, landowners receive annual rental payments and other payments for implementing long-term conservation practices on their land, for among other things, management of wetlands. The Farmable Wetlands Program, which began as a pilot in six states in fiscal year 2001 and is part of the Conservation Reserve Program, provides payments to farmers who voluntarily restore farmable wetlands. The program expanded nationwide in fiscal year 2002 when it was authorized by the Farm Security and Rural Investment Act of 2002. Funding for this program comes from the Commodity Credit Corporation—a government-owned and -operated corporation established in 1933 to stabilize, support, and protect farm income and prices. Among its goals, the Forest Service promotes ecosystem health and conservation in part by improving and protecting watershed conditions to provide the water quality and quantity necessary to support ecological functions. In the United States, forests cover approximately one-third of the land area from which about 66 percent of freshwater originates. Under the Organic Act of 1897, one of the primary reasons for establishing national forests was to maintain and restore watersheds to protect freshwater resources. The agency fulfills these efforts by supporting programs for watershed management activities, such as conducting water quality monitoring and watershed restoration. The agency also secures water rights to protect and use freshwater on Forest Service lands. The mission of the Natural Resources Conservation Service is to provide leadership in a partnership effort to help landowners conserve, maintain, and improve natural resources, including freshwater. As a part of these efforts, the agency supports watershed management, flood control, and water conservation programs. The agency provides financial support for freshwater activities through a variety of programs, including Watershed Protection and Flood Prevention Operations and the Watershed Rehabilitation Program. Most financial support for freshwater activities reported by the agency was through the Wetlands Reserve Program, which was mandated by the Food Security Act of 1985 and reauthorized by the Farm Security and Rural Investment Act of 2002. The program is voluntary and offers landowners financial and technical assistance for the restoration, protection, and enhancement of wetlands. To be eligible for the program, a landowner must have owned the land for at least 12 months and the wetland must be restorable and suitable for wildlife benefits. As with the Farmable Wetlands Program administered by the Farm Service Agency, the source of funds for the program is through the Commodity Credit Corporation. Part of the mission of the Rural Utilities Service is to improve the quality of life in rural communities by administering drinking water supply and wastewater treatment programs. The agency provides the most financial support for freshwater programs of any Agriculture agency. The agency primarily uses grant and loan programs to provide financial support for developing water systems in rural areas and reducing water costs for rural users. The main authority to administer these programs is through the Consolidated Farm and Rural Development Act of 1961. Municipalities, counties, special-purpose districts, Indian tribes, and nonprofit organizations are eligible for the programs. In addition, the agency provides technical assistance and training grants to nonprofit organizations to assist rural communities with drinking water and wastewater issues. Furthermore, the agency also administers a loan guarantee program for drinking water supply and wastewater treatment loans it provides. In addition to executing its own grants during fiscal years 2000 through 2004, the agency, under a memorandum of understanding, administered grants funded by Commerce’s Economic Development Administration and the Appalachian Regional Commission for drinking water supply and wastewater treatment projects. The Department of Commerce’s strategic goals include, among other things, encouraging economic growth that benefits Americans and observing, protecting, and managing the Earth’s resources to promote environmental stewardship. The department’s freshwater programs include water infrastructure and water availability activities. The Economic Development Administration, established by the Public Works and Economic Development Act of 1965, as amended, leads the federal government’s economic development efforts by facilitating growth in America. Through its Public Works and Economic Adjustment Assistance Programs, the agency provides grants to communities and entities in regions experiencing economic decline and distress. These grants are used for, among other things, revitalizing, expanding, and upgrading the physical infrastructure, including water and sewer systems. Under the Partnership Planning Program, the agency also provides grants for planning and technical assistance. The agency can track the majority of its freshwater activities using standard industrial codes, such as water system and water treatment. The agency also supports other programs that can provide funds for freshwater-related activities. In addition to executing its own grants during fiscal years 2000 through 2004, the agency, under a memorandum of agreement, administered grants funded by the Appalachian Regional Commission and Defense’s Office of Economic Adjustment to carry out freshwater infrastructure activities. Among its missions, the National Oceanic and Atmospheric Administration researches and gathers data related to changes in the weather and availability of water. Several offices within the agency provide financial support for freshwater activities. The National Weather Service— authorized under the Organic Act of October 1, 1890, as amended, and the Flood Control Act of 1938, as amended—provided the largest portion of the agency’s budget for freshwater programs. Within its hydrology program, the National Weather Service supports watershed management and flood control activities, such as forecasting water availability activities on rivers, lakes, and streams and inland water research. In addition, the Office of Oceanic and Atmospheric Research provides financial support for drinking water supply, water conservation, watershed management, and navigation in the Great Lakes region. These activities are carried out under various legal authorities, such as the National Climate Program Act (15 U.S.C. §§ 2901-2908) and the Great Lakes Water Quality Agreement of 1978, as amended. Other offices within the National Oceanic and Atmospheric Administration also provide financial support for, among other things, fisheries and aquatic species management and freshwater-related activities in coastal, estuarine, and marine environments. During fiscal years 2000 through 2004, the agency performed work in other countries, but funding for these efforts typically originated with State and USAID. The Department of Defense provides services for military and civilian purposes. The Corps provides financial support for the vast majority of the department’s freshwater programs for civilians through its Civil Works program. In addition, the Army and Navy provide financial support for science and technology research on drinking water supply, water conservation, wastewater treatment, and desalination. Some of these technologies have the potential to be transitioned from a military function to benefit civilians. The Army Corps of Engineers provides engineering services for military and civilian purposes. In addition to designing and managing construction of military facilities, the Corps plans, designs, builds, and operates water resources and other civil works projects through its Civil Works program. The Corps carries out water infrastructure and environmental management and restoration projects under various legal authorities, including numerous river and harbors acts, flood control acts, and Water Resources Development Acts. Activities supported include navigation, flood protection, dam and reservoir projects, and drinking water and wastewater projects. The Corps’ annual appropriations are primarily directed to specific projects. These appropriations are received through Energy and Water Development appropriation acts rather than the Department of Defense appropriations acts. In addition to annual appropriations from the general fund, the Corps receives appropriations from dedicated funding sources, such as the Harbor Maintenance Trust Fund and the Inland Waterway Trust Fund, which receive revenue through receipts of, respectively, a tax on port use and a tax on fuel used by vessels in commercial waterway transportation. The Corps also performs domestic work funded by agencies, such as EPA, Agriculture, and Interior. The Army Material Command conducts research related to, among other things, the logistics of providing water for combat operations on land. The agency performs in-house research, manages contract efforts, and supports other military and government agencies with drinking water treatment, monitoring, storage, and distribution; wastewater treatment; desalination; and water conservation programs. Within the agency, the Army Tank- Automotive Research, Development and Engineering Center researches, develops, engineers, and integrates advanced technology into ground operations. For example, the Water Purification and Recovery Technology program seeks to reduce the logistical burden of providing water for ground troops. According to Army officials, the Army carries out these activities under its general mission to provide and equip combat operations on land. The Office of Naval Research coordinates, executes, and promotes the science and technology programs of the U.S. Navy and Marine Corps. The agency provides financial support through direct federal spending and grant programs for drinking water supply, wastewater treatment, and desalination projects. In addition to private contractors, the agency collaborates with schools, universities, government laboratories, and nonprofit organizations to execute its science and technology programs. More recently in fiscal year 2003, in response to appropriations committee direction, the Office of Naval Research began funding the Expeditionary Unit Water Purification project, which will develop prototype demonstrators to produce drinking water from brackish or saline water. The agency leads this effort, which involves other federal agencies, such as the Army, Interior’s Reclamation, and the Department of Energy. The Office of Naval Research conducts its research efforts under 10 U.S.C. §§ 5022- 5023. The Office of the Secretary of Defense administers programs that can provide financial support for freshwater activities, although that is not the focus of the programs. For example, the Strategic Environmental Research and Development Program and the Environmental Security Technology Certification Program fund research for technologies related to, among other things, water quality and wastewater treatment. In addition, the Legacy Program supports watershed rehabilitation and freshwater conservation efforts, although these efforts are not specifically tracked. The Office of Economic Adjustment provides technical expertise and financial support to state and local governments in planning community adjustments. The agency may acquire the services of Commerce’s Economic Development Administration to execute grants that implement plans to convert former military bases to civilian uses. According to a senior agency official, the agency does not track its implementation projects by category, some of which could be water infrastructure projects. Created by the Department of Energy Organization Act of 1977, Energy’s mission is to advance the national, economic, and energy security of the United States; promote scientific and technological innovation in support of that mission; and ensure the environmental cleanup of the nation’s nuclear weapons complex. Although freshwater-related issues are not a primary focus, the department, through grants and direct federal spending, provides financial support for groundwater cleanup projects and for the research and development of desalination technologies. As provided for in the Uranium Mill Tailings Radiation Control Act of 1978, the department engages in groundwater cleanup and remediation activities at several former uranium mill sites. The department also provides funding for alternative clean water supplies, a distillation water treatment plant, and flood control activities at these sites. In addition, Energy and Interior’s Reclamation collaborated to produce a national road map in 2003 for the research and development of desalination and water purification technologies. The Conference Committee report accompanying the 2004 Energy and Water Development Appropriations Act directed Energy’s Office of Environmental Management’s Technology Development and Deployment Program to provide $3 million to support the research and development of desalination research technologies. Sandia National Laboratories coordinated these activities for the department. The Department of Health and Human Services is the government’s principal agency for protecting the health of Americans and providing essential human services. The department’s freshwater programs primarily provide financial support for water supply and wastewater treatment systems. The Administration for Children and Families promotes the economic and social well-being of families, children, individuals, and communities. As part of these efforts, the Office of Community Services provides training and technical assistance grants on a regionwide basis to help small, rural, low-income communities construct, repair, and rehabilitate water and wastewater systems. The agency’s Rural Community Facilities Development Program provides these services, authorized under section 680 of the Community Services Block Grant Act included in the Coats Human Services Reauthorization Act of 1998. Project activities under this program include (1) providing training and technical assistance to low- income rural communities to develop expertise needed to establish and manage water facilities; (2) improving the coordination of federal, state, and local agencies with rural water and wastewater management; and (3) educating local, rural community leaders about available federal resources. According to agency officials, the program enables rural communities to comply with the requirements of the Clean Water Act and the Safe Drinking Water Act. Entities eligible for the program include multistate, regional, private, and nonprofit 501(c)(3) tax-exempt organizations. The Indian Health Service originated in 1955 when Interior transferred administration of the American Indian and Alaska Native health programs to the Department of Health and Human Services. However, the Indian Health Service continues to receive annual appropriations from a budgetline within Interior’s annual appropriations. The Office of Environmental Health and Engineering provides technical and financial support to Indian tribes and Alaska Native communities to promote a healthy environment through the cooperative development and continuing operation of safe water, wastewater, and solid waste systems. The Sanitation Facilities Construction Program is the Indian Health Service’s primary water infrastructure program. Program services are authorized under Public Law No. 86-121, Indian Health Care Improvement Act (Pub. L. No. 94-437), as amended, and the Indian Self-Determination and Education Assistance Act (Pub. L. No. 93-638), as amended. The Indian Health Service may provide sanitation facilities directly or in partnership with other federal agencies or with nonfederal entities. The agency’s nonfederal project partners could include tribes, tribal-designated housing entities, tribal enterprises, states, counties, and municipalities. The agency’s federal partners include Housing’s Office of Native American Programs, Interior’s Bureau of Indian Affairs and Reclamation, Agriculture’s Rural Utilities Service Program, and EPA. The agency may also administer projects funded by those federal agencies to provide sanitation facilities services to tribes. The Department of Housing and Urban Development supports community development through partnerships with states and local governments. The department’s primary tool for providing financing for public improvements is the Community Development Block Grant program, authorized under title 1 of the Housing and Community Development Act of 1974, as amended. The program is managed by (1) the Office of Community Planning and Development and (2) the Office of Public and Indian Housing. The Office of Community Planning and Development runs the Community Development Block Grant program for states, entitled cities and counties, and insular areas (U.S. territories). Grant recipients design and administer their projects, and the Office of Community Planning and Development provides project oversight. In order to qualify for a grant, a project must meet one of the following national objectives: (1) principally benefit low- and moderate-income families, (2) aid in the prevention or elimination of slums or blight, or (3) meet other urgent community development needs. The agency can track some of its freshwater projects on the basis of general categories, such as water and sewer and flood and drainage activities. In addition to the Community Development Block Grant program, local governments may apply for a loan guarantee program, authorized under section 108 of the Housing and Community Development Act of 1974, as amended. The local government pledges its future year Community Development Block Grant funds as security for a federally guaranteed loan, which provides funds to pursue neighborhood revitalization projects. Freshwater projects are not specifically tracked in this program. In addition to executing its own grants during fiscal years 2000 through 2004, the agency, through a memorandum of agreement, administered grants funded by the Appalachian Regional Commission. The Office of Public and Indian Housing provides programs for Indian tribes and Alaska Native communities, which are similar to those provided by the Office of Community Planning and Development. These programs are the Indian Community Development Block Grant program governed by title I of the Housing and Community Development Act of 1974, as amended, and the Indian Housing Block Grant program governed by the Native American Housing Assistance and Self Determination Act of 1996. The programs can provide financial support for water infrastructure projects; however, the freshwater component is not specifically tracked. Indian Housing Block Grant recipients are eligible to secure financing for affordable housing activities using a 95 percent federal loan guarantee under title VI of the Native American Housing Assistance and Self- Determination Act. Infrastructure projects that support freshwater and sanitary waste disposal for low- and moderate-income households are eligible activities under the title VI program. Part of the mission of the Department of the Interior is to protect natural areas through scientific research and to foster sound use of land and water resources. Consequently, the department supports multiple types of freshwater programs. Specifically related to freshwater resources, the Bureau of Indian Affairs is responsible for protecting water and land rights and developing and maintaining infrastructure, such as dams and drinking water facilities, on 55.7 million acres of land held in trust by the United States for Indian tribes and Alaska Native communities. As part of these efforts, the agency provides financial support for drinking water supply, wastewater treatment, irrigation, dam safety, water rights litigation and negotiation, and Indian land and water claim settlements programs. The agency’s authority to support these programs comes primarily from the Snyder Act of 1921 and the Indian Dams Safety Act of 1993. The mission of the Bureau of Land Management is to sustain the health, diversity, and productivity of federally owned lands that are located primarily in 12 western states. As such, the agency supports watershed management and water dispute management activities and carries out these efforts primarily under the Federal Land and Policy Management Act of 1976. Among its programs, the agency supports water resource inventories, watershed assessments, wetland and stream projects, and the monitoring of lake and stream ecosystems. The mission of the Bureau of Reclamation is to manage, develop, and protect water and related resources in an environmentally and economically sound manner. As a part of these efforts, the agency is the nation’s largest supplier of water—managing 457 dams and 348 reservoirs in 17 western states—and delivers water to irrigate 10 million acres of land and to supply more than 31 million municipal, rural, and industrial water users. Financial support for freshwater programs is provided under a number of different legal authorities, but primarily under the Reclamation Act of 1902. The agency provides financial support for freshwater activities primarily through the Water and Related Resources program, through direct federal spending and through grant, loan, and loan guarantee programs. The terms of funding vary and are dictated by project authorization, legislation, or other authorizations. In addition to annual appropriations, Reclamation receives funding from a variety of sources, such as the Reclamation Fund, Central Valley Project Restoration Fund, and funds collected from surcharges placed on the use of water and power. Reclamation also receives funds from other federal agencies to conduct various freshwater programs, such as collecting data for the U.S. Geological Survey’s watershed management and flood control activities. Moreover, Reclamation—under section 607 of the Foreign Assistance Act of 1961—periodically received funds from USAID and State to conduct freshwater projects in other countries during fiscal years 2000 through 2004. The mission of the Fish and Wildlife Service is, working with others, to conserve, protect, and enhance fish, wildlife, and plants and their habitats for the continuing benefit of the American people. As such, the agency’s freshwater programs primarily support watershed protection, restoration, and management through direct federal spending and grant programs. For example, under the Landowner Incentive Program, the agency provides grants to state agencies with primary responsibility for fish and wildlife to establish or supplement landowner incentive programs that protect and restore habitats on private lands. Freshwater programs carried out by the agency are conducted under a number of different legal authorities— generally under the North American Wetlands Conservation Act of 1989, the Pittman-Robertson Wildlife Restoration Act of 1937, and the Dingell- Johnson Sport Fish Restoration Act of 1950. According to agency officials, the majority of the funds come from dedicated funding sources—the Pittman-Robertson Wildlife Restoration Act and the Dingell-Johnson Sport Fish Restoration Act, which includes excise taxes collected on sporting arms, ammunition, bows and arrows, and fishing equipment. The remaining funds come from annual appropriations from the general fund. The mission of the National Park Service is to preserve unimpaired natural and cultural resources and values of the national park system. Through the National Park Service Organic Act of 1916, the agency mainly supports watershed protection, restoration, and management programs as well as drinking water supply and water rights management programs. Each of the 388 national parks is responsible for management activities in the park. Information on the amount of financial support provided for freshwater programs at individual parks is not tracked centrally. However, the agency did provide information on major freshwater projects as well as the financial support provided by the Water Resources Division. The agency conducts freshwater work directly through wetlands restoration efforts at the Everglades National Park and stream restoration activities affiliated with the removal of the Elwha and Glines Canyon Dams in Olympic National Park. In addition, the Water Resources Division within the agency provides direct financial and technical support for freshwater resources management and policy and operation support to units of the national park system. The mission of the U.S. Geological Survey is to provide reliable scientific information to, among other things, describe and understand the Earth and manage freshwater resources. The agency conducts programs that support the planning and operation of freshwater resources, primarily through technical assistance and research activities. Specifically, the agency collects basic data on stream flow, groundwater levels, and water quality and conducts interpretive studies designed to answer specific questions about water resources. These activities are primarily carried out through the Cooperative Water Program, National Water Quality Assessment, and Hydrologic Networks and Analysis Program. The agency conducts its freshwater programs under many different legal authorities—most generally under the U.S. Geological Survey Organic Act. The agency also receives funds from other federal agencies—such as EPA, the Corps, State, and Interior’s National Park Service—to gather data on water resources and water contaminant studies. The Department of Transportation develops and coordinates policies to provide for an efficient and economical national transportation system. The department’s freshwater programs include watershed management and navigation. The Federal Highway Administration is charged with carrying out highway safety projects and administering the Federal-Aid Highway Program. Funded by the Highway Trust Fund as authorized by the Transportation Equity Act for the 21st Century and other acts, the program provides financial resources and technical assistance to state and local governments for constructing, preserving, and improving highways. The program provides financial support for selecting, planning, designing, and building highways. Funds may also be used for reducing water pollution due to highway runoff and are included in the project’s overall costs. States are responsible for project oversight and may voluntarily report data spent on individual project costs to the Federal Highway Administration. Another component of the Federal-Aid Highway Program is the wetland mitigation program, which replaces an average of 1.5 acres of wetlands for every acre affected by highway construction activities. Financial information on this program is not included in the agency’s total because it is incomplete and primarily available on an acreage basis, which compares acres of wetlands replaced and the acres affected by highway construction activities. A wholly owned government corporation within the Department of Transportation, the Saint Lawrence Seaway Development Corporation was created by the Wiley-Dondero Act of May 13, 1954. In cooperation with the Canadian St. Lawrence Seaway Management Corporation, the agency serves the marine transportation industries by providing a safe, secure, reliable, efficient, and competitive international waterway. The agency constructs, operates, and maintains the part of the St. Lawrence Seaway between Montreal and Lake Erie that is within the territorial limits of the United States. The majority of the agency’s activities are related to lock infrastructure and waterway operations, maintenance, and security. The agency receives the vast majority of its annual budget from an appropriation from the Harbor Maintenance Trust Fund. The rest of its budget is derived from other sources, such as interest income, rent payments, and the collection of noncommercial tolls. The Appalachian Regional Commission, established by the Appalachian Regional Development Act of 1965, as amended, seeks to foster economic and community development across the 13-state Appalachian Region. The region includes all of West Virginia and portions of Alabama, Georgia, Kentucky, Maryland, Mississippi, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, and Virginia. The agency provides grants, which are primarily funded by annual appropriations, for basic infrastructure services to public and nonprofit entities through its Area Development Program. According to a senior agency official, approximately 33 percent of the agency’s budget goes toward grants for drinking water, wastewater, and related activities. The agency has other programs that may also provide financial support for freshwater activities, although that is not their primary purpose. The Appalachian Regional Commission grants are administered either by the agency or by another federal agency, such as Agriculture’s Rural Utilities Service; Commerce’s Economic Development Administration; Housing’s Office of Community Planning and Development; and, occasionally, EPA. Within the Environmental Protection Agency, the Office of Water supports most of the agency’s freshwater efforts, primarily through the Drinking Water and Clean Water State Revolving Funds. Under these programs, EPA provides grants to states for below-market loans to municipalities for drinking water or surface water protection and restoration projects. The Clean Water State Revolving Fund supports the construction of municipal wastewater facilities and nonpoint source pollution control and estuary protection projects. The Drinking Water State Revolving Fund provides financial support to community water systems for installing, upgrading, or replacing infrastructure. EPA also administers other grant programs, such as the Public Water System Supervision Grants Program, Water Quality Cooperative Agreements, Non-Point Source Grants, and Wetland Program Grants. These programs are conducted primarily under the Clean Water Act and the Safe Drinking Water Act. EPA also receives funds from other federal agencies to carry out freshwater programs. For example, EPA received funds from the Corps for wetland restoration activities in coastal Louisiana. Established by the National Science Foundation Act of 1950, the National Science Foundation is an independent federal agency whose mission is to promote the progress of science; advance the nation’s health, prosperity, and welfare; and secure the nation’s defense. Using annual appropriations from the general fund, the agency provides grants for research activities across scientific and engineering disciplines to address issues related to, among other things, the preservation, management, and enhancement of the environment. With regard to freshwater issues, the agency provides financial support for research on, among other things, drinking water treatment, desalination, flood control, wastewater treatment, and watershed management. For example, during fiscal years 2000 through 2004, the agency—through the Science and Technology Centers Program— provided financial support to the Center for Sustainability of semi-Arid Hydrology and Riparian Areas. The center will carry out multidisciplinary research into the hydrology and management of freshwater resources in semi-arid regions. The Small Business Administration, created by the Small Business Act of 1953, seeks to assist the interests of small businesses. The agency does not lend money; rather, it provides loan guarantees to small businesses. The agency’s Basic 7(a) and Basic 504 loan programs can be used for water infrastructure projects, such as water supply and irrigation systems, sewage treatment facilities, and dredging and surface cleanup activities. Under both programs, the agency guarantees a portion of the loans and shares the risk with a commercial lender if a borrower defaults on its loan. A senior agency official notes that unlike the Basic 7(a) program, the Basic 504 program receives no appropriations. Eligibility for the programs varies slightly, but under both programs recipients must be a for-profit corporation. The Small Business Administration initially obligated about $327,000 during fiscal years 2000 to 2004 to cover potential default on water-related loans in its loan guarantee program. This small amount of obligations is not included in our governmentwide totals. Our review focuses on the federal agencies that provide the majority of the federal financial support for selected freshwater programs. Additional agencies that support freshwater activities, as described by agency officials, reports, and documents, are discussed below. The Bonneville Power Administration is a self-financing agency that markets wholesale electrical power and operates and markets transmission services in the Pacific Northwest. The agency pays for its costs through power and transmission sales and funds the region’s efforts to protect and rebuild fish and wildlife populations in the Columbia River Basin. Within the Department of Homeland Security, the Federal Emergency Management Agency leads the nation’s efforts to prepare for hazards and manages the federal response and recovery efforts following any national event. In terms of freshwater activities, the agency manages the National Flood Insurance Program—a program that assists with flood insurance, floodplain management, and flood hazard mapping activities. Regional commissions provide assistance to some of the nation’s most chronically poor and distressed communities. Since its creation in 1965, the Appalachian Regional Commission has been providing assistance to counties affected by severe and chronic economic distress. More recently, Congress created additional regional economic development entities. The Delta Regional Authority, created in 2000, serves parts of Alabama, Arkansas, Illinois, Kentucky, Louisiana, Mississippi, Missouri, and Tennessee. The Northern Great Plains Regional Authority, created in 2002, includes the states of Iowa, Minnesota, Nebraska, North Dakota, and South Dakota. Additionally, Congress created a wholly intrastate commission in 1998, the Denali Commission, to provide infrastructure and economic development throughout Alaska. The Water Resources Planning Act of 1965 authorized the President to establish river basin commissions to serve as the principal agencies for coordinating the development of water and related land resources in river basins. In 1981, an executive order terminated six of these commissions. Many river basin commissions established by interstate compacts, such as the Delaware River Basin Commission, still exist. Information on federal financial assistance provided to river basin commissions is not readily available because individual commissions primarily maintain these data. The Tennessee Valley Authority is a self-financing corporation of the federal government that supplies affordable and reliable power and operates fossil fuel, nuclear, and hydropower plants in the Tennessee Valley. The agency receives its revenues through power sales and the sale of bonds in the financial market and uses its own funds for a variety of stewardship and watershed activities. These activities include reservoir operations, navigation, watershed improvement activities, aquatic plant management, and land planning and use. During fiscal years 2000 through 2004, the United States provided financial contributions to three binational commissions for freshwater projects along U.S. borders. The United States typically made contributions to the budgets of these commissions, and a portion of these contributions supported freshwater projects along U.S. borders. Since the commissions coordinate activities along U.S. borders with State and EPA, financial information on the U.S. contributions for freshwater projects during fiscal years 2000 through 2004 was available. These commissions, as described by federal agencies and commission officials and documents, are discussed below. Created as a side agreement to the North American Free Trade Agreement, the Border Environment Cooperation Commission is a binational international organization that works to conserve, protect, and enhance the environment along the U.S.-Mexico border. The commission’s annual budget for drinking water and wastewater treatment activities comes mainly from EPA. The commission’s annual budget for irrigation, air quality, solid waste, and other projects comes from State and Mexico’s Secretariat of Environment and Natural Resources. Managed by a binational Board of Directors, composed of five members from each country, the commission identifies, supports, evaluates, and certifies various environmental infrastructure projects. Once the commission certifies that sustainability and public participation requirements are met, project sponsors may qualify for funding from the North American Development Bank or from other sources requiring such certification. Additionally, through its Project Development Assistance Program, which is EPA-funded, the commission provides technical grants to qualifying border communities for the development of water and wastewater projects. The International Boundary and Water Commission is a binational, treaty- based organization comprised of a U.S. section, headquartered in El Paso, Texas, and a Mexican section, headquartered in Ciudad Juarez, Mexico. Through binational cooperation, the commission seeks to preserve the international land and river boundary between Mexico and the United States in a manner that balances economic, environmental, and sovereignty needs; carry out the conservation, flood control, national ownership, and utilization of international waters; and improve the quality and utilization of international waters in a manner that supports ecological needs and regional development. The U.S. section of the commission operates under State’s foreign policy guidance and receives an annual appropriation through the Departments of Commerce, State, Justice, and other related agencies’ appropriation bills. The U.S. section of the commission operates and maintains two international wastewater treatment plants, multiple diversion dams, and numerous smaller in-river structures for flood control and water distribution. In addition, the U.S. section and the Mexican section of the commission jointly operate two international dams. Additionally, the U.S. section of the commission is responsible for the operation and maintenance of five flood control projects with over 500 miles of levees and related structures, which protect approximately 3 million residents and 1.5 million acres of adjoining farmland in the United States and Mexico. Established by the Boundary Waters Treaty of 1909, the International Joint Commission is an independent binational organization that assists the governments of the United States and Canada in addressing water quality and quantity issues and air pollution problems along the U.S.-Canadian border. The U.S. section of the commission receives an annual appropriation through State for these purposes. The commission is made up of six commissioners, three that are appointed by the President of the United States and three appointed by the Governor in Council of Canada, on the advice of the Prime Minister. Although the commission does not build or maintain any water-related infrastructure, it undertakes research efforts and analyses of binational water issues and of the operations of selected water works affecting both countries. For example, a current research and development effort is a 5-year, $20-million study to determine changes to the operation of infrastructure that affect water levels and flows on Lake Ontario and the St. Lawrence River. The U.S. section and the Canadian government are sharing equally the cost of this study. This appendix on U.S. financial support of freshwater programs abroad has three sections. First, we provide a general overview of the agencies responsible for the majority of the U.S. government’s direct financial support for freshwater programs abroad, along with the total amount each agency provided for freshwater programs during fiscal years 2000 through 2004, excluding aid provided to Afghanistan and Iraq. Second, we briefly discuss other agencies that can also support freshwater programs abroad. Third, we provide information on international organizations to which the United States contributes annually and that support freshwater projects around the world. The following agencies provide the majority of the U.S. government’s financial support specifically for freshwater programs abroad. These programs, as described by agency officials, documents, and reports, are discussed below. With the Department of Agriculture, the Foreign Agricultural Service works to improve foreign market access for U.S. products and provides food aid and technical assistance to foreign countries. As the primary agency responsible for Agriculture’s international work, the agency administers international research and technical assistance activities in coordination with developing and transitional countries. These activities include international cooperative research efforts on freshwater quality and availability, irrigation, and watershed management issues. The majority of the agency’s financial support for freshwater activities abroad is provided by the Export Credit Guarantee Program using various food aid agreements, as authorized by several statutes, including the Food for Progress Act of 1985; section 416(b) of the Agriculture Act of 1949, as amended; and the McGovern-Dole International Food for Education and Child Nutrition Program of 2003. Through these food aid agreements, the Export Credit Guarantee Program provides U.S. agricultural commodities to a recipient country for resale. The proceeds from the sale may be used to support various agricultural, economic, and infrastructure development projects in the recipient country. Recently, this program funded the development of, among other things, water supply, irrigation, and water treatment projects in countries such as Guatemala, Pakistan, Vietnam, and Uzbekistan. Additionally, as authorized by section 1543 of the Food, Agriculture, Conservation, and Trade Act of 1990, as amended, the agency manages the Cochran Fellowship Program, which provides opportunities for professionals from foreign countries to attend training programs on a number of agricultural issues, including drought mitigation and irrigation management. Although freshwater is not a focus area for the Department of Defense, the department provides financial support for several efforts related to freshwater projects abroad. In addition to providing engineering services for military and civilian purposes in the United States, the Corps, through its Civil Works program, provides technical assistance to foreign countries on a number of freshwater issues, such as wastewater treatment, flood control, and irrigation. The Corps is authorized by section 234 (Interagency and International Support Authority) of the Water Resources Development Act of 1996 to provide up to $250,000 of its own funding for technical assistance on freshwater projects to international organizations and federal agencies in foreign countries. However, other federal agencies, such as USAID, and the governments of foreign countries often provide the funding that supports the Corps’ freshwater work in foreign countries. Recently, the Corps received funds from USAID to provide technical assistance for the reconstruction of drinking water infrastructure and wastewater treatment systems in Iraq. The Corps also received funds from USAID for similar work in Afghanistan. In addition to the Corps, the Defense Security Cooperation Agency, within the Office of the Secretary of Defense, coordinates and oversees humanitarian assistance programs abroad. Under 10 U.S.C. § 2561, U.S. military units are authorized to perform humanitarian activities (primarily carried out through contracts) in communities and areas where military units are deployed abroad. These activities include, among other activities, digging wells and improving sanitation facilities. For example, military commanders in Afghanistan and Iraq from the United States Central Command have used these funds to repair and augment drinking water and wastewater systems. As part of its mission to conserve fish, wildlife, and plants and their habitats for the American people, the Fish and Wildlife Service—within the Department of the Interior—protects wetlands used by waterfowl and migratory birds, in Canada and Mexico, as well as in countries in Central and South America and the Caribbean. The agency provides financial support to protect, restore, and enhance these wetlands through several programs, including the North American Wetlands Conservation Act of 1989 and the Neotropical Migratory Bird Conservation Act of 2000. Through its North American Wetlands Conservation program, the agency coordinates with representatives from Canada and Mexico to provide grants for long-term acquisition, restoration, and/or enhancement of critical wetlands used by waterfowl and migratory birds in the three countries. Grant recipients are required to provide matching funds at a 1:1 ratio. Under the Neotropical Migratory Bird Conservation Act of 2000, the Fish and Wildlife Service established a matching grants program that, among other things, supports the maintenance, protection, and restoration of the habitats of birds in the United States, Latin America, and the Caribbean. Government agencies, individuals, corporations, and other private entities may apply for a grant from this program, if they provide $3 for every $1 they receive from the agency. As recognized in the Foreign Assistance Act of 1961, as amended, the Department of State is the lead agency responsible for the continuous supervision and general direction of foreign assistance. The department’s mission is to create a more secure, democratic, and prosperous world for the benefit of the American people and the international community. Several bureaus within State provide financial support for freshwater programs abroad; however, the level of their involvement varies. The Bureau of Oceans and International Environmental and Scientific Affairs coordinates the department’s overall policy for international environmental issues, including freshwater issues. The bureau provides grants to support international initiatives on watershed management, access to safe drinking water, flood control, and water dispute management. The Bureau of Near Eastern Affairs coordinates U.S. foreign policy and diplomatic relations with countries in the Middle East and North Africa. In support of the United State’s role in the Middle East Peace Process, this bureau promotes technical cooperation between Arabs and Israelis on numerous freshwater issues. Typically, the bureau provides funds to other federal agencies, such as those within Agriculture and Interior, for projects designed to promote water cooperation and the building of strong working relationships between water officials and experts in the region. The Bureau of Population, Refugees, and Migration coordinates the department’s efforts related to global population, refugees, and migration issues. This bureau provides financial support for various water and sanitation projects in foreign countries through grants and cooperative agreements issued to international and nongovernmental organizations. These projects provide water for drinking, irrigation, and sanitation to temporary refugee camps. Although State does not specifically track freshwater expenditures for refugee camps, a bureau official estimates that less than $5 million a year is obligated to support these activities. The Bureau of International Organization Affairs develops and implements the U.S. government’s policies and efforts with the United Nations, its affiliated agencies, and other international organizations. This bureau manages the U.S. government’s assessed and voluntary contributions to various organizations. Most of these contributions are made directly to the core budgets of these organizations and are not designated for specific activities. The African Development Foundation is a public corporation and federal agency established by the African Development Foundation Act of 1980, as amended. The agency supports community-based, self-help initiatives to alleviate poverty and to promote sustainable development in Africa. Currently working in 15 sub-Saharan African countries, the agency provides grants of usually $250,000 or less directly to community-based, nongovernmental organizations and enterprises administered by Africans. The agency’s grants enable community-based groups to expand their production capacity and increase incomes, thereby improving the community’s economic security. Although freshwater is not a program focus, the agency supported various irrigation, drinking water, and wastewater projects in countries such as Botswana, Guinea, Mali, and Niger. Established by the National Science Foundation Act of 1950, as amended, the National Science Foundation is an independent U.S. government agency whose mission is to promote the progress of science; advance the national health, prosperity, and welfare; and secure the national defense. Using annual appropriations from the general fund, the agency provides grants for scientific research for various freshwater issues abroad, including watershed management, drinking water treatment, and wastewater management. The statute creating the agency authorizes it to promote the interchange of scientific information between scientists and engineers in the United States and foreign countries. For example, the agency’s Office of International Science and Engineering provides grants to U.S. scientists and engineers to participate in international collaborative research partnerships with their foreign colleagues. The U.S. Agency for International Development is an independent federal agency created under the Foreign Assistance Act of 1961, as amended. Operating under guidance from the Secretary of State, USAID provides economic, development, and humanitarian assistance around the world in support of U.S. foreign policy goals. As part of the agency’s efforts to promote an integrated approach to water resources management, USAID provides technical assistance, educational and outreach opportunities, emergency relief assistance, and international leadership on a variety of water issues in over 76 countries. USAID reports that over the 5-year period covered in this analysis (fiscal years 2000 through 2004), approximately 5 percent of its annual appropriation has been used to support freshwater- related activities. USAID provides financial support for freshwater programs through partnerships with nongovernmental organizations; government entities (host country and U.S. government agencies); and public international organizations, such as United Nations agencies. During fiscal years 2000 through 2004, USAID missions in Egypt, Jordan, and the West Bank/Gaza received the most financial support from USAID for freshwater projects. Most funds for freshwater projects are distributed as grants, although some funds are provided in the form of loan guarantees through USAID’s Development Credit Authority. The majority of USAID’s freshwater funds are used to support water supply projects. USAID also supports sanitation and wastewater management projects. Additionally, USAID provides financial support for, among other things, watershed management, irrigation, and flood and drought forecasting and preparedness activities in foreign countries. During fiscal years 2002 through 2004, USAID provided about $670 million for the reconstruction of drinking water infrastructure and wastewater treatment systems in Afghanistan and Iraq. USAID primarily drew on funds from the President’s Supplemental Appropriation for the Reconstruction and Development of Iraq and Afghanistan to carry out this work. Created under the Foreign Assistance Act of 1961, as amended, the U.S. Trade and Development Agency is an agency that provides financial support to promote U.S. private sector participation in development projects in developing and middle-income countries. The agency offers early planning support to overseas development projects by funding technical assistance activities, feasibility studies, conferences, and other activities. The agency provides grants directly to host-country project sponsors (i.e., local, regional, and national governments; private sector; and nongovernmental organizations) that agree to select U.S. companies to perform the work associated with project planning. Agency-supported projects span a variety of sectors, including drinking water supply, wastewater treatment, irrigation, and flood control. For example, the agency supported the development of water treatment and supply projects in Ghana, wastewater treatment facilities in India, and an irrigation project in China. Our review focused on the federal agencies that provide the majority of the federal financial support for selected freshwater programs abroad. Additional agencies that support freshwater programs, as described by agency officials, reports, and documents, are discussed below. In conjunction with the Pan American Health Organization, the Centers for Disease Control and Prevention developed a Safe Water System program to help provide safe drinking water to developing countries. To date, this project has supplied inexpensive, adaptable, and flexible drinking water technologies to communities in at least 19 countries. The Export-Import Bank is an independent agency that assists in financing the export of U.S. goods and services to international markets by providing U.S. businesses with working capital guarantees, export credit insurance, loan guarantees, and direct loans. Although freshwater-related goods and services are not a primary focus, the bank provides financial support for the export of water purification devices, wastewater treatment systems, and technologies designed to prevent and mitigate water pollution. The Inter-American Foundation is an independent foreign assistance agency that provides grants to nongovernmental organizations in Latin America and the Caribbean for economic and social development projects. The agency primarily relies on congressional appropriations to fund development projects. The agency also has access to additional funding from the Social Progress Trust Fund, which consists of payments on loans made by the United States to Latin American and Caribbean countries under the Alliance for Progress program. According to a senior agency official, the agency does not specifically track freshwater expenditures, which are a small part of the agency’s total portfolio; activities supported include improving access to drinking water, developing irrigation systems, and protecting watersheds. The Overseas Private Investment Corporation is a self-sustaining U.S. government corporation created to facilitate U.S. private investment in developing countries and emerging market economies, primarily by offering political risk insurance, loan guarantees, and direct loans. The agency provides financial support for various development projects worldwide, including some freshwater projects. Created in late 1961, the Peace Corps promotes world peace and friendship by helping developing countries meet their need for trained workers while promoting mutual understanding between Americans and citizens of foreign countries. The Peace Corps supports volunteers who commit to 2- year assignments in host communities where they work on various community development projects, such as teaching English, strengthening farmer cooperatives, or building sanitation systems. Although some volunteers work on water sanitation systems and agricultural projects, the Peace Corps does not directly fund any freshwater projects. During fiscal years 2000 through 2004, the United States indirectly supported freshwater programs abroad through its financial contributions to various international organizations that support freshwater programs abroad. The United States typically made contributions to the general budgets of these organizations, although some contributions were directed for a particular project or program. Some portion of the general budgets of these organizations was used to support freshwater projects around the world. These organizations can be split into two groups: (1) multilateral development banks and international financial institutions and (2) other international organizations. Contributions to multilateral development banks and international financial institutions are typically coordinated by the Department of the Treasury, while contributions to the other organizations are typically coordinated by State. These organizations’ freshwater projects and programs, as described by officials and documents, are discussed below. The African Development Bank Group—the African Development Bank and the African Development Fund—provides loans and offers technical assistance to African countries for development projects. The bank’s priority areas include rural water supply, irrigation, and other agricultural and development projects. The bank’s membership includes all countries in Africa as well as some countries in the Americas, Asia, and Europe. The Asian Development Bank—the Ordinary Capital Resources group and the Asian Development Fund—provides loans and loan guarantees and offers technical assistance to low- and middle-income countries in Asia and the Pacific for a variety of economic and social development projects. Traditionally, the bank supports projects related to agriculture (including irrigation) and rural development. Recently, the bank placed additional importance on its Social Infrastructure Sector, which supports projects to improve water supply. Since its inception, the bank has lent the most money to Indonesia; China, Pakistan, and India were also major borrowers. Created in 1991, the European Bank for Reconstruction and Development seeks to foster the transition toward open market-oriented economies in countries in Central and Eastern Europe and the Commonwealth of Independent States. The bank provides financial support for, among other issues, the development and improvement of water and wastewater systems. The Global Environment Facility is a multilateral, financial institution that provides grants and other financial assistance, particularly in low- and middle-income countries, for projects that protect the global environment. One of the organization’s top priority areas is protecting freshwater resources. Working in partnership with the U.N. Environment Programme, the U.N. Development Programme, and the World Bank, the organization funds projects to improve protection of safe drinking water supply, manage water disputes, and reduce water-borne pollutants. The Inter-American Development Bank provides financial and technical support for various development projects in Latin America and the Caribbean. The bank funds projects in several sectors, including agriculture, fisheries, water and sanitation, and the environment. Since 1961, the bank has lent most of its money to Brazil, Argentina, Mexico, and Colombia. Established in 1977, the International Fund for Agricultural Development works to alleviate poverty and improve nutrition around the world, with a special focus on low-income countries. As an international financial institution associated with the United Nations, the organization provides loans and grants for technical assistance, research, and activities in several areas related to freshwater. These areas include agricultural development, irrigation, and water infrastructure development in rural areas. Created in a side agreement to the North American Free Trade Agreement, the North American Development Bank is jointly funded by Mexico and the United States. Working closely with the Border Environment Cooperation Commission, the bank finances projects related to drinking water supply, wastewater treatment, and other environmental infrastructure projects along the U.S.-Mexico border. The United States provides its annual contributions—through EPA—to the bank as a grant. The bank may use a portion of these contributions to support freshwater projects along the U.S. borders. The bank’s Board of Directors consists of members from the United States and Mexico, with the chairmanship of the board annually alternating between U.S. and Mexican representatives. The World Bank Group, the largest multilateral development bank, is made up of the International Bank for Reconstruction and Development, International Development Association, International Finance Corporation, and Multilateral Investment Guaranty Agency. These institutions provide loans and loan guarantees and offer technical assistance for various economic and social development projects. The bank has several water- related program areas that focus on, among other things, extending water supply and sanitation services to the urban poor, increasing rural access to water supply and sanitation, and improving water resources management. The bank has funded the construction of dams, flood control infrastructure, and drinking water and wastewater treatment facilities in countries around the world. Together with the U.N. Development Programme, the World Bank created the Water and Sanitation Program, which works with donors, governments of foreign countries, and nongovernmental organizations to support the development of cost- effective water delivery technologies and implementation of strategies for providing safe water and sanitation to the world’s poor. Created more than six decades ago, the Inter-American Institute for Cooperation on Agriculture is a specialized international agency that promotes agricultural development, food security, and rural economic development in North, Central, and South America and the Caribbean. The institute’s focal areas include rural development, agribusiness development, and agricultural and food safety. Within these areas, the institute provides technical assistance for the management of, among other things, water resources, watershed management, and efforts to combat desertification. The Organization of American States brings together 35 countries in North, Central, and South America and the Caribbean to strengthen cooperation and advance common interests. Through its Office for Sustainable Development and Environment, the organization provides technical assistance to countries and promotes cooperation on various issues, including integrated water resources management. For example, the organization provides financial support to help countries in South and Central America manage transboundary water resources in several major river basins. The organization works in partnership with the United Nations’ Environment Programme, the World Bank, and the Global Environment Facility on these projects. The organization also serves as the technical secretariat for the Inter-American Water Resources Network, which was created in 1993 to foster cooperation on water management issues in the hemisphere. The Organisation for Economic Co-operation and Development brings together 30 industrialized nations in a forum to discuss, develop, and refine economic and social policies. The organization is funded by contributions from member countries, with the United States providing 25 percent of the general budget. Among other issues, the organization carries out research and analysis on water management policies, water use for agriculture, water pricing, water and wastewater infrastructure financing, and technologies to improve water quality. The Pan American Health Organization is an international public health agency that works to improve the health and living standards of residents in North, Central, and South America and the Caribbean. Affiliated with the United Nations and the World Health Organization, the organization provides technical support to various countries on a number of health- related issues. Among the organization’s top priorities are improving supplies of clean water and adequate sanitation facilities. The Ramsar Convention on Wetlands is an inter-governmental treaty adopted in 1971 that addresses various wetlands-related issues worldwide. A small secretariat carries out the work of the convention using financial contributions from the 138 countries that are signatories to the convention. The United States makes voluntary contributions to the organization’s general budget and also funds a grant program for wetlands-related training. The organization has also used U.S. contributions to support the protection of key wetland habitats for migratory birds in Latin American and the Caribbean. Among other purposes, the United Nations seeks to achieve international cooperation in solving economic, social, cultural, and humanitarian problems. Funded through assessed dues paid by its 191 member nations, the United Nations consists of a number of commissions, funds, organizations, and other entities developed to respond to global needs. The United States has been the largest contributor of funds since the organization’s inception in 1945, making annual contributions that amount to about 22 percent of the United Nations’ general budget. In addition to its assessed contributions, the United States provides voluntary contributions to the United Nations and its affiliated organizations. Most of these contributions are used for humanitarian and development programs. With respect to freshwater projects, the United Nations provides advisory and technical assistance to governments on various freshwater issues, such as water resources management and infrastructure improvements. Within the United Nations’ system, 26 entities support water-related projects. However, according to a U.N. official who works on freshwater issues, the entities do not currently work together to track the financial support provided for freshwater projects. Created in 1948, the World Conservation Union is an international environmental organization with members from the government of foreign countries, governmental agencies, and nongovernmental entities. The organization has over 1,000 members from 140 countries, including 6 U.S. federal agencies—USAID, Agriculture’s Forest Service, Commerce’s National Oceanic and Atmospheric Administration, Interior’s Fish and Wildlife Service and National Park Service, and EPA. The organization’s mission is to encourage the conservation of nature and ensure that the use of natural resources is ecologically sustainable. In addition to providing technical assistance on the management and restoration of wetlands and water resources, the organization supports several freshwater projects, such as the Water for Nature Initiative. This initiative works with partners in approximately 40 countries to improve water resources management in 10 river basins. In addition to those named above, Diana Cheng, Richard P. Johnson, Nathan Morris, Lynn Musser, Jonathan Nash, and Kim Raheb made significant contributions to this report. Denise Fantone, Jessica Fast, Carol Herrnstadt Shulman, and John Hutton also contributed to this report. | As the world's population tripled during the past century, demand for the finite amount of freshwater resources increased six-fold, straining these resources for many countries, including the United States. The United Nations estimates that, worldwide, more than 1 billion people live without access to clean drinking water and over 2.4 billion people lack the basic sanitation needed for human health. Freshwater supply shortages--already evident in the drought-ridden western United States--pose serious challenges and can have economic, social, and environmental consequences. Multiple federal agencies share responsibility for managing freshwater resources, but consolidated information on the federal government's financial support of these activities is not readily accessible. GAO was asked to determine for fiscal years 2000 through 2004 how much financial support federal agencies provided for freshwater programs in the United States and abroad. For the purposes of this report, freshwater programs include desalination, drinking water supply, flood control, irrigation, navigation, wastewater treatment, water conservation, water dispute management, and watershed management. Of the over $52 billion in total financial support provided by federal agencies for freshwater programs during fiscal years 2000 through 2004, about $49 billion was directed to domestic programs and about $3 billion supported programs abroad. Domestic program activities involved 27 federal agencies, but 3 agencies--the Environmental Protection Agency, the Army Corps of Engineers, and the Department of Agriculture's (Agriculture) Rural Utilities Service--accounted for over 70 percent of the financial support. Eighteen agencies supported domestic drinking water supply programs and 16 supported domestic wastewater treatment and watershed management programs. Grant programs of over $22 billion and direct federal spending of about $22 billion accounted for most of the domestic financial support. In addition to the about $49 billion that directly support freshwater activities in the United States, some agencies also have programs that may indirectly support such activities, but it is difficult to determine the dollar value of this indirect support. For example, Agriculture's Conservation Reserve Program supports multiple activities, including irrigation, but information on each activity supported by the program is not readily available. Also included in the domestic program is about $175 million that the United States provided to three commissions that conduct freshwater activities along U.S. borders with Mexico and Canada. Of the estimated $3 billion in total financial support directed toward freshwater programs abroad between fiscal years 2000 through 2004, about $1 billion was recently provided for freshwater projects in Afghanistan and Iraq. Most of the financial support for international freshwater programs was provided by the U.S. Agency for International Development. Foreign wastewater treatment and watershed management programs were the ones that most of the agencies supported. The vast majority of the U.S. support for international programs was provided through grants. Not included in the $3 billion for international support are the contributions that the United States made to the general budgets of numerous international organizations, such as the United Nations and the World Bank. The international organizations used some portion of the U.S. contributions to support freshwater activities around the globe. |
Education administers and oversees federal student aid programs authorized by HEA, monitors participants’ activities, and establishes program requirements. Among these financial aid programs is FFELP. The five principal entities involved in FFELP are students, schools, lenders, guaranty agencies,and Education. At schools participating in FFELP, students apply to a participating lender for a loan. The school verifies the student’s eligibility and determines the loan amount the student is eligible to receive. The student then receives the loan from the lender. The guaranty agency guarantees the loan against default. The guaranty agency is the intermediary between Education and the lender, insuring the loan against default and making certain that the lender and the school meet program requirements. The lender is responsible for servicing and collecting the loan, and, if the student defaults on the loan, the lender files a claim with the guaranty agency for reimbursement of most of its loss.Education reimburses guaranty agencies for most of their claims paid to lenders for defaulted loans and for some of their administrative costs. Most lenders contract with a third-party entity to service the loan and collect payments from borrowers. HEA designates which entities are eligible to make FFELP loans to students. In general, an eligible lender is defined as under certain circumstances, a national or state chartered bank, a mutual savings bank, a savings and loan association, a stock savings bank, or a credit union; a pension fund as defined in the Employee Retirement Income Security an insurance company that is subject to examination and supervision by an agency of the United States or a state; in any state, a single agency of the state or a single nonprofit private agency designated by the state; under certain circumstances, an eligible institution; under certain circumstances, the Student Loan Marketing Association or the Holding Company of the Student Loan Marketing Association, including any subsidiary of the Holding Company, or an agency of any state functioning as a secondary market; under certain circumstances, a guaranty agency; a Rural Rehabilitation Corporation; under certain circumstances, any nonprofit private agency functioning in any state as a secondary market; and a consumer finance company subsidiary of a national bank. The majority of organizations making loans to students fall into one of these eligible lender categories. However, organizations that do not meet HEA criteria may still participate in FFELP by contracting with an eligible lender to serve as its trustee.These ineligible lenders fall into two categories: (1) secondary marketsthat have not been designated as an eligible lender for a state and (2) private companies that wish to make and hold student loans. Ineligible lenders generally contract with trustees for three purposes. First, ineligible lenders contract with trustees to allow the ineligible to originate student loans or to purchase them from another originating lender. Second, ineligibles use trustees to securitize portfolios of student loans.Third, trustees are used when ineligibles need to raise the capital necessary to make or purchase student loans without securitizing other loans. For example, some secondary markets raise capital by selling tax-exempt bonds to investors. This report is concerned primarily with trustee arrangements used to enable ineligible lenders to make and hold student loans. At FFELP’s outset the government expected to share the program’s financial risks with state-designated guaranty agencies. However, when states failed to establish such agencies, the Congress enacted legislation with several incentives to increase lender and guaranty agency participation. For example, the Congress provided federal funds for guaranty agencies to use as seed money to pay claims to lenders. In providing these incentives, the Congress kept the financial risk almost entirely with the federal government. The Congress has since shifted some risk back to the guaranty agencies and lenders by reducing the maximum reimbursement and insurance rates on defaulted loans. These actions were intended to encourage both lenders and guaranty agencies to work with borrowers to prevent them from defaulting on their loans. Apart from the defaults that occur when borrowers fail to repay their loans, some loans lose their federal guarantee because lenders, servicers, or guaranty agencies fail to follow the Department’s requirements for making, servicing, and collecting loans or because of fraudulent activity at these organizations. When these latter circumstances occur, Education may assess penalties, refuse to make future payments, or recover payments already made to lenders and agencies for such things as interest subsidies and insurance claims. In 1998, HEA was amended to include a provision stating that eligible lenders that act as trustees are responsible for meeting statutory and regulatory requirements for the loans they hold as trustees. Loss of the federal guarantee due to servicer problems has occurred under the auspices of trustee arrangements in the past. For example, in the late 1980s, Bank of America served as trustee for the California Student Loan Finance Corporation—a California secondary market. Education determined that the loan servicer, United Education and Software Company, failed to transfer certain information to its new computer system and then created a computer program that falsely added collection activity information for a large volume of loans held by the secondary market. As a result, student loans totaling approximately $400 million lost their federal guarantee. However, servicer problems that may result in loan guarantee loss can also occur on loans held by eligible lenders and are not unique to trustee arrangements. The Department of Education reports that approximately 125 trustee arrangements exist between eligible and ineligible lenders. These arrangements account for $25.3 billion in outstanding loans— approximately 19 percent of the outstanding balance of all FFELP loans as of December 1999.The 125 arrangements represent liaisons between 16 eligible lender trustees and 31 ineligible lenders. Ineligible lenders can be further grouped as 17 secondary markets and 14 other ineligibles. Table 1 shows the outstanding balance of loans held via trustee arrangements for the two categories of ineligible lenders. Costs of trustee arrangements fall into two categories—payments to initiate the agreement and annual fees to maintain it. The ineligible lenders we interviewed reported that initiation costs were generally a flat fee ranging between $2,500 and $20,000. Some of these ineligible lenders, such as some secondary markets, reported that their trustee (who serves as trustee for raising capital as well as for originating and purchasing loans) charged this initiation fee each time they issued bonds to raise capital. Annual trustee fees reported by ineligible lenders ranged between $4,500 and $75,000, depending on a number of factors. These lenders reported that the fees could be a flat fee or calculated as a percentage of outstanding loan balances or on the outstanding balance of bond issues. Ineligible lenders reported that annual costs are used to maintain the arrangement and cover the necessary services provided by the trustee, such as transferring documents between the trustee and the ineligible lender for signature, handling paperwork associated with the guarantors and servicers involved in the loan transactions, and carrying out administrative activities associated with obtaining financing to originate or purchase loans. Some eligible and ineligible lenders reported that a variety of other factors could influence the size of both initiation and annual fees. Some of these factors included the structure of the trustee arrangement and the complexity of the the geographic region in which the trustee and ineligible lender are the trustee’s decision to charge administrative fees on a per-loan basis rather than as a flat fee, the additional services the trustee will perform for the ineligible lender, such as payroll services, and the strength of the trustee’s relationship with the ineligible lender. Although some ineligible lenders feared that the provision of HEA that makes trustees fully responsible for trustee-held loans might have an impact on costs, they reported that costs did not significantly increase or decrease after the 1998 amendments were enacted and that they did not see a change in the role of the trustee.According to ineligible lenders we interviewed, current initiation and annual trustee fees do not prohibit them from conducting business in the student loan market. The trustee arrangements we reviewed shared several characteristics. Trustees and ineligible lenders reported similar (1) criteria used by trustees to evaluate ineligibles before they entered into arrangements, (2) elements in trustee arrangement contracts, (3) amounts of review performed of ineligible lenders, and (4) amounts of day-to-day interaction between the trustee and the ineligible lender. For example, trustees reported that most important in deciding whether to enter an arrangement is the business reputation of the ineligible lenderand the reputation of the loan servicing organization it has chosen. In addition, some trustees review the structure of the financing method used by the ineligible lender to raise the capital necessary to originate or purchase loans. This review might include examining a rating agency’s report on the transaction. Trustees and ineligible lenders also reported that their contracts included similar clauses covering trustee and ineligible lender resignations and requirements placed on the ineligible lender. For example, they reported that trustee resignation clauses allow the trustee to resign from the arrangement by giving a specific number of days’ notice to the ineligible lender, such as 60 or 90 days. If the ineligible lender is unable to locate another trustee within that time period, the resignation clause sometimes provides for additional time to obtain a new trustee. In any event, most trustees and ineligible lenders reported that the trustee would probably remain in place until a new one can be engaged, even if the designated time periods have elapsed. The trustee arrangements we reviewed were also similar in the requirements placed on the ineligible lender and the amount of monitoring the trustee performed. For example, trustees generally require ineligible lenders to abide by HEA requirements and to oversee the loan servicer, but the trustees perform limited monitoring activities. One trustee reported that the monitoring is not specific or regular and mostly involves a review of annual reports and audits. Other trustees reported limiting their monitoring to a review of financial statements. Eligible lender trustees and ineligible lenders reported limited day-to-day interaction. For example, some trustees reported that they interact with the ineligible lender only when it is necessary to sign forms, such as quarterly reports that must be submitted to the Department of Education. One trustee reported no daily interaction. Trustee arrangements come with some protections to ensure the federal government’s investment in FFELP is secure while allowing ineligible lenders to participate in the program. The most direct protection comes from an HEA provision that holds eligible lenders fully responsible for any loans they hold as trustees should the loans lose the federal guarantee. For example, although Education officials and other FFELP participants believe the probability of large-scale problems is low, a loan may lose its guarantee because of servicing errors or because of negligence on the part of the servicer or the lender. When these problems occur, the federal government will not reimburse the guarantor or the lender for the associated dollar loss. However, because some problems may not be found until after the federal government has already provided reimbursement, the government may have to recover these monies from the parties involved. According to Education officials, the HEA provision that holds trustees responsible for an ineligible lender’s loans allows the federal government to recoup the losses from the eligible lender trustee rather than the ineligible lender. Education officials further stated the ability to recoup losses from a trustee is important since they believe they do not have direct oversight authority of ineligible lenders. These officials believe the federal government is likely to recover its losses from trustees for two reasons. First, most financial institutions that serve as eligible lender trustees are subject to federal (and in many instances, state) oversight. For example, bank regulatorspromulgate regulations and policies that prescribe safeand sound banking activities and examine banks to assess their safety and soundness. Among the most important of these regulations are those dealing with minimum capital standards. These capital standards provide benchmarks against which regulators can assess the safety and soundness of a bank’s operations as well as its financial condition. Education officials stated that because ineligible lenders are generally not subject to financial safety and soundness reviews by government agencies, Education lacks assurance that these lenders would be able to meet their financial obligations in the program. Second, because most eligible lender trustees also hold student loans in their own name and receive regular FFELP- related payments from the government for those loans, the federal government has additional sources from which to recover any repayments due the government on ineligible lenders’ loans that lose their guarantee. For example, lenders usually receive special allowance payments (SAP) and interest subsidiesfor the FFELP loans they hold. If the government were owed monies on an ineligible lender’s loans, it could recover those funds by withholding the SAP on the eligible lender trustee’s self-originated loans. Because ineligible lenders do not receive any payments directly from Education, it does not have an ability to collect liabilities by offset. Both eligible and ineligible lenders reported that trustee arrangements have had a positive effect on the student loan market. For example, both participant groups believe that market participation and loan availability are positively affected since trustee arrangements allow lenders that do not meet HEA’s eligible lender definition to make and hold loans, thus increasing the amount of loan capital available to students. These trustee arrangements allow ineligible lenders not only to originate loans, but also to purchase loans that other lenders have originated. This purchasing role—the primary role for many ineligible secondary markets—permits originating lenders to free up capital to make new loans to students. Eligible and ineligible lenders agree that participation by ineligible lenders increases competition among lenders and can, in turn, contribute to improved service and lower costs for student borrowers. Some ineligible lenders, however, believe that two factors—HEA requirements and the general evolution of financial markets—could affect their participation in the student loan market in the future. HEA currently states that a bank functioning as an eligible lender cannot have as its primary consumer credit function the making or holding of student loans. Education, in its FFELP regulations, interprets the act to mean that the lender’s FFELP loans—or in the case of a bank holding company, the FFELP loans of the company’s wholly owned subsidiaries as a group— cannot total more than one-half of the lender’s combined consumer credit loan portfolio, including home mortgages held by the lender or its subsidiaries. This provision is commonly known as the 50-percent rule. Strict application of this rule would require eligible lenders to include loans they hold as a trustee for an ineligible lender in this calculation, although the regulations do not specifically address these loans. Education officials told us that although Education has inconsistently applied the regulations in the past, it currently interprets the provision as excluding trustee-held loans in determining the lender’s primary consumer credit function, but has not formally promulgated its interpretation. Lenders have expressed confusion regarding application of the rule. For example, some ineligible lenders and other student loan market participants believe that the 50- percent rule applies to loans held as a trustee and thus believe that only a handful of large banks have sufficient consumer credit capacity to serve as a trustee. Similarly, while a few eligible lender trustee representatives interpreted the rule as not applying to trustee-held loans, some other representatives were not sure how the rule should be interpreted. Of those who were unsure, one trustee representative was not concerned because either calculation—including or excluding the loans—would place the bank in compliance with the act. Representatives of one bank incorrectly interpreted an exception to the 50-percent rule.On the other hand, a representative of another bank stated he had turned down eligible lender trustee business because the additional trustee-held loans would put the bank above its 50-percent allocation of student loans. A second factor that is perceived by ineligible lenders as having the potential to limit their participation in FFELP is evolution in the financial markets. Most ineligible lenders we interviewed stated that the number of available trustees has decreased as eligible lender banks have merged with each other in recent years. Recent mergers include First National Bank of Chicago and Bank One, Norwest and Wells Fargo, and Firstar Bank with Star Bank and Mercantile Bank (St. Louis); financial market analysts expect this consolidation activity to continue. Four ineligible lenders reported they had a difficult time obtaining their current trustee arrangement, one taking over a year to do so. Several others said that finding a trustee in the future would become more problematic. Ineligible lender representatives also expressed concern that because trustees also originate student loans for themselves, trustees may decide to end their eligible lender trustee services rather than continue to provide the mechanism for a competitor to do business. A few ineligible lenders stated they were uncomfortable with trustees having the final say as to which ineligible lenders can participate in the student loan market. Given these issues, some ineligible lenders believe their market presence could diminish in the future. Given current law and the federal regulations that govern FFELP, trustee arrangements between eligible and ineligible lenders serve an important role in enabling ineligible lenders to participate in FFELP, and in protecting the federal government’s investment in the program. The presence of an eligible lender from whom the government can recoup its financial losses is critical since Education has no direct oversight authority to ensure that ineligible lenders are operating their programs in accordance with HEA. Obtaining a trustee arrangement does not appear to be a widespread problem among ineligible lenders to date. However, if eligible lenders remain unclear and therefore continue to interpret the 50-percent rule as applying to loans they hold as a trustee, and if the financial industry continues to consolidate, the number of trustees could decline. This decline could, in turn, reduce competition in the student loan market. To clarify eligible lenders’ capacity to serve as trustees for ineligible lenders, the Secretary of Education should formally clarify Education’s interpretation of how the HEA provision prohibiting banks or their subsidiaries from holding FFELP loans that total more than one-half of their combined consumer credit loan portfolio applies to loans held by the trustee for an ineligible lender. Education agreed with our recommendation that it clarify its interpretation of how the HEA provision applies to loans held by the trustee for an ineligible lender. Education said that it has not yet decided which policy to establish nor how to formalize the decision once it is made. Education did not say when it expected to implement our recommendation. Education also provided technical comments, which we incorporated where appropriate. (See app. II for a copy of Education’s comments.) We are sending copies of this report to the Honorable Richard Riley, Secretary of Education; eligible lender trustees; ineligible lenders; and other interested parties. Copies will also be made available to others on request. If you or your staff have any questions about this report, please call me at (202) 512-7215. Other major contributors to this report are listed in appendix III. We focused our review on trustee arrangements created to allow ineligible lenders to originate or purchase student loans. To determine the number of existing trustee arrangements, we obtained data from Education’s lender database, including the names of the parties involved in each trustee arrangement, the lender identification numberassociated with each arrangement, and the outstanding loan balances. These data represent the best information available, and an Education official acknowledged that the list might be incomplete. To verify the parties involved in the identified arrangements and to determine the type of student loan transactions that made up the outstanding balance associated with a particular lender identification number, we surveyed eligible and ineligible lenders. For each lender identification number on the trustee arrangement list, we asked the lenders to verify whether the account was used for originations and/or purchases of student loans, securitizations, or some combination of these transactions. We also asked lenders to provide us with information on any additional identification numbers used for trustee arrangements. We conducted follow-up interviews as necessary to ensure that we received accurate information. In determining the total outstanding balance of loans associated with trustee arrangements, we included information only for arrangements that allow ineligible lenders to originate or purchase student loans and securitizations of these loans. Further, we included balance information only for the Student Loan Marketing Corporation’s privatized business and the ineligible lenders it owns, such as Nellie Mae Corporation, because these organizations must use a trustee to originate or purchase loans. To obtain detailed information on the identified trustee arrangements, the costs associated with them, and the key characteristics they shared, we interviewed Education officials, eligible lender trustee representatives, and representatives of ineligible secondary markets and other ineligible lenders. In addition, we conducted joint interviews with three sets of trustees and ineligible lenders involved in trustee arrangements and obtained a written response from another. Further, we obtained information through interviews about the impact of the 1998 amendments to HEA on the number of eligible lenders willing to serve as trustees and on the costs of trustee arrangements. To determine the protections offered the federal government by trustee arrangements, we identified and synthesized applicable legislation and regulations and reviewed sample documents for trustee arrangements. We also interviewed eligible and ineligible lenders, guaranty agencies, servicers, Education officials, and other student loan market participants on the responsibilities of an eligible lender trustee and the process Education uses to oversee market participants and to recoup any losses it may incur on defaulted loans. To determine the effects of trustee arrangements on participation in the student loan market and the availability of student loans, we interviewed eligible and ineligible lenders, secondary markets, guaranty agencies, servicers, Education officials, and other student loan market participants. We obtained the perspectives of participants on whether trustee arrangements have a positive or negative effect on lenders’ participation in the student loan market and identified past, current, and potential problems lenders faced in using trustee arrangements. Because of the competitive nature of the student loan market, information obtained from the lender community was sometimes general in nature. We also obtained Education officials’ interpretation of applicable laws and their views on the effects of past rulings regarding trustee arrangements. MajorManagementChallengesandProgramRisks:Departmentof Education(GAO/OCG-99-5, Jan. 1999). Risk-BasedCapital:RegulatoryandIndustryApproachestoCapitaland Risk(GAO/GGD-98-153, July 20, 1998). High-RiskSeries:StudentFinancialAid(GAO/HR-97-11, Feb. 1997). StudentLoanLenders:InformationontheActivitiesoftheFirst IndependentTrustCompany(GAO/HRD-90-183FS, Sept. 25, 1990). The first copy of each GAO report is free. Additional copies of reports are $2 each. A check or money order should be made out to the Superintendent of Documents. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. Ordersbymail: U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Ordersbyvisiting: Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Ordersbyphone: (202) 512-6000 fax: (202) 512-6061 TDD (202) 512-2537 Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. Web site: http://www.gao.gov/fraudnet/fraudnet.htm e-mail: fraudnet@gao.gov 1-800-424-5454 (automated answering system) | Pursuant to a congressional request, GAO provided information on student loan trustee arrangements, focusing on the: (1) number and cost of trustee arrangements and their shared characteristics; (2) benefits and protections afforded the federal government through use of trustee arrangements; and (3) effect of trustee arrangements on market participation and the availability of student loans. GAO noted that: (1) the Department of Education reports that approximately 125 trustee arrangements exist between 16 eligible lender trustees and 31 ineligible lenders for the purpose of originating or purchasing student loans; (2) these arrangements account for $25.3 billion in outstanding loans--approximately 19 percent of the outstanding balance of all Federal Family Education Loan Program (FFELP) loans as of December 1999; (3) costs of trustee arrangements fall into two categories--costs to initiate the arrangement and annual costs to maintain it; (4) ineligible lenders GAO interviewed said that the costs did not prohibit them from conducting business in the student loan market; (5) the amount charged by an eligible lender for its trustee services varied and was based on the volume of loans the ineligible lender was anticipated to originate and on the number and kind of other services the trustee provided; (6) both eligible and ineligible lenders reported little, if any, change in the availability of lenders to serve as trustees or the costs of these arrangements since 1998; (7) several characteristics were common among the trustee arrangements GAO reviewed, including the criteria used by trustees to evaluate ineligible lenders before they entered into trustee arrangements, the various elements of the trustee arrangement contracts, and the day-to-day interaction between the trustee and the ineligible lender; (8) trustee arrangements come with some protections to ensure the federal government's investment in FFELP is secure while allowing ineligible lenders to participate in the program; (9) most financial institutions that serve as eligible lender trustees are subject to federal oversight; (10) because most eligible lender trustees also hold student loans in their own name and receive regular FFELP-related payments from the government for those loans, the federal government has recourse for recovering any repayments due the government on ineligible lenders' loans that lose the federal guarantee; (11) Education officials stated that because ineligible lenders are generally not subject to financial safety and soundness reviews by government agencies, Education lacks assurance that these lenders would be able to meet their financial obligations in the program; and (12) both eligible and ineligible lenders said they believe that market participation and loan availability are positively affected by trustee arrangements which allow lenders to make and hold loans. |
CBP has developed a layered security strategy that provides multiple opportunities to mitigate threats and allows CBP to focus its limited resources on cargo containers that are the most likely to pose a risk to the United States. Risk management is a strategy called for by federal law and presidential directive and is meant to help policymakers and program officials most effectively mitigate risk while allocating limited resources under conditions of uncertainty. This layered security strategy is composed of different but complementary initiatives and programs, such as CSI and C-TPAT, which build on each other and work with other federal security programs, such as the Department of Energy’s (DOE) Megaports Initiative. This layered strategy attempts to address cargo container security comprehensively while ensuring that security attention is directed toward the highest-risk containers within the supply chain. CBP’s CSI program aims to identify and examine U.S.-bound cargo that pose a high risk of concealing WMD or other terrorist contraband by reviewing advanced cargo information sent by ocean cargo carriers. As of January 2008, CBP operated CSI in 58 foreign seaports, which, at the time, accounted for 86 percent of all U.S.-bound cargo containers. As part of the CSI program, CBP officers, usually stationed at foreign seaports, seek to identify high-risk U.S.-bound cargo containers by using information from cargo carriers as well as reviewing databases and interacting with host government officials. When requested by CBP, host government customs officials examine the high-risk container cargo by scanning it using various types of nonintrusive inspection (NII) equipment, such as large-scale X-ray machines, or by physically searching a container’s contents before it is sent to the United States. Initiated in November 2001, the C-TPAT program aims to secure the flow of goods bound for the United States by developing a voluntary antiterrorism partnership with stakeholders from the international trade community. To join C-TPAT, a company submits a security profile, which CBP compares to its minimum security requirements for the company’s trade sector. CBP then reviews the company’s compliance with customs laws and regulations and any violation history that might preclude the approval of benefits—which includes reduced scrutiny or expedited processing of the company’s shipments. CBP data show that from 2004 through 2006, C-TPAT members were responsible for importing about 30 percent of U.S.-bound cargo containers, specifically importing 29.5 percent of the 11.7 million oceangoing cargo containers off-loaded in the United States in the first 9 months of 2007. As of May 2008, there were over 8,400 C-TPAT members from the import trade community that had various roles in the supply chain. To more effectively implement the components of its layered security strategy, CBP has worked to promote international partnerships to enhance security so that high-risk cargo can be identified before it arrives in the United States. For the CSI program, CBP has negotiated and entered into nonbinding, reciprocal arrangements with foreign governments, specifying the placement of CBP officials at foreign seaports and the exchange of information between CBP and foreign customs administrations. These arrangements allow participating foreign governments the opportunity to place their customs officials at U.S. seaports and request inspection of cargo containers departing from the United States that are bound for their respective countries. CBP also works with other customs organizations to enhance international supply chain security. For example, CBP has taken a lead role in working with foreign customs administrations and the WCO to establish and implement international risk-based management principles and standards, similar to those used in the CSI and C-TPAT programs, to improve the ability of member customs administrations to increase the security of the global supply chain while facilitating international trade. The member countries of the WCO, including the United States, adopted such risk-based principles and standards through the WCO Framework of Standards to Secure and Facilitate Global Trade (commonly referred to as the SAFE Framework), in June 2005. To improve maritime container security, the SAFE Port Act was enacted in October 2006 and requires, among other things, that CBP conduct a pilot program to determine the feasibility of scanning 100 percent of U.S.-bound containers. It also specifies that the pilot should test integrated scanning systems that combine the use of radiation portal monitors and NII equipment, building upon CSI and the Megaports Initiative. To fulfill this and other requirements of the SAFE Port Act, CBP and DOE jointly announced the formation of SFI in December 2006. The first phase of SFI is the International Container Security project—commonly known as the SFI pilot program. The SFI pilot program tests the feasibility of 100 percent scanning of U.S.-bound container cargo at seven overseas seaports and involves the deployment of advanced cargo scanning equipment and an integrated examination system. The advanced cargo scanning equipment—NII and radiation detection equipment—produce data to indicate the presence of illicit nuclear and radiological material in containers. The integrated examination system then uses software to make this information available to CBP for analysis. According to CBP, it will review the scan data at the foreign seaport or at CBP’s National Targeting Center–Cargo (NTCC) in the United States. If the scanning equipment indicates a potential concern, both CSI and host government customs officials are to simultaneously receive an alert and the specific container is to be further inspected before it continues on to the United States. As shown in table 1, under the SFI pilot program, three SFI seaports are to scan 100 percent of U.S.-bound container cargo that passes through those seaports, while the other four seaports are to deploy scanning equipment in a more limited capacity. As required by the SAFE Port Act, CBP was to issue a report in April 2008 on the lessons learned from the SFI pilot program and the need and feasibility of expanding the 100 percent scanning system to other CSI seaports, among other things. As we prepared this statement, CBP had not yet issued this report. Every 6 months after the issuance of this report, CBP is to report on the status of full-scale deployment of the integrated scanning systems at foreign seaports to scan 100 percent of U.S.-bound cargo. We identified challenges in nine areas that are related to the continuation of the SFI pilot program and the longer-term 100 percent scanning requirement: (1) workforce planning, (2) the lack of information about host government cargo examination systems,(3) measuring performance outcomes, (4) undefined resource responsibilities for the cost and labor for implementation, (5) logistical feasibility for scanning equipment and processes, (6) technological issues, (7) the use and ownership of scanning data, (8) a perceived disparity between 100 percent scanning and the risk management approach of CBP’s international partners, and (9) potential requests for reciprocity from foreign governments. In our prior work examining the CSI and C-TPAT programs, we reported that CBP faced challenges identifying an appropriate number of positions for the programs and finding enough qualified people to fill these positions. For example, we reported in 2005 and again in 2008 that CBP’s human capital plan did not systematically determine the optimal number of officers needed at each CSI seaport to carry out duties that require an overseas presence (such as coordinating with host government officials or witnessing the examinations they conduct) as opposed to duties that could be performed remotely in the United States (such as reviewing databases). Determining optimal staffing levels is particularly important since CBP reports facing ongoing challenges identifying sufficient numbers of qualified employees to staff the program. For example, CBP officials reported that 9 qualified applicants applied for 40 permanent positions at CSI seaports. We also reported that according to CBP officials, to fill open CSI positions, officers have in some cases been deployed who have not received all required training. We recommended in April 2005 that CBP revise the CSI staffing model to consider (1) what functions need to be performed at CSI seaports and what can be performed in the United States, (2) the optimum levels of staff needed at CSI seaports to maximize the benefits of targeting and inspection activities in conjunction with host nation customs officials, and (3) the cost of locating targeters overseas at CSI seaports instead of in the United States. CBP agreed with our recommendation on CSI’s staffing model and said that modifications to the model would allow program objectives to be achieved in a more cost- effective manner. CBP said that it would evaluate the minimum level of staff needed at CSI seaports to maintain ongoing dialogue with host nation officials, as well as assess the staffing levels needed domestically to support CSI activities. However, as of January 2008, CBP’s human capital plan did not systematically make these determinations. The ability of the SFI pilot program—and by extension the 100 percent scanning requirement of the SAFE Port and 9/11 Acts—to operate effectively and enhance maritime container security depends, in part, on the success of CBP’s ability to manage and deploy staff in a way that ensures that critical security functions are performed. Under the CSI program, CBP operated and conducted cargo container scanning at 58 foreign seaports as of January 2008; however, given that additional scanning equipment will be used in the SFI pilot program, and fulfilling the 100 percent scanning requirement will naturally increase the number of containers to be scanned at the more than 700 seaports that ship cargo to the United States, the SFI pilot program and 100 percent scanning requirement will generate an increased quantity of scan data. According to European customs officials, for there to be value added in these additional scans, the scan data must be reviewed. Therefore, in implementing the 100 percent scanning requirement, CBP will face staffing challenges because more CBP officers will be required to review and analyze these data from participating seaports. As we reported in January 2008, CBP does not systematically collect information on CSI host governments’ examination equipment or processes. We noted that CBP must respect the sovereignty of countries participating in CSI and, therefore, cannot require that a country use specific scanning equipment or follow a set of prescribed examination practices. Thus, while CBP has set minimum technical criteria to evaluate the quality and performance of equipment being considered for use at domestic seaports, it has no comparable standards for scanning equipment used at foreign seaports. In addition, CBP officials stated that there are no plans to evaluate examination equipment at foreign seaports against the domestic criteria. CBP officials added, however, that the capabilities of scanning equipment are only one element for determining the effectiveness of examinations that take place at CSI seaports. It is better, in their view, to make assessments of the processes, personnel, and equipment that collectively constitute the host governments’ entire examination systems. However, in January 2008, we reported that CBP does not gather this type of information and recommended that CBP, in collaboration with host government officials, improve the information gathered at each CSI port by (1) establishing general guidelines and technical criteria regarding the minimal capability and operating procedures for an examination system that can provide a basis for determining the reliability of examinations and related CSI activities; (2) systematically collecting data for that purpose; and (3) analyzing the data against the guidelines and technical criteria to determine what, if any, mitigating actions or incentives CBP should take to help ensure the desired level of security. CBP partially concurred with this recommendation in terms of improving the information gathered about host governments’ examination systems. In particular, CBP agreed on the importance of an accepted examination process and noted that it continues to improve the information it gathers. CBP did not indicate that it would systematically pursue information on these host government examination systems. It did state that it was working through the WCO to address uniform technical standards for equipment. We reported that while CBP engaged with international trade groups to develop supply chain security requirements, these requirements do not specify particular equipment capabilities or examination practices. Both the SAFE Port and 9/11 Acts require DHS to develop technical and operational standards for scanning systems; therefore, the challenges that CSI has faced in obtaining information about host governments’ examination systems are relevant to the SFI pilot program and the 100 percent scanning requirement. However, as noted earlier in this statement, the United States cannot compel foreign governments to use specific equipment for the SFI pilot program or the 100 percent scanning requirement, thus challenging CBP’s ability to set and enforce standards. In addition, because CBP does not systematically collect information on the efficacy of host governments’ examinations systems, it lacks reasonable assurance that these examinations could reliably detect and identify WMD unless it implements our January 2008 recommendation to determine actions to take to ensure the desired level of security. This is particularly important since currently, under CSI, most high-risk cargo containers examined at international seaports are not re-examined upon arrival at domestic seaports. In our reviews of the CSI and C-TPAT programs, we identified challenges with CBP’s ability to measure program performance because of, among other things, the difficulty in determining whether these programs were achieving their desired result of increasing security for the United States. In the past, we and the Office of Management and Budget (OMB) have acknowledged the difficulty in developing outcome-based performance measures for programs that aim to deter or prevent specific behaviors. In the case of C-TPAT, we noted in our March 2005 and April 2008 reports that CBP had not developed a comprehensive set of performance measures and indicators for the programs, such as outcome-based measures, to monitor the status of program goals. A senior CBP official stated that developing these measures for C-TPAT, as well as other CBP programs, has been difficult because CBP lacks the data necessary to determine whether a program has prevented or deterred terrorist activity. We recommended that CBP complete the development of performance measures, to include outcome-based measures and performance targets, to track the program’s status in meeting its strategic goals. CBP agreed with our recommendation on developing performance measures, and had developed initial measures relating to membership, inspection percentages, and validation effectiveness. However, as we reported in April 2008, CBP had yet to develop measures that assess C-TPAT’s progress toward achieving its strategic goal to ensure that its members improve the security of their supply chains pursuant to C-TPAT security criteria. Given that, as with CSI and C-TPAT, the purpose of the SFI pilot program and the 100 percent scanning provision is to increase security for the United States, the same challenges related to defining and measuring performance could also apply to the SFI pilot program and the 100 percent scanning provision. Without outcome-based performance measures, it will be difficult for CBP and DHS managers and Congress to effectively provide program oversight and determine whether 100 percent scanning achieves the desired result—namely increased security for the United States. While CBP and DOE have purchased security equipment for foreign seaports participating in the SFI pilot program, it is unclear who will pay for additional resources—including increased staff, equipment, and infrastructure—and who will be responsible for operating and maintaining the equipment used for the statutory requirement to scan 100 percent of U.S.-bound cargo containers at foreign seaports. According to CBP, the average cost of initiating operations at CSI seaports was about $395,000 in 2004 and $227,000 in 2005. By comparison, CBP reported that it and DOE have spent approximately $60 million, collectively, to implement 100 percent scanning at the three foreign seaports fully participating in the SFI pilot program. The SAFE Port and 9/11 Acts did not require nor prohibit the federal government from bearing the cost of scanning 100 percent of U.S.-bound cargo containers. According to customs officials in the UK who participated in the SFI pilot program at the Port of Southampton, resource issues will inhibit their ability to implement permanently the 100 percent scanning requirement. For example, these customs officials commented that to accommodate the SFI pilot program at the Port of Southampton, existing customs staff had to be reallocated from other functions. The UK customs officials further stated that this reallocation was feasible for the 6-month pilot program, but it would not be feasible on a permanent basis. Similarly, a customs official from another country with whom we met told us that while his country does not scan 100 percent of exports, its customs service has increased its focus on examining more exported container cargo, and this shift has led to a 50 percent increase in personnel. European government officials expressed concerns regarding the cost of equipment to meet the 100 percent scanning requirement, as well as the cost of additional personnel necessary to operate the new scanning equipment, view and transmit the images to the United States, and resolve false alarms. Though CBP and DOE have provided the bulk of equipment and other infrastructure necessary to implement the SFI pilot program, they have also benefited from host nation officials and port operators willing to provide, to varying degrees, the resources associated with additional staffing, alarm response protocols, construction, and other infrastructure upgrades. However, according to CBP, there is no assurance that this kind of mutual support is either sustainable in the long term or exists in all countries or at all seaports that export goods to the United States. Logistical issues, such as crowded terminal facilities and the variety of transportation modes at terminals, could present challenges to the SFI pilot program and implementation of 100 percent scanning. Seaports may lack the space necessary to install additional scanning equipment needed to comply with the 100 percent scanning requirement. For example, we observed that scanning equipment at some seaports is located several miles away from where cargo containers are stored, which could add to the time and cost requirements for scanning these containers. Similarly, while some seaports have natural bottlenecks that allow for container scanning equipment to be placed such that all outgoing containers would have to pass through the equipment, not all seaports are so configured, and the potential exists for containers to be shipped to the United States without being scanned. Another potential logistical vulnerability is related to the transportation modes by which cargo containers arrive and pass through seaports. For example, cargo containers that arrive at a seaport by truck or rail are generally easier to isolate, whereas transshipment cargo containers—those moved from one vessel to another—are only available for scanning for a comparatively short time and may be more difficult to access. For example, UK customs officials stated that it was not possible to route transshipment containers that arrived by sea through the SFI equipment. As a consequence, the scanning of transshipment containers has not begun at the Port of Southampton. CBP and European customs officials evaluating the SFI pilot program stated that while the pilot has been comparatively successful at relatively lower-volume seaports, such as Southampton, implementing 100 percent scanning would be significantly more challenging at seaports with a higher volume of cargo container traffic or greater percentages of transshipment cargo containers. The SFI pilot program currently faces technology challenges, such as mechanical breakdowns of scanning equipment because of environmental factors, inadequate infrastructure for the transmission of electronic information, and difficulties in integrating different generations of scanning equipment. Environmental conditions at some sites can compromise the effectiveness of radiation detection equipment used in the SFI pilot program. For example, two of the three seaports fully participating in the SFI pilot program experienced weather-related mechanical breakdowns of scanning equipment. Specifically, at the Port of Southampton, a piece of radiation scanning equipment failed because of rainy conditions and had to be replaced, resulting in 2 weeks of diminished scanning capabilities. Additionally, Port Qasim in Pakistan has experienced difficulties with scanning equipment because of the extreme heat. Because of the range of climates at the more than 700 other international seaports that ship cargo to the United States, these types of technological challenges could be experienced elsewhere. The limited infrastructure at some foreign seaports poses a challenge to the installation and operation of radiation detection equipment, as well as to the electronic transmission of scan data to CBP officers in the United States. Many seaports are located in remote areas that often do not have access to reliable supplies of electricity or infrastructure needed to operate radiation portal monitors and associated communication systems. For example, at Port Salalah in Oman, a key challenge has been the cost of data transmission, because of low bandwidth communications infrastructure, to send data to the CBP officers who review the scans. Prior to SFI, the CSI office in Port Salalah already used transmission technology that cost annually about 10 times that of other SFI seaports. To participate in SFI, CBP originally planned to procure additional technology costing approximately $1.5 million each year to transmit the SFI data from Port Salalah. However, CBP was able to devise a lower-cost option that involved sharing communications infrastructure with existing CSI operations at the port because U.S.-bound container volume is relatively low in Oman. While CBP reported that this solution could keep data transmission costs down at other low-volume seaports, it is unclear whether this could be accomplished at higher-volume seaports. In addition to compatibility with existing infrastructure, SFI seaports have experienced compatibility issues with equipment from different generations. According to CBP, there are various manufacturers of equipment used at CSI seaports, and although the integration of equipment and technology at SFI pilot program seaports has generally been successful, it has not been without challenges. For example, at Port Salalah integration of a large number of new pieces of equipment by new vendors caused operational delays. The legislation that mandated the SFI pilot program and 100 percent scanning does not specify who will have the authority or responsibility to collect, maintain, disseminate, view, or analyze scan data collected on cargo containers bound for the United States. While the SAFE Port Act specifies that SFI pilot program scan data should be available for review by U.S. government officials, neither it nor the 9/11 Act establishes who is to be responsible for managing the data collected at foreign seaports. Other unresolved questions include ownership of data, how proprietary information is to be treated, and how privacy concerns are to be addressed. For example, officials from UK Customs stated that UK privacy legislation barred sharing information on cargo containers with CBP unless a specific risk was associated with the containers. To comply with UK laws, while still allowing CBP to obtain scan data on container cargo, UK Customs and CBP negotiated working practices to allow CBP to use its own handheld radiation scanning devices to determine whether cargo containers emitted radiation, but this was only for purposes of the SFI pilot program. According to the European Commission, for 100 percent scanning to go forward, the transfer of sensitive information would have to take place systematically, which would only be possible if a new international agreement between the United States and the European Union (EU) was enacted. In the absence of agreements with the host governments at the more than 700 seaports that ship cargo to the United States, access to data on the results of container scans could be difficult to ensure. Some of CBP’s international partners, including foreign customs services, port operators, trade groups, and international organizations, have stated that the 100 percent scanning requirement is inconsistent with widely accepted risk management principles, and some governments have expressed to DHS and Congress that 100 percent scanning is not consistent with these principles as contained in the SAFE Framework. Similarly, some European customs officials have told us that the 100 percent scanning requirement is in contrast to the risk-based strategy behind CSI and C-TPAT, and the WCO has stated that implementation of 100 percent scanning would be “tantamount to abandonment of risk management.” In addition, some of CBP’s international partners have stated that the requirement could potentially reduce security. For example, the European Commission noted that there has been no demonstration that 100 percent scanning is a better means for enhancing security than current risk-based methods. Further, CBP officials have told us that the 100 percent scanning requirement may be a disincentive for foreign countries or companies to adopt risk-based security initiatives, such as CSI, C-TPAT, or the SAFE Framework. Similarly, in April 2008, the Association of German Seaport Operators released a position paper that stated that implementing the 100 percent scanning requirement would undermine mutual, already achieved, security successes and hinder maritime security by depriving resources from areas that present a more significant threat and warrant closer scrutiny. Implementation of the 100 percent scanning requirement could result in calls for reciprocity of scanning activities from foreign officials and be viewed as a barrier to trade. European customs officials, as well as officials from the WCO, have objected to the unilateral nature of the 100 percent scanning requirement, noting that this requirement contrasts with prior multilateral efforts CBP has implemented. Similarly, the Association of German Port Operators published a position paper stating that the legislative requirement inherently ignores the international character of global maritime trade and is a classic example of an issue that should have been discussed with and passed by the legislative body of an international organization, such as the WCO. In its report on the SFI pilot program, the European Commission expressed concern that it would be difficult for EU customs administrations to implement a measure designed to protect the United States that would divert resources away from strengthening EU security. Customs officials from Europe, as well as members of the World Shipping Council and the Federation of European Private Port Operators, indicated that should implementation of the 100 percent scanning requirement be pursued, foreign governments could establish similar requirements for the United States, forcing U.S. export cargo containers to undergo scanning before being loaded at U.S. seaports. According to CBP officials, the SFI pilot program, as an extension of CSI, allows foreign officials to ask the United States to reciprocate and scan 100 percent of cargo containers bound for their respective countries. CBP officials told us that CBP does not have the personnel, equipment, or space to scan 100 percent of cargo containers being exported to other countries, should it be requested to do so. In addition to the issue of reciprocity, European and Asian government officials, as well as officials from the WCO, have stated that 100 percent scanning could constitute a barrier to trade. For example, the Association of German Seaport Operators stated that the 100 percent mandate would amount to an unfair nontariff trade barrier between the United States and foreign seaports. Similarly, senior officials from the European Commission expressed concern that a 100 percent scanning requirement placed on foreign seaports could disrupt the international trading system. Further, the WCO passed a unanimous resolution in December 2007, expressing concern that implementation of 100 percent scanning would be detrimental to world trade and could result in unreasonable delays, port congestion, and international trading difficulties. Mr. Chairman and members of the subcommittee, this concludes my prepared statement. We look forward to working with CBP and the Congress to track progress of the SFI pilot and to identify the way forward for supply chain security. I would be happy to respond to any questions you may have. For information about this testimony, please contact Stephen L. Caldwell, Director, Homeland Security and Justice Issues, at (202) 512-9610 or caldwells@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. This testimony was prepared under the direction of Christopher Conrad, Assistant Director. Other individuals making key contributions to this testimony include Frances Cook, Stephanie Fain, Emily Hanawalt, Valerie Kasindi, Robert Rivas, and Sally Williamson. Supply Chain Security: U.S. Customs and Border Protection Has Enhanced Its Partnership with Import Trade Sectors, but Challenges Remain in Verifying Security Practices. GAO-08-240. Washington, D.C.: April 25, 2008. Supply Chain Security: Examinations of High-Risk Cargo at Foreign Seaports Have Increased, but Improved Data Collection and Performance Measures Are Needed. GAO-08-187. Washington, D.C.: January 25, 2008. Maritime Security: The SAFE Port Act: Status and Implementation One Year Later. GAO-08-126T. Washington, D.C.: October 30, 2007. Maritime Security: One Year Later: A Progress Report on the SAFE Port Act. GAO-08-171T. Washington, D.C.: October 16, 2007. Maritime Security: The SAFE Port Act and Efforts to Secure Our Nation’s Seaports. GAO-08-86T. Washington, D.C.: October 4, 2007. Combating Nuclear Smuggling: Additional Actions Needed to Ensure Adequate Testing of Next Generation Radiation Detection Equipment. GAO-07-1247T. Washington, D.C.: September 18, 2007. Maritime Security: Observations on Selected Aspects of the SAFE Port Act. GAO-07-754T. Washington, D.C.: April 26, 2007. Customs Revenue: Customs and Border Protection Needs to Improve Workforce Planning and Accountability. GAO-07-529. Washington, D.C.: April 12, 2007. Cargo Container Inspections: Preliminary Observations on the Status of Efforts to Improve the Automated Targeting System. GAO-06-591T. Washington, D.C.: March 30, 2006. Combating Nuclear Smuggling: Efforts to Deploy Radiation Detection Equipment in the United States and in Other Countries. GAO-05-840T. Washington, D.C.: June 21, 2005. Homeland Security: Key Cargo Security Programs Can Be Improved. GAO-05-466T. Washington, D.C.: May 26, 2005. Maritime Security: Enhancements Made, but Implementation and Sustainability Remain Key Challenges. GAO-05-448T. Washington, D.C.: May 17, 2005. Container Security: A Flexible Staffing Model and Minimum Equipment Requirements Would Improve Overseas Targeting and Inspection Efforts. GAO-05-557. Washington, D.C.: April 26, 2005. Preventing Nuclear Smuggling: DOE Has Made Limited Progress in Installing Radiation Detection Equipment at Highest Priority Foreign Seaports. GAO-05-375. Washington, D.C.: March 31, 2005. Cargo Security: Partnership Program Grants Importers Reduced Scrutiny with Limited Assurance of Improved Security. GAO-05-404. Washington, D.C.: March 11, 2005. Homeland Security: Process for Reporting Lessons Learned from Seaport Exercises Needs Further Attention. GAO-05-170. Washington, D.C.: January 14, 2005. Port Security: Better Planning Needed to Develop and Operate Maritime Worker Identification Card Program. GAO-05-106. Washington, D.C.: December 10, 2004. Maritime Security: Substantial Work Remains to Translate New Planning Requirements into Effective Port Security. GAO-04-838. Washington, D.C.: June 30, 2004. Homeland Security: Summary of Challenges Faced in Targeting Oceangoing Cargo Containers for Inspection. GAO-04-557T. Washington, D.C.: March 31, 2004. Container Security: Expansion of Key Customs Programs Will Require Greater Attention to Critical Success Factors. GAO-03-770. Washington, D.C.: July 25, 2003. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | U.S. Customs and Border Protection (CBP), within the Department of Homeland Security (DHS), is responsible for preventing weapons of mass destruction from entering the United States in cargo containers that are shipped from more than 700 foreign seaports. The Security and Accountability for Every (SAFE) Port Act calls for testing the feasibility of scanning 100 percent of U.S.-bound cargo containers, and the Implementing Recommendations of the 9/11 Commission Act (9/11 Act) requires scanning 100 percent of U.S.-bound cargo containers by 2012. To fulfill these requirements, CBP created the Secure Freight Initiative (SFI) and has initiated a pilot program at seven seaports. This testimony discusses challenges related to the SFI pilot program and implementation of the requirement to scan 100 percent of U.S.-bound container cargo. This testimony is based on GAO products issued from July 2003 through April 2008 and ongoing work. To conduct this work, GAO reviewed reports from CBP and international partners on SFI and other container security programs, and interviewed CBP and foreign customs officials. GAO identified challenges in nine areas that are related to the continuation of the SFI pilot program and the longer-term 100 percent scanning requirement: (1) Workforce planning: The SFI pilot program could generate an increased quantity of scan data. Therefore, more CBP officers will be required to review and analyze data for participating seaports. (2) Host nation examination practices: The SAFE Port and 9/11 Acts require DHS to develop standards for the scanning systems, but CBP lacks information on host nation equipment and practices. (3) Measuring performance: CBP has had difficulties defining performance measures for its container security programs; therefore, it will be difficult to assess if 100 percent scanning achieves increased security. (4) Resource responsibilities: Neither the SAFE Port Act nor the 9/11 Act specifies whether the United States would bear the costs of implementing 100 percent scanning. (5) Logistics: Space constraints can require seaports to place scanning equipment miles from where cargo containers are stored, and some containers are only available for scanning for a short period of time and may be difficult to access. (6) Technology and infrastructure: Environmental conditions can damage equipment and cause delays, and infrastructure capacity and equipment compatibility have presented difficulties in the SFI pilot program. (7) Use and ownership of data: Legislation specifies that scan data should be available to CBP officials, but the data are often generated and collected by foreign seaports and, in some cases, will require international agreements for transfer to CBP officials. (8) Consistency with risk management: International partners state that 100 percent scanning is inconsistent with accepted risk management principles and diverts resources away from other security threats. (9) Reciprocity and trade concerns: Foreign governments could call for reciprocity of 100 percent scanning, requiring the United States to scan cargo containers, and some view this requirement as a barrier to trade. |
The Business Opportunity Development Reform Act of 1988 amended the Small Business Act to establish an annual governmentwide goal of awarding not less than 20 percent of prime contract dollars to small businesses. The Small Business Reauthorization Act of 1997 further amended the Small Business Act to increase this goal to not less than 23 percent. To help meet this goal, SBA annually establishes prime contract goals for various categories of small businesses for each federal agency. Although SBA is responsible for coordinating with executive branch agencies to move the federal government toward this mandated goal, agency heads are responsible for achieving the small business goals with their agencies. A 1978 report by the Senate Select Committee on Small Business noted that officials who were responsible for advocating for small business participation in federal government procurement often did not hold positions that were high enough in the agency structure to be effective. The 1978 amendment to the Small Business Act that established section 15(k)(3) addressed this issue by establishing a direct reporting relationship between the OSDBU director and the agency head or deputy head. The statute, as amended, specifies that the director would have supervisory authority over OSDBU staff, implement and execute the functions and duties under the relevant sections of the Small Business Act, and identify proposed solicitations that involved the bundling of contract requirements. (The Small Business Reauthorization Act of 1997 defines the bundling of contract requirements as the consolidation of two or more procurement requirements for goods or services previously provided or performed under separate smaller contracts into a solicitation of offers for a single contract that is likely to be unsuitable for award to a small business concern for various reasons.) Section 15(k) of the Small Business Act lists eight functions of OSDBU directors, as follows: Identifying proposed solicitations that involve significant bundling of contract requirements. Working with agency procurement officials to revise such proposed solicitations to increase the probability of participation by a small business. Facilitating the participation of small businesses as subcontractors if solicitations for bundled contracts are to be issued. Assisting small businesses in obtaining payments from an agency with which it has contracted. Helping small businesses acting as subcontractors to obtain payments from prime contractors. Making recommendations to contracting officers as to whether particular requirements should be set-aside for small businesses. Maintaining supervisory authority over OSDBU personnel. Cooperating and consulting on a regular basis with SBA in carrying out OSDBU functions and duties under the Small Business Act and assigning a small business technical advisor to each office with an SBA-appointed procurement center representative. (A procurement center representative is an SBA staff member assigned to federal buying activities with major contracting programs to carry out SBA policies and programs.) Acquisition Regulation Supplement for the Department of Defense. F purpose of our 2003 and 2010 surveys, we divided the procurement process into four steps: Acquisition planning involves developing an overall management strategy in for the procurement process for a potential contract. It takes place well advance of a contract’s award date and generally involves both a close partnership between the program and procurement offices and the involvement of other key stakeholders. Solicitation development to submit offers, or bids. is the process of preparing requests for vendors Proposal evaluation occurs after potential contractors submit proposa that outline how they will fulfill the solicitation requirements. Agency personnel evaluate proposals for award and contracting officers award the contract. Monitoring, also known as surveillance, helps to determine a c progress and identify any factors that may delay performance. To carry out the functions listed in section 15(k) of the Small Business OSDBU directors provide advice on small business matters and collaborate with the small business community. Some of the primary duties of OSDBU directors include advising agency leadership on small business matters; providing direction for developing and implementing policies and initiatives to help ensure that small businesses have the opportunity to compete for and receive a fair share of agency procurement; representing the agency at meetings, workgroups, and conferences related to small business; initiating and building partnerships with the small business community; providing agency acquisition and program personnel with leadership and oversight of education and train related to small business contracting; conducting reviews of small business programs; and serving as the agency liaison to SBA, including providing annual reports on agency activities, performance, and efforts to improve performance. OSDBU directors are not the only officials responsible for helping small businesses participate in federal procurement. At the agency level, the heads of procurement departments are responsible for implementing thesmall business programs at their agencies, including achieving program goals. Generally, staff within agency procurement departments who are assigned to work on small business issues (small business specialist coordinate with OSDBU directors on their agencies’ small business programs. s) We found that 9 of the 16 agencies we reviewed were in compliance with the Small Business Act’s requirement that OSDBU directors be responsible only to and report directly to the agency or deputy agency head (se e table 1). We determined that the remaining seven agencies were not in compliance. These agencies, which use various reporting structures, a were not in compliance in 2003, when we last assessed the reporting structure. The OSDBU directors at the compliant agencies cited benefit the reporting relationship. OSDBU directors at agencies that were not complying with section 15(k)(3) differed in their of reporting to the agency head or deputy head. Business Act. However, none of the legal arguments that the agencies raised caused us to revise our conclusions or recommendations. For example, the Department of the Interior stated that the Assistant to whom the OSDBU director reported was the agency head for acquisition matters, in accordance with the FAR. We responded that Interior’s designation of the Assistant Secretary as its “agency head” for procurement powers did not mean that the person thereby became its agency head for purposes of section 15(k)(3). Several agencies also commented on the effectiveness of their small business programs and reporting structure. During our interviews for this report, officials generally did not state that their agencies were complying with the requirement. Rather, they commented on how their current reporting structures were working. For example, officials at five agencies stated small business matters were not suffering as a result of the structure. At NASA, the Administrator is the head of the agency. OSDBU directors reported to the Deputy Secretary. For instance, at the Departments of the Air Force, Army, and Navy, the OSDBU directors reported to the Under Secretary for their respective departments, and the Under Secretary signed their performance appraisals. At Energy and HHS, the OSDBU directors reported to the Deputy Secretary, who also assessed their performance. In our 2003 report, we concluded that the Department of Education was not in compliance with section 15(k)(3). At that time, the OSDBU director did not report only to the Deputy Secretary but also reported to the Deputy Secretary’s Chief of Staff. Since then, the agency has changed its reporting structure to ensure that the OSDBU director is responsible only to the Deputy Secretary. The OSDBU director stated that she now had direct access to the Deputy Secretary, meeting with him for routine discussions about small business activities and related issues. In addition, the Deputy Secretary signed her performance appraisals. We also concluded in 2003 that EPA was not in compliance because the OSDBU director was not responsible only to the Administrator or Deputy Administrator. At that time, the OSDBU director told us she reported to the Deputy Chief of Staff, who also evaluated the OSDBU director’s performance. For this review, the OSDBU director told us that, while some matters were routed through the Deputy Chief of Staff, she ultimately reported to the Deputy Administrator. The Deputy Administrator signed her two most recent performance appraisals. As shown in table 1, seven agencies were not in compliance with section 15(k)(3). All of these agencies were also not in compliance in 2003. As in our prior report, we found that a variety of reporting structures were in place. OSDBU directors either reported to lower-level officials than the agency head or deputy or had delegated their OSDBU director responsibilities to officials who did not report to either the agency head or the deputy head. However, these arrangements do not meet the law’s intent and undermine the purpose of section 15(k)(3). To help ensure that the OSDBU’s responsibilities are effectively implemented, the act mandates that the OSDBU director—that is, the person actually carrying out the responsibilities—have direct access and be responsible only to the agency head or deputy. Appendix II provides details of the reporting arrangements at each agency. At the Departments of Commerce, the Interior, and Justice and the Social Security Administration (SSA), the OSDBU directors reported to officials at lower levels than the agency head or deputy head. For example, at the Department of Commerce, the organization chart showed that the OSDBU director reported to two lower-level officials—the Deputy Assistant Secretary for Administration and the Assistant Secretary for Administration. At the Department of the Interior, the OSDBU director reported to the Deputy Assistant Secretary for Budget, Finance, Performance, and Acquisition and to the Assistant Secretary, Policy, Management and Budget. At the Justice Department, OSDBU officials told us that the current reporting structure was the same as in 2003. The OSDBU is located within the Justice Management Division, with the director under the supervision of the Deputy Assistant Attorney General for Policy, Management and Planning. SSA also had the same reporting structure that we documented in our 2003 report, with the OSDBU director reporting to the Deputy Commissioner, Office of Budget, Finance and Management, who is one of nine deputy commissioners managing programs and operations. The designated OSDBU directors at the Departments of Agriculture, State, and the Treasury delegated their responsibilities to officials who did not directly report to either the Secretaries or Deputy Secretaries. These arrangements were the same as those we determined in 2003 were not in compliance with the Small Business Act. At these agencies, an Assistant Secretary who managed the agency’s administrative functions was designated as the statutory OSDBU director. The Assistant Secretaries then delegated nearly all of their OSDBU responsibilities to lower-ranking officials who reported directly to the Assistant Secretaries. The lower- ranking officials thus became the de facto OSDBU directors. At the Department of Agriculture, for example, the designated OSDBU director was the Assistant Secretary for Administration, who reported to the Secretary and Deputy Secretary. However, the Assistant Secretary had delegated nearly all of his OSDBU responsibilities to a lower-level official who did not have direct access to the agency head or deputy head. At the Department of State, the Assistant Secretary for Administration was the designated OSDBU director. The Assistant Secretary, who reported to one of the department’s two Deputy Secretaries on small business matters, had delegated his OSDBU responsibilities to the Operations Director for the OSDBU, who reported directly to him. At Treasury, the Assistant Secretary of the Treasury for Management/Chief Financial Officer/Chief Performance Officer was the designated OSDBU director. However, the Director of the Office of Small Business Programs, an official who did not directly report to either the Secretary or the Deputy Secretary, was responsible for the day-to-day management of Treasury’s small business programs. The Director of the Office of Small Business Programs told us she spent 100 percent of her time on small business issues. OSDBU directors’ opinions varied on the importance of reporting only to the agency head or deputy head. The OSDBU directors at the nine agencies that were complying with section 15(k)(3) cited positive elements to this reporting relationship. Five of the nine OSDBU directors stated that reporting to the agency head or deputy showed top-level support for small business efforts that sent a message to the rest of the agency. For example, one OSDBU director explained that reporting directly to the agency head or deputy head helped ensure that he was viewed as equal to other senior managers. He noted that this relationship was important because it allowed him to participate in senior management meetings where decisions were made. Another OSDBU director stated that she had a strong relationship with senior management and did not hesitate to invite senior leaders to participate in small business outreach events. She added that if she did not report to the agency head or deputy, she would lose this rapport with senior leadership. OSDBU directors at the seven agencies that were not complying with section 15(k)(3) differed on the importance of reporting to the agency head or deputy head. For example, two of these OSDBU directors thought that not reporting to the agency head or deputy was a problem. One director stated that reporting to the agency head or deputy could provide the OSDBU with more authority and enable it to collaborate more effectively with other offices. The other director noted that being too far down the reporting structure meant that she could not independently voice her opinion, especially when it differed from her supervisor’s. The OSDBU directors at the other five agencies did not see problems with the existing structure, stating that small business matters were not suffering as a result of the structure. For instance, one director stated that his agency’s structure worked well and that the agency’s small business initiatives were resulting in high marks on the SBA scorecard and effective relationships with other agency officials. This official noted that if he were to report directly to the Secretary or Deputy Secretary, small business efforts would compete against significant national foreign policy priorities. Another director stated that the OSDBU was getting all of the support it needed under the current reporting relationship. The director further explained that the office did not have a problem with resources and that he had a strong relationship with his supervisor. Additionally, he noted that any areas needing attention were communicated to higher management. Yet another director stated that his agency had a successful small business procurement program. He also cited accomplishments such as meeting small business contracting goals and increasing the number of small businesses with which the agency interacted. However, the Small Business Act requires that the OSDBU director have direct access to the agency head or deputy to help ensure that the OSDBU’s responsibilities are effectively implemented. As such, the statements made by the OSDBU directors at these five agencies do not justify their noncompliance with section 15(k)(3). SBA officials said the agency had also raised concerns about compliance with the reporting requirement during its surveillance reviews of federal agencies. These reviews are evaluations of small business contracting that assess (1) management of the small business programs, (2) compliance with regulations and published policies and procedures, (3) outreach programs focusing on small businesses, and (4) procurement documentation. When SBA finds that an agency does not have the required reporting relationship, it identifies this as a deficiency in the review report. Ongoing noncompliance with section 15(k)(3) undermines the intent of the act and may prevent some OSDBU directors from having direct access to top agency management. Given how long these agencies have not been in compliance with the requirement, at a minimum they have an obligation to explain their noncompliance to Congress and provide support for their need, if any, for greater statutory flexibility in establishing a reporting structure for their OSDBU director. Like the results of our 2003 survey, the responses of the 25 OSDBU directors we surveyed in 2010 indicated that they generally focused their procurement activities on the functions listed in section 15(k) of the Small Business Act. Most OSDBU directors reported they viewed five of the eight functions identified in section 15(k) as among their offices’ current duties, but the extent to which the individual OSDBUs carried out each activity varied. Directors who did not view a section 15(k) function as their responsibility generally reported that contracting, acquisition, or program staff performed it. Section 15(k) lists the functions of OSDBU directors but does not necessarily require them to personally carry out these activities themselves. Few OSDBU directors viewed non-15(k) procurement activities such as developing solicitations and evaluating proposals as roles of their OSDBU. For this report, we asked 25 OSDBU directors which of the responsibilities listed in the Small Business Act they saw as responsibilities of their offices. As shown in figure 1, at least 19 of the 25 OSDBU directors that we surveyed reported they viewed five of the eight functions identified in section 15(k) of the Small Business Act as current duties of their office. These five functions included (1) having supervisory authority over OSDBU staff, (2) three functions involving contract bundling (that is, the consolidation of two or more procurement requirements for goods or services previously provided under separate smaller contracts), and (3) assisting small businesses to obtain payments from agencies. Fewer OSDBU directors (10 to 18) viewed the remaining three functions— reviewing individual acquisitions for small business set-asides, assisting small businesses to obtain payments from prime contractors, and assigning a small business technical advisor to offices with an SBA representative—as their responsibilities. The data show little change from the responses to our 2003 survey. We also asked OSDBU directors about the extent to which they carried out six of the eight 15(k) functions, and their responses varied. Over half of those OSDBU directors who responded to the contract bundling questions reported that they carried out these functions to either a great or very great extent (see table 2). In contrast, six OSDBU directors reported having assisted small businesses to obtain payments from their agencies to a great or very great extent. Even fewer (two) reported having assisted small businesses to obtain payments from prime contractors to a great or very great extent. Of the 18 OSDBU directors who reported that reviewing or determining individual contracts that should be set aside for a small business was a function of their OSDBU, 13 stated they reviewed proposed small business set-asides for individual acquisitions in all or most cases. In their written comments, nine directors noted that they reviewed all acquisitions exceeding a certain amount for small business set-aside determinations. For instance, one of these directors explained that the agency had a regulation that prescribed policies, responsibilities, and procedures for clearing contracts over the simplified acquisition threshold ($150,000) that were not set aside or reserved for small business participation, including bundled contracts. We also asked the OSDBU directors to indicate the extent to which they cooperated and consulted with SBA in carrying out their responsibilities. Twenty-one directors reported that they cooperated and consulted with SBA to a great or very great extent. In their written comments, more than half of the directors noted they participated in SBA-sponsored activities and initiatives. For instance, 13 reported attending or sending staff to monthly SBA Small Business Procurement Advisory Council meetings. The number of OSDBU directors surveyed who did not view a section 15(k) function as their current responsibility varied, depending on the specific function. The number ranged from 1 who did not view maintaining supervisory authority over OSDBU personnel as a function to 11 who did not view assisting small businesses to obtain payments from prime contractors as a responsibility. In their written comments and follow-up interviews, the directors who did not view a section 15(k) function as their responsibility generally stated that contracting, acquisition, or program staff performed it. It was not clear from our survey results the extent to which OSDBU directors are involved in those functions carried out by other agency staff. Appendix III provides details on the agency personnel other than OSDBU staff who carry out certain section 15(k) functions. In 2010, a smaller number of OSDBU directors than in 2003 viewed additional procurement activities such as developing solicitations, evaluating proposals, developing factors for evaluating solicitations, and monitoring small businesses as roles of the OSDBU (see fig. 2). For example, 3 directors reported that developing proposed solicitations was a role of the OSDBU in 2010, compared with 9 directors in 2003. The majority of the 22 directors who reported they did not carry out this function commented that their agencies’ contracting offices performed this role. Of these 22 directors, 6 reported that the OSDBU played a collaborative role, such as reviewing solicitation language. Additionally, in 2010, 11 directors viewed developing evaluation factors for solicitations at their agencies as a role of the OSDBU, compared with 15 in 2003. However, of the 14 directors who said they did not perform the function in 2010, 4 reported having some involvement in the process. For example, 1 director commented that the OSDBU had provided examples of solicitation evaluation criteria for agency procurements. OSDBU directors reported they collaborated with acquisition officials and conducted outreach to small businesses to promote small business contracting. For example, nearly all of the 25 OSDBU directors we surveyed indicated they were involved in acquisition planning. At the seven agencies with major contracting activity, OSDBU officials told us that top agency leaders also participated in outreach events and issued agency policy statements supporting small business efforts. Agencies were held accountable for fostering small business contracting through SBA’s small business goals, and SBA had initiated efforts to identify promising practices OSDBUs can take to further small business contracting. Some OSDBU directors we surveyed reported that inadequate staffing levels and limited budgetary resources were challenges to carrying out their responsibilities. OSDBU directors use a variety of methods—including internal and external collaboration, outreach to small businesses, and oversight of agency small business contracting—to facilitate small business contracting. The OSDBU directors we surveyed and spoke with said that both internal and external collaboration were important to their efforts to promote small business contracting. Twenty-three of the 25 OSDBU directors surveyed viewed involvement in acquisition planning as a role or function of their OSDBUs. Of this number, 15 directors reported they carried out this function to a great or very great extent. For instance, OSDBU directors reported acquisition planning activities such as preparing annual forecasts of contracting opportunities, being a voting member of the agency’s senior procurement council, participating in contracting agreement and review board meetings, and reviewing and providing feedback on draft acquisition plans. Further, the OSDBU directors we interviewed at the seven agencies with major contracting activity all viewed relationships with acquisition staff as important in promoting small business contracting. For example, the OSDBU director at DLA told us that establishing relationships with acquisition management was the most important part of promoting small business contracting. In addition, the OSDBU director at HHS described small business contracting as a “three-legged stool,” with acquisition staff, program staff, and OSDBU staff working together to support it. Four of the seven OSDBU directors stressed the importance of working with acquisition officials from the start of a project rather than trying to integrate them after a project was under way. For instance, the director at HHS stated that the OSDBU spent a great deal of time on early acquisition planning to help ensure that small businesses received consideration throughout the decision-making process. Officials at the Air Force OSDBU also stated that early involvement gave the office time to review acquisitions and discuss small business involvement. These officials noted that reviewing proposed actions late in the process placed the OSDBU in a defensive position. Six of the seven directors we interviewed also told us they participated in acquisition teams as advocates for small business. For instance, Energy has established a monthly Advanced Planning Acquisition Team comprising OSDBU, procurement, acquisition, and SBA officials. The purpose of the team is to review proposed strategies for new and existing acquisitions to identify prime and subcontracting opportunities for small businesses. The OSDBU directors at both Army and DLA are members of acquisition review boards for contracts of more than $500 million and $20 million, respectively. The seven OSDBU directors we interviewed also stated that small business staff in the field, often referred to as small business specialists, reviewed proposed acquisitions for opportunities. For example, the Army OSDBU director told us that the OSDBU reviewed all acquisitions of more than $500 million but that these specialists reviewed all acquisitions over $150,000 that were not set-aside for small businesses and worked with acquisition staff to ensure that small businesses were considered. In addition, the OSDBU directors we interviewed collaborated with other federal agencies to maximize small business contracting. All seven participate in the Federal OSDBU Directors Interagency Council, an informal organization that meets to exchange information on initiatives and processes related to small businesses and outreach events that promote small business contracting. Among other things, the council seeks to identify best practices and share ideas and experiences among federal agencies and private industry to help leverage resources and develop solutions for promoting small business involvement. Nearly all of the OSDBU directors we surveyed saw outreach activities to small businesses as a function of their offices. For instance, 23 of the 25 OSDBU directors viewed hosting conferences for small businesses as one of their responsibilities, and 23 had hosted such an event in the last 2 years. More specifically, these 23 agencies had hosted an average of 20 conferences in the past 24 months. For instance, one OSDBU director reported the agency had sponsored conferences of varying sizes to address contractual requirements that program offices noticed or forecasted. The same director noted the conferences were typically conducted in a networking or “business fair” format, allowing vendors to engage program and contracting officials. In addition, most of the OSDBU directors surveyed (20 of 25) saw sponsoring training programs for small businesses as one of their responsibilities, and 18 had hosted such an event in the last 2 years. For example, OSDBU directors described sponsoring trainings for small businesses in specific socioeconomic groups, providing one-on-one trainings, and offering workshops focused on specific skills such as writing proposals and teaming with larger businesses. The seven OSDBU directors we interviewed provided examples of outreach to the small business community. For example, the OSDBU director at DLA said the OSDBU worked closely with the American Legion to promote contracting to small businesses owned by service-disabled veterans. According to the NASA OSDBU director, small business staff held over 100 outreach events in fiscal year 2010. Other OSDBU officials described their efforts to disseminate information to small businesses through Web sites. For instance, Navy’s OSDBU director stated the agency was working to standardize the Web site formats used by its various units for outreach to small businesses. He explained that OSDBU staff had reviewed the Web sites to determine whether a small business could access information easily, and found retrieving information on small businesses difficult because each of the sites was set up differently. The Air Force maintains a small business Web site that provides small businesses with information on the agency’s contracting opportunities. An OSDBU official stated the site was comprehensive and included the contact information for small business staff, long-range acquisition data, information on various outreach efforts, links to the Air Force’s quarterly newsletter, and articles on small business issues. OSDBUs also provide oversight of agencies’ small business contracting. The FAR requires annual assessments of the extent to which small businesses are receiving a fair share of federal procurements. We found that these assessments varied across the seven agencies we interviewed. For example, the OSDBU at Energy conducts quarterly reviews of program offices. Offices are rated using a color-coded system—green for reaching 95 percent of small business goals and yellow or red for lower percentages. The NASA OSDBU produces monthly reports on the agency’s individual space and research centers’ progress in meeting small business goals and collects data on the centers’ outreach and training efforts. The OSDBU follows up with individual centers that do not meet their goals. At Army, the OSDBU director tracks agency performance using Federal Procurement Data System—Next Generation data. She noted that if the data indicated decreases in small business contracting from prior-year trends, agency leadership would be informed. All seven agencies we interviewed also conducted program management reviews (PMR) or similar reviews to help ensure that small businesses were being considered for contracts and internal controls for ongoing contracts followed. During PMRs, officials review a sample of existing contracts to determine whether proper procedures and internal controls, including those related to small business, have been followed. The OSDBU directors at four agencies (Army, DLA, HHS, and NASA) reported that OSDBU and acquisition officials jointly conducted these reviews. At the Air Force, small business staff in field offices conducted the PMRs, and at Energy, procurement staff conducted them. The Navy OSDBU had replaced PMRs with procurement performance management assessments, which review contracting offices and commands on 11 categories, including involvement of small business in acquisition planning and procurement planning, marketing strategies and approaches, number of set-asides, and inclusion of small business clauses in all contracts. OSDBU officials at the seven agencies we contacted told us that top agency leaders supported small business contracting by participating in outreach events and issuing policy statements. For instance, these OSDBU directors cited the following examples of their agency heads’ participation in outreach events: The DLA Director took part in public forums such as conferences and emphasized small business contracting when interviewed for a magazine article. The NASA Administrator attended the agency’s second annual small business symposium and award ceremony to personally recognize the achievements of small businesses, while the Deputy Administrator handed out awards. In fiscal year 2010, top Army management attended five major outreach events, including the National Veterans Conference, an event that the OSDBU had hosted for 6 years. Top management also promoted small business contracting by issuing agency policy statements or memos. For instance: The Secretary of the Air Force issued a joint memo with the Chief of Staff in January 2009 that encouraged officials to “aggressively seek” small businesses owned by service-disabled veterans for Air Force contracts and encouraged using the OSDBU to help identify strategies and capable firms. The Secretary of Energy issued a memo in November 2009 asking all program offices to work with the OSDBU to promote prime contracting, subcontracting, and financial assistance opportunities for small businesses, including small disadvantaged, Historically Underutilized Business Zone (HUBZone), 8(a), women-owned, and service-disabled veteran-owned small businesses. In June 2009, the Secretary of the Navy issued a memo establishing the Department of the Navy Small Business Award to recognize individuals and teams who have made outstanding contributions in promoting competition and innovation in the Navy acquisition process. The Deputy Secretary at HHS issued a memo in January 2009, which stated that component heads were expected to provide their full support to the small business program at every juncture within the acquisition process. Agencies as a whole are held accountable for furthering small business contracting through SBA’s small business goaling program and scorecard. SBA produces an annual “goaling” report showing the percentage of contracting dollars awarded to small businesses. SBA’s goaling report for fiscal year 2009 shows that 22 percent of eligible contract dollars governmentwide were awarded to small businesses, an amount that was just short of the statutory requirement of 23 percent. As shown in table 3, the percentage awarded to small businesses by the agencies we surveyed ranged from 6 percent to 56 percent of eligible contract dollars. SBA sets annual goals for small business contracting at each agency, basing these goals on the agencies’ past performance, total spending, and purchases of goods and services from small businesses. These goals are set for the agency as a whole, not just for the OSDBU. As we reported in 2009, OSDBU officials at some agencies reported challenges with the goal- setting process, including limitations in negotiating and appealing their goals. At that time, agencies told GAO the goal-setting process was not a negotiation and that SBA did not factor in changes to agencies’ contracting priorities in setting the goals. Two of the seven agencies that we interviewed (Energy and HHS) said that meeting small business goals was difficult because the goods and services the agency purchased were not well suited for small business contracts. For example, 85 percent of Energy’s contracts are facility management contracts that have traditionally been awarded to large businesses and universities to manage operations at sites such as Los Alamos National Laboratory. According to SBA officials, the agency has revised the goal-setting process to make it more collaborative. SBA also assessed all of the agencies we surveyed using its annual small business scorecard to ensure greater accountability. According to SBA, the scorecard fulfills its statutory requirement to report to the President and Congress on achievements by federal agencies against their annual goals. The SBA scorecard evaluates factors such as goals met, progress shown, agency small business strategies, and top-level commitment to meeting goals for 24 agencies and offices and the government as a whole. In fiscal year 2009, SBA updated the scorecard and now assigns a letter grade to each agency. Eighty percent of an agency’s grade is based on its progress in meeting its prime contracting goal, 10 percent on its progress in meeting its subcontracting goal, and 10 percent on a performance rating assigned to the agency by a panel of OSDBU directors. The agencies submit reports on small business achievements that the panel uses to determine performance ratings. The scorecard does not specifically consider the performance of the OSDBU or its director, although the OSDBU director may be involved in some of the activities evaluated. All seven of the OSDBU directors we interviewed are held accountable for promoting small business contracting primarily through internal performance standards and appraisals. Performance standards identify goals and set objectives that are used as key indicators of achievement during annual or midpoint performance appraisals. For example, the NASA OSDBU director explained that, as part of his performance appraisal process, he was reviewed against measurable and achievable goals. These included developing a small business improvement plan for the agency with specific initiatives to meet its small business goals and implementing a training course for agency staff on a standard method to evaluate proposals submitted to the agency. To help OSDBUs improve their capabilities, SBA and others have initiated efforts to identify promising practices that OSDBUs can use to facilitate small business contracting. A White House Interagency Task Force on Federal Contracting Opportunities for Small Businesses, which was established April 2010, identified the need for best practices. The task force issued a report that identified challenges to small businesses such as inadequate training for agency staff. Among other things, it recommended that the executive branch facilitate the identification and rapid adoption of successful practices for increasing opportunities for small businesses. To implement this recommendation, the task force suggested that SBA (1) develop a Web site to share best practices and (2) organize an event for OSDBUs to present best practices for ensuring greater small business participation and catalog and publicize the results. According to SBA officials, they have taken steps to plan these efforts. SBA had already begun to highlight on its Web site best practices for making opportunities available to small businesses and has identified three to date. In addition, during the scorecard process, the panel of OSDBU directors identified best practices at 12 agencies, although these have not been published with this list on SBA’s Web site. The Federal OSDBU Directors Interagency Council also seeks to identify best practices and has identified some that focus on interactions between OSDBUs and small businesses. Most of the 25 OSDBU directors surveyed indicated that inadequate staffing levels and limited budgetary resources were challenges to carrying out their responsibilities to at least some extent (see table 4). These issues were also reported as challenges in 2003. Twenty agencies reported that inadequate staffing levels were a challenge to at least some extent in 2010, compared with 17 agencies in 2003. Seventeen agencies reported that limited budgetary resources were a challenge to at least some extent in 2010 and in 2003. Six agencies reported that inadequate staffing levels were a challenge to a great or very great extent, and seven agencies reported that limited budgetary resources were a challenge to a great or very great extent. In a follow-up interview, OSDBU officials at one agency explained that while they were able to perform their mission, inadequate staffing levels had resulted in increased staff workloads, longer work days, and the need to cross-train staff. They noted that in addition to the functions listed in 15(k) of the Small Business Act, they were also responsible for reviewing grant programs for small business opportunities. The OSDBU director at another agency indicated that because she had only one staff person, her office was unable to review the majority of procurement actions, including those involving contract bundling. She also noted that limited budgetary resources restricted the hiring of additional staff. Additionally, in follow- up interviews, four OSDBU directors noted that limited budgetary resources hindered their efforts to reach out to small businesses. Almost half of the OSDBU directors surveyed also reported that their lack of influence in the procurement process was a challenge to carrying out their responsibilities to at least some extent. Three agencies indicated that this issue represented a challenge to a great or very great extent. In a follow-up interview, OSDBU officials at one agency told us that better coordination was needed between the OSDBU and acquisition officials on issues related to small businesses. While the OSDBU participated in acquisition planning, the officials noted that their role in acquisition decisions was not clear. Another OSDBU director stated that as part of the acquisition approval process, he could disagree with a contract recommendation. However, the director noted that his decision could also be appealed and overruled by an acquisition executive, thus limiting the OSDBU’s influence. At the seven agencies with major contracting activity, trends in staffing and funding levels varied over the past 5 years. From fiscal year 2006 to 2010, OSDBU staffing generally decreased at Energy, increased slightly at DLA, the Navy, and NASA, and stayed the same at the Army and HHS (see table 5). At the Air Force, staffing fluctuated in the interim, but was the same in fiscal year 2010 as it was in fiscal year 2006. Additionally, from fiscal year 2006 to 2010 OSDBU funding generally decreased at the Air Force, the Army, and Energy and increased at DLA, HHS, the Navy, and NASA (see table 6). The OSDBU directors at the seven agencies we interviewed noted a relationship between changes in staffing and funding levels. For example, recent decreases in the Air Force and Energy ODSBU budgets were due to decreases in staffing. Air Force officials told us that the OSDBU had streamlined operations by eliminating some redundant contractor positions and converting some contractors into civilian positions. They noted that these changes had resulted in greater efficiencies and increased quality. Energy officials stated that since the change in administrations in 2008, three political appointees working at the Energy ODSBU had left, halving the number of staff responsible for the same workload. However, they noted that hiring interns and contractors had mitigated staffing constraints and the agency anticipated hiring additional OSDBU staff in the future. Recent increases in the ODSBU funding for Navy and DLA had resulted in increases in staffing. The Navy ODSBU director explained that in addition to funding training and outreach programs, the recent increases in funding were used to create an OSDBU industry analyst position, and two additional positions were being developed. The DLA OSDBU director noted that recent budget increases provided for an increase in staffing but funding for other expenses such as travel had not increased, causing her to have to turn down speaking and outreach events due to lack of funds. Six of the seven OSDBU directors we interviewed told us they would increase the breadth of the activities they currently performed if more resources were available. For example, the OSDBU directors at the Army and Air Force stated they would increase their outreach activities. Additionally, a few OSDBU directors noted that if more resources were available, they would perform additional analysis and offer more training. For example, the DLA OSDBU director stated she would analyze DLA commodity purchases and research trends, assist with more market research, provide more training to small business and other staff, and fill skill gaps within the office. With additional resources, the HHS OSDBU director stated she would increase internal training opportunities for acquisition program staff and provide online training for small businesses. Section 15(k)(3) of the Small Business Act seeks to help ensure that small business advocates within federal agencies have direct access to the highest levels of the agency. However, 7 of the 16 federal agencies that we reviewed were not in compliance with the act. Moreover, all of the agencies that are currently not in compliance were not in compliance in 2003 and have maintained the same or a similar reporting structure, even though at the time we recommended changing them. The OSDBU directors at the nine agencies that were in compliance with section 15(k)(3) cited positive benefits to the reporting relationship, while the OSDBU directors at the seven noncompliant agencies differed on the importance of reporting to the agency head or deputy head. Two thought that not reporting to the agency head or deputy head was a problem, while the other five asserted that the lack of a direct reporting relationship with the agency head had not adversely affected their efforts to advocate for small business contracting. However, we did not find that these arguments justified noncompliance with section 15(k)(3). Continued noncompliance with the requirement undermines the intent of the provision. If the agencies believe their reporting structures are sufficient to ensure that small business contracting receives attention from top management at federal agencies, at a minimum they have the obligation to explain their noncompliance to Congress and provide support for their views, including requesting any statutory flexibilities to permit exceptions as appropriate. One potential mechanism for making this information available to Congress would be through SBA’s annual scorecard process. The noncompliant agencies could include their rationale for their reporting structure in the annual reports that they submit to SBA, and SBA could include such information in its scorecard report to Congress. Given the ongoing requirement in the Small Business Act that OSDBU directors report to agency heads or deputy heads, we recommend that the heads of the Departments of Agriculture, Commerce, the Interior, Justice, State, and the Treasury and the Social Security Administration take steps as necessary to comply with the requirement or report to Congress on why they have not complied. Such information could be included in SBA’s annual scorecard report to Congress. Moreover, agencies that have not complied with the requirement could seek any statutory flexibilities or exceptions they believe may be appropriate. We sent a draft of this report to 26 agencies for their review and comment. Of the nine agencies that we concluded were complying with section 15(k)(3) of the Small Business Act, only the Department of Education provided written comments. In those comments, the agency noted that since our 2003 report it had taken definitive corrective steps to achieve and remain in compliance with all applicable requirements and stated that it was pleased with our recognition of its changes to the reporting structure. (See appendix IV for Education’s written comments.) None of the remaining eight agencies found to be in compliance with section 15(k)(3)—the Departments of the Air Force, the Army, Energy, and the Navy; DLA; EPA; HHS; and NASA—provided any comments on the draft report. Of the seven agencies that we concluded were not complying with section 15(k)(3), the Departments of Commerce, the Interior, Justice, State, and the Treasury and SSA provided written comments. These written comments are reproduced in appendixes V through X. The Department of Agriculture did not comment on the draft report. Of the six agencies that provided written comments, the Departments of Commerce, Justice, State, and the Treasury disagreed with our conclusion that their reporting relationships did not comply with section 15(k)(3) of the Small Business Act. The Social Security Administration agreed to revise its reporting structure, while the Department of the Interior stated it planned to evaluate our recommendation and options to resolve the issue. None of the agencies’ comments caused us to revise our conclusions or recommendations. The six agencies’ specific comments and our responses are summarized below. The Department of Commerce stated that the agency was in compliance with section 15(k)(3) because the OSDBU director reported directly to the Deputy Secretary on all legislative and policy issues and to the Chief Financial Officer and Assistant Secretary for Administration on administrative matters such as personnel and budget. The comment letter also cited the agency’s small business achievements, such as exceeding small business goals. As noted in the draft report, the OSDBU director stated that she reported to the Deputy Assistant Secretary for Administration for all small business matters and to the Assistant Secretary for Administration for administrative matters such as budget and personnel. Agency documents, such as the organization chart and the OSDBU director’s two most recent performance appraisals, confirmed this reporting relationship. Further, the OSDBU director stated she had never met the Secretary or previous Deputy Secretary. She met the new Acting Deputy Secretary in December 2010, but did not have a reporting relationship with her. Therefore, we did not revise our conclusion or recommendation. The Department of the Interior stated that it did not have comments on the draft report, but would be evaluating our recommendation and options to resolve the issue. The letter also indicated that the agency would report back to GAO once it had finalized its plans. The Department of Justice agreed that its OSDBU was located within the Justice Management Division. However, its comment letter stated that this arrangement was in place for administrative purposes, and that the OSDBU director reported directly to the Deputy Attorney General on matters of substance. The letter stated that through this organizational structure, the Deputy Attorney General ensured that small businesses were provided the maximum practicable opportunity to participate in contracting opportunities throughout the agency. The letter also commented that the current placement of the OSDBU allowed for the efficient management and implementation of the small business contracting programs that were vital to the agency in satisfying its mission. As we noted in the draft report, agency documentation such as the OSDBU director’s position description and performance appraisals indicated that the director reported to the Deputy Assistant Attorney General for Policy, Management and Planning. Further, the OSDBU director told us during our review that he had never met the Deputy Attorney General. Therefore, we did not revise our conclusion or recommendation. The Department of State disagreed with the report’s conclusions. The letter pointed out that the Assistant Secretary of State for Administration, who is the designated OSDBU director, reports to the Deputy Secretary concerning small business activities. This information was included in the draft report. However, we concluded that the agency was not in compliance with section 15(k)(3) because the Assistant Secretary had delegated his OSDBU responsibilities to a lower-level official who did not report to the Secretary or Deputy Secretary. Regarding this conclusion, State commented that section 15(k)(3) permitted the delegation of functions from the Assistant Secretary to the OSDBU Operations Director and stated that while OSDBU directors were responsible for implementing and executing the specific functions and duties assigned under sections 8 and 15 of the Small Business Act, section 15(k)(3) contained no requirement that the director personally perform any specific functions. The letter further commented that executive branch authority was typically exercised through delegation, with an agency’s basic authority being vested in the agency head and subsequently redelegated. It cited the case of Fleming v. Mohawk Wrecking & Lumber Co., 331 U.S. 121 (1947) as an example in which the authority to redelegate is implied. However, as we stated in our 2003 report, the Fleming case recognizes that the delegation of authority may be withheld by implication, and we believe section 15(k)(3) does exactly that. As explained in this report, to ensure that the OSDBU responsibilities are effectively implemented, the statute mandates that the OSDBU director (i.e., the person carrying out the responsibilities) have immediate access and be responsible only to the agency head or deputy. The legislative history reveals that the reason for this requirement is that Congress believed that agency officials responsible for promoting procurements for small and disadvantaged businesses were often too far down the chain of command to be effective. The reporting requirement of section 15(k)(3) was intended to remedy this situation. The Department of State’s letter also highlighted its small and disadvantaged business goal achievements and commented that reorganizing the OSDBU so that its Operations Director reported directly to the Secretary or her Deputy would decrease efficiency. The letter concluded by stating that the OSDBU Operations Director currently had direct access whenever necessary to decision makers in the Department programs that utilize small and disadvantaged businesses. As indicated in the recommendation, agencies that believe that they should have greater flexibility should pursue this with Congress. The Department of the Treasury disagreed with the report’s conclusion and commented that the statute did not require that the OSDBU director be an official assigned to small business issues full time. The letter stated that the Assistant Secretary for Management, in the capacity of OSDBU director, had a direct reporting relationship to the Deputy Secretary of the Treasury and provided oversight and direction to the Office of Small Business Programs as well as to bureau heads and their procurement officials in executing the agency’s responsibilities under the Small Business Act. The letter noted that the Assistant Secretary exerted considerable influence over acquisition and budget officials at all levels in regards to attaining small business goals. Finally, the letter commented that the OSDBU director assigning day-to-day operations pertaining to small business functions to subordinate officials did not establish any such other officials as the “de facto OSDBU director.” As previously discussed, we believe that section 15(k)(3) includes an implied prohibition against delegating the OSDBU director’s authority. As a result, we did not revise our conclusion or recommendation. The Social Security Administration’s letter did not provide comments on the draft report but indicated that the agency had reevaluated the reporting relationship in light of our draft. The letter stated that in the future the OSDBU director would be reporting to the Deputy Commissioner, the deputy agency head for SSA. Of the remaining 10 agencies that received a copy of the draft report, 9 agencies—the Departments of Homeland Security, Housing and Urban Development, Labor, Transportation, and Veterans Affairs; the General Services Administration; the Office of the Secretary of Defense; the Office of Personnel Management; and the U.S. Agency for International Development—did not provide any comments on the report. The Small Business Administration provided technical comments, which have been incorporated as appropriate. We also provided a copy of our survey results, which will be published in a separate product (GAO-11-436SP), to the 25 agencies we surveyed and SBA for their review and comment. Only one agency had a comment. The Department of State provided a comment via email stating that the survey questions focused on the daily duties of the OSDBU, but did not provide an opportunity to explain the delegation of these duties to the OSDBU Operations Director. We note that while the survey was designed to capture the OSDBU director’s activities (and not the delegation of such duties per se), there were numerous open-ended questions that allowed respondents to add explanations or qualifications to their responses. State also had the opportunity to explain the delegation of duties during our interviews with OSDBU officials. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies of this report to the Ranking Member of the Senate Committee on Small Business and Entrepreneurship, the Chairman and Ranking Member of the House Committee on Small Business, and other interested congressional committees. The report will also be available at no charge on the GAO Web site at http://www.gao.gov. If you or your office have any questions about this report, please contact me at (202) 512-8678 or shearw@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix XI. We reviewed Office of Small and Disadvantaged Business Utilization (OSDBU) practices for carrying out the requirements of the Small Business Act at federal agencies with major contracting activity. Specifically, we (1) assessed whether the director reported directly to the agency head or the deputy head; (2) determined the functions conducted by the OSDBUs; and (3) examined the actions taken by the OSDBUs and other officials at the various agencies to further small business contracting opportunities and the effects of funding and staff levels on these efforts. To determine which federal agencies engage in major contracting activity, we reviewed fiscal year 2009 data from the Federal Procurement Data System-Next Generation. (These were the most recent data available at the time of our review.) This dataset includes information on 20 defense agencies and 62 civilian agencies. Using these data, we determined that seven agencies each procured more than $15 billion in goods and services in fiscal year 2009. Table 7 shows the seven agencies with major contracting activity as well as the other agencies covered in our review. To assess whether the OSDBU director reports directly to the agency head or the deputy head as required by section 15(k)(3) of the Small Business Act, we focused on the seven agencies with major contracting activity and nine additional agencies that we reported in September 2003 were not complying with this requirement. These nine agencies were the Departments of Agriculture, Commerce, Education, the Interior, Justice, State, and the Treasury; the Environmental Protection Agency; and the Social Security Administration. We considered agencies to be in compliance if the designated OSDBU directors exercised the OSDBU responsibilities and reported directly to and were responsible only to the agency head or the agency head’s deputy. To determine compliance, we reviewed organization charts to identify where the OSDBU was located in relation to the agency head or deputy head; OSDBU directors’ performance appraisals for the previous 2 years to identify the agency official who evaluated the OSDBU director’s performance; the most recent position description of the OSDBU director to identify the OSDBU director’s supervisor; and various other agency documents, such as reports and memoranda discussing the agency’s small business programs. We also interviewed the designated OSDBU directors at each agency to identify the official(s) they had reported to during the past year and asked them to provide information characterizing the reporting relationship, such as the extent to which small business issues were discussed. In addition, we reviewed and analyzed section 15(k)(3). To obtain information on the functions conducted by the OSDBU, actions taken by the OSDBU to further small business contracting opportunities, and the effects of funding and staff levels on these efforts, we surveyed the OSDBU directors at 25 federal agencies using a Web-based survey. The survey was similar to one we administered in 2003 and asked the OSDBU directors about their roles and functions in three areas: participation in the agency procurement process, facilitation of small business participation in agency contracting, and interaction with the Small Business Administration (SBA). The survey questions covered the OSDBU functions listed in 15(k) of the Small Business Act as well as additional functions the OSDBUs might perform. In addition, the survey asked OSDBU directors about challenges—including limited budgetary resources and lack of adequate staffing levels—they face in carrying out their responsibilities. We selected 25 agencies to include in our survey of OSDBU directors. These agencies included all 20 civilian agencies that procured more than $800 million in goods and services in fiscal year 2009, as well as the Department of Defense (DOD)—Office of the Secretary; the Departments of the Air Force, Army, and Navy; and the Defense Logistics Agency (DLA). The 20 agencies were responsible for more than 98 percent of civilian agency obligations in fiscal year 2009. We selected the Air Force, Army, Navy, and DLA because they were the four components within DOD that procured the most goods and services in fiscal year 2009; together, they were responsible for more than 90 percent of DOD’s obligations. The 25 agencies we selected included all 24 agencies we surveyed in 2003 and the Department of Homeland Security. To have comparable data with the 2003 survey of OSDBU directors, our survey instrument listed the same questions and response choices as the 2003 survey. Updates to the 2003 survey were limited to making minor word changes, reordering several questions, and deleting several questions that were no longer relevant. We obtained input from GAO experts on survey design. We also pretested the survey instrument with two OSDBU directors to help ensure that the questions were still applicable and would be correctly interpreted by respondents. Agency officials, including the OSDBU directors, were notified about the survey before it was launched on November 1, 2010. OSDBU directors were asked to complete the survey by November 22, 2010, and by December 29, 2010, we had a 100 percent response rate. From January to March 2011, we conducted follow-up with 22 OSDBU directors who answered that one or more of the functions listed in 15(k) of the Small Business Act was not a function of their office or who did not provide responses to these questions. The purpose of the follow-up was to determine which office, if not the OSDBU, carried out these functions at their agency or to collect answers from OSDBU directors who did not provide them initially. To do this, we conducted interviews with 11 ODSBU directors and corresponded via e-mail with 11 others. Based on this follow-up, we changed 16 of the original survey answers related to whether the OSDBU director viewed a section 15(k) responsibility as an OSDBU role. Answers were changed only if at least one of the following criteria were satisfied: the OSDBU directors explicitly stated that they wished to change their answer and provided an explanation for the change; the director misunderstood the question; or the director provided a response to an initially unanswered question. Additionally, we asked 11 of the 22 OSDBU directors with whom we were conducting follow-up for further explanation of challenges they had identified as affecting their office to a great or very great extent. Two of the respondents requested that we change their responses to several challenges due to initially misunderstanding the question. We agreed and these adjustments are reported in our findings. Also, as necessary, we followed-up with OSDBU directors to clarify their responses to our open- ended questions. While the OSDBU directors at 25 agencies were asked to participate in the survey and the survey results are therefore not subject to sampling errors, not all respondents answered every question. Nonresponse, including item nonresponse, and the practical difficulties of conducting any survey, may introduce error in survey results. We took steps to minimize such errors by conducting follow-up discussions with respondents who failed to answer specific questions, and by checking and verifying survey responses and analysis. The survey contained closed-ended questions that we asked OSDBU directors to answer by selecting from a finite number of response categories. For example, some questions asked OSDBU directors to select “Yes—an OSDBU role or function” or “No—not an OSDBU role or function” based on if a certain function was performed by their office. Other questions asked OSDBU directors to identify the extent to which they performed a certain function or the extent to which a certain factor was a challenge in carrying out the responsibilities of their office. Our analysis involved reviewing the frequency of responses to a given question using aggregate survey data. In the report, there are instances in which we identify all of the responses and other instances in which we identify the most common response. This report does not contain all the results from the survey; the survey and a more complete tabulation of the results are provided in a supplement to this report (GAO-11-436SP). The survey also contained open-ended questions that asked OSDBU directors to provide a narrative response. Most of these open-ended questions provided respondents the opportunity to explain answers provided to close-ended questions. For example, some closed-ended questions asked the respondents if a certain activity was a function of their office and the subsequent open-ended question asked them to elaborate on which office carries out this role or function if they had responded “No— not an OSDBU role or function” to the prior question. We used these open- ended responses to provide context to close-ended questions, and some of these narrative responses were included in our findings. To examine the actions taken by the OSDBU and other agency officials to further small business contracting opportunities and the effects of funding and staff levels, we interviewed agency officials and reviewed documents at the seven agencies with major contracting activity. Also, we reviewed agency documentation, such as policy statements issued by agency leadership on ODSBU practices or small business efforts, small business manuals or strategic plans, and budget and staffing documentation. We interviewed the OSDBU directors at the seven agencies on the actions they and other agency officials had taken to further small business contracting; methods used by SBA and the agency to hold officials accountable for furthering small business contracting; and challenges that affected these efforts, such as funding and staffing levels. Furthermore, we reviewed documentation and data related to the SBA small business goaling program and spoke with SBA officials about this program and reviewed OSDBU council documentation and spoke with council leadership. We conducted our work from June 2010 to May 2011 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Appendix II: Re Agen 15(k)(3) As discussed in the body of our report, seven agencies were not in compliance with section 15(k)(3) of the Small Business Act, which requires that the director of the Office of Small and Disadvantaged Business Utilization (OSDBU) be responsible only to and report directly to the agency head or deputy head. We found that a variety of reporting structures were in place. OSDBU directors either reported to lower-level officials than the agency head or deputy or had delegated their OSDBU director responsibilities to officials who did not report to either the agency head or the deputy head. At the Departments of Commerce, the Interior, and Justice and the Social Security Administration (SSA), the OSDBU directors reported to officials at lower levels than the agency head or deputy head. A document outlining the organization structure of the Department of Commerce’s Office of the Chief Financial Officer and Assistant Secretary for Administration stated that the OSDBU director reported to the Deputy Secretary on matters of policy and to the Assistant Secretary and Deputy Assistant Secretary for Administration on administrative matters. However, the OSDBU director at Commerce stated that she reported to the Deputy Assistant Secretary for Administration for all small business matters and to the Assistant Secretary for Administration for administrative matters such as budget and personnel. The organization chart also showed a direct link from the OSDBU to the Deputy Assistant Secretary for Administration and the Assistant Secretary for Administration (see fig. 3). In fiscal year 2009, the Assistant Secretary for Administration evaluated the OSDBU director’s performance, while the Deputy Assistant Secretary for Administration signed her performance appraisal in fiscal year 2010. The OSDBU director stated that she had never met the Secretary or previous Deputy Secretary and only met the new Acting Deputy Secretary in December 2010. The OSDBU director at the Department of the Interior reported to the Deputy Assistant Secretary for Budget, Finance, Performance, and Acquisition and to the Assistant Secretary, Policy, Management and Budget (see fig. 4). The organization chart had a notation that the OSDBU director reported to the Secretary and received administrative support from the Deputy Assistant Secretary for Budget, Finance, Performance, and Acquisition, who in turn reported to the Assistant Secretary, Policy, Management and Budget. However, the OSDBU director told us he had never met with the Secretary or Deputy Secretary and they had not provided direct input into his performance appraisals. Instead, the OSDBU director told us he met frequently with the Deputy Assistant Secretary for Budget, Finance, Performance, and Acquisition on administrative and small business matters. The director’s position description indicated that the director reports to the Deputy Assistant Secretary. His two most recent performance appraisals were signed by the Deputy Assistant Secretary and the Assistant Secretary for Policy, Management and Budget, respectively. As shown in figure 5, the OSDBU director at Justice reported to the Deputy Assistant Attorney General for Policy, Management and Planning. OSDBU officials told us the current reporting structure was the same structure that was in place in 2003. The organization chart showed that the OSDBU was located within the Justice Management Division, with the director under the supervision of the Deputy Assistant Attorney General for Policy, Management and Planning. The Justice Management Division was headed by the Assistant Attorney General for Administration, who reports to the Deputy Attorney General. While the organization chart showed that the OSDBU was located within the Justice Management Division for administrative purposes, the OSDBU director’s position description listed his immediate supervisor as the Deputy Assistant Attorney General for Policy, Management and Planning. The Deputy Assistant Attorney General signed the OSDBU director’s two most recent performance appraisals. The OSDBU director at SSA reported to the Deputy Commissioner, Office of Budget, Finance and Management, who was one of nine deputy commissioners managing various programs and operations (see fig. 6). Both the organization chart and the OSDBU director’s position description confirmed this reporting relationship. The OSDBU director told us that he reported to this Deputy Commissioner for small business matters. Additionally, the director’s most recent performance appraisal was signed by the Deputy Commissioner, Office of Budget, Finance and Management. The director confirmed that this same structure was in place in 2003. The OSDBU director also told us he had never met either the Commissioner or Deputy Commissioner of SSA. As we found in our 2003 report, the designated OSDBU directors at the Departments of Agriculture, State, and the Treasury delegated their responsibilities to officials who did not directly report to either the Secretaries or Deputy Secretaries. At these agencies, the Assistant Secretary who managed the agency’s administrative functions was designated as the statutory OSDBU director. The Assistant Secretaries then delegated nearly all of their OSDBU responsibilities to lower-ranking officials who reported directly to the Assistant Secretaries. The lower- ranking officials thus became the de facto OSDBU directors. The designated OSDBU director at the Department of Agriculture was the Assistant Secretary for Administration, who reported to the Secretary and Deputy Secretary. However, the Assistant Secretary had delegated nearly all of his OSDBU responsibilities to a lower-level official (see fig. 7). This structure was the same one that we determined in 2003 was not in compliance with the Small Business Act. The delegated OSDBU director told us that he did not report to the Secretary or Deputy Secretary on matters involving policy, budget, and personnel. The agency’s organization chart confirmed that the delegated OSDBU director reported to the Assistant Secretary for Administration. The Assistant Secretary for Administration and the Secretary signed the director’s most recent performance appraisal. Other evidence showed that the delegated OSDBU director carried out the day-to-day implementation of the agency’s OSDBU. The delegated director told us he handled the day-to-day duties and functions of Agriculture’s OSDBU and that he spent 100 percent of his time on OSDBU duties and functions. Moreover, his position description indicated that he was the official responsible for carrying out the duties and functions listed under section 15(k). The position description stated, among other things, that the delegated director was responsible for (1) establishing short- and long- range program objectives, time schedules, and courses of action for the accomplishment of small business goals; (2) formulating, recommending, and implementing broad policies and procedures that provide the structural framework for all OSDBU functions; and (3) keeping abreast of all OSDBU activities and initiating any corrective actions deemed necessary. In contrast, OSDBU officials stated that the Assistant Secretary for Administration spent time on OSDBU activities on an as-needed basis but estimated that it averaged about 3 to 4 hours per week. The Assistant Secretary for Administration was the designated OSDBU director at the State Department. The Assistant Secretary, who reported to one of the department’s two Deputy Secretaries on small business matters, had delegated his OSDBU responsibilities to the Operations Director for the OSDBU (see fig. 8). In fiscal year 2010, the Operations Director’s performance appraisal was signed by the Acting Assistant Secretary for Administration. The position description for the Operations Director indicated that he carried out the functions of the OSDBU director. For example, it showed that his duties included (1) providing overall direction for policies and programs governing the agency’s procurement and financial assistance actions in accordance with the Small Business Act and (2) developing small business goals. The Assistant Secretary of the Treasury for Management/Chief Financial Officer/Chief Performance Officer was the designated OSDBU director. He stated that he was responsible for meeting the agency’s small business goals and interacted with the Secretary and Deputy Secretary regularly, including providing updates on small business matters. However, the Director of the Office of Small Business Programs—an official who did not directly report to either the Secretary or the Deputy Secretary—was responsible for day-to-day management of Treasury’s small business programs. According to Treasury, the Director of the Office of Small Business Programs reported to the Director of the Office of Minority and Women Inclusion, who in turn reported to the Assistant Secretary (see fig. 9). The Director of the Office of Small Business Programs stated that she spent 100 percent of her time on small business matters, which included all of the functions described in section 15(k) of the Small Business Act. Her position description confirmed this statement, indicating that her responsibilities included (1) planning, developing, issuing, and providing overall direction for policies and programs governing Treasury procurement and financial assistance action in accordance with the Small Business Act and (2) directing Treasury’s annual goal-setting process. The number of Office of Small and Disadvantaged Business Utilization (OSDBU) directors surveyed who did not view a section 15(k) function as their current responsibility varied, depending on the specific function. The number ranged from 1 who did not view maintaining supervisory authority over OSDBU personnel as a function to 11 who did not view assisting small businesses to obtain payments from prime contractors as a responsibility. In their written comments and follow-up interviews, the directors who did not view a section 15(k) function as their responsibility generally stated that contracting, acquisition, or program staff performed it. The OSDBU director at the Social Security Administration (SSA) reported that maintaining supervisory authority over OSDBU personnel was not a function of his office because he did not have staff. The OSDBU director at the Office of Personnel Management (OPM) reported that attempting to identify proposed solicitations that involved bundling of contract requirements was not a function of his office. He commented that the contracting office within his agency performed this function. The OSDBU director at SSA reported that working with agency acquisition officials to revise procurement strategies for bundled contract requirements was not a function of his office. He commented that no office carried out this role. Rather, he noted that when contract bundling was identified, the acquisition official prepared a bundling justification for the head of the procuring activity to sign. In a follow-up interview, he clarified that nothing in the agency’s policies required coordination with the OSDBU on contract bundling or gave the OSDBU the opportunity to revise procurement strategies. Five OSDBU directors reported that facilitating small businesses’ participation as subcontractors to bundled contracts was not a function of their office. At the Department of Agriculture and OPM, the OSDBU directors commented that their agencies had not bundled any contracts. The Agriculture OSDBU director also stated that his office evaluates proposed contract actions to ensure that there are no bundled contracts. The OSDBU director at the Office of the Secretary of Defense reported that this function was generally performed by the contracting offices at the Department’s individual components, such as the Army, Navy, Air Force, and Defense Logistics Agency (DLA). In a follow-up interview, the OSDBU director at SSA stated that his role was limited by agency policy to reviewing subcontracting plans to ensure that certain clauses required by the Federal Acquisition Regulation were included. He noted that he would need additional resources to advocate for increased small business participation in subcontracting. The OSDBU director at the Department of Commerce explained that the OSDBU did not have the staff to review subcontracting plans. Six OSDBU directors reported that assisting small businesses to obtain payments from their agencies was not a function of their office. All six directors—Departments of the Air Force, Education, and the Interior; the Environmental Protection Agency (EPA); Office of the Secretary of Defense; and SSA—reported that payment issues were addressed by agency officials in the contracting, acquisition, or program offices. Seven OSDBU directors reported that determining a small business set- aside for an individual contract was not a function of their OSDBU. Five of these directors (Departments of the Army, Education, and Housing and Urban Development (HUD); the Office of the Secretary of Defense; and OPM) commented that their agency’s contracting or program offices performed this function. The OSDBU directors at the Departments of Transportation and Energy commented that they reviewed acquisitions over a certain threshold level ($150,000 at Transportation and $3 million at Energy). Ten OSDBU directors reported that assigning a small business technical advisor to each office with an SBA procurement center representative was not a function of their office. In follow-up communication, the Acting OSDBU director at the Department of Veterans Affairs (VA) explained that his office had not assigned a small business technical advisor to each office with a procurement center representative but noted that OSDBU staff performed duties similar to those of a technical advisor. The Acting OSDBU director at the Department of Energy explained that because the agency had already implemented several levels of review by various technical and procurement staff, it had delayed hiring and assigning a technical advisor to contracting offices. However, he stated that the office was reassessing their review processes and resources to determine when such a hire would be feasible. The remaining eight OSDBU directors at the Departments of the Air Force, the Army, Commerce, the Interior, Justice, the Navy, and Transportation and DLA reported that the contracting offices within their agencies assigned small business technical advisors. For instance, Air Force officials commented that small business technical advisors were assigned only to field sites where they could assist in identifying specific opportunities for small businesses. Eleven OSDBU directors reported that assisting small businesses to obtain payments from prime contractors was not one of their functions. Seven of these directors (Departments of Agriculture, the Air Force, and Education; EPA; HUD; Office of the Secretary of Defense; and SSA) commented that contracting, acquisition, or program officials carried out this function at their agencies. Two of the seven directors clarified in follow-up interviews that they were not privy to subcontractor information. The OSDBU director at OPM commented that because the payment of invoices by a prime contractor to its subcontractors is part of a contractual arrangement to which the government is not a party, this function should not be performed by anyone at the agency. The OSDBU director at the Department of Transportation commented that the office provides counseling on progress payments and prompt payment guidance to small businesses. The OSDBU directors at Interior and the U.S. Agency for International Development commented that if a small business were to contact the OSDBU for payment assistance, the OSDBU would facilitate communication with the contracting officer responsible for payments. In addition to the contact named above, Paige Smith (Assistant Director), Farah Angersola, Tania Calhoun, Emily Chalmers, Janet Fong, Colleen Moffatt, Marc Molino, Kelly Rubin, Rebecca Shea, Andrew Stavisky, and William Woods made key contributions to this report. | Section 15(k) of the Small Business Act requires that all federal agencies with procurement powers establish an Office of Small and Disadvantaged Business Utilization (OSDBU) to advocate for small businesses. Section 15(k)(3) requires that OSDBU directors be responsible only to and report directly to agency or deputy agency heads. GAO was asked to assess agencies' compliance with the reporting structure and identify the functions OSDBUs performed. GAO reviewed compliance with section 15(k)(3) at 16 agencies--the 7 agencies that each procured more than $15 billion in goods and services in 2009 and 9 that it had previously reported were not complying with this requirement. GAO also surveyed the OSDBU directors at 25 agencies that represented more than 98 percent of civilian obligations and 90 percent of DOD obligations in 2009. Nine of the 16 federal agencies that GAO reviewed were in compliance with section 15(k)(3) of the Small Business Act, which requires OSDBU directors to be responsible only to and report directly to the agency or deputy agency head. The remaining seven agencies were not in compliance with the provision, and their OSDBU directors reported to lower-level officials or had delegated OSDBU responsibilities to officials who did not meet the reporting requirement. These agencies were not in compliance when GAO last examined them in 2003. During GAO's current review, directors who reported to agency heads cited benefits to the relationship, while those who did not had mixed views. GAO concluded that the views expressed by the directors at noncompliant agencies did not justify noncompliance and that these agencies should comply or provide support to Congress of their need, if any, for statutory flexibilities. Ongoing noncompliance with section 15(k)(3) undermines the intent of the act and may prevent some OSDBU directors from having direct access to top agency management. Consistent with its 2004 report, GAO's current work found that the 25 OSDBU directors surveyed focused their procurement activities on certain functions listed in section 15(k). At least 19 directors listed the five functions related to contract bundling, maintaining supervisory authority over staff, and helping small businesses obtain payments from agencies as among their duties. Fewer directors viewed the remaining three functions, such as reviewing acquisitions for small business set-asides and assisting small businesses to obtain payments from prime contractors, as duties. Directors who did not view these functions as their responsibility generally noted that contracting or program staff performed them. Whether OSDBU directors who do not perform certain functions listed in 15(k) are complying with the statute is not clear. GAO recommends that agencies not in compliance with section 15(k)(3) take steps to comply with this statutory requirement or report to Congress on why they have not complied, including any requests for statutory reporting flexibility as appropriate. SSA agreed with the recommendation, and Interior agreed to reevaluate its reporting structure. Commerce, Justice, State, and the Treasury disagreed, believing they were in compliance. GAO maintains its position on agencies' compliance status, as discussed further in the report. Agriculture did not comment. |
FECA provides cash benefits to eligible federal employees who suffer temporary or permanent disabilities resulting from work-related injuries or diseases. Labor’s Division of Federal Employees’ Compensation in the Office of Workers’ Compensation Programs (OWCP) administers the FECA program and charges agencies for whom injured employees worked for benefits provided. These agencies subsequently reimburse Labor’s Employees’ Compensation Fund from their next annual appropriation. FECA benefits are adjusted annually for cost-of-living and are neither subject to age restrictions nor taxed. USPS increaseshas large FECA program costs. At the time of their injury, 43 percent of FECA beneficiaries in 2010 were employed by USPS, as shown in table 1. One way to measure the adequacy of FECA benefits is to consider wage replacement rates, which are the proportion of pre-injury wages that are replaced by FECA. Wage replacement rates that do not account for missed career growth capture the degree to which a beneficiary is able to maintain his or her pre-injury standard of living whereas wage replacement rates that account for missed career growth capture the degree to which a beneficiary is able to maintain his or her foregone standard of living (i.e., standard of living absent an injury). Data limitations can preclude calculating wage replacement rates that account for missed career growth; however, doing so provides a more complete story of the comparison between an injured worker and his or her counter-factual of having never been injured. Wage replacement rates can be targeted by policy-makers; however, there is no consensus on what wage replacement rate policies should target. FECA beneficiaries receive different benefits past retirement age than workers who retire under a federal retirement system. Specifically, under FERS, federal retirees have a benefit package comprised of three components: the FERS annuity, which is based on years of service; the TSP, which is similar to a 401(k); and Social Security benefits. FECA benefits do not change at retirement age and beneficiaries cannot receive a FERS annuity and FECA benefits simultaneously. In addition, FECA beneficiaries cannot contribute to their TSP accounts post-injury, but they can receive benefits accrued from contributions made to their TSP accounts prior to being injured. In addition, Social Security benefits attributable to federal service are offset by FECA. If an individual has an extended disability and no current capacity to work, OWCP determines that he or she is a total-disability beneficiary and calculates long-term FECA benefits as a proportion of the beneficiary’s entire income at the time of injury. In 2010, 31,880 FECA beneficiaries received long-term total-disability cash benefits. Alternatively, if an individual recovers sufficiently to return to work in some capacity, OWCP determines that he or she is a partial-disability beneficiary and reduces his or her FECA benefits from the total-disability amount. For such partial-disability beneficiaries, OWCP calculates long- term benefits based on any loss of wage earning capacity (LWEC), as compared to their pre-injury wages. A beneficiary’s LWEC may be based on the difference between their pre-injury wages and their actual post-injury earnings if the beneficiary has found employment that OWCP determines to be commensurate with their rehabilitation. Alternatively, OWCP constructs a beneficiary’s LWEC based on the difference between pre-injury wages and OWCP’s estimate of what the FECA beneficiary could earn in an appropriate job placement (constructed earnings). In 2010, 10,594 FECA beneficiaries received long-term partial-disability cash benefits. In addition to our work on FECA benefit levels, we have also conducted work on program integrity and management. We have identified several weaknesses in these areas. Most recently, in April 2013, we found examples of improper payments and indicators of potential fraud in the FECA program, which could be attributed, in part, to oversight and data- access issues. We also found that FECA program requirements allow claimants to receive earnings, and earnings increases, without necessarily resulting in adjustment of FECA compensation. We recommended that the Secretary of Labor assess the feasibility of developing a cost-effective mechanism to share FECA compensation information with states. Labor agreed with the recommendation and stated that it will undertake a review to determine whether such data sharing and reporting is feasible. Our simulations of the effects of compensating non-USPS and USPS total-disability beneficiaries at the single rate (regardless of the presence of dependents) of either 66-2/3 or 70 percent of wages at injury, reduced the median wage replacement rates. Median wage replacement rates overall, and within the subgroups we examined, were generally lower under the 66-2/3 percent compensation proposal. Compared to the current FECA program, both proposals reduced 2010 median wage replacement rates for total-disability non-USPS and USPS beneficiaries, as shown in figure 1. The decreases in the overall median wage replacement rates were due to the greater proportion of beneficiaries who had a dependent—73 percent of non-USPS beneficiaries and 71 percent of USPS beneficiaries. Beneficiaries with a dependent received lower compensation under both proposals whereas beneficiaries without a dependent saw their compensation increase or stay the same. As shown in the middle group of bars in figure 1, the results of our simulation indicate that median wage replacement rates for USPS beneficiaries were generally higher than those for non-USPS beneficiaries. In both cases, the wage replacement rates account for missed income growth, as they are simulated based on 2010 take-home pay. All else equal, FECA beneficiaries who would have experienced more income growth—from the time of injury through 2010—had lower wage replacement rates than did those beneficiaries who would have experienced less income growth absent their injury. In general, USPS beneficiaries missed less income growth due to their injury than did non- USPS beneficiaries. Consequentially, USPS beneficiaries had higher wage replacement rates than non-USPS beneficiaries. For example, 4 out of 5 USPS beneficiaries in our analysis would have had less than 10 percent income growth had they never been injured. In contrast, 2 out of 5 non-USPS beneficiaries would have had less than 10 percent income growth, absent an injury. Under our simulations, both proposals increased the difference in wage replacement rates between beneficiaries with and without a dependent, increasing the magnitude and reversing the direction of the difference in median wage replacement rates, as shown in figure 2. Had we been able to account for the actual number of dependents, beneficiaries with dependents would have had lower wage replacement rates and thus the difference between median wage replacement rates would have been smaller under FECA and larger under both proposals. For other beneficiary subgroups we examined, the proposals did not reduce wage replacement rates disproportionately to the reduction in the overall median. However, we found that under current FECA policy and both proposals, wage replacement rates for some beneficiaries, such as those who, due to injury earlier in their careers, missed out on substantial income growth, were substantially lower than the overall median. FECA was not designed to account for missed income growth and thus total- disability beneficiaries who missed substantial income growth had lower wage replacement rates—outweighing the cumulative effect of FECA’s annual cost of living adjustments—as shown in figure 3. According to our retirement simulation comparing current FECA benefits to FERS benefits, we found that the overall median FECA benefit package (FECA benefits and TSP annuity) for both USPS and non-USPS FECA beneficiaries was greater than the current median FERS retirement benefit package (FERS annuity, TSP annuity, and Social Security). Specifically, the median FECA benefit package for non-USPS beneficiaries was 32 percent greater than the current median FERS—and 37 percent greater for USPS FECA beneficiaries. This implies that in retirement, FECA beneficiaries generally had greater income from FECA and their TSP in comparison to the FERS benefits they would have received absent an injury. Although the overall median FECA benefit was substantially higher than the median FERS benefit for 2010 annuitants, the difference between the two varies based on years of service. Our simulations showed that median FECA benefit packages were consistently greater than median FERS benefit packages across varying years of service; however, the gap between the two benefits narrowed as years of service increased. This occurred in large part because FERS benefits increase substantially with additional years of service. For example, under our simulation non- USPS beneficiaries whose total federal career would have spanned less than 10 years had a median FECA benefit that was about 46 percent greater than the corresponding FERS benefit. In contrast, non-USPS beneficiaries whose total federal career would have spanned 25-29 years had a median FECA benefit that was 16 percent greater than the corresponding FERS benefit. For USPS beneficiaries, those whose total federal career would have spanned less than 10 years had a median FECA benefit that was about 65 percent greater than the corresponding FERS benefit, while beneficiaries whose total federal career would have spanned between 20 and 24 years had a median FECA benefit that was 23 percent greater than the corresponding FERS benefit. Based on our simulation, we found that reducing FECA benefits once beneficiaries reach retirement age to 50 percent of wages at the time of injury would result in an overall median for the reduced FECA benefit package (reduced FECA plus the TSP) that was about 6 percent less than the median FERS benefit package for non-USPS annuitants. Under our simulation, for USPS annuitants, the reduced FECA benefit package would be approximately equal to the median 2010 FERS benefit package. This implies that under the proposed reduction, both USPS and non- USPS FECA beneficiaries would have similar income from their FECA benefit package in comparison to their foregone FERS benefit. In addition, under our simulation reduced FECA benefits were similar or less than FERS benefits across varying years of service. However, as years of service increase, the gap between the two benefits widened. For example, we found that non-USPS beneficiaries whose total federal career would have spanned less than 10 years had a median reduced FECA benefit that was about 2 percent greater than the corresponding FERS benefit. In contrast, those non-USPS beneficiaries whose total federal career would have spanned 25-29 years had a median reduced FECA benefit that was 19 percent less than the corresponding FERS benefit. Similarly, USPS beneficiaries whose total federal career would have spanned less than 10 years had a median reduced FECA benefit that was about 13 percent greater than the corresponding FERS benefit. In contrast, USPS beneficiaries whose total federal career would have spanned 25 to 29 years had a median reduced FECA benefit that was 20 percent less than the corresponding FERS benefit. Because FERS had only been in place for 26 years in 2010, our simulation did not capture the “mature” FERS benefit that an annuitant could accrue with more years of service. Consequently, it is likely that our analysis understates the potential FERS benefit when we consider 2010 benefit levels. As a result, we conducted a simulation of a “mature” FERS that was coupled with the assumption that individuals have 30-year federal careers. Based on this simulation, we found that the median current FECA benefit packages for non-USPS beneficiaries were on par or less than the median FERS benefit package—depending on the amount an individual contributes toward their TSP account for retirement. As shown on the right sides of figures 4 and 5, under the default scenario where there is no employee contribution and the employing agency contributes 1 percent to TSP, the median FECA benefit package is about 1 percent greater than the median FERS benefit package. However, under a scenario where each employee contributes 5 percent—and receives a 5 percent agency match—the median FECA benefit package is about 10 percent less than the median FERS benefit package. Similarly, our simulation showed that for USPS annuitants, the median FECA benefit package was about 13 percent greater than the median FERS benefit package under the 1 percent agency contribution scenario, and about 4 percent less than the median FERS benefit package under the 10 percent contribution scenario. Our simulation also found that, for both non-USPS and USPS annuitants, the median reduced FECA benefit package under the proposed changes was less than the median FERS benefit package—regardless of the simulated contributions to TSP accounts. Specifically, under a scenario where there is no employee contribution—and a 1 percent contribution from the employing agency—the median reduced FECA benefit package is about 31 percent less than the median FERS benefit package for non- USPS annuitants and 22 percent less than the median FERS benefit package for USPS annuitants. Under a scenario where each employee contributes 5 percent—and receives a 5 percent agency match—the median reduced FECA benefit package is about 35 percent less than the FERS benefit package for non-USPS annuitants and about 29 percent less than the FERS benefit package for USPS annuitants. Effects of Proposed FECA Revisions on Partial-disability Beneficiaries Depend on Post-Injury Earning Capacity and Employment Over Time We found partial-disability beneficiaries to be fundamentally different from total-disability beneficiaries, as they receive reduced benefits based on their potential to be re-employed and have work earnings. However, there is limited information available about the overall population of partial- disability beneficiaries. They do not all find work and their participation in the workforce may change over time, and their individual experiences will determine how they would fare under the proposed revisions. Partial-disability beneficiaries in the case studies we examined fared differently under both FECA and the proposed revisions to pre-retirement compensation, depending on the extent to which they had work earnings in addition to their FECA benefits. To consider this larger context, we conducted total income comparisons for the partial-disability case studies we examined. We defined the post-injury total income comparison to be the sum of post-injury FECA benefits and any gross earnings from employment at the time of the LWEC decision, as a percentage of pre- injury gross income. Among the seven partial-disability case studies we examined, those beneficiaries with constructed earnings LWECs had post-injury total income comparisons that were substantially less than those with actual earnings LWECs. As shown in table 2, the beneficiaries in case studies 5- 7 had constructed earnings LWECs and had post-injury total incomes that ranged from 29 to 65 percent of their pre-injury income under current FECA policy. This range was substantially lower than the total income comparisons for the beneficiaries in case studies 1-4 with actual earnings LWECs (77-96 percent). We found that by definition, at the time of their LWEC decision, those beneficiaries with constructed earnings LWECs earned less than the income OWCP used to calculate their LWECs. Consequently, their total income comparisons—FECA benefits plus earnings, as a percentage of pre-injury wages—are necessarily lower than those with actual earnings LWECs. We also found that beneficiaries in our case studies were affected differently by the proposed revisions to pre-retirement benefits. As expected, the beneficiaries who did not have a dependent (case studies 2, 4, and 7) experienced either slight increases or no change in their post- injury total income comparisons under the proposed revisions to pre- retirement benefits. Under both proposals, the beneficiaries in our case studies who had a dependent (case studies 1, 3, 5, and 6) experienced declines in their post-injury total income comparisons.decreases in total income comparisons were relatively small compared to the impact of not having actual earnings. For instance, the beneficiary with a constructed earnings LWEC in case study 6 experienced declines in total income comparisons of about 3 to 4 percentage points between current FECA policy and the proposals. However, the beneficiary’s total income comparisons under current FECA policy and the proposals were over 30 percentage points lower than those of the beneficiary in case study 3 who had the lowest total income comparisons of those beneficiaries with actual earnings LWECs. Due to the importance of actual work earnings on partial-disability beneficiaries’ situations, we have previously concluded that a snapshot of post-injury total income comparisons is insufficient to predict how beneficiaries fare over the remainder of their post-injury careers. Employment at the time of OWCP’s LWEC decision does not necessarily imply stable employment over time, as beneficiaries can find, change, or lose jobs over time. We have also found that the proposals to reduce FECA benefits at retirement age would primarily affect those partial-disability beneficiaries who continue to receive FECA benefits past retirement age. As we reported in December 2012, among those partial-disability beneficiaries who stopped receiving FECA benefits in 2005-2011, 68 percent did so due to their election of OPM retirement or other benefits, such as Veterans Affairs disability benefits. At that time, Labor officials told us that because many variables affect retirement benefits, they cannot predict why partial-disability beneficiaries would potentially choose to retire instead of continuing to receive FECA benefits. Only 17 percent of partial- disability beneficiaries who stopped receiving FECA benefits were beneficiaries who died (i.e., received benefits from injury until death). These aggregate numbers do not track individual beneficiaries’ decisions to elect retirement or to continue receiving FECA benefits past retirement age, but they suggest that there is a substantial percentage of partial- disability beneficiaries that elects other benefits instead of FECA at some point post-injury. Since those beneficiaries who elect FERS retirement would not be affected by the proposed revisions to FECA compensation at retirement age, the overall effects of the proposals on partial-disability beneficiaries should be considered in the larger context of retirement options. To do so, in our December 2012 report, we used data from the seven partial- disability case studies to simulate and compare FERS and FECA benefits and to highlight various retirement options these partial-disability beneficiaries may face. As shown in table 3, we found: The beneficiaries in case studies 2, 4, and 6 had potential FERS benefit packages that were higher than their FECA benefits under current policy and the proposed revision—they would likely not be affected by the proposed revision. The beneficiaries in case studies 1, 3, and 7 had potential FERS benefit packages that were lower than their FECA benefits under current policy and the proposed revision—they would likely face a reduction in FECA benefits in retirement under the proposed revision. The beneficiary in case study 5 had a potential FERS benefit package that was lower than his FECA benefits under current policy, but higher than his benefits under the proposed FECA reduction—he would likely face a reduction in FECA benefits in retirement under the proposed revision. Based on our prior work, we have concluded that the differences in retirement options that individual beneficiaries face stem from two key factors: (1) OWCP’s determination of their earning capacities, and (2) their total years of federal service. Partial-disability beneficiaries with greater potential for earnings from work receive relatively lower FECA benefits to account for their relatively lower loss of wage earning capacity, all else equal. In table 2, beneficiaries with: low earning capacities post-injury (case studies 1, 3, and 5) had FECA benefits that were more favorable than FERS benefits; high earning capacities post-injury (case studies 2 and 4) had FECA benefits that were less favorable than FERS benefits; and mid-range earning capacities post-injury (case studies 6 and 7) had FECA benefits whose favorability depended on their total years of federal service. Fewer years of federal service resulted in a lower FERS annuity and lower Social Security benefits attributable to federal service, all else equal. We have also found that partial-disability beneficiaries who choose to remain on FECA past retirement age currently face lower FECA benefits in retirement as compared with total-disability beneficiaries, and would experience a reduction in benefits under the proposals. Partial-disability beneficiaries receive FECA benefits that are lower than those of otherwise identical total-disability beneficiaries to account for their potential for work earnings. As long as they work, their income is comprised of their earnings and their FECA benefits. However, once they choose to retire, partial-disability beneficiaries who choose to stay on FECA likely no longer have any work earnings and are not eligible to simultaneously receive their FERS annuity. Thus, we found that because of the way FECA benefits are currently calculated, such partial- disability beneficiaries may have less income in retirement than otherwise identical total-disability beneficiaries, and the proposals would reduce benefits in retirement without differentiating between partial and total- disability beneficiaries. The proposed reduction may serve as a long- term incentive for partial-disability beneficiaries to return to work,particularly because their initial FECA benefits are lower than those of total-disability beneficiaries. In conclusion, FECA continues to play a vital role in providing compensation to federal employees who are unable to work because of injuries sustained while performing their federal duties and FECA benefits generally serve as the exclusive remedy for being injured on the job. Our simulations of the potential effects of proposed changes to FECA benefit levels incorporated the kinds of approaches used in the literature on assessing benefit adequacy for workers’ compensation programs, such as taking account of missed career growth. More specifically, we assessed the proposed changes by simulating the level of take-home pay or retirement benefits FECA beneficiaries would have received if they had not been injured, which provides a realistic basis for assessing how beneficiaries may be affected. However, we did not recommend any particular level of benefit adequacy. As policymakers assess proposed changes to FECA benefit levels, they will implicitly be making decisions about what constitutes an adequate level of benefits for FECA beneficiaries before and after they reach retirement age. While our analyses focused on how the median FECA beneficiary might be affected by proposed changes, it also highlighted how potential effects may vary for different subpopulations of beneficiaries, which can assist policymakers as they consider such changes to the FECA program. This concludes my statement and I would be happy to answer any questions. For further information regarding this testimony, please contact Andrew Sherrill at (202) 512-7215 or sherrilla@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Individuals who made key contributions to this testimony include: Nagla’a El-Hodiri, Assistant Director; James Bennett, Jessica Botsford, Sherwin Chapman, Michael J. Collins, Melinda Cordero, Holly Dye, Michael Kniss, Gene Kuehneman, Kathy Leslie, James Rebbe, Jeff Tessin, Walter Vance, and Rebecca Woiwode. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | In 2012, the FECA program provided more than $2.1 billion in wage-loss compensation to federal workers who sustained injuries or illnesses while performing federal duties. Total-disability beneficiaries with an eligible dependent are compensated at 75 percent of gross wages at the time of injury and those without are compensated at 66-2/3 percent. Benefits are adjusted for inflation and are not taxed nor subject to age restrictions. Some policymakers have raised questions about the level of FECA benefits, especially compared to federal retirement benefits. Proposals to revise FECA for future total- and partial- disability beneficiaries include: setting initial FECA benefits at a single rate (66-2/3 or 70 percent of applicable wages at time of injury), regardless of whether the beneficiary has eligible dependents; and converting FECA benefits to 50 percent of applicable wages at time of injuryadjusted for inflationonce beneficiaries reach full Social Security retirement age. This testimony presents results of GAOs four recent reports on FECA issues. It summarizes (1) potential effects of the proposals to compensate total-disability FECA beneficiaries at a single rate; (2) potential effects of the proposal to reduce FECA benefits to 50 percent of applicable wages at full Social Security retirement age for total-disability beneficiaries; and (3) how partial disability beneficiaries might fare under the proposed changes. To do this work, GAO conducted simulations comparing FECA benefits to income (take-home pay or retirement benefits) a beneficiary would have had absent an injury, and conducted seven case studies of partial disability beneficiaries. GAO's simulation found that under the current Federal Employees' Compensation Act (FECA) program, the median wage replacement rate--the percentage of take-home pay replaced by FECA--for total-disability beneficiaries was 88 percent for U.S. Postal Service (USPS) beneficiaries and 80 percent for non-USPS beneficiaries in 2010. GAO also found that proposals to set initial FECA benefits at a single compensation rate would reduce these replacement rates by 3 to 4 percentage points under the 70-percent option and 7 to 8 percentage points under the 66-2/3 percent option. Beneficiaries with dependents would receive reduced FECA benefits under both options. The decreases in wage replacement rates were due to the greater proportion of beneficiaries who had a dependent--over 70 percent of both USPS and non-USPS beneficiaries. In GAO's simulation comparing FECA benefits to retirement benefits, GAO found that under the current FECA program, the median FECA benefit package for total-disability retirement-age beneficiaries was 37 and 32 percent greater than the median 2010 retirement benefit package for USPS and non-USPS beneficiaries, respectively. This analysis focused on individuals covered under the Federal Employees Retirement System (FERS), which generally covers employees first hired in 1984 or later, and covered about 85 percent of the federal workforce in 2009. GAO also found that the proposal to reduce FECA benefits at the full Social Security retirement age would result in a median FECA package roughly equal to the median FERS retirement package in 2010. However, the median years of service for the FERS annuitants GAO analyzed was about 16 to 18 years, so these simulations did not capture a fully mature retirement system and likely understated the future FERS benefit level. Consequently, GAO also simulated a mature FERS system--intended to reflect future benefits of workers with 30-year careers--and found that the median FECA benefit package under the proposed change would be from 22 to 35 percent less than the median FERS retirement package. Partial-disability beneficiaries are fundamentally different from total-disability beneficiaries, as they receive reduced FECA benefits based on a determination of their earning capacity. GAO's seven case studies of partial-disability beneficiaries showed that how they might fare under the proposed FECA changes can vary considerably based on their individual circumstances, such as their earning capacity and actual levels of earnings. For example, among GAO's case studies, those beneficiaries with high earning capacities may elect to retire under FERS and would likely not be affected by the proposed FECA reduction at retirement age because their potential retirement benefits were substantially higher than their current or proposed reduced FECA benefit levels. In contrast, those beneficiaries with low earning capacities had potential retirement benefits that were lower than their current FECA benefits and the proposed FECA reduction at retirement age would reduce their FECA benefits. |
In June 2002, the Coast Guard awarded a contract to ICGS to begin the acquisition phase of the Deepwater program. Rather than using the traditional approach of replacing classes of ships or aircraft through a series of individual acquisitions, the Coast Guard chose to employ a “system of systems” acquisition strategy that would replace its aging assets of ships, aircraft, and communication capabilities with a single, integrated package of new or modernized assets and capabilities. The focus of the program is not just on new ships and aircraft, but rather on an integrated approach to modernizing existing or “legacy” assets while transitioning to newer, more capable assets, with improved command, control, communications and computers, intelligence, surveillance, and reconnaissance (C4ISR) capabilities. At full implementation, the Deepwater program will replace the Coast Guard’s entire fleet of current deepwater surface and air assets, comprising three classes of new cutters and their associated small boats, a new fixed-wing manned aircraft fleet, a combination of new and upgraded helicopters, and both cutter-based and land-based unmanned air vehicles (UAVs). In addition, all of these assets will be linked with state-of-the-art C4ISR capabilities and computers and will be supported by an integrated logistics management system. ICGS, a business entity jointly owned by Northrop Grumman and Lockheed Martin, acts as the system integrator to develop and deliver the Deepwater program. They are the two first-tier subcontractors for the program and either provides Deepwater assets themselves or award second-tier subcontracts for the assets. The Coast Guard’s contract with ICGS has a 5-year base period with five additional 5-year options. The Coast Guard is scheduled to decide on whether to extend ICGS’s contract by June 2006, which is 1 year prior to the end of the first 5-year contract term. Although ICGS is responsible for designing, constructing, deploying, supporting, and integrating Deepwater assets, the Coast Guard maintains responsibility for oversight and overall management of the program. To help fulfill this responsibility, the Coast Guard uses several management tools to track progress in the program, such as monthly status reports called quad reports, a performance-based management tool called the Earned Value Management System (EVMS), and a schedule management system called IMS. The quad reports are monthly summaries of the work performed on the various deepwater assets and were developed to give managers a monthly progress report on the performance of the program in such areas as cost, schedule, and contract administration. EVMS is used to track cost and schedule for certain delivery orders that have been placed to manage the risk of the major assets and activities. IMS is a three-tiered, calendar-based schedule used to track the completion of tasks and milestones of the individual Deepwater delivery task orders. However, data reliability with IMS has been an area of concern for the Coast Guard, and it is currently working with a private contractor and ICGS to address these concerns. Currently, the Coast Guard only maintains the lowest and most detailed level of IMS with monthly updates. These updates reflect the status of active contracts of individual assets and feed into EVMS’s monthly cost performance reports of those individual contracts, which allow program management to monitor the cost, schedule, and technical performance of ongoing work at these lowest and most detailed levels. In recent months, we have issued two reports that have raised concern about the Coast Guard’s initial management of the Deepwater program and the potential for escalating costs. In March 2004, we reported that key components needed to manage the Deepwater program and oversee the system integrator’s performance have not been effectively implemented. We recommended that the Secretary of Homeland Security direct the Commandant of the Coast Guard to take a number of actions to improve Deepwater program management and contractor oversight. In April 2004, we testified that the most significant challenge the Coast Guard faces as it moves forward in the Deepwater program is keeping the program on schedule and within planned budget estimates. We noted that the Coast Guard was at risk of having to expend funds to repair deteriorating legacy assets that otherwise had been planned for Deepwater modernization initiatives, which could potentially further delay the program and increase total program costs. We noted further that the Coast Guard’s current estimate for completing the acquisition program within the 20-year schedule has risen to $17.2 billion, an increase of $2.2 billion over the original $15 billion estimate. The degree to which the Deepwater program is on track with regard to its original 2002 acquisition schedule is difficult to determine, because the Coast Guard has not maintained and updated the acquisition schedule. The original acquisition schedule included acquisition phases (such as concept technology and design, system development and demonstration, and fabrication), interim phase milestones (such as preliminary and critical design reviews, installation, and testing), and the critical paths integrating the delivery of individual components to particular assets. However, while the Coast Guard has what is called an integrated master schedule, it currently uses it only at the lowest and most detailed levels and have not updated it to demonstrate whether individual components and assets are being integrated and delivered on schedule and in the critical sequence. There are a number of reasons why tracking the current integrated schedule on a major government investment or acquisition is important, but two stand out. First, our work has shown that it is important to identify potential risks in major acquisitions as early as possible so problems can be avoided or minimized. The ability to achieve scheduled events is a key indicator of risk. Second, cost, schedule, and performance are fundamental to the Congress’s oversight of major acquisitions. In fact, by law, DOD’s major defense acquisition programs have to report cost, schedule, and performance updates to the Congress at least annually and whenever cost and schedule thresholds are breached. In practice, schedules on DOD’s major defense acquisitions are continually monitored and reported to management on a quarterly basis. While Coast Guard officials indicated that the management tools they use are sufficient for tracking delivery dates throughout the acquisition, we found that these tools could not provide a ready and reliable picture of the current integrated Deepwater acquisition schedule. For example, IMS has had problems with data reliability and the monthly status reports, while assessing the progress of individual assets, do not translate those individual assessments into an overall integrated Deepwater acquisition schedule. In fact, the February 2004 status report noted that because there was no accurate overall integrated schedule for the command and control design phase, the linkages with the development of other assets were largely unknown. Although maintaining a current integrated schedule is a best practice for ensuring adequate contract oversight and management and is a requirement for DOD acquisitions, Coast Guard officials said they have not done so because of the numerous changes the Deepwater Program experiences every year and because of the cost, personnel, and time involved in crafting a revised master plan with the systems integrator on an annual basis. The officials said they have not planned to update the original acquisition schedule until just prior to deciding whether to extend the award of the next 5-year Deepwater contract in 2007. However, we have found that in acquisitions of similar scope—and in particular, acquisitions made by DOD—maintaining a current schedule is a fundamental practice that the department considers necessary. While the Coast Guard could not provide us with an updated integrated schedule, we developed the current acquisition status of a number of selected Deepwater assets from documents provided by the Coast Guard. Our analysis indicates that several Deepwater assets and capabilities have experienced delays and are at risk of being delivered later than anticipated in the original implementation plan. For example, the delivery of the first two maritime patrol aircraft is behind schedule by about 1 year, and the delivery and integration of the vertical-take-off-and-land unmanned air vehicle to the first national security cutter has been delayed 18 months. The $168 million appropriated in fiscal year 2004 for the Deepwater program above the President’s request of $500 million will allow the Coast Guard to conduct a number of projects that had been delayed or would not have been funded in fiscal year 2004, but it will not fully return the program to its original 2002 acquisition schedule. Our analysis indicates there are several reasons for this result. First, even with the additional amount, the program still had a cumulative funding shortfall of $39 million through fiscal year 2004, as compared to the Coast Guard’s planned funding amount. Second, because part of the additional $168 million was needed to start work delayed from fiscal year 2003, it did not provide enough funding for the delayed work and for all the work that the Coast Guard had originally planned for fiscal year 2004. As a result, some of this work will have to be delayed to fiscal year 2005. Third, the delivery of some assets has fallen so far behind schedule that it is impossible to ensure their delivery according to the original 2002 implementation schedule by simply providing more money. For example, according to the 2002 implementation schedule, nine maritime patrol aircraft were to be delivered by the end of 2005; according to the current contract, none will be delivered in 2005 and two will be delivered by the end of 2006. Similarly, the original schedule called for completing 18 123-foot patrol boat conversions by the end of 2005; according to the current contract, 8 will be completed by the end of 2005. Fourth, the acquisition of some assets has been delayed for reasons other than funding. For example, greater than anticipated hull corrosion in the 110 ft. patrol boats has delayed the conversion of those into 123 footers and delays in the availability of faster satellite service for legacy cutters has delayed the delivery of those upgraded legacy cutters. Finally, in keeping with appropriations conference committee directives for how the additional amount was to be spent, part of the additional funding went for design of the offshore patrol cutter, work that, under the original schedule, would not begin for another 6 to 8 years. This work may speed up acquisition of these assets, but they were not on the original schedule this early in the program. The Coast Guard is currently revising Deepwater’s mission needs statement in response to increased homeland security requirements resulting from the terrorist attacks of September 11, 2001. According to the Coast Guard, in May 2004 it submitted the revised statement to the Department of Homeland Security’s (DHS) Joint Requirements Council, which conditionally accepted it until the Deepwater contractor, ICGS, completes a new estimate of the costs needed to acquire the necessary functional capabilities resulting from the revised mission needs statement. The council further directed the Coast Guard to complete the new cost estimate in order to brief the DHS Investment Review Board in October 2004. Acting as an agent of the DHS, the board will approve any increases in the Deepwater total ownership cost associated with the revised mission needs statement and any growth in the Deepwater budget. Specifications for some specific Deepwater assets—most notably the National Security Cutter—will be changed in light of the Coast Guard’s added homeland security responsibilities. Coast Guard officials said the main impact of the revision would be to close the gaps in Deepwater asset capabilities created by the expansion of Coast Guard mission requirement after September 11. Regarding DOD’s assistance to the Coast Guard for Deepwater, the Navy budgeted $25 million for the Deepwater program in fiscal years 2002-04, mainly to help outfit the National Security Cutter. The Navy budgeted these funds to provide capabilities and equipment it regarded as important for the Coast Guard to have for potential national defense missions that would be carried out in conjunction with Navy operations. The 2002 Deepwater acquisition schedule showed not only the individual planned phases and interim milestones for designing, testing, and fabricating assets but also the integrated schedules of critical linkages between assets. The absence of an up-to-date integrated acquisition schedule for the Deepwater program is a concern, because it is a symptom of the larger issues we reported in March 2004 that are related to whether this complicated acquisition is being adequately managed and whether the government’s interests are being properly safeguarded. Since the inception of the unique contracting approach to this project, we have pointed out that it poses risks, in that it would be expensive to alter and, because of the unique “systems integrator” approach, does not operate like a conventional acquisition. The recent disclosure that, just 3 years into the acquisition, costs have risen by $2.2 billion points to the need for a clear understanding of what assets are being acquired, when they are being acquired, and at what cost. This lack of such a current schedule lessens the Coast Guard’s ability to monitor the contractor’s performance and to take early action on potential risks before they become problems later in the program. The Coast Guard has so far maintained that it is not worth the time and cost involved to update the acquisition schedule until just before the award of the next 5-year Deepwater contract, but we disagree. We recognize that there are costs involved with keeping an acquisition schedule updated. However, for Department of Defense acquisitions of such scope, maintaining a current schedule is a fundamental and necessary practice. Deepwater remains a program in transition, in that the Coast Guard is currently revising the program’s mission requirements to include increased homeland security requirements. The major modifications that may result to key assets make keeping the program on track that much harder. As evolving mission requirements are translated into decisions about what assets are needed and what capabilities they will need to have, it becomes even more imperative that Coast Guard officials update the acquisition schedule on a more timely basis, so that budget submissions by the Coast Guard can allow DHS and Congress to base decisions on accurate information. We recommend that the Secretary of Homeland Security direct the Commandant of the Coast Guard to update the original 2002 Deepwater acquisition schedule in time to support the fiscal year 2006 Deepwater budget submission to DHS and Congress and at least once a year thereafter to support each budget submission. The updated schedule should include the current status of asset acquisition phases (such as concept technology and design, system development and demonstration, and fabrication), interim phase milestones (such as preliminary and critical design reviews, installation, and testing), and the critical paths linking the delivery of individual components to particular assets. We provided a draft of this report to DHS and the Coast Guard for their review and comment. In written comments, which are reproduced in appendix II, the Coast Guard generally concurred with the findings and recommendations in the report. The Coast Guard also provided technical comments, which we incorporated as appropriate. The Coast Guard, however, did not concur with three areas of discussion. First, the Coast Guard said that our draft report did not accurately portray its current acquisition tools and disagreed with our determination that data within one of those tools, the IMS, was not reliable enough for use in our report. As we noted in the report, at the time of our review Coast Guard officials expressed concerns about the reliability of IMS data, such as out-dated data and less than adequate data input and adjustment, and told us that they had hired a consultant to address those concerns. For that reason and with Coast Guard and ICGS agreement on our methodology, we based our analysis on the same source documents the Coast Guard uses as inputs to IMS’s lowest, most detailed level. The Coast Guard stated in its written comments that while not in the form of an acquisition schedule, IMS’s lower level data are reliable. According to its written comments, the Coast Guard is making improvements in updating IMS at all levels. Second, the Coast Guard said that we did not highlight that delays in the delivery schedule have been caused by a lack of funding and the subsequent need to sustain legacy assets rather than a lack of acquisition schedule updates. The scope of our review was to determine the impact of the additional $168 million in fiscal year 2004 on returning Deepwater to its original 2002 schedule, not to determine effects of receiving appropriations that were less than requested in previous years. However, in explaining why the $168 million will not return the Deepwater to its original schedule, we did note that part of the $168 million was needed to start work delayed from fiscal year 2003, and that the acquisition of some assets had been delayed due to greater than anticipated hull corrosion in legacy 110-foot patrol boats causing the delay in converting them to 123- footers. Finally, we did not intend to imply that the lack of acquisition schedule updates caused delays in delivery schedule. However, we continue to believe that not updating an integrated schedule to show the acquisition linkages between critical assets is a management concern. We believe in the importance of updating acquisition schedules at least annually not only as a best practice for managing and overseeing a complex integrated acquisition such as Deepwater but also as a up-to-date source of information on which DHS and the Congress can base annual budget decisions. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 15 days from its issue date. At that time, we will send copies of this report to the Secretary of Homeland Security, the Commandant of the Coast Guard, appropriate congressional committees, and other interested parties. The report will also be available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staffs have any questions about this report, please contact me at (415) 904-2200 or by email at wrightsonm@gao.gov, or Steve Calvo, Assistant Director, at (206) 287-4839 or by email at calvos@gao.gov. Other key contributors to this report were Shawn Arbogast, Leo Barbour, David Best, Michele Fejfar, Paul Francis, Sam Hinojosa, David Hooper, Michele Mackin, and Stan Stenersen. the Coast Guard’s status in revising Deepwater’s mission to meet increased homeland security requirements and the amount of funding support the Department of Defense (DOD) has provided for the Deepwater program for fiscal years 2002-2004. Interviewed Coast Guard and contractor officials. Examined delivery task orders and associated statements of work for individual Deepwater assets to establish what work was actually started, when it began, and the period of performance for that work; and monthly status reports of Deepwater program managers assessing the performance of the work. Compared the actual start and end dates of work with what was planned in the original 2002 acquisition schedule. Compared the original plan’s listing of the work scheduled to begin for each fiscal year with the work that was actually begun. Interviewed program officials regarding the reliability of data in the Deepwater integrated master schedule. Interviewed DOD officials and examined Navy budget documents to determine DOD funding support for Deepwater. The degree to which the Deepwater program is on track with regard to its original 2002 acquisition schedule is difficult to determine, because the Coast Guard has not kept the acquisition schedule updated. Several assets are experiencing delays and are at risk for being delivered later than originally planned. The additional $168 million appropriated in fiscal year 2004 will allow the Coast Guard to conduct a number of projects that had been delayed or would not have been funded in fiscal year 2004, but it will not fully return the program to its original 2002 acquisition schedule. Additional information on the Deepwater Program. The Coast Guard submitted a revised Mission Needs Statement to the Department of Homeland Security in May 2004. Navy budgeted $25 million in fiscal years 2002-04 for intermediate gun and combat systems for cutters. of over 90 cutters and 200 aircraft used for missions that generally occur beyond 50 miles from shore but may start at coasts and extend seaward to wherever the Coast Guard is required to take appropriate action. In June 2002, the Coast Guard awarded a 5-year base contract to Integrated Coast Guard Systems (ICGS), Inc., to begin the acquisition phase of the Deepwater program, with five additional 5-year options. The Coast Guard is scheduled to decide on whether to extend ICGS’s contract by June 2006, 1 year prior to the end of the first 5-year contract term. The Coast Guard chose to employ a "system of systems" acquisition strategy that would replace its aging Deepwater assets with a single, integrated package of new or modernized assets. ICGS acts as the system integrator to develop and deliver an improved integrated system of ships, aircraft, unmanned air vehicles, command, control, communications, computers, intelligence, surveillance and reconnaissance (C4ISR), and supporting logistics. Background (cont’d) ICGS is responsible for designing, constructing, deploying, supporting, and integrating Deepwater assets. The Coast Guard maintains responsibility for oversight and overall management of the program. The Coast Guard uses several management tools to track progress in the program, such as monthly status reports, Earned Value Management System (EVMS), and an integrated master schedule. However, because data reliability with the integrated master schedule has been an area of concern for the Coast Guard, they are currently working with a private contractor and ICGS to address these concerns. Background (cont’d) GAO recently reported concerns with the Coast Guard’s initial management of the Deepwater program and the potential for escalating costs. Key components needed to manage the Deepwater program and oversee the system integrator’s performance have not been effectively implemented. Contract Management: Coast Guard’s Deepwater Program Needs Increased Attention to Management and Contractor Oversight, GAO-04-380 (Washington, D.C.: Mar. 9, 2004). The most significant challenge to the Coast Guard is keeping the Deepwater program on schedule and within planned budget estimates through a well-managed and adequately funded effort. The Coast Guard estimates that the project’s cost is now $2.2 billion more than the initial estimate. Coast Guard: Key Management and Budget Challenges for Fiscal Year 2005 and Beyond, GAO-04-636T (Washington, D.C.: Apr. 7, 2004). acquisition schedule is difficult to determine, because the Coast Guard does not maintain and update the acquisition schedule. The Coast Guard cannot readily and reliably present the integrated acquisition status of the Deepwater program. The Coast Guard’s reasons for not updating the schedule include: Updating more regularly is impractical and unnecessary because Deepwater experiences numerous program changes each year and keeping track of those changes would consume too much funding, time, and personnel. Monthly status reports, EVMS, and other management tools are adequate for tracking Deepwater’s acquisition. The Coast Guard intends to update the schedule prior to deciding whether to extend the first 5-year option to the contractor in 2007. Status of Acquisition Schedule (cont’d) Tracking the current integrated schedule on a major government investments or acquisitions similar to the Deepwater program is important because identifying potential risks in major acquisitions as early as possible is important so problems can be avoided or minimized. The ability to achieve scheduled events is a key indicator of risk. cost, schedule, and performance are fundamental to the Congress’s oversight of major acquisitions. By law, DOD’s major defense acquisition programs have to report cost, schedule, and performance updates to the Congress at least annually and whenever cost and schedule thresholds are breached. 10 U.S.C. Sections 2430, 2432, 2433, and 2435, on major defense acquisition programs. In practice, schedules on major defense acquisitions are continually monitored and reported to DOD management on a quarterly basis. DOD Practice: Interim Defense Acquisition Guidebook (October 2002). GAO best practices reports: Best Practices: Better Matching of Needs and Resources will Lead to Better Weapon System Outcomes, GAO- 01-288 (Mar. 8, 2001); Defense Acquisitions: DOD’s Revised Policy Emphasizes Best Practices, but More Controls are Needed, GAO 04- 53 (Nov. 10, 2003); and Defense Acquisitions: Assessments of Major Weapons Programs, GAO-04-248 (Mar. 31, 2004). They only only track the schedules of individual assets at the lowest, most detailed level and not at the integrated level. The integrated master schedule has data reliability problems. The monthly status reports do not translate the progress of individual assets into an overall Deepwater acquisition schedule. A February 2004 status report noted that because there was no accurate overall integrated schedule for the command and control design, the linkages with the development of other assets were largely unknown. schedule, our analysis showed that a number of key Deepwater assets would be delivered later than originally scheduled. Status of Acquisition Schedules for Selected Assets (cont’d) Part of the additional $168 million in funding allowed the Coast Guard to start work delayed from fiscal year 2003 or to start work originally scheduled for fiscal year 2004. Atlantic area Pacific area Greater antilles section, San Juan, P.R. 2004 will not fully return Deepwater to its original 2002 acquisition schedule. Reason 1: A cumulative shortfall of $39 million still exists. Reason 2: The additional amount did not fund all work planned for fiscal year 2004; some will be delayed to fiscal year 2005 or beyond. Type of asset or work delayed to fiscal year 2005 or beyond C4ISR Modification of 270 Class Cutter ships 10,13 CAMS CC-2 Low Rate of Initial Production COMMSTA CC-2 Low Rate of Initial Production (CSTA-01,02,03,04,05,06) Production and Deployment for Major Modification of 110/123 Class Patrol Cutter Lot 5 (follow ships 13-20) Coast Guard Air Station (CGAS) Production and Deployment Aircraft Repair and Supply Center (AR&SC) Production and Deployment Aircraft Training Center (ATC) Production and Deployment Aviation Technical Training Center (ATTC) Production and Deployment Production and Deployment of MPA Low rate of initial production (aircraft 7-9) Impact of Additional Funding (cont’d) Reason 3: Some assets are so far behind schedule that it is impossible to return them to their original schedule. Nine maritime patrol aircraft were originally planned for delivery by the end of 2005; current plans show 2 to be delivered by the end of 2006. Eighteen conversions of 110-foot patrol boats to 123 feet were originally planned for delivery by the end of 2005; current plans show 8 to be completed by the end of 2005. than funding. Greater than anticipated hull corrosion in the 110-foot patrol boats has delayed their conversion to 123 feet and delivery several months. Legacy cutters are scheduled to receive a C4ISR upgrade of higher speed International Maritime Satellite service. However, the availability of the higher speed service has been delayed several times from early 2003 to the Coast Guard’s current projection of June 2004. scheduled to begin for several years. In keeping with fiscal year 2004 appropriations conference committee directives for how the additional amount was to be spent, $20 million went to begin the design phase for the Offshore Patrol Cutter (OPC). The Coast Guard had planned to begin the design phase about 2010 for delivery of the first OPC in 2012; current estimate is that the acquisition schedule has been accelerated by several years. Incorporation of new homeland security requirements into the Deepwater mission needs statement is still under way. In May 2004, the DHS Joint Requirements Council conditionally accepted it and directed the Coast Guard to develop new total ownership cost estimates for submission to and approval by the DHS Investment Review Board in October 2004. Coast Guard officials said the revision’s main impact will be an increase in number of assets. RAND concluded in an April 2004 study that the Coast Guard probably needs twice as many cutters and 50 percent more aircraft. In the interim, some requirements changes have been made to individual assets, mainly the National Security Cutter. Operating requirement for chemical-biological-radiological-nuclear defense capabilities was expanded to enable the cutter to operate in a contaminated environment, not just pass through it. Size of flight deck enlarged to accommodate Navy, Army, and Customs and Border Protection Agency models of the H-60 helicopter. Shipboard sensitive compartmented information facility added for collection and use of intelligence. used for Deepwater program. $12.6 million in fiscal years 2002-03 was for testing/evaluating an intermediate gun system for the National Security Cutter (and possibly other cutters). $12.5 million in fiscal year 2004 was for combat systems suites for the National Security Cutter. The Navy did not transfer the $25.1 million to the Coast Guard but will transfer the resulting equipment. Through a 1987 agreement, the Navy provides to the Coast Guard all Navy-owned, military readiness equipment and associated support materials that the Navy deems necessary to enable the Coast Guard to carry out assigned missions while operating with the Navy. Upon declaration of war, or when the President directs, the Coast Guard operates as a service of the Navy and is subject to the orders of the Secretary of the Navy. not worth the time and cost involved. Absence of an up-to-date integrated acquisition schedule for Deepwater is a concern, because it is a symptom of the larger issues related to whether this complicated acquisition is being adequately managed. The recent disclosure by the Coast Guard that Deepwater costs have risen by an estimated $2.2 billion points to the need for a clear understanding of what is being acquired, when it is being acquired, and at what cost. Maintaining a current integrated schedule for a DOD acquisition of such scope is a fundamental and necessary practice. Evolving mission requirements and the need to evaluate contractor performance for contract renewal heighten need for timely and accurate information so the Coast Guard and Congress can base decisions on accurate information. We recommend the Secretary of Homeland Security direct the Commandant of the Coast Guard to update the original 2002 Deepwater acquisition schedule in time to support the fiscal year 2006 Deepwater budget submission to the Department of Homeland Security and Congress and at least once a year thereafter to support each budget submission. The updated schedule should include the current status of asset acquisition phases (such as concept technology and design, system development and demonstration, and fabrication); interim phase milestones (such as preliminary and critical design reviews, installation, and testing); and the critical paths linking the delivery of individual components to particular assets. | In 2002, the Coast Guard began its $17 billion, 20-year Integrated Deepwater System acquisition program to replace or modernize its cutters, aircraft, and communications equipment for missions generally beyond 50 miles from shore. During fiscal years 2002-03, Deepwater received about $125 million less than the Coast Guard had planned. In fiscal year 2004, Congress appropriated $668 million, $168 million more than the President's request. GAO has raised concern recently about the Coast Guard's initial management of Deepwater and the potential for escalating costs. GAO was asked to review the status of the program against the initial acquisition schedule and determine the impact of the additional $168 million in fiscal year 2004 funding on this schedule. The degree to which the Deepwater program is on track with its original 2002 integrated acquisition schedule is difficult to determine because the Coast Guard has not updated the schedule. Coast Guard officials said they have not updated it because of the numerous changes Deepwater experiences every year and the cost, personnel, and time involved. However, in similar acquisitions--those of the Department of Defense (DOD)--cost, schedule, and performance updates are fundamental to congressional oversight. DOD is required to update the schedule at least annually and whenever cost and schedule thresholds are breached. In practice, DOD continually monitors and reports schedules for management on a quarterly basis. Updating the acquisition schedule--including phases such as design and fabrication, interim phase milestones, and critical paths linking assets-- on a more timely basis is imperative so that annual Coast Guard budget submissions can allow Congress to base decisions on accurate information. GAO used available data to develop the current acquisition status for a number of selected Deepwater assets and found that they have experienced delays and are at risk of being delivered later than anticipated. The additional $168 million in fiscal year 2004, while allowing the Coast Guard to conduct a number of Deepwater projects that had been delayed or would not have been funded in fiscal year 2004, will not fully return the program to its original 2002 acquisition schedule. Reasons include: all work originally planned for fiscal year 2004 was not funded and some will have to be delayed to fiscal year 2005; delivery of some assets has fallen so far behind schedule that ensuring their original delivery dates is impossible; and nonfunding reasons have caused delays, such as greater than expected hull corrosion of patrol boats delaying length extension upgrades. |
Geoengineering proposals to deliberately alter the climate in response to the greenhouse effect have appeared in scientific advisory reports since at least the 1960s. Until recently, these proposals generally remained outside the mainstream discussions of climate policy, which focused either on strategies to reduce emissions or adapt to climate change impacts. However, there is growing concern among many scientists that the lack of progress on emissions reductions will lead to gradual increases in atmospheric concentrations of CO beyond a threshold that could prevent substantial impacts to human health and environmental systems. Furthermore, there is also concern about the existence of “tipping points,” where the earth’s climate system reaches a threshold that unexpectedly results in abrupt and severe changes. One example would be the rapid collapse of the West Antarctic Ice Sheet—which would lead to a large and sudden contribution to sea level rise. These concerns have led to increased interest in geoengineering as a potential tool to help reduce the impacts of climate change, although the NRC study noted that few, if any, individuals are promoting geoengineering as a near-term alternative to emissions reductions. While both CDR and SRM are intended to reduce global temperatures, there are substantial differences in how CDR and SRM operate on the climate system, the timescales required for results, and their associated risks and trade-offs. Consequently, CDR and SRM are often discussed separately. The Royal Society identified several CDR approaches that would directly remove CO in the ocean or on land. Examples of ocean-based CDR approaches include: Enhanced removal by physical processes. Enhanced upwelling/downwelling—altering ocean circulation patterns to bring deep, nutrient rich water to the ocean’s surface (upwelling), to promote phytoplankton growth—which removes CO-rich water from the surface of the ocean to the deep-sea (downwelling). Enhanced removal by biological processes. Ocean fertilization— introducing nutrients such as iron, phosphorus, or nitrogen to the ocean surface to promote phytoplankton growth. The phytoplankton remove CO is transported to the deep ocean as detritus. Enhanced removal by chemical processes. Ocean-based enhanced weathering— accelerating chemical reactions between certain minerals and CO to a nongaseous state. Methods include adding chemically reactive alkaline minerals, such as limestone or silicates, to the ocean to increase the ocean’s natural ability to absorb and store COExamples of land-based CDR approaches include: Physical removal by industrial processes. Direct air capture— technology-based processing of ambient air to remove CO from the atmosphere. The resulting stream of pure CO Biomass energy with CO capture and geological sequestration— harvesting vegetation and using it as a fuel source with capture and storage of the resulting emissions in geological formations (geological sequestration). Biomass for sequestration—harvesting of vegetation and sequestering it as organic material by burying trees or crop wastes, or as charcoal (biochar). Afforestation and land-use management—the planting of trees on lands that historically have not been forested, or otherwise managing vegetation cover to maximize CO, which convert the CO into a geological formation of reactive minerals. The Royal Society identified several SRM approaches that would reflect a small percentage of incoming sunlight back to space, as shown in figure 2. SRM approaches are generally discussed in terms of which sphere they would act upon—space, the atmosphere, or the earth’s surface. Examples of SRM approaches include: Space-based methods. Reflecting or deflecting incoming solar radiation using space-based shielding materials, such as mirrors. Atmosphere-based methods. Stratospheric aerosol injection—injecting reflective aerosol particles into the stratosphere to scatter sunlight back into space. Although it is possible that a wide range of particles could serve this purpose, most attention has been on sulfur particles—partly because temporary global cooling has been produced in the past by volcanic eruptions. Cloud-brightening—adding sea salt or other cloud condensation surfaces to low-level marine clouds to increase their ability to reflect sunlight before it reaches the earth’s surface. Surface-based methods. Increasing the reflectivity of the earth’s land or ocean surfaces through activities such as painting roofs white, planting more reflective crops or other vegetation, or covering desert or ocean surfaces with reflective materials. According to the NRC and Royal Society studies, geoengineering is one of several potential tools to limit the impact and consequences of climate change. However, these studies state that geoengineering is a potential complement to, rather than a substitute for, sharp reductions in greenhouse gas emissions. For example, while geoengineering includes a range of approaches with varying degrees of potential effectiveness and consequences—no geoengineering approach can provide an easy or risk- free alternative solution to the problem of climate change, according to the Royal Society study. For example, compared to current CDR proposals, using SRM to divert incoming sunlight would relatively quickly produce a cooling effect to counteract the warming influence of increased atmospheric concentrations of greenhouse gases. However, SRM does not address the rising atmospheric concentration of greenhouse gases produced by human activity, and therefore would not reduce other serious climate change impacts such as ocean acidification. Furthermore, according to these studies, both CDR and SRM involve additional environmental risks or other trade-offs. For example, ocean- based CDR approaches, such as ocean fertilization, could have unanticipated negative impacts on ocean ecosystems. Additionally, the large-scale deployment of certain land-based CDR approaches, such as afforestation, land-use management for sequestration, and biomass for energy or burial, create trade-offs for land use. In general, the Royal Society study found that compared to CDR, most SRM approaches are associated with a higher risk of negative environmental effects, such as negative impacts on regional temperature or precipitation. For example, one study found that combining a reduction of incoming solar radiation with high levels of atmospheric CO could have substantial impacts on regional precipitation—potentially leading to reductions that could create droughts in some areas. Additionally, the Royal Society study said that increasing the reflectivity of desert or ocean surfaces could have major impacts on desert or ocean ecosystems. Moreover, this study indicated that the artificial balance between increased greenhouse gas concentrations and reduced solar radiation created by large-scale deployment of an SRM approach would need to be maintained over decades and possibly centuries or longer. Experts said that geoengineering is an emerging field, with relatively few experiments or other studies conducted and with major uncertainties remaining. We found that more is known about certain CDR approaches, since related laboratory and field experiments have been conducted, whereas there is limited understanding of other CDR approaches and SRM. Moreover, major uncertainties remain regarding the scientific, legal, political, economic, and ethical implications of researching or deploying geoengineering. : Update and Recommendations (Paris: 2007). energy requirements, and scalability. Similarly, the Royal Society study found that both land- and ocean-based enhanced weathering CDR approaches could potentially store a large amount of carbon, but face barriers to deployment such as scale, cost, and possible environmental consequences. This report also found that while some other land-based CDR approaches—such as afforestation, land-use management techniques, and biomass for energy or burial—can remove CO on a global scale is low. Other CDR approaches have been the focus of relatively few laboratory and field experiments, and fundamental questions remain about their potential efficacy. For example, according to the Royal Society and NRC studies, while ocean fertilization has received some sustained research activity, its potential to remove CO For example, according to researchers who designed a 2009 joint German and Indian iron fertilization experiment, their experiment was designed to test a range of scientific hypotheses pertaining to the structure and functioning of Southern Ocean ecosystems and their potential impact on global cycles of biologically-generated elements, such as carbon and nitrogen. Furthermore, these researchers noted that future long-term experiments to study phytoplankton blooms and their effect on the deep ocean and underlying sediments would have to be much larger than experiments to date. According to our review of relevant studies and expert interviews, understanding of SRM is more limited than that of CDR because there have been few laboratory experiments, field experiments, or computer modeling efforts. Two of the most frequently discussed SRM approaches are stratospheric aerosol injection and cloud-brightening, according to many of the scientific experts we spoke with. For stratospheric aerosol injection, some of the experts said that research to date consisted primarily of a few modeling analyses. They also said that more work would need to be done to assess whether this approach could reduce incoming solar radiation without serious consequences. For example, one study identified the potential for regional impacts on precipitation— potentially leading to drought in some areas. Based on our literature review and interviews with experts, to date only one study has been published for a field experiment related to SRM technologies—a 2009 Russian experiment that injected aerosols into the middle troposphere to measure their reflectivity. Similarly, in the case of cloud-brightening, several experts said that there currently is not enough research to assess its effectiveness or impacts. According to the 2010 NRC study, other methods for SRM, including using space-based reflectors and increasing the solar reflectivity of buildings or plants, have limited potential, either due to the cost of deployment or the limited potential to affect the climate. Experts we interviewed and relevant studies identified several major uncertainties in the field of geoengineering that are in need of further investigation. These uncertainties ranged from important scientific questions for CDR and SRM, to political, ethical, and regulatory issues. Areas that merit further investigation include: Technical feasibility and effectiveness of SRM and certain CDR approaches. Experts we interviewed and the Royal Society and NRC studies agreed that SRM approaches generally were not researched sufficiently to be considered well-understood or technically feasible. Additionally, questions remain regarding the effectiveness of certain CDR approaches, such as ocean fertilization and some land-based methods, to significantly reduce atmospheric concentrations of CO on a global scale, or sequester CO The Royal Society study also noted that large-scale deployment of CDR approaches, such as methods requiring substantial mineral extraction—including land- or ocean-based enhanced weathering— may have unintended and significant impacts within and beyond national borders. For example, the study noted that impacts from enhanced weathering approaches could include localized environmental damage caused by increased mineral extraction activity, as well as changes to soil and ocean surface water pH that could affect vegetation and marine life. Several of the experts that we spoke with agreed that potential unintended consequences of geoengineering approaches require further study. Better understanding of the climate and a way to determine when a “climate emergency” is reached. The NRC study recommended additional basic climate science research, including (1) improved detection and attribution of climate change to distinguish the effects of intentional intervention in the climate system from other causes of climate change, and (2) information on climate system thresholds, reversibility, and abrupt changes to inform societal debate and decision-making over what would constitute a “climate emergency” and whether deployment of a geoengineering approach would be merited. How best to regulate geoengineering internationally. Several of the experts we interviewed as well as the NRC study emphasized the potential for international tension, distrust, or even conflict over geoengineering deployment. The NRC study also stated that global- scale geoengineering deployment creates the potential for uneven positive and negative regional outcomes, and this raises questions of decision-making and national security. Further research can help clarify what type of governance might be useful and when, both for deployment and for field experiments that may involve risks of negative consequences. Political, economic, and ethical concerns. Some experts we interviewed and relevant studies said that geoengineering introduces important political, economic, and ethical questions. For example, several experts said that pursuing geoengineering research could unintentionally reduce interest in reducing CO emissions and that social science research would be needed to assess this potential effect. The NRC studies stated that major questions remain regarding the economic viability of certain CDR approaches, such as direct air capture and enhanced weathering. Additionally, one expert raised concerns over the potential economic costs associated with unintended impacts from deploying SRM. Furthermore, NRC reported that public involvement is critical to making decisions about whether to pursue testing and deployment of geoengineering and that research is needed to determine how best to involve the public in such a decision-making process. USGCRP agencies reported funding at least 52 research activities relevant to geoengineering in fiscal years 2009 and 2010. We found that, of these 52 activities, 43 were either related to conventional mitigation strategies or were fundamental scientific research, whereas 9 directly investigated a particular geoengineering approach. We identified approximately $100.9 million in geoengineering-related funding across USGCRP agencies in fiscal years 2009 and 2010, with about $1.9 million of this amount related to research directly investigating a particular geoengineering approach. The other roughly $99 million was related to research concerning conventional mitigation strategies that could be applied directly to a particular geoengineering approach or basic science that could be applied generally to geoengineering. However, there is no coordinated federal strategy or operational definition for geoengineering, so agencies and policymakers may not know the full extent of relevant federal research. The 13 agencies participating in USGCRP identified 52 research activities relevant to geoengineering—accounting for approximately $100.9 million in federal funding for fiscal years 2009 and 2010. Twenty-eight of these activities—funded at approximately $54.4 million—were related to conventional mitigation strategies that are directly applicable to a particular CDR approach, such as enhancing land-based biological removal of CO, according to our analysis. Fifteen of the reported activities—funded at approximately $44.6 million—were fundamental scientific efforts that could be generally applied to geoengineering, such as modeling the interactions between the atmosphere and the climate and basic research into processes to separate gas streams into their individual components, such as CO Of the 43 activities related to fundamental research or mitigation efforts relevant to geoengineering but not designed to address it directly, the Department of Commerce (Commerce) reported the most funding— approximately $41.6 million. This was largely due to the National Oceanic and Atmospheric Administration’s (NOAA) climate modeling and monitoring of biological emissions and absorption of greenhouse gases, which NOAA officials said could be relevant for assessing the impacts and efficacy of various geoengineering approaches. The U.S. Department of Agriculture (USDA), DOE, the Department of the Interior (Interior), and EPA reported similar levels of funding—from about $11.3 million to $13.9 million. These efforts were largely directed at measuring and monitoring carbon sequestration potential in soils and biomass and assessing the impacts and storage potential for geological sequestration of CO. Although these activities are associated with efforts to reduce or offset emissions, agency officials identified them as relevant to certain CDR approaches—such as large-scale afforestation, and direct air capture—based on the working definition we provided. Table 2 summarizes the approximately $99 million in reported funding for the 43 relevant activities related to conventional mitigation efforts and fundamental scientific research, by agency. The National Science Foundation (NSF), DOE, and Commerce were the only agencies that reported funding for activities directly supporting geoengineering research during fiscal years 2009 and 2010. Of these agencies, NSF reported the most funding—approximately $1.1 million— directed to three research activities: a study on the potential impacts of ocean iron fertilization, a study to examine the moral challenges associated with SRM, and a modeling effort investigating stratospheric aerosol injection and space-based SRM approaches. DOE reported funding research—approximately $700,000—for two studies about direct air capture technologies, a modeling activity for stratospheric aerosol injection and cloud-brightening, as well as a study investigating the unintended consequences of climate change responses, including CDR and SRM approaches. Commerce reported funding two relevant research efforts—for about $70,000—examining the unintended impacts of SRM approaches, with one study focused on climate-related impacts and the other study exploring potential effects on solar electricity generation. Table 3 summarizes the approximately $1.9 million in reported funding for the nine relevant activities directly supporting geoengineering research, by agency. During our review, we also found examples of other relevant activities sponsored by USGCRP agencies that were outside the scope of our data request, mostly because they occurred prior to 2009. These activities included: DOE sponsored studies on ocean-based carbon sequestration approaches, such as ocean fertilization and direct injection of CO Additionally, scientists at NASA’s Ames Research Center held a conference on SRM approaches in 2006, in conjunction with the Carnegie Institution of Washington. NASA also funded atmospheric modeling studies, which were used by independent researchers, in part, to assess the potential impact of stratospheric aerosols on the ozone layer. In 2008, NSF sponsored studies examining the long-term carbon storage potential of soils and the impact of increased nitrogen on biological carbon sequestration. A Department of Defense (DOD) advisory group sponsored a 1-day workshop at Stanford University on geoengineering in 2009; however, DOD officials said that no funded research projects resulted from this workshop. In 2007, EPA funded research relevant to the economic implications of SRM approaches through its National Center for Environmental Economics. Furthermore, federal officials also noted that a large fraction of the existing federal research and observations on basic climate change and earth science could be relevant to improving understanding about proposed geoengineering approaches and their potential impacts. For instance, federal officials said that basic research conducted by USGCRP agencies into oceanic chemistry could help address uncertainty about the potential effectiveness and impacts of CDR approaches, such as ocean fertilization. Similarly, ongoing research conducted by USGCRP agencies related to understanding atmospheric circulation and aerosol/cloud interactions could help improve understanding about the potential effectiveness and impacts of proposed SRM approaches. We found that it was difficult to determine the full extent of federal geoengineering research activities because there is no coordinated federal strategy for geoengineering, including guidance on how to define federal geoengineering activities or efforts to identify and track federal funding related to geoengineering. Officials from federal offices coordinating federal responses to climate change—CEQ, OSTP, and USGCRP—stated that they do not currently have a coordinated geoengineering strategy or position. For example, a USGCRP official stated that there is no group coordinating federal geoengineering research and that such a group is not currently necessary because of the small amount of federal funding involved. However, while USGCRP agencies reported about $1.9 million in funding for activities directly investigating geoengineering, federal officials also told us that a large fraction of the existing federal research and observations on basic climate change and earth science could be relevant to understanding geoengineering. According to the USGCRP’s most recent report to Congress, USGCRP agencies requested roughly $2 billion in budget authority for climate change and earth science activities in fiscal year 2010. Consequently, the actual funding for research that could be applied either generally or directly to understanding geoengineering approaches is likely greater than the roughly $100.9 million reported in response to our data request. However, without the guidance of an operational definition for what constitutes geoengineering or a strategy to capitalize on existing research efforts, federal agencies may not recognize or be able to report the full extent of potentially relevant research activities. For example, some agency officials indicated that, without a clear governmentwide definition, in their determination of which federal research activities were relevant to geoengineering, our data request was subject to different interpretations— particularly for CDR approaches, since there is extensive overlap with existing mitigation efforts. In particular, EPA and USDA officials said that there is a large body of research regarding biological sequestration but that these officials would not consider it to be geoengineering. However, officials from other agencies, such as Interior and DOE, included certain research on biological sequestration as relevant to geoengineering based on the definition we provided. Similarly, we found that from NSF officials’ perspectives, the distinction between existing efforts to develop carbon capture and storage technologies, such as membranes to separate CO from other gases, and geoengineering direct air capture technologies is also not well-defined. This definitional issue is not unique to these agencies. In its recent study Advancing the Science of Climate Change, NRC acknowledged the lack of consensus regarding what constitutes geoengineering in relation to widely accepted practices that remove CO The NRC study included other findings about the nation’s climate change science efforts that may be relevant to a potential federal geoengineering strategy. The study emphasized the importance of providing decision makers with scientific information on a range of available options, including geoengineering, to limit future climate change and its impacts. According to this study, this information would help policymakers use adaptive risk management to update response strategies as new information on climate change risks and response strategies becomes available. NRC recommended an integrative, interdisciplinary research effort to improve understanding of available response options, as well as of climate change and its impacts. The study indicated that this effort should be led by a single coordinating body, and NRC identified USGCRP’s capacity to play a role in such an effort. Similarly, several of the experts we interviewed recommended that federal geoengineering research should be an interdisciplinary effort across multiple agencies and led by a coordinating body, such as OSTP or USGCRP. Our recent work offers insights on key considerations for establishing governmentwide strategies, which could be relevant to a future geoengineering strategy. Specifically, our review of federal efforts related to crosscutting issues, such as climate change adaptation and global food security, highlighted key practices for enhancing collaboration across agencies. These practices include establishing a commonly accepted operational definition for relevant activities; leveraging existing resources to support common outcomes and address identified needs; and developing mechanisms to monitor, evaluate, and report on results. Furthermore, our review of DOE’s FutureGen project—a program to help build the world’s first coal-fired, zero-emissions power plant—identified important factors to consider when developing a strategy for technology- based research. Specifically, we found that it is important to comprehensively assess the costs, benefits, and risks of each technological option and to identify potential overlap between proposed and existing programs. For example, the NRC study acknowledged the importance of improving understanding of SRM and its consequences, without replacing or reducing existing research on climate change science or other approaches to limiting climate change or adapting to its impacts. As the study noted, much of the research needed to advance scientific understanding of SRM, such as studying the climate effects of aerosols and cloud physics, is also necessary to advance understanding of the climate system, and could therefore contribute more broadly to climate change science. Similarly, an OSTP official said that ongoing fundamental research to investigate the relationship of cloud/aerosol interactions could also be applied to improve understanding of certain SRM approaches. In the absence of a coordinated federal strategy for geoengineering, decisions about whether a particular research activity is relevant to geoengineering may not necessarily be consistent across the federal government. In addition, agencies generally do not collect and share information on such research activities in the context of geoengineering. While EPA officials told us that certain agencies, such as EPA, State, and NOAA, share information about ocean fertilization and direct injection of CO into deep sub-seabed geological formations as part of a working group for international regulation of the ocean, a USGCRP official said there is no working group to share information or coordinate geoengineering research more broadly, because such an action would require a decision from the administration to pursue geoengineering research on a larger scale. However, without a coordinated effort to identify relevant research and share information across agencies, policymakers and agency officials may lack key information needed to inform their decisions on geoengineering research. For example, if policymakers and officials do not know the full extent of the relevant federally funded research that is under way, they may not have sufficient information to leverage existing research on climate change science to also improve understanding of geoengineering. Legal experts we interviewed and EPA and Department of State (State) officials said that the extent to which existing laws and international agreements apply to geoengineering is unclear, largely because detailed information on geoengineering approaches and effects is not available. EPA has taken steps to regulate one CDR approach and has determined that an existing law provides sufficient authority to regulate two other approaches. EPA officials provided their preliminary thoughts on how other laws might apply to geoengineering activities. However, according to EPA officials, they have not fully assessed (1) whether the agency has the authority to regulate or (2) how to regulate most geoengineering approaches, because the research is still in its initial stages. Similarly, legal experts and State officials stated that many international agreements could apply to geoengineering; however, most agreements’ applicability is unclear because they were not intended to address geoengineering and parties to the agreements have not determined whether or how the agreements should apply to relevant geoengineering activities. This uncertainty and inaction is due, in part, to the limited general understanding of geoengineering and a lack of geoengineering activity. Legal experts and federal officials identified challenges for establishing governance of geoengineering, such as the potential for unintended and uneven impacts, although their views differed on the most effective governance approach. EPA officials stated that the extent to which existing federal environmental laws apply to geoengineering is unclear, largely because detailed information on most geoengineering approaches and effects is not available. However, EPA officials said that there is relatively more information available about geological sequestration of CO from the air and sequester it underground or in the sub-seabed. EPA has taken steps to regulate geological sequestration under the Safe Drinking Water Act, and EPA officials said that the Marine Protection, Research, and Sanctuaries Act of 1972 provides the agency with authority to regulate (1) certain sub-seabed geological sequestration activities, and (2) ocean fertilization activities. Specifically: EPA has authority under the Safe Drinking Water Act to regulate underground injections of various substances and is using this authority to develop a rule to govern the underground injection of CO captured from an emission source, such as a power plant or industrial facility, the rule’s definition of geological sequestration is broad enough to include long-term sequestration of CO injection and geological sequestration, which is scheduled to be finalized in the fall of 2010. Under the Marine Protection, Research and Sanctuaries Act of 1972, as amended, certain persons are generally prohibited from dumping material, including material for ocean fertilization, into the ocean without a permit from EPA. EPA officials said that certain sub-seabed geological sequestration of CO Resource Conservation and Recovery Act of 1976 (RCRA). RCRA regulates the management of hazardous waste from generation of the waste to its disposal. An EPA official stated that EPA has been examining questions of RCRA’s applicability to geological sequestration of CO and is currently considering a proposed rule to clarify how RCRA hazardous waste requirements apply in that context. This official also noted that RCRA’s applicability to other geoengineering approaches where materials are applied to the land or oceans would depend on whether there was intent to discard the materials and whether the materials are a hazardous waste. Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA). CERCLA authorizes EPA to clean up hazardous substance releases at contaminated sites and then seek reimbursement from the parties responsible for contaminating them or compel the responsible parties to clean up these sites. Responsible parties include current and former site owners and operators, as well as those who transport or arrange for the disposal of the hazardous substances. Although a stream of pure CO streams could contain hazardous substances, thus subjecting the parties injecting the CO enters groundwater, it might also cause hazardous substances, such as some metals, to be dissolved by the groundwater from enclosing strata. If that constitutes a release of hazardous substances from a “facility,” such as the strata, then the owner of that facility could be liable for any cleanup costs caused by that release. This official was not aware of CERCLA’s applicability to any other geoengineering activity. Clean Air Act. This law authorizes EPA to regulate emissions of certain air pollutants from mobile and stationary sources into the ambient air, including those that destroy the stratospheric ozone layer. EPA officials said that the act could apply to geoengineering activities that emitted air pollutants into the atmosphere—either as the purpose of the activity or as a side effect—depending on where the pollutant was released and the delivery mechanism. Officials also noted that although the act regulates emissions into the ambient air, substances injected into the upper atmosphere that eventually cycle down to ground level could also be subject to regulation, depending on the definition of ambient air. EPA officials stated that they would require further information on the specific technology and activity to determine exactly how the law might apply. In addition, EPA and DOE officials noted that geoengineering activities undertaken, funded, or authorized by federal agencies would be subject to the National Environmental Policy Act of 1969 (NEPA), the Endangered Species Act, and the conformity provision of the Clean Air Act. NEPA requires federal agencies to evaluate the likely environmental effects of certain major federal actions using an environmental assessment or, if the projects likely would significantly affect the environment, a more detailed environmental impact statement. Under the Endangered Species Act, geoengineering activities taken or authorized by federal agencies would require consultation among federal agencies, including the Fish and Wildlife Service and NOAA, to ensure that the activity is not likely to jeopardize the continued existence of any endangered or threatened species or adversely modify habitat critical for the species. Under the Clean Air Act’s conformity provision, no federal agency may approve or provide financial assistance for any activity that does not conform with a state implementation plan, which is a plan required by the act to ensure that national ambient air quality standards are met. Acknowledging the lack of an existing international agreement that comprehensively addresses geoengineering, State officials and legal experts we interviewed said that many agreements could perhaps apply to a geoengineering activity and its impacts, depending on the activity’s nature, location, and actors. For example, some international agreements with broad scopes, such as the United Nations Framework Convention on Climate Change, could apply generally to geoengineering activities, whereas other agreements specifically addressing the atmosphere, oceans, and space could apply only if the activity occurred in or impacted that particular area. However, international agreements legally bind only those countries that have become parties to the particular agreement. Therefore, the number of parties to a particular agreement determines, in part, where the agreement applies, and countries that are not parties are not legally bound to abide by the agreement. Table 4 summarizes certain agreements identified by legal experts and relevant studies as potentially applicable to geoengineering and the number of parties to a particular agreement. Almost all the legal experts and State officials we spoke with noted that, of all the potential geoengineering approaches, sub-seabed geological sequestration of CO Ocean fertilization. The parties to the London Convention and the London Protocol have decided that the scope of these agreements includes ocean fertilization activities for legitimate scientific research and that ocean fertilization activity other than legitimate scientific research should be considered contrary to the aims of the agreements and should not be allowed. The treaties’ scientific bodies are developing an assessment framework for countries to use in evaluating whether research proposals are legitimate scientific research. Additionally, the parties to the Convention on Biological Diversity issued a decision in 2009 urging countries to ensure that ocean fertilization activities, except for certain small-scale scientific research within coastal waters, do not take place until there is an adequate scientific basis on which to justify them and a global, transparent, and effective control and regulatory mechanism in place. The parties to the London Convention and London Protocol are considering an additional resolution or amendment concerning ocean fertilization, and the parties to the Convention on Biological Diversity continue to discuss the issue. Geological sequestration. In 2006, the parties to the London Protocol agreed to amend the protocol to include, in certain circumstances, CO streams for sequestration in sub-seabed geological formations in the protocol’s list of wastes and other matter that could be dumped. Under the amendment, CO stream consists overwhelmingly of CO(3) no wastes or other matter are added for the purpose of disposing of those wastes or other matter. The parties also developed specific guidelines for countries to use when assessing whether applications for disposal of streams into sub-seabed geological formations is consistent with the protocol. In late 2009, the parties to the London Protocol adopted an amendment to allow the export of CO streams for disposal in certain circumstances. The parties are developing specific guidance for these exports and issues related to the management of transboundary movement of CO sequestration in sub-seabed geological formations. However, legal experts and State officials stated that although parties to three agreements have taken action to clarify the agreements’ applicability to particular geoengineering approaches, most agreements’ applicability is unclear because they were not intended to address geoengineering and the parties had not yet addressed the issue. In addition, legal experts and federal officials generally said that more detailed information on geoengineering approaches and their effects would be needed for officials to develop a regulatory and governance framework. For example, aside from ocean fertilization and other marine-focused geoengineering approaches, State officials said that many of the ideas remain too theoretical and distant from implementation to consider addressing them through international law. Legal experts and EPA and State officials identified various challenges to establishing domestic and international governance of geoengineering. For example, the legal experts and EPA and State officials we interviewed generally agreed that there needs to be further research on most geoengineering approaches and their potential effects to inform—and in federal officials’ views to warrant—discussions regarding regulation. Similarly, some of these experts and federal officials said that a general lack of significant efforts to pursue geoengineering is a contributing factor to why geoengineering governance has not been pursued further to date. For example, a State official noted that geoengineering has not received much attention within international negotiations related to climate change, and there isn’t enough geoengineering-related activity to drive interest in expanding international governance at this time. Legal experts and State officials had differing views about an international governance framework for geoengineering. Specifically, several legal experts recommended including all geoengineering activities with transboundary impacts in a single comprehensive agreement. Some of these experts said an existing comprehensive international agreement could be adapted to address geoengineering. Some of them specifically identified the United Nations Framework Convention on Climate Change as an appropriate agreement because it addresses climate change and geoengineering is intended to diminish climate change or its impacts. Other legal experts said a new international agreement was needed because of the difficulty reaching consensus within the United Nations Framework Convention on Climate Change. Experts in favor of a single comprehensive agreement said that it would be preferable to the patchwork of existing agreements, which were not designed to address geoengineering, because these agreements do not create comprehensive governance frameworks that could be used to address geoengineering. Additionally, some experts said that certain existing agreements rely on the parties to regulate activities under their jurisdiction without the international community’s participation in decision-making, which may not be the best structure for regulating geoengineering research or deployment. State officials we interviewed said that it would be better to rely on existing treaties to the extent they are adequate and appropriate and consider developing new international instruments if needed, since there is limited knowledge and practice of geoengineering. State officials said this approach would enable greater rigor and flexibility than trying to address all geoengineering activities within a single comprehensive agreement. They cited the London Convention and London Protocol as examples. While these agreements might not have addressed ocean fertilization several years ago, the parties took action when ocean fertilization reached a state of development where an agreed approach to regulation was considered necessary, and the agreements now unquestionably address it. In contrast, State officials said that parties to other agreements have not addressed other geoengineering approaches because they have not reached a similar stage of development. State officials said it was hard to imagine a single agreement appropriately covering geoengineering activities with all potential transboundary effects. State officials also said that while some countries have called for a broader inquiry into marine geoengineering more generally under the London Convention and the London Protocol, the parties deemed those calls premature at best. Legal experts and EPA officials we interviewed generally agreed that the federal government should take a coordinated, interagency approach to domestic geoengineering regulation. For example, the legal experts who spoke about domestic regulation generally agreed that the federal government should play a role in governing geoengineering research— either by developing research norms and guidelines or applying existing regulations and guidelines. One expert noted that it was important that regulators stay abreast of research on the most mature technologies so that the regulatory framework would be in place prior to field experiments. Some experts and EPA officials also agreed that because there is a wide variety of geoengineering activities, research and regulation would fall under multiple agencies’ purview and expertise. For example, one expert said that there should be a coordinated interagency effort led by OSTP or USGCRP. Another said that the federal government should focus on a comprehensive policy for climate change, including geoengineering, and that that policy would determine what new regulations would be necessary to guide and govern research. EPA and State officials both said that agencies such as NOAA, NASA, and DOE should be involved in regulatory discussions due to their jurisdictional or scientific expertise. As an example, EPA officials noted that the Interagency Task Force on Carbon Capture and Storage, co-chaired by DOE and EPA, was created to propose a plan to overcome the barriers to widespread deployment of these technologies, which include geologic sequestration. The plan addresses, among other issues, how to coordinate existing administrative authorities and programs, legal barriers to deployment, and identifies areas where additional statutory authority may be necessary. Legal experts we interviewed generally agreed that governance for geoengineering research should be addressed separately from governance for deployment of geoengineering approaches. For example, experts said that discussions of governance of deployment were premature, and one expert cautioned that discussing deployment could raise the level of controversy surrounding the subject, leading to a general gridlock that could disrupt discussions about research and lower interest in a coordinated and transparent approach. Both State and EPA officials cited the need for further research into geoengineering prior to engaging in discussions of domestic regulation or a governance framework at the international level. State officials said that, in practice, the United States and other countries have already effectively separated geoengineering research and deployment governance for ocean fertilization under the London Convention and London Protocol, because the parties decided that any ocean fertilization activities other than those for legitimate scientific research should not be allowed at this time. However, the legal experts we spoke with also agreed that some type of regulation of geoengineering field research was necessary in the near future, particularly for those approaches where large-scale experiments could have transboundary impacts. According to these experts, any framework governing research should include several elements, such as transparency, coordination, flexibility, a review process for experiments, the use of environmental risk thresholds, and an emphasis on modeling prior to field studies. A few legal experts said that these elements could start as voluntary norms and guidelines within the research community and then evolve into formal regulations prior to field trials. As one expert said, transparent decision-making and guidelines are necessary to ensure that research does not pose unacceptable risk to the environment. State officials said that, generally, the United States supports careful consideration of research implications rather than a full ban on research. In addition, they said that some geoengineering research could be fostered most effectively through international cooperation and coordination rather than governance, or that domestic regulation is more appropriate than international regulation. Legal experts and EPA and State officials cited other challenges related to geoengineering governance, particularly for those approaches with uneven or unintended environmental effects. For example, some legal experts said that controversy surrounding certain geoengineering approaches, as well as a lack of understanding and acceptance, could make domestic and international governance difficult. In addition, State officials said that if large-scale experiments or activities have unknown consequences or effects borne by nations other than the nation conducting the experiment or activity, this could risk undermining existing agreements on climate change strategies. Furthermore, legal experts and EPA officials agreed that liability for unintended consequences was an important issue that would need to be addressed. Specifically, one expert suggested that there should be a mechanism to compensate individuals or nations for damages resulting from geoengineering activities. Moreover, some legal experts were concerned about the ability of parties to enforce certain international agreements related to geoengineering. Major scientific bodies such as the NRC and Royal Society have identified geoengineering as one of several potential tools to limit the impact and consequences of climate change. However, these bodies have stated that geoengineering is a potential complement to, rather than a substitute for, sharp reductions in greenhouse gas emissions. While the NRC and Royal Society have identified geoengineering as a potential tool, what role geoengineering might play in a domestic and international response strategy will likely be shaped by resolving unanswered scientific questions surrounding the technical feasibility, unintended consequences, effectiveness, cost, and risks associated with each approach. Answers to these questions will also inform the public debate concerning whether geoengineering is an acceptable response given the ethical and social implications of deliberate interventions in the earth’s climate system. The federal government is already engaging in research that could help address some of the uncertainties surrounding geoengineering and inform policy decisions about research priorities. While agencies identified about $100.9 million in research funding relevant to geoengineering in fiscal years 2009 and 2010, federal officials also said that a substantial portion of the existing federal climate change and earth science research could be relevant to understanding geoengineering—roughly $2 billion in requested budget authority for 2010 alone. However, because there is no coordinated federal geoengineering strategy, it is difficult to determine the extent of relevant research. At present, while some agencies are sharing information on two geoengineering approaches to inform negotiations relevant to international regulation of ocean dumping and address barriers to geological sequestration as a mitigation strategy, agencies generally are not collecting and sharing information more broadly on research relevant to other geoengineering approaches. Without a definition of geoengineering for agencies to use, and without coordination among agencies to identify the full extent of available research efforts relevant to geoengineering as well as to identify research priorities, policymakers and agency officials may lack sufficient information to leverage existing research resources to their full benefit. In turn, this lack of information may hinder policy decisions and governance at the domestic and international level. Even if policymakers decide that geoengineering should not be pursued domestically, knowledge of geoengineering approaches and their potential effects will be essential to inform international negotiations regarding other countries’ consideration of, or actions related to, geoengineering research and deployment. GAO recommends that the appropriate entities within the Executive Office of the President (EOP), such as the Office of Science and Technology Policy (OSTP), in consultation with relevant federal agencies, develop a clear, defined, and coordinated approach to geoengineering research in the context of a federal strategy to address climate change that (1) defines geoengineering for federal agencies; (2) leverages existing resources by having federal agencies collect information and coordinate federal research related to geoengineering in a transparent manner; and if the administration decides to establish a formal geoengineering research program, (3) sets clear research priorities to inform decision-making and future governance efforts. We provided a draft of this report to the Office of Science and Technology Policy (OSTP) within the Executive Office of the President (EOP) for review and comment. OSTP also circulated the report to the 13 participating USGCRP agencies. In response to the draft, OSTP, the Council on Environmental Quality, U.S. Department of Agriculture (USDA), Department of State (State), National Oceanic and Atmospheric Administration (NOAA), and National Science Foundation (NSF) neither agreed nor disagreed with our findings and recommendation; rather, they provided technical and other comments, which we incorporated as appropriate. General comments and our response are summarized below. In their comments, USDA, NSF, and OSTP raised various concerns about how geoengineering should be defined. For example, OSTP and USDA cited concerns that the definition used in this report is too broad because it overlaps with certain land-based practices, such as biological sequestration of CO emissions. For the purposes of this report, we used the Royal Society study’s definition and descriptions of geoengineering approaches because this study was the most comprehensive review of geoengineering science available at the time of our request. Other scientific organizations, such as the National Research Council (NRC), the American Meteorological Society, and the American Geophysical Union have also either reported on or issued position statements regarding geoengineering, and used a similarly broad definition. However, as we note in the report, discussions about how to define geoengineering and what activities should be considered geoengineering remain active. Variations in agencies’ interpretation of our data request, as well as the comments noted above, support our recommendation that additional clarity and guidance regarding the federal approach to geoengineering is needed, and that further discussion of what types of activities should be included in a federal operational definition of geoengineering may be warranted. Accordingly, we recommended that the appropriate entities within the EOP consult with the relevant federal agencies to develop a clear, defined, and coordinated approach to geoengineering research in the context of a federal strategy to address climate change. Additionally, NOAA and NSF noted that because the global nature of climate change requires an international response, international coordination and collaboration would be important for geoengineering activities and oversight efforts. As we noted in our report, the applicability of international agreements to geoengineering remains unclear; however, parties to three agreements have issued decisions regarding the agreements’ applicability to ocean fertilization and sub-seabed geological sequestration. Furthermore, the legal experts we spoke with generally agreed that some type of regulation of geoengineering field research is necessary in the near future, particularly for those approaches where large-scale experiments could have transboundary impacts. According to these experts, any framework governing research should include several elements, such as transparency, coordination, flexibility, a review process for experiments, the use of environmental risk thresholds, and an emphasis on modeling prior to field studies. NOAA emphasized the importance of fully understanding unintended consequences and risks associated with geoengineering approaches. In particular, NOAA commented that sufficient resources should be directed specifically towards identifying possible unintended consequences and risks. As we note in the report, relevant studies indicate that there are additional environmental risks and trade-offs associated with both CDR and SRM approaches. Furthermore, our discussions with experts and review of relevant studies identified unintended consequences associated with geoengineering approaches as a key uncertainty requiring further study. In addition to these comments, CEQ, OSTP, and the agencies provided technical changes and corrections which we incorporated where appropriate. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies of this report to the appropriate congressional committees, the Office of Science and Technology Policy within the Executive Office of the President, and other interested parties. In addition, the report will be available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staff members have any questions about this report, please contact Frank Rusco at (202) 512-3841 or ruscof@gao.gov, or John Stephenson at (202) 512-3841 or stephensonj@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in Appendix V. This report examines (1) the general state of the science regarding geoengineering approaches and their potential effects; (2) the extent to which the federal government is sponsoring or participating in geoengineering research or deployment; and (3) the views of legal experts and federal officials about the extent to which federal laws and international agreements apply to geoengineering activities, and associated challenges, if any, to geoengineering governance. To determine the general state of the science regarding geoengineering approaches and their potential effects, we summarized the results of semi- structured interviews with scientific and policy experts as well as the findings from relevant literature. First, we identified 95 potential experts based on five criteria indicating recognition from their peers as geoengineering experts. These criteria included having (1) presented at the Asilomar International Conference on Climate Intervention Technologies, (2) presented at the geoengineering panels held at the American Association for the Advancement of Science 2010 Annual Meeting, (3) served as a witness at one of the three hearings on geoengineering held by the House Science and Technology Committee, (4) recommendations from other recognized experts that we had interviewed during our work for the March testimony for the committee, and (5) participating in smaller panels or working groups that specifically focused on geoengineering. To identify the most active experts in the field, we scored the experts from the initial list based on their participation in the five previously noted activities. Based on this process, we selected the 10 highest-scoring experts and contacted them for interviews. We selected 10 experts to ensure we could collect a range of views from experts associated with academia, nongovernmental organizations, and government. To assess potential conflicts of interest, we asked the 10 experts to submit a conflict of interest form. These forms included questions about potential financial or other interests that might bias an expert’s opinions related to the state of geoengineering science. We conducted a content analysis to summarize expert responses and grouped responses into overall themes. The views expressed by experts do not necessarily represent the views of GAO. Not all of the experts provided their views on all issues. In addition to gathering expert views, we selected and reviewed collaborative peer-reviewed studies that addressed geoengineering, such as the National Research Council’s Advancing the Science of Climate Change study as well as the Royal Society’s study Geoengineering and the climate: Science, governance and uncertainty. The Royal Society is the United Kingdom’s national academy of sciences. provided. We then categorized the remaining activities into three broad types: (1) activities related to conventional carbon mitigation efforts that are directly applicable to a proposed geoengineering approach, although not designated as such; (2) activities related to improving basic scientific understanding of earth systems, processes, or technologies that could be applied generally to geoengineering; and (3) activities designed specifically to address a proposed geoengineering approach that does not overlap with a conventional carbon mitigation strategy. We sent the results of our analysis and categorization of agency-reported activities to each agency for their review and verification in July 2010. Specifically, we asked agency officials to ensure that the data were complete and accurate, and that our categorization of the data was appropriate. Each agency verified our analysis. In addition, we met with officials and staff from interagency bodies coordinating federal responses to climate change, including the Office of Science and Technology Policy (OSTP), Council on Environmental Quality (CEQ), and USGCRP, as well as the Department of Energy (DOE), which coordinates the Climate Change Technology Program—a multiagency research and development program for climate change technology. We assessed the reliability of the data and found the data to be sufficiently reliable for the purposes of this report. , (5) water desalinization project using solar energy, (6) internationally collaborating with China to foster emissions mitigation research, (7) developing technology to facilitate the conversion of methane gas to liquid fuel, and five activities to develop technologies related to biofuels. Based on their description, we determined that these 12 activities did not appear relevant to identified CDR or SRM approaches. each expert from the initial list based on the three criteria noted above. Based on this process, we selected the 8 highest scoring experts and contacted them for interviews. We selected 8 experts because the scoring process created a natural break between the 8 highest scoring experts and the remaining experts. To assess potential conflicts of interest, we asked each expert to submit a conflict of interest form. These forms included questions about potential financial or other interests that might bias an expert’s opinions related to the applicability of federal laws and international agreements to geoengineering. We conducted a content analysis to summarize expert responses and grouped responses into overall themes. The views expressed by experts do not necessarily represent the views of GAO. Not all of the experts provided their views on all issues. We also met with federal officials from the Environmental Protection Agency (EPA) and the Department of State (State) to collect their views on the applicability of domestic laws and international agreements to geoengineering, and governance challenges, if any. In addition to gathering experts’ and federal officials’ views, we selected and reviewed collaborative reports that addressed geoengineering governance, such as the Royal Society’s study Geoengineering and the climate: Science, governance and uncertainty, and the United Kingdom House of Commons Science and Technology Committee report The Regulation of Geoengineering, among others. To corroborate the legal information provided to us by our experts, we utilized these collaborative reports as well as select articles from relevant journals to support specific key details from the interviews. We conducted this performance audit from December 2009 through September 2010 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. We identified and selected scientific and policy experts to provide their views on the general state of the science regarding geoengineering approaches and their potential effects. We also identified and selected legal experts to provide their views on the applicability of federal laws and international agreements to geoengineering, and associated challenges, if any, to geoengineering governance. This appendix lists the experts we selected and contacted for interviews. In two cases, experts we contacted did not participate in our review, either due to schedule conflicts or because they did not respond to our request. To help federal officials identify relevant activities, we provided them with a data collection instrument that defined geoengineering and described proposed geoengineering approaches, as outlined below. The definition and descriptions were based on the Royal Society study—which was the most authoritative review of geoengineering at the time of our data request. This appendix reflects the language and more technical descriptions we provided to the agencies and, as such, will not be an exact match to the more generalized language used to describe these approaches in the background section of this report. We have provided additional explanations of some scientific terms in footnotes to the text. These footnotes were not part of the data collection instrument sent to the agencies. Cloud condensation nuclei are small particles in the air that become surfaces on which water vapor can condense and form cloud droplets. Sources of cloud condensation nuclei can be both natural and human-caused. Natural sources of cloud condensation nuclei include volcanic dust, sea spray salt, and bacteria. Humans also release unnatural chemicals into the air from the burning of fossil fuels and from industrial sources. In response to our data collection instrument, the 13 agencies participating in the USGCRP reported the following research activities relevant to geoengineering. Our request was limited to activities funded during fiscal years 2009 and 2010; however, in some cases, reported activities were initiated prior to fiscal year 2009 and continued beyond fiscal year 2010, as noted in the “Dates of research” column in tables 5, 6, and 7. To be consistent with the tables in the report, the activities are organized by agency and geoengineering approach. According to agency officials, none of the activities listed below received funding in the American Recovery and Reinvestment Act of 2009. The Departments of Health and Human Services and State, as well as the U.S. Agency for International Development, the National Aeronautics and Space Administration, and the Smithsonian Institution, all reported no relevant activities during fiscal years 2009 and 2010. In addition to the contacts named above, Tim Minelli (Assistant Director), Ana Ivelisse Aviles, Judith Droitcour, Lorraine Ettaro, Cindy Gilbert, Eli Lewine, Madhav Panwar, Timothy Persons, Katherine Raheb, Benjamin Shouse, Jeanette Soares, Ardith Spence, Kiki Theodoropoulos, J. D. Thompson, and Lisa Van Arsdale made key contributions to this report. | Policymakers have raised questions about geoengineering--large-scale deliberate interventions in the earth's climate system to diminish climate change or its impacts--and its role in a broader strategy of mitigating and adapting to climate change. Most geoengineering proposals fall into two categories: carbon dioxide removal (CDR), which would remove carbon dioxide (CO2) from the atmosphere, and solar radiation management (SRM), which would offset temperature increases by reflecting sunlight back into space. GAO was asked to examine (1) the state of geoengineering science, (2) federal involvement in geoengineering, and (3) the views of experts and federal officials about the extent to which federal laws and international agreements apply to geoengineering, and any governance challenges. GAO examined relevant scientific and policy studies, relevant domestic laws and international agreements, analyzed agency data describing relevant research for fiscal years 2009 and 2010, and interviewed federal officials and selected recognized experts in the field. Few geoengineering experiments or modeling studies have been conducted, and major uncertainties remain on the efficacy and potential consequences of geoengineering approaches. GAO's review of relevant studies and discussions with selected experts indicated that relatively more laboratory and field research relevant to certain CDR approaches exists, although most of this research was not designed to apply to geoengineering. In contrast, few modeling studies or field experiments have focused on SRM approaches, according to experts and recent studies. Experts identified only one SRM field experiment with published results--a 2009 Russian experiment that injected aerosols into the middle troposphere to measure their reflectivity. Experts, as well as relevant studies, identified several major uncertainties in need of further investigation for CDR and SRM. Federal agencies identified 52 research activities, totaling about $100.9 million, relevant to geoengineering during fiscal years 2009 and 2010. GAO's analysis found that 43 activities, totaling about $99 million, focused either on mitigation strategies or basic science. Most of the research focused on mitigation efforts, such as geological sequestration of CO2, which were identified as relevant to CDR approaches but not designed to address them directly. GAO found that nine activities, totaling about $1.9 million, directly investigated SRM or less conventional CDR approaches. Officials from interagency bodies coordinating federal responses to climate change indicated that their offices have not developed a coordinated strategy, and believe that, due to limited federal investment, it is premature to coordinate geoengineering activities. However, federal officials also noted that a large share of existing federal climate science research could be relevant to geoengineering. Agencies requested roughly $2 billion for such activities in fiscal year 2010. Without a coordinated federal strategy for geoengineering, it is difficult for agencies to determine the extent of relevant research, and policymakers may lack key information to inform subsequent decisions on geoengineering and existing climate science efforts. According to legal experts and federal officials, the extent to which federal laws and international agreements apply to geoengineering is unclear. The Environmental Protection Agency (EPA) has taken steps to regulate one CDR approach and has determined that it has sufficient authority to regulate two other approaches. EPA officials said EPA has not assessed the applicability of other laws because geoengineering research is in its initial stages. Similarly, legal experts and Department of State officials said that, except for three instances, parties to international agreements have not addressed their agreements' applicability to geoengineering, largely due to limited geoengineering activity and awareness of the issue. Legal experts' and officials' views differed on the best approach for international governance, but generally agreed that the federal government should take a coordinated, interagency approach on domestic regulation. Experts and officials also identified governance challenges, such as the need to address liability. GAO recommends that within the Executive Office of the President, the appropriate entities, such as the Office of Science and Technology Policy (OSTP), establish a clear strategy for geoengineering research in the context of the federal response to climate change to ensure a coordinated federal approach. OSTP neither agreed nor disagreed with our recommendation, but provided technical comments. |
The 1935 Social Security Act mandated coverage for most workers in commerce and industry, which at that time comprised about 60 percent of the workforce. State and local government employees were excluded because they had their own retirement systems and there was concern over the question of the federal government’s right to impose a tax on state governments. Subsequently, the Congress extended mandatory Social Security coverage to most of the excluded groups, including state and local employees not covered by a public pension plan. The Congress also extended voluntary coverage to state and local employees covered by public pension plans. Since 1983, however, public employers have not been permitted to withdraw from the program once they are covered. SSA estimates that 96 percent of the workforce, including 70 percent of the state and local government workforce, is now covered by Social Security. Social Security provides retirement, disability, and survivor benefits to insured workers and their dependents. Insured workers are eligible for full retirement benefits at age 65 and reduced benefits at age 62. Social security retirement benefits are based on the worker’s age and career earnings, are fully indexed for inflation after retirement, and replace a relatively higher proportion of the final year’s wages for low earners. Social Security’s primary source of revenue is the Old Age, Survivors, and Disability Insurance portion of the payroll tax paid by employers and employees. The payroll tax is 6.2 percent of earnings each for employers and employees, up to an established maximum. SSA estimates that 5 million state and local government employees, excluding students and election workers, are not covered by Social Security. SSA also estimates that annual wages for noncovered employees total about $132.5 billion. Seven states—California, Colorado, Illinois, Louisiana, Massachusetts, Ohio, and Texas—account for more than 75 percent of the noncovered payroll. A 1995 survey of public pension plans found that police, firefighters, and teachers are more likely to occupy noncovered positions than other employees. specified benefit rate for each year of service and the member’s final average salary over a specified time period, usually 3 years. For example, plans with a 2-percent rate replace 60 percent of a member’s final average salary after 30 years of service. In addition to retirement benefits, a 1994 Department of Labor survey found that all members have a survivor annuity option, 91 percent have disability benefits, and 62 percent receive some cost-of-living increases after retirement. As part of our study, we examined nine state and local defined benefit plans covering over 2 million employees. For those plans, employer contributions ranged from 6 to 14.5 percent of payroll and employee contributions ranged from 6.4 to 9.3 percent of payroll. (See the appendix.) Extending mandatory Social Security coverage to states and localities with noncovered workers would reduce the trust funds’ long-term financial shortfall, increase program participation, and simplify program administration. SSA estimates that mandatory coverage would reduce Social Security’s financial shortfall by about 10 percent—from 2.19 percent of payroll (a present discounted value of $3.1 trillion) to 1.97 percent of payroll (a present discounted value of $2.9 trillion)—over a 75-year period. Figure 1 shows that mandatory coverage would also extend the program’s solvency by about 2 years, from 2032 to 2034. As with most other elements of the reform proposals put forward by the 1994-1996 Social Security Advisory Council, such as raising the retirement age, extending mandatory coverage to newly hired state and local employees would resolve only a part of the trust funds’ solvency problem. A combination of adjustments will be needed to extend the program’s solvency over the entire 75-year period. Council stated that mandatory coverage is basically “an issue of fairness.” The Advisory Council report stated that “an effective Social Security program helps to reduce public costs for relief and assistance, which, in turn, means lower general taxes. There is an element of unfairness in a situation where practically all contribute to Social Security, while a few benefit both directly and indirectly but are excused from contributing to the program.” Mandatory coverage would also simplify program administration in the long run. SSA’s Office of Research, Evaluation, and Statistics estimates that 95 percent of noncovered state and local employees become entitled to Social Security as either workers, spouses, or dependents. SSA’s Office of the Chief Actuary estimates that 50 to 60 percent of noncovered employees will be fully insured by age 62 from covered employment. The Congress has established the government pension offset and windfall elimination provisions to reduce the unfair advantage that workers who are eligible for pension benefits based on noncovered employment might have when they apply for Social Security benefits. The earnings histories for workers with noncovered earnings may appear to qualify them for the higher earnings replacement rates that Social Security assigns to lower earners, when in fact they have substantial income from public pension plans. With some exceptions, the government pension offset and windfall elimination provisions require SSA to use revised formulas to calculate benefits for workers with noncovered employment. However, a separate GAO study for the Chairman of this Subcommittee indicates that SSA is often unable to determine whether applicants should be subject to the government pension offset or windfall elimination provisions. We estimate that failure to reduce benefits for federal, state, and local employees caused $160 million to $355 million in overpayments between 1978 and 1995. In response, SSA plans to perform additional computer matches with the Office of Personnel Management and the Internal Revenue Service (IRS) to get noncovered pension data in order to ensure that these provisions are applied. Mandatory coverage would reduce benefit adjustments by gradually reducing the number of employees in noncovered jobs. Eventually, all state and local employees, with the exception of a few categories of workers, such as students and election workers, would be in covered employment. Additionally, in 1995, SSA asked its Inspector General to undertake a review of state and local government employers’ compliance with Social Security coverage provisions. In December 1996, SSA’s Office of the Inspector General reported that Social Security provisions related to coverage of state and local employees are complex and difficult to administer. The report stated that few resources were devoted to training state and local officials and ensuring that administration and enforcement roles and responsibilities are clearly defined. The report concluded that there is a significant risk of sizeable noncompliance with state and local coverage provisions. In response, SSA and IRS, which is responsible for collecting Social Security payroll taxes, initiated an effort to educate employers and ensure compliance with legal requirements for withholding Social Security payroll taxes. If all newly hired public employees were to receive mandated Social Security coverage, they would have the income protection afforded by Social Security. Also, they and their employers would pay the combined Social Security payroll tax of 12.4 percent of payroll. Each state and locality with noncovered workers would decide how to respond to the increase in retirement costs and benefits. They could absorb the added cost and leave current pension plans unchanged or eliminate plans completely. From discussions with state and local representatives, however, we believe states and localities with noncovered workers would likely adjust their pension plans to reflect Social Security’s costs and benefits. To illustrate the implications of mandatory coverage to employers and employees, we examined three possible responses: States and localities could maintain similar benefits for current and newly hired employees. This response would likely result in an increase in total retirement costs and some additional benefits for many newly hired employees. States and localities could examine other pension plans that are already coordinated with Social Security and provide newly hired employees with similar benefits. This response would also likely increase costs and benefits for newly hired employees. States and localities could maintain level retirement spending. This response could require a reduction in pension benefits. According to pension plan representatives, each of these responses to mandatory coverage would result in reduced contributions to current plans, which could affect long-term financing of the plans. States and localities with noncovered workers could opt to provide newly hired employees with Social Security and pension benefits that, in total, approximate the pension benefits of current employees. Studies indicate that such an option could increase retirement costs by 7 percent of new-employee payroll. Using SSA’s data and its assumption that mandatory coverage would start January 1, 2000, a 7 percent of payroll increase in retirement costs for newly hired employees would mean additional costs to states and localities with noncovered workers of about $9.1 billion over the first 5 years. A 1980 study of the costs of providing Social Security coverage for noncovered workers provides support for the estimated 7 percent of payroll increase. The Universal Social Security Coverage Study Group developed options for mandatory coverage of employees at all levels of government and analyzed the fiscal effects of each option. The study group used two teams of actuaries to study over 40 pension plans. The study estimated that costs, including Social Security taxes and pension plan contributions, would need to increase an average of 2 to 7 percent of payroll to maintain level benefits for current and newly hired employees. The study assumed that most newly hired employees would have salary replacement percentages in their first year of retirement that would be comparable to those provided to current employees. For example, employees retiring before age 62 would receive a temporary supplemental pension benefit to more closely maintain the benefits of the current plan. Since Social Security benefits are fully indexed for inflation and many pension plans have limited or no cost-of-living protection, total lifetime benefits for many newly hired employees would be greater than those provided to current employees. Existing pension plan disability and survivor benefits were also adjusted to reflect Social Security disability and survivor benefits. example, a December 1997 study for a plan in Ohio indicated that providing retirement and other benefits for future employees that, when added to Social Security benefits, approximate benefits for current employees would require an increase in contributions of 6 to 7 percent of new-employee payroll. A 1997 study for a pension plan in Illinois indicated the increased payments necessary to maintain similar total benefits for current and future employees would be about 6.5 percent of new-employee payroll. The 1980 study stated that the causes of the cost increase cannot be ascribed directly to specific Social Security or pension plan provisions. The study also states, however, that certain Social Security and pension plan provisions are among the most important factors contributing to the cost increase. Social Security is fully indexed for cost-of-living increases, is completely portable, and provides substantial additional benefits for spouses and dependents. In addition, pension plans would need to provide special supplemental benefits for employees who retire before age 62, especially in police and firefighter plans. The study also found that the magnitude of the cost increase would depend on the pension plan’s current benefits. Cost increases would be less for plans that already provide disability, survivor, and other benefits similar to those provided by Social Security because those plans would be able to eliminate duplicate benefits. About 70 percent of the state and local workforce is already covered by Social Security. If coverage is mandated, states and localities with noncovered employees could decide to provide newly hired employees with pension plan benefits similar to those provided to currently covered employees. The 1980 study examined this option and concluded that implementation would increase costs by 6 to 14 percent of payroll—or 3 to 11 percent of payroll after eliminating the Medicare tax. The study also found that most pension plans for covered employees did not provide supplemental retirement benefits for employees who retire before Social Security benefits are available. For most of the examined pension plans, the present value of lifetime benefits for employees covered by Social Security would be greater than the value of benefits for current noncovered employees. Our analysis of 1995 Public Pension Coordinating Council data also indicates that retirement costs for states and localities covered by Social Security are higher than the costs for noncovered states and localities. For the pension plans that responded to the survey, the average employee cost rate was about 9 percent of pay in covered plans, including Social Security taxes, and 8 percent of pay in noncovered plans. The average employer cost rate, excluding the cost of unfunded liabilities, was about 12 percent of payroll for employers in covered plans, including Social Security taxes, and 8 percent of payroll for employers in noncovered plans. These data also indicate that many employees in covered and noncovered plans, especially police and firefighters, retire before age 65, when covered employees would be eligible for full Social Security benefits. Our analysis indicates that covered employees who retire before age 65 initially have a lower salary replacement rate than noncovered employees. The average salary replacement rate with 30 years of service was 53 percent for members of Social Security covered plans and 64.7 percent for members of noncovered plans. At age 65, however, Social Security covered employees have a higher total benefit than noncovered employees. According to the Department of Labor’s 1994 survey, for example, an employee age 65 with 30 years of service, final earnings of $35,000, and Social Security coverage had 87 percent of earnings replaced—51 percent by a pension plan and 36 percent by Social Security. The same employee with no Social Security coverage had 63 percent of earnings replaced by a pension plan. We did not compare the expected value of total lifetime benefits for covered and noncovered employees because amounts would vary depending on the benefits offered by each plan. original plan and the lower costs and benefits of the revised plan. Subsequently, however, newly hired general employees were limited to the reduced benefits. Several employee, employer, and plan representatives stated that spending increases necessary to maintain level retirement income and other benefits for current and future members would be difficult to achieve. They indicate that states and localities might decide to maintain current spending levels, which could result in reduced benefits under state and local pension plans for many employees. A June 1997 actuarial evaluation of an Ohio pension plan examined the impact on benefits of mandating Social Security coverage for all employees, assuming no increase in total retirement costs. The study concluded that level spending could be maintained if (1) salary replacement rates for employees retiring with 30 years of service were reduced from 60.3 percent to 44.1 percent, (2) current retiree health benefits were eliminated for both current and future employees, and (3) the funding period for the plan’s unfunded accrued liability were extended from 27 years to 40 years. Most states and localities use a reserve funding approach to finance their pension plans. In reserve funding, employers—and frequently employees—make systematic contributions toward funding the benefits earned by active employees. These contributions, together with investment income, are intended to accumulate sufficient assets to cover promised benefits by the time employees retire. However, many public pension plans have unfunded liabilities. The nine plans that we examined, for example, have unfunded accrued liabilities ranging from less than 1 percent to over 30 percent of total liabilities. Unfunded liabilities occur for a number of reasons. For example, public plans generally use actuarial methods and assumptions to calculate required contribution rates. Unfunded liabilities can occur if a plan’s actuarial assumptions do not accurately predict reality. Additionally, retroactive increases in plan benefits can create unfunded liabilities. Unlike private pension plans, the unfunded liabilities of public pension plans are not regulated by the federal government. States or localities determine how and when unfunded liabilities will be financed. Mandatory coverage and the resulting changes to plan benefits for newly hired employees are likely to result in reduced contributions to the current pension plan. The impact of reduced contributions on plan finances would depend on the actuarial method and assumptions used by each plan, the adequacy of current plan funding, and other factors. For example, plan representatives are concerned that efforts to provide adequate retirement income benefits for newly hired employees would affect employers’ willingness or ability to continue amortizing their current plans’ unfunded accrued liabilities. Mandatory coverage presents several legal and administrative issues, and states and localities could require several years to design, legislate, and implement changes to current pension plans. Mandating Social Security coverage for state and local employees could elicit a constitutional challenge. We believe that mandatory coverage is likely to be upheld under current Supreme Court decisions. Several employer, employee, and plan representatives with whom we spoke stated that they believe mandatory Social Security coverage would be unconstitutional and should be challenged in court. However, recent Supreme Court cases have affirmed the authority of the federal government to enact taxes that affect the states and to impose federal requirements governing the states’ relations with their employees. A plan representative suggested that the Court might now come to a different conclusion. He pointed out that a case upholding federal authority to apply minimum wage and overtime requirements to the states was a 5 to 4 decision and that until then, the Court had clearly said that applying such requirements to the states was unconstitutional. States and localities also point to several recent decisions of the Court that they see as sympathetic to the concept of state sovereignty. However, the facts of these cases are generally distinguishable from the situation that would be presented by mandatory Social Security coverage. Unless the Court were to reverse itself, which it seldom does, mandatory Social Security coverage of state and local employees is likely to be upheld. Current decisions indicate that mandating such coverage is within the authority of the federal government. The federal government required approximately 3 years to enact legislation to implement a new federal employee pension plan after Social Security coverage was mandated for federal employees in 1983. According to the 1980 Universal Social Security Coverage Study Group, transition problems for state and local employers would be different from those faced by the federal government. For example, benefit provisions vary among the thousands of public employee retirement plans, as do the characteristics of the employees covered by those plans. Additionally, state governments and many local governments have laws regulating pensions. The study group estimated that 4 years would be required to redesign pension formulas, legislate changes, adjust budgets, and disseminate information to employers and employees. Our discussions with employer, employee, and pension plan representatives also indicate that up to 4 years would be needed to implement a mandatory coverage decision. Additionally, constitutional provisions or statutes in some states may prevent employers from reducing benefits for employees once they are hired. These states may need to immediately enact legislation to draw a line between current and future employees until decisions are made concerning the pension benefits for new employees who would be covered by Social Security. According to the National Conference of State Legislators, legislators in seven states, including Texas and Nevada, meet only biennially. Therefore, the initial legislation could require 2 years in those states. In deciding whether to extend mandatory Social Security coverage to all newly hired state and local employees, the Congress will need to weigh several factors. First, the Social Security program would benefit from mandatory coverage. The long-term actuarial deficit would be reduced, and the trust funds’ solvency would be extended for about 2 years. However, there are other considerations besides this relatively small contribution to the program’s solvency. Mandatory coverage would also increase participation in an important national program and simplify program administration. approximate the benefits provided to current workers. At the same time, Social Security would provide newly hired employees with benefits that are not available, or are available to a lesser extent, under current state and local pension plans. In addition, mandatory coverage would present legal and administrative issues. States and localities might attempt to halt mandatory Social Security coverage in court, although such a challenge is unlikely to be upheld. Finally, states and localities could require up to 4 years to implement mandatory coverage. Mr. Chairman, this concludes my prepared statement. At this time, I will be happy to answer any questions you or the other Subcommittee Members may have. SSA estimates that about 4 million of the approximately 5 million state and local employees not covered by Social Security are in the seven states with the largest number of noncovered workers. (See table I.1.) Number of noncovered employees (in thousands) The nine public pension plans included in our study have about 2 million members. For the most part, members of these plans are not covered by Social Security. (See table I.2.) The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO discussed extending mandatory Social Security coverage to all newly hired state and local government employees, focusing on: (1) the implications of mandating such coverage for the Social Security program, public employers, newly hired employees, and the affected pension plans; and (2) potential legal and administrative issues associated with implementing mandatory coverage. GAO noted that: (1) mandating coverage for all newly hired public employees would reduce Social Security's long-term financial shortfall by about 10 percent, increase participation in an important national program, and simplify program administration; (2) the impact on public employers, employees, and pension plans would depend on how states and localities with noncovered employees would react to these new coverage provisions; (3) one often-discussed option would be for public employers to modify their pension plans in response to mandatory Social Security coverage; (4) costs would likely increase for those states and localities that wanted to keep their enhanced benefits for newly hired employees; (5) alternatively, states and localities that wanted to maintain level spending for retirement would likely need to reduce some pension benefits; (6) regardless, mandating coverage for public employees would present legal and administrative issues that would need to be resolved; (7) in deciding whether to extend mandatory Social Security coverage to all newly hired state and local employees, Congress would need to weigh several factors: (a) the Social Security program would benefit from mandatory coverage; (b) the long-term actuarial deficit would be reduced; and (c) the trust funds' solvency would be extended for about 2 years; (8) states and localities with noncovered workers would likely need to increase total retirement spending to provide future workers with pension benefits that, when combined with Social Security benefits, approximate the benefits provided to current workers; (9) at the same time, Social Security would provide newly hired employees with benefits that are not available, or are available to a lesser extent, under current state and local pension plans; (10) states and localities might attempt to halt mandatory Social Security coverage in court, although such a challenge is unlikely to be upheld; and (11) states and localities could require up to 4 years to implement mandatory coverage. |
The Great Lakes Basin is a large area that extends well beyond the five lakes proper to include their watersheds, tributaries, connecting channels, and a portion of the St. Lawrence River. The basin encompasses nearly all of the state of Michigan and parts of Illinois, Indiana, Minnesota, New York, Ohio, Pennsylvania, Wisconsin, and the Canadian province of Ontario. The lakes form the largest freshwater system on earth, accounting for 20 percent of the world’s fresh surface water and over 95 percent of the U.S. fresh surface water supply for the contiguous 48 states. Millions of people in the United States and Canada rely on the five Great Lakes—Superior, Michigan, Erie, Huron, and Ontario—as a principal source of their drinking water, recreation, and economic livelihood. Over time, industrial, agricultural, and residential development on lands adjacent to the lakes have seriously degraded the lakes’ water quality, posing threats to human health and the environment, and forcing restrictions on activities, such as swimming and fish consumption. To protect the Great Lakes Basin and to address water quality problems, the governments of the United States and Canada entered into the bilateral Great Lakes Water Quality Agreement in 1972. In the agreement, the United States and Canada agreed to restore and maintain the chemical, physical, and biological integrity of the Great Lakes Basin. A new agreement with the same name was reached in 1978, and amended in 1983 and 1987. The agreement prescribes prevention and cleanup measures to improve environmental conditions in the Great Lakes. The agreement obligates the International Joint Commission (IJC), an international body, to assist and to report on the implementation of the agreement. The Clean Water Act directs EPA to lead efforts to meet the goals of the Great Lakes Water Quality Agreement and establishes GLNPO within EPA, charging it with, among other things, cooperating with federal, state, tribal, and international agencies to develop action plans to carry out the U.S. responsibilities under the agreement. GLNPO is further responsible for coordinating the agency’s actions both in headquarters and in the regions to improve Great Lakes’ water quality. In addition to GLNPO, numerous federal, state, binational, and nonprofit organizations conduct activities that focus on improving the overall Great Lakes Basin environment or some specific environmental issue within the basin. About 200 programs—148 federal and 51 state—fund restoration activities within the Great Lakes Basin. Most of these programs, however, involve the localized application of national or state environmental initiatives and do not specifically focus on basin concerns. Officials from 11 federal agencies identified 115 of these broadly scoped federal programs, and officials from seven of the eight Great Lakes states identified 34 similar state programs. EPA administers the majority of the federal programs that provide a broad range of environmental activities involving research, cleanup, restoration, and pollution prevention. For example, EPA’s nationwide Superfund program funds cleanup activities at contaminated areas throughout the basin. While these broadly scoped federal and state programs contribute to basin restoration, program officials do not track or try to isolate the portion of funding going to specific areas like the basin, making it difficult to determine their contribution to total Great Lakes spending. However, basin-specific information was available on some of these programs. Specifically, basin related expenditures for 53 of the 115 broadly scoped federal programs totaled about $1.8 billion in fiscal years 1992 through 2001. Expenditures for 14 broadly scoped state funded programs totaled $461.3 million during approximately the same time period. Several federal and state programs were specifically designed to focus on environmental conditions across the Great Lakes Basin. Officials from seven federal agencies identified 33 Great Lakes-specific programs that had expenditures of $387 million in fiscal years 1992 through 2001. Most of these programs funded a variety of activities, such as research, cleanup, or pollution prevention. An additional $358 million was expended for legislatively directed Corps of Engineers projects in the basin, such as a $93.8 million project to restore Chicago’s shoreline. Officials from seven states reported 17 Great Lakes specific programs that expended about $956 million in 1992 through 2001, with Michigan’s programs accounting for 96 percent of this amount. State programs focused on unique state needs, such as Ohio’s program to control shoreline erosion along Lake Erie, and Michigan’s program to provide bond funding for environmental activities. Besides federal and state government agencies, other organizations, such as foundations, fund a variety of restoration activities in the Great Lakes Basin by approving grants to nonprofit and other organizations. Other governmental and nongovernmental organizations fund restoration activities. For example, individual municipalities, township governments, counties, and conservation districts are involved in various restoration activities. Restoration of the Great Lakes Basin is a major endeavor involving many environmental programs and organizations. The magnitude of the area comprising the basin and the numerous environmental programs operating within it require the development of one overarching strategy to address and manage the complex undertaking of restoring the basin’s environmental health. The Great Lakes region cannot hope to successfully receive support as a national priority without a comprehensive, overarching plan for restoring the Great Lakes. In lieu of such a plan, organizations at the binational, federal, and state levels have developed their own strategies for the Great Lakes, which have inadvertently made the coordination of various programs operating in the basin more challenging. The Great Lakes Basin needs a comprehensive strategy or plan similar to those developed for other large ecosystem restoration efforts, such as the ones for the South Florida ecosystem and the Chesapeake Bay. In South Florida, federal, state, local and tribal organizations joined forces to participate on a centralized task force formalized in the Water Resource Development Act of 1996. The strategic plan developed for the South Florida ecosystem by the task force made substantial progress in guiding the restoration activities. The plan identifies the resources needed to achieve restoration and assigns accountability for specific actions for the extensive restoration effort estimated to cost $14.8 billion. The Chesapeake Bay watershed also has an overarching restoration strategy stemming from a 1983 agreement signed by the states of Maryland, Virginia, and Pennsylvania; the District of Columbia; the Chesapeake Bay Commission; and EPA. This agreement was the basis for a program to protect and restore this ecosystem. The implementation of this strategy has resulted in improvements in habitat restoration and aquatic life, such as increased forested buffer zone and shad population. Several organizations have developed strategies for the basin at the binational, federal, or state levels that address either the entire basin or the specific problems in the Great Lakes. EPA’s Great Lakes Strategy 2002, developed by a committee of federal and state officials, is the most recent of these strategies. While this strategy identified restoration objectives and planned actions by various federal and state agencies, it is largely a description of existing program activity relating to basin restoration. State officials told us that the states had already planned the actions described in it, but that these actions were contingent on funding for specific environmental programs. The strategy included a statement that it should not be construed as a commitment for additional funding or resources, and it did not provide a basis for prioritizing activities. In addition, we identified other strategies that addressed particular contaminants, restoration of individual lakes, or cleanup of contaminated areas. Ad hoc coordination takes place among federal agencies, states, and other environmental organizations in developing these strategies or when programmatic activity calls for coordination. Other Great Lakes strategies address unique environmental problems or specific geographical areas. For example, a strategy for each lake addresses the open lake waters through Lakewide Management Plans (LaMP), which EPA is responsible for developing. Toward this end, EPA formed working groups for each lake to identify and address restoration activities. For example, the LaMP for Lake Michigan, issued in 2002, includes a summary of the lake’s ecosystem status and addresses progress in achieving the goals described in the previous plan, with examples of significant activities completed and other relevant topics. However, EPA has not used the LaMPs to assess the overall health of the ecosystem. The Binational Executive Committee for the United States and Canada issued its Great Lakes Binational Toxics Strategy in 1997 that established a collaborative process by which EPA and Environment Canada, in consultation with other federal departments and agencies, states, the province of Ontario, and tribes, work toward the goal of the virtual elimination of persistent toxic substances in the Great Lakes. The strategy was designed to address particular substances that bioaccumulate in fish or animals and pose a human health risk. Michigan developed a strategy for environmental cleanup called the Clean Michigan Initiative. This initiative provides funding for a variety of environmental, parks, and redevelopment programs. It includes nine components, including Brownfields redevelopment and environmental cleanups, nonpoint source pollution control, clean water, cleanup of contaminated sediments, and pollution prevention. The initiative is funded by a $675 million general obligation bond and as of early 2003; most of the funds had not been distributed. Although there are many strategies and coordination efforts ongoing, no one organization coordinates restoration efforts. We found that extensive strategizing, planning, and coordinating have not resulted in significant restoration. Thus, the ecosystem remains compromised and contaminated sediments in the lakes produce health problems, as reported by the IJC. In addition to the absence of a coordinating agency, federal and state officials cited a lack of funding commitments as a principal barrier impeding restoration progress. Inadequate funding has also contributed to the failure to restore and protect the Great Lakes, according to the IJC biennial report on Great Lakes water quality issued in July 2000. The IJC restated this position in a 2002 report, concluding that any progress to restore the Great Lakes would continue at a slow incremental pace without increased funding. In its 1993 biennial report, the IJC concluded that remediation of contaminated areas could not be accomplished unless government officials came to grips with the magnitude of cleanup costs and started the process of securing the necessary resources. Despite this warning, however, as we reported in 2002, EPA reduced the funding available for ensuring the cleanup of contaminated areas under the assumption that the states would fill the funding void. States, however, did not increase their funding, and restoration progress slowed or stopped altogether. Officials for 24 of 33 federal programs and for 3 of 17 state programs reported insufficient funding for federal and state Great Lakes specific programs. Ultimate responsibility for coordinating Great Lakes restoration programs rests with GLNPO; however, GLNPO has not fully exercised this authority. Other organizations or committees have formed to assume coordination and strategy development roles. The Clean Water Act provides GLNPO with the authority to fulfill the U.S. responsibilities under the GLWQA. Specifically, the act directs EPA to coordinate the actions of EPA’s headquarters and regional offices aimed at improving Great Lakes water quality. It also provides GLNPO authority to coordinate EPA’s actions with the actions of other federal agencies and state and local authorities for obtaining input in developing water quality strategies and obtaining support in achieving the objectives of the GLWQA. The act also provides that the EPA Administrator shall ensure that GLNPO enters into agreements with the various organizational elements of the agency engaged in Great Lakes activities and with appropriate state agencies. The agreements should specifically delineate the duties and responsibilities, time periods for carrying out duties, and resources committed to these duties. GLNPO officials stated that they do not enter into formal agreements with other EPA offices, but rather fulfill their responsibilities under the act by having federal agencies and state officials agree to the restoration activities contained in the Great Lakes Strategy 2002. However, the strategy does not represent formal agreements to conduct specific duties and responsibilities with committed resources. EPA’s Office of Inspector General reported the absence of these agreements in September 1999. The report stated that GLNPO did not have agreements as required by the act and recommended that such agreements be made to improve working relationships and coordination. To improve coordination of Great Lakes activities and ensure that federal dollars are effectively spent, we recommended that the Administrator, EPA, ensure that GLNPO fulfills its responsibility for coordinating programs within the Great Lakes Basin; charge GLNPO with developing, in consultation with the governors of the Great Lakes states, federal agencies, and other organizations, an overarching strategy that, clearly defines the roles and responsibilities for coordinating and prioritizing funding for projects; and submit a time-phased funding requirement proposal to the Congress necessary to implement the strategy. The Great Lakes Water Quality Agreement, as amended in 1987, calls for establishing a monitoring system to measure restoration progress and assess the degree that the United States and Canada are complying with the goals and objectives of the agreement. However, implementation of this provision has not progressed to the point that overall restoration progress can be measured or determined based on quantitative information. Recent assessments of overall progress, which rely on a mix of quantitative data and subjective judgments, do not provide an adequate basis for making an overall assessment. The current assessment process has emerged from a series of biennial State of the Lakes Ecosystem Conferences (SOLEC) initiated in 1994 for developing indicators agreed upon by conference participants. Prior to the 1987 amendments to the GLWQA, the 1978 agreement between the two countries also contained a requirement for surveillance and monitoring and for the development of a Great Lakes International Surveillance Plan. The IJC Water Quality Board was involved in managing and developing the program until the 1987 amendments placed this responsibility on the United States and Canada. This change resulted in a significant reduction in the two countries’ support for surveillance and monitoring. In fact, the organizational structure to implement the surveillance plan was abandoned in 1990, leaving only one initiative in place—the International Atmospheric Deposition Network (IADN), a network of 15 air-monitoring stations located throughout the basin. With the surveillance and monitoring efforts languishing, IJC established the Indicators for Evaluation Task Force in 1993 to identify the appropriate framework to evaluate progress in the Great Lakes. In 1996, the task force proposed that nine desired measurements and outcomes be used to develop indicators for measuring progress in the Great Lakes. Shortly before the task force began its work, the United States and Canada had agreed to hold conferences every 2 years to assess the environmental conditions in the Great Lakes in order to develop binational reports on the environmental conditions to measure progress under the agreement. Besides assessing environmental conditions the conferences were focused on achieving three other objectives, including providing a forum for communication and networking among stakeholders. Conference participants included U.S. and Canadian representatives from federal, state, provincial, and tribal agencies, as well as other organizations with environmental restoration or pollution prevention interests in the Great Lakes Basin. The 1994 SOLEC conference culminated in a “State of the Great Lakes 1995” report, which provided an overview of the Great Lakes ecosystem at the end of 1994 and concluded that overall the aquatic community health was mixed or improving. The same assessment was echoed in the 1997 state of the lakes report. Meanwhile the IJC agreed that the nine desired outcome areas recommended by the task force would help assess overall progress. It recommended that SOLEC, during the conference in 2000, establish environmental indicators that would allow the IJC to evaluate what had been accomplished and what needed to be done for three of the nine indicators—the public’s ability to eat the fish, drink the water, and swim in the water without any restrictions. However, the indicators developed through the SOLEC process and the accomplishments reported by federal and state program managers do not provide an adequate basis for making an overall assessment for Great Lakes restoration progress. The SOLEC process is ongoing and the indicators still being developed are not generally supported by sufficient underlying data for making progress assessments. The number of indicators considered during the SOLEC conferences has been pared down from more than 850 indicators in 1998 to 80 indicators in 2000, although data were available for only 33 of them. After the SOLEC 2000 conference, IJC staff assessed the indicators supported by data that measured the desired outcomes of swimmability, drinkability, and the edibility of fish in the Great Lakes. Overall, the IJC commended SOLEC’s quick response that brought together information regarding the outcomes and SOLEC’s ongoing efforts. The IJC, however, recognized that sufficient data were not being collected throughout the Great Lakes Basin and that the methods of collection, the data collection time frames, the lack of uniform protocols, and the incompatible nature of some data jeopardized their use as indicators. Specifically, for the desired outcome of swimmability, the IJC concurred that it was not always safe to swim at certain beaches, but noted that progress for this desired outcome was limited because beaches were sampled by local jurisdictions without uniform sampling or reporting methods. At the 2002 SOLEC conference, the number of indicators assessed by conference participants increased from 33 to 45. The IJC expressed concern that there are too many indicators, insufficient supporting backup data, and a lack of commitment and funding from EPA to implement and make operational the agreed upon SOLEC baseline data collection and monitoring techniques. The IJC recommended in its last biennial report in September 2002 that any new indicators should be developed only where resources are sufficient to access scientifically valid and reliable information. The information from the 2002 SOLEC conference culminated in the “State of the Great Lakes 2003” report, which concluded that the chemical, physical, and biological integrity of the basin is mixed based on assessments of 43 indicators. This conclusion was based on five positive signs of recovery, such as persistent toxic substances are continuing to decline, and seven negative signs, such as phosphorous levels are increasing in Lake Erie. The ultimate successful development and assessment of indicators for the Great Lakes through the SOLEC process are uncertain because insufficient resources have been committed to the process, no plan provides completion dates for indicator development and implementation, and no entity is coordinating the data collection. Even though the SOLEC process has successfully engaged a wide range of binational parties in developing indicators, the resources devoted to this process are largely provided on a volunteer basis without firm commitments to continue in the future. GLNPO officials described the SOLEC process as a professional, collaborative process dependent on the voluntary participation of officials from federal and state agencies, academic institutions, and other organizations attending SOLEC and developing information on specific indicators. Because SOLEC is a voluntary process, the indicator data resides in a diverse number of sources with limited control by SOLEC organizers. GLNPO officials stated that EPA does not have either the authority or the responsibility to direct the data collection activities of federal, state, and local agencies as they relate to surveillance and monitoring of technical data elements that are needed to develop, implement, and assess Great Lakes environmental indicators. Efforts are underway for the various federal and state agencies to take ownership for collecting and reporting data outputs from their respective areas of responsibility and for SOLEC to be sustained and implemented; each indicator must have a sponsor. However, any breakdown in submission of this information would leave a gap in the SOLEC indicator process. EPA supports the development of environmental indicators as evidenced by the fact that, since 1994, GLNPO has provided about $100,000 annually to sponsor the SOLEC conferences. Additionally, GLNPO spends over $4 million per year to collect surveillance data for its open-lake water quality monitoring program, which also provides supporting data for some of the indicators addressed by SOLEC. A significant portion of these funds, however, supports the operation of GLNPO’s research vessel, the Lake Guardian, an offshore supply vessel converted for use as a research vessel. GLNPO also supports activities that are linked or otherwise feed information into the SOLEC process, including the following: collecting information on plankton and benthic communities in the Great Lakes for open water indicator development; sampling various chemicals in the open-lake waters, such as phosphorus for the total phosphorus indicator; monitoring fish contaminants in the open waters, directly supporting the indicator for contaminants in whole fish and a separate monitoring effort for contaminants in popular sport fish species that supports the indicator for chemical contaminants in edible fish tissue; and operating 15 air-monitoring stations with Environment Canada comprising the IADN that provides information for establishing trends in concentrations of certain chemicals and loadings of chemicals into the lakes. EPA uses information from the network to take actions to control the chemicals and track progress toward environmental goals. To better coordinate monitoring activities GLNPO and Environment Canada began developing a web-based inventory of monitoring activities in the Great Lakes Basin. The first workshop on developing this system was held in January 2002. Once development of this system is complete, organizations conducting monitoring activities will be requested to provide descriptive information about these monitoring activities and contact points for obtaining specific monitoring data. We are currently conducting a review for 20 members of Congress serving on the Great Lakes Task Force that further examines monitoring activities in the Great Lakes Basin. In this review we hope to identify some of the major challenges to developing a Great Lakes Basin monitoring system. Program officials frequently cite output data as measures of success rather than actual program accomplishments in improving environmental conditions in the basin. As a rule, program output data describe activities, such as projects funded, and are of limited value in determining environmental progress. For example, in reporting the accomplishments for Michigan’s Great Lakes Protection Fund, officials noted that the program had funded 125 research projects over an 11-year period and publicized its project results at an annual forum and on a Web site. Similarly, the Lake Ontario Atlantic Salmon Reintroduction Program administered by the Department of Interior’s Fish and Wildlife Service listed under its accomplishments the completion of a pilot study and technical assistance provided to a Native American tribe. Of the 50 federal and state programs created specifically to address conditions in the basin, 27 reported accomplishments in terms of outputs, such as reports or studies prepared or presentations made to groups. Because research and capacity building programs largely support other activities, it is particularly difficult to relate reported program accomplishments to outcomes. The federal and state environmental program officials responding to our evaluation generally provided output data or, as reported for 15 programs, reported that the accomplishments had not been measured for the programs. Only eight of the federal or state Great Lakes-specific programs reported outcome information, much of which generally described how effective the programs’ activity or action had been in improving environmental conditions. For example, EPA’s Region II program for reducing toxic chemical inputs into the Niagara River, which connects Lake Erie to Lake Ontario, reported reductions in priority toxics from 1986 through 2002 from ambient water quality monitoring. Other significant outcomes reported as accomplishments for the Great Lakes included (1) reducing phosphorus loadings by waste treatment plants and limiting phosphorus use in household detergents; (2) prohibiting the release of some toxicants into the Great Lakes, and reducing to an acceptable level the amount of some other toxicants that could be input; (3) effectively reducing the sea lamprey population in several invasive species infested watersheds; and (4) restocking the fish-depleted populations in some watersheds. To fulfill the need for a monitoring system called for in the GLWQA and to ensure that the limited funds available are optimally spent, we recommended that the Administrator, EPA, in coordination with Canadian officials and as part of an overarching Great Lakes strategy, (1) develop environmental indicators and a monitoring system for the Great Lakes Basin that can be used to measure overall restoration progress and (2) require that these indicators be used to evaluate, prioritize, and make funding decisions on the merits of alternative restoration projects. Mr. Chairman, this completes my prepared statement. I would be happy to answer any questions that you or other members of the Subcommittee may have at this time. For further information, please contact John B. Stephenson at (202) 512-3841. Individuals making key contributions to this testimony were Willie Bailey, Greg Carroll, Karen Keegan, Jonathan McMurray, and John Wanska. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The five Great Lakes, which comprise the largest system of freshwater in the world, are threatened on many environmental fronts. To address the extent of progress made in restoring the Great Lakes Basin, which includes the lakes and surrounding area, GAO (1) identified the federal and state environmental programs operating in the basin and funding devoted to them, (2) evaluated the restoration strategies used and how they are coordinated, and (3) assessed overall environmental progress made in the basin restoration effort. There are 148 federal and 51 state programs funding environmental restoration activities in the Great Lakes Basin. Most of these programs are nationwide or statewide programs that do not specifically focus on the Great Lakes. However, GAO identified 33 federal Great Lakes specific programs, and 17 additional unique Great Lakes specific programs funded by states. Although Great Lakes funding is not routinely tracked for many of these programs, we identified a total of about $3.6 billion in basin-specific projects for fiscal years 1992 through 2001. Several disparate Great Lakes environmental strategies are being used at the binational, federal, and state levels. Currently, these strategies are not coordinated in a way that ensures effective use of limited resources. Without such coordination it is difficult to determine the overall progress of restoration efforts. The Water Quality Act of 1987 charged EPA's Great Lakes National Program Office with the responsibility for coordinating federal actions for improving Great Lakes' water quality; however, the office has not fully exercised this authority to this point. With available information, current environmental indicators do not allow a comprehensive assessment of restoration progress in the Great Lakes. Current indicators rely on limited quantitative data and subjective judgments to determine whether conditions are improving, such as whether fish are safe to eat. The ultimate success of an ongoing binational effort to develop a set of overall indicators for the Great Lakes is uncertain because it relies on the resources voluntarily provided by several organizations. Further, no date for completing a final list of indicators has been established. |
Following a disaster, and upon the request of a state governor, the President may issue a major disaster declaration that triggers a range of assistance from federal agencies. Under the provisions of the Robert T. Stafford Disaster Relief and Emergency Assistance Act, the federal government will assist disaster-stricken states and communities in their efforts to help those in need, remove debris, and rebuild damaged structures, among other things. The costs for this federal disaster assistance have grown significantly since the late 1980s. During the 12-year period ending in 1989, the expenditures from FEMA’s disaster relief fund totaled about $7 billion (in fiscal year 2001 dollars). However, during the following 12-year period ending in 2001, as the number of large, costly disasters has grown and the activities eligible for federal assistance have increased, expenditures from the disaster relief fund increased nearly fivefold to over $39 billion (in fiscal year 2001 dollars). Figure 1 shows the annual amounts spent for disaster relief and the number of disasters from fiscal years 1978 to 2001. Disaster assistance costs are expected to remain high in 2002, in part as a result of the September 11, 2001, terrorist attacks. According to FEMA’s projections, disaster assistance expenditures from the disaster relief fund in fiscal year 2002 will total more than $4 billion. FEMA has been designated the lead agency for the nation’s emergency management system, and traditionally the agency has directed its efforts towards disaster response and recovery. It also helps state and local governments prepare for possible disaster events. However, as costs for disaster assistance have increased, the agency has placed increasing emphasis on disaster mitigation, defined by FEMA as sustained action that reduces or eliminates long-term risk to people and property from hazards and their effects. In fact, FEMA has made disaster mitigation a primary goal in its efforts to reduce the long-term cost of disasters. Among the most significant of these programs are the HMGP and the Project Impact program. These programs are FEMA’s sole multihazard mitigation programs, helping states and communities address the natural hazards and risks—such as earthquakes, floods, tornadoes, and hurricanes—they deem most significant. Together, these programs represent FEMA’s most substantial mitigation efforts in terms of expenditures, state and community involvement, and scope of activities funded. Other mitigation programs FEMA conducts, although not insignificant, address specific concerns, such as dam safety, fires, and flooding, and are funded at relatively low levels. The Congress has also recognized the benefits of mitigation, and as recently as 2000 passed legislation to establish a national mitigation program. The Disaster Mitigation Act of 2000 sought to (1) reduce the loss of life and property, economic disruption, and disaster assistance cost resulting from natural disaster and (2) provide a source of predisaster mitigation funding that will assist states and local governments in implementing effective hazard mitigation measures. The act provided authorization legislation for Project Impact’s predisaster mitigation activities, and established a funding formula under which communities in all states would participate and receive funding. The act also placed an emphasis on mitigation planning: it authorized the use of HMGP funds for planning purposes and increased by one-third the HMGP funding for states that meet enhanced planning criteria. Recently, however, proposals have been made that may significantly affect the mitigation programs conducted by FEMA. The administration has proposed a substantial change to FEMA’s multihazard mitigation programs. The proposal, as contained in FEMA’s fiscal year 2003 budget request, would eliminate the HMGP and establish a new $300 million program for predisaster mitigation. The program would also award grants on a nationwide, competitive basis that is significantly different from the formula-based grant process in the existing programs. The House and Senate have both proposed creating a Department of Homeland Security that would subsume FEMA as a part of the department. If enacted as currently proposed, all the activities of FEMA—both those that are security related and those, such as natural hazard mitigation programs, which are nonsecurity related—would transfer to the department. Further, the federal government is taking a more expanded role in state and local government disaster prevention and preparedness efforts, and it is initiating more activities—and providing more funding—for predisaster assistance, with a substantial focus on terrorism. In this regard, numerous legislative proposals are being considered that increase planning requirements and funding to prepare for and prevent future terrorist attacks. FEMA’s multihazard mitigation programs differ substantially and have many successful attributes according to state and local officials. The HMGP, FEMA’s oldest and largest multihazard mitigation program, is a postdisaster program that has provided the bulk of mitigation assistance provided to states and communities. Through the HMGP, states and communities that have experienced a presidentially declared disaster receive funds to implement cost-effective mitigation projects. States and communities have used these funds primarily to implement “brick and mortar” projects such as retrofitting structures or acquiring and relocating structures from hazard-prone areas. The HMGP is viewed as highly effective because it provides funding in the aftermath of a disaster—when state and local governments as well as individuals have a heightened interest in participating in mitigation activities. As a result, states and local communities have been able to fund critical mitigation projects in these periods. FEMA has used its more recent and smaller predisaster Project Impact program to provide mitigation assistance directly to communities in every state, regardless of whether the state had recently experienced a disaster. Communities have used Project Impact in large measure on planning and outreach activities designed to (1) help educate the public and promote mitigation, (2) assess risks and identify potential mitigation projects, and (3) build partnerships and leverage resources. State and local officials also said that Project Impact has been successful in increasing awareness of and community support for mitigation efforts due to its funding of these types of activities. The HMGP was created in 1988 to assist states and communities in implementing long-term hazard mitigation measures following a major disaster declaration. The purpose of the program is to enable mitigation measures to be implemented during the immediate recovery period following a disaster so that future risk to lives and property from severe natural hazards can be significantly reduced or permanently eliminated. To accomplish this objective, the HMGP provides funding to states affected by presidentially declared disasters to undertake mitigation activities identified in state or local hazard mitigation plans. FEMA has provided a significant amount of funds for mitigation activities through the HMGP. During fiscal years 1996 through 2001, over $2.2 billion from FEMA’s disaster relief fund was obligated to states for this program. The maximum amount of HMGP funding available to any state following a presidential disaster declaration had been up to 15 percent of the total estimated amount of federal assistance provided on a declared disaster; however, the Disaster Mitigation Act of 2000 increased this amount to 20 percent for states that meet enhanced planning criteria. Appendix II contains a listing of HMGP obligations by year and state. While FEMA provides the funding for the program, the responsibility for administering the HMGP rests with states. To this end, states review, prioritize, and select projects based upon state mitigation priorities and available funds. State and local governments, Native American tribes, and certain nonprofit organizations are eligible to develop project applications. To be considered for selection by states, projects must meet minimum eligibility requirements. For example, projects must conform to the state hazard mitigation plan, comply with environmental laws and regulations, and be cost-effective. FEMA will provide up to 75 percent of the cost of mitigation projects selected under HMGP; applicants must provide the remaining project costs. Further, while states are responsible for selecting projects, FEMA conducts the final eligibility review of projects to ensure they meet all program requirements. HMGP funds have primarily been used by states and communities to implement “brick and mortar” projects. These types of projects include the following: acquiring properties in hazard-prone areas and either demolishing the associated structure or relocating the structure to a site outside the hazard-prone area; performing modifications to structures, such as reinforcing roofs, walls, and foundations, to protect them from floods, high winds, or other natural hazards; constructing new storm water drainage systems and other flood control building protective structures, such as safe rooms inside schools in tornado-prone areas, to better ensure the safety of individuals. Figure 2 shows projects undertaken with HMGP funding. Hazard mitigation officials from the 24 states we contacted said that the HMGP has been effective in stimulating action to mitigate the impacts of natural hazards, primarily because it takes advantage of a “window of opportunity” that exists in the postdisaster environment. The state hazard mitigation officials said that the states’ and localities’ commitment to fund and implement mitigation measures is most likely to occur soon after they have experienced the devastation caused by a major disaster. These officials emphasized that states, local communities, and citizens affected by a disaster recognize the need for effective mitigation and are willing to share costs and take the steps necessary to remove themselves from harm’s way in the immediate postdisaster environment; but as time passes they are less willing to do so. Even officials from states that have traditionally received little funding under this program, such as Wyoming and Utah, expressed support for the program’s postdisaster approach to funding mitigation activities. Below are some examples of significant mitigation projects that states told us they were able to undertake with HMGP funds because of the state, local, and citizen support for mitigation that existed in the immediate postdisaster environment. Following the devastation of Hurricane Floyd in 1999, North Carolina undertook a program to remove homes from flood-prone areas. In the immediate aftermath of Floyd, the state legislature passed a disaster recovery bill that not only provided $73.4 million in matching funds but also an additional $160 million to buy out flood-prone properties. The state used these funds, along with nearly $228 million in HMGP and other federal postdisaster mitigation funds, to target 4,500 properties whose owners were willing to accept buyouts. As of June 2002, the state had completed 70 percent of these buyouts. In the aftermath of the May 1999 tornadoes that damaged nearly 3,350 structures and left 6 people dead in the Wichita area, Kansas utilized HMGP funding to make schools more tornado-resistant, a critical need identified by the event. Inspections of damaged schools revealed that some designated refuge areas had suffered significant damage and that if children had been present, injuries would have likely occurred. According to state officials, the event and the immediate availability of HMGP funds were key in convincing local citizens and school district officials to approve a school district bond that included funds to construct tornado shelters inside schools. Funding from this bond provided the local match needed to use HMGP funds to construct 24 shelters to protect approximately 7,800 students and staff. In response to a 2000 tornado that left 1 dead, injured more than 100, and damaged nearly 200 homes and businesses in the city of Xenia, the state of Ohio utilized HMGP funding for the construction of safe rooms in this tornado-prone community. Since 1900, Xenia has been hit by 6 tornadoes including a 1974 tornado that killed 26 people. In the wake of the 2000 tornado’s devastation, the state and the community provided 3 times the required HMGP match to undertake the construction of residential safe rooms in the homes of 50 families. According to mitigation officials from these states, it is unlikely that these mitigation projects would have been undertaken without the infusion of HMGP funding in the postdisaster environment. Studies that have examined community action after a disaster support state officials’ claims that a window of opportunity exists and is critical for accomplishing mitigation activities. For example, a 1997 university study that examined public response after hurricanes and earthquakes found that communities and local decision makers were more willing to undertake mitigation soon after a disaster than at other times. Similarly, a FEMA sponsored case study of natural disasters in South Florida noted that the focus on mitigation dissipates after cleanup and recovery are completed as public attention moves elsewhere. Further, according to the director of the Natural Hazards Research and Applications Information Center located at the University of Colorado, research generally suggests that local political support for mitigation is strongest in the approximately 6 months following a disaster, after which funding becomes more difficult to obtain as other state and local issues take precedent. He added that research suggests that public support for mitigation lasts for about 1 year, during which time citizens are more willing to participate in mitigation activities. Whereas the HMGP has focused on implementing projects in a postdisaster environment, the Project Impact program focused on developing broad community support for mitigation activities before a disaster strikes. To accomplish this end, the Project Impact program provided small, one-time grants directly to communities, which, among other things, were designed to develop mitigation plans, build effective partnerships, and encourage private sector financial participation. During fiscal years 1997 through 2001, Project Impact provided a total of $77 million to communities within every state and certain U.S. territories. Unlike the HMGP, the amount of Project Impact funding available to communities within a state was not predicated upon the occurrence of a disaster; in fact, the program was structured so that under its funding formula, communities in every state participated in the program. By 2001, there were nearly 250 communities participating in the program, with Project Impact communities receiving grants between $60,000 and $1,000,000. Appendix III lists Project Impact grants by year and community. While states selected which communities received Project Impact grants, communities were responsible for selecting the mitigation measures to fund with these grants. Similar to the HMGP, however, communities were required to provide 25 percent of the costs for the mitigation measures. While the mitigation measures could be “brick and mortar” projects, they could also fund other activities such as establishing community partnerships, supporting public awareness of mitigation, identifying hazards, and conducting risk assessments. Additionally, Project Impact funds could be used to promote the concept of personal and community responsibility for mitigation measures. For example, FEMA encouraged communities to establish committees composed of local officials, business professionals, and other stakeholders to develop and implement mitigation activities of importance to the community. Figure 3 shows projects undertaken with Project Impact. The state emergency management officials from the 24 states, as well as the Project Impact communities we visited, believe Project Impact has also been successful in improving mitigation efforts throughout the country. They stated that the program’s focus on planning and developing broad community support for mitigation in a predisaster environment has been very beneficial in building grass root support for mitigation. The state officials identified four specific features of the Project Impact program as being most beneficial, namely the program’s funding of mitigation planning activities, development of partnerships to address mitigation needs, providing “seed money” to attract additional funding, and heightening of mitigation awareness resulting from education and outreach activities. A primary benefit of Project Impact was its emphasis on developing private and public sector partnerships as a means for communities to address their mitigation needs. According to state and community officials, effective hazard mitigation requires the involvement of not only governments but also of the private sector—both business and individuals—to fully identify and address concerns. They stated that Project Impact had been very successful in creating partnerships that identify, and in most cases fund, mitigation activities. For example, a major corporation in Deerfield Beach, Florida, became a Project Impact partner and, at its own expense, installed impact resistance glass and concrete roofs in all of its structures to make them more disaster resistant. This corporation also donated shutters for the homes of some low-income, elderly residents. Similarly, in Wilmington, North Carolina, a local home improvement store used its facilities to distribute hurricane preparedness and mitigation brochures and was a major financial contributor to the local Hurricane Preparedness Expo that featured speakers and demonstrations to assist residents with their mitigation efforts. A second benefit of Project Impact was its focus on planning as a critical phase in implementing mitigation projects. According to state and local mitigation officials, Project Impact assisted communities in identifying vulnerabilities, assessing risks, and developing and prioritizing mitigation projects to address their needs. Some states and communities pointed out that the development of the mitigation plan would not have been done without Project Impact funding. For example, Chattooga County, Georgia, hazard mitigation officials stated that the Project Impact program provided funding and technical assistance that enabled them to assess their risks and develop a local mitigation plan that prioritized projects to address these risks. As a result, the community is developing a project to connect six separate water systems within the county to address their drought risk. Third, Project Impact was important for obtaining additional funding from the private sector to promote and implement mitigation. State and community officials pointed out that they utilized their Project Impact grant as “seed money” to attract additional funding from businesses, nonprofits and other government agencies. For example, Centerville, Utah, received $500,000 in Project Impact funds in 1998 that it utilized in part to host several meetings and outreach sessions with local businesses and government officials to solicit additional funding. The outreach effort allowed them to leverage an additional $2,134,447 through partnerships with the private and public sector. This additional funding enabled the city to address many of its mitigation needs such as upgrading the city’s storm drainage system, constructing a debris basin to eliminate the downstream flood hazard, and retrofitting buildings to better stabilize them against earthquakes. Lastly, the state and community mitigation officials also stated that the education and outreach aspects of the program were instrumental in prompting the public and private sector to undertake mitigation activities. They told us that this was one of the strongest points of the program, as it increased the public’s awareness and concern about mitigation and in the view of some officials, became an impetus for achieving mitigation efforts without requiring government funding. For instance, according to information provided by Deerfield Beach, Florida, one citizen credited the outreach efforts of the local Project Impact program for motivating him to utilize his own funds to construct Marina One, a disaster-resistant structure for housing boats. FEMA’s fiscal year 2003 budget request proposes eliminating the HMGP and establishing a new $300 million program for predisaster mitigation. This proposed program would award mitigation grants on a competitive basis— instead of the current formula-based awards—to better ensure that funding goes to the most cost-beneficial projects. The proposal has raised concerns, however, about whether participation in mitigation activities might decrease and about how FEMA might implement the program. Concerns have been raised about demonstrating the cost-effectiveness of some mitigation projects. For example, in August 1999, we reported that although established procedures existed for selecting HMGP projects, FEMA exempted four categories of projects from benefit-cost analysis, including the purchase of substantially damaged properties in 100-year floodplains. These projects were exempt because program officials believe that, in the case of damaged properties in the floodplains, they were being consistent with the policies of the National Flood Insurance Program that allows the purchase of damaged properties in floodplains without benefit- cost analysis, or in the other cases because determination of benefits was difficult. Nevertheless, for these categories of projects—the number of which FEMA could not identify—the cost-effectiveness was unknown. Similarly, FEMA’s Office of Inspector General reported in March 1998 and again in February 2001 concerns about the cost-effectiveness of mitigation projects. The office pointed out that analyses had often not been done and techniques for conducting them were poorly understood. The Inspector General’s office also reported that many projects had been exempted from analysis. The administration has also had concerns about the cost-effectiveness of mitigation projects, and in FEMA’s fiscal year 2003 budget request, proposes eliminating the HMGP and establishing a $300 million predisaster mitigation program that would award grants on a nationally competitive basis. According to the budget request, the administration has concluded that 45 percent of HMGP projects undertaken from 1993 to 2000 were either minimally cost effective or not cost effective at all. Consequently, the administration proposed substantial changes to FEMA’s multihazard mitigation programs. Under the proposed new program, mitigation grants would be awarded on a nationally competitive basis—instead of the current formula-based awards—to better ensure that funding goes to the most cost-beneficial projects. According to Office of Management and Budget officials, future mitigation efforts funded by the federal government need to be those that provide the most benefit from a nationwide perspective and to not be limited primarily to states affected by disasters. The officials said that only through a program that does not allocate funds in any formula—but is instead based on objective criteria such as cost-effectiveness—can the government be best assured that it maximizes the value of its mitigation program funding. The administration stated that the program would ensure more stability to disaster mitigation efforts since a consistent level of mitigation assistance would be available to states and communities, and they would no longer be dependent on disaster declarations to obtain mitigation grants. Further, according to the administration, a consistent level of funding would allow states and communities to develop more comprehensive proposals and projects to reduce their overall risks, consistent with state and local mitigation plans and would also strengthen states’ capability to pursue their mitigation priorities. From our analysis of the proposed program and discussions with FEMA and state hazard mitigation officials, concerns have been raised about whether participation in mitigation activities might decrease and about how FEMA might implement the program. Specifically, there are concerns that (1) FEMA and states may not be able to take advantage of interest in participating in mitigation activities that often emerges after a disaster has struck; (2) some states might be entirely excluded from mitigation funding; (3) outreach and planning activities that help increase participation in mitigation might be curtailed; and (4) FEMA might face challenges, such as establishing a process for comparing the costs and benefits of projects, in implementing the new program. The proposed program may limit the ability of emergency management officials to take advantage of mitigation opportunities. State officials with whom we spoke maintained that the postdisaster environment is the most conducive for implementing mitigation efforts, and that it can be difficult to maintain public or private sector support for mitigation in a predisaster environment. To illustrate this point, officials from Ohio noted how the public interest in constructing safe rooms has diminished since a tornado struck the community of Xenia in 2000. Despite the current availability of predisaster funds, businesses that expressed interest in constructing public safe rooms in the immediate aftermath of the disaster have now, 2 years later, shown little interest in doing so. Similarly, North Carolina officials noted how state support for mitigation has diminished since the devastation of Hurricane Floyd in 1999. In June 2002, in an attempt to address serious budgetary issues, the state legislature began an effort to reallocate some of the funds that had already been obligated to mitigation after Floyd. As a result, the remaining 30 percent of the planned buyouts are in jeopardy, according to state officials. The National Emergency Management Association (NEMA) has expressed similar views. Its official position paper on the budget proposal notes, “in the tight fiscal environment that states and communities are facing, the commitment of funding is most likely to occur only shortly after they have experienced devastation.” All states might not participate in mitigation activities under the new proposal. Many states rely on federal funding to support their mitigation programs, and without the current formula-based programs to provide a minimal level of funding support, their mitigation programs may not continue. According to NEMA, at least 10 states derive all funding for managing the state’s hazard mitigation program from the current federal mitigation programs, and officials from other states told us that state legislatures that currently provide mitigation program funding often require a track record of federal funding for a program before they will provide additional or continual funding for staff working on such programs. State officials said that without a base level of support from the federal government, a number of state mitigation programs will no longer exist, because the states will no longer employ the staff needed to implement and support a competitive program. Several state officials said that such diminished funding will not achieve the new program’s objectives of developing better projects or strengthening their ability to pursue mitigation priorities. Moreover, they added that this deviates from the manner in which the Congress recently mandated that predisaster funds be allocated, as the Disaster Mitigation Act of 2000 directed a program for predisaster mitigation that involved all states. The public outreach and planning activities that were widely conducted under the Project Impact program may be jeopardized under a competitive predisaster mitigation program. Both FEMA and state officials said that such activities are essential to creating a positive environment for mitigation, because these activities create grassroot support and interest in conducting mitigation. However, both groups stated that establishing the financial benefit of these activities is difficult. For example, North Carolina officials pointed to the Project Impact efforts in Wilmington that involved distributing hurricane maps to all schoolchildren showing flood and storm surge areas, hurricane preparedness actions, and possible mitigation measures. The officials said this activity is very beneficial in building support for mitigation—and ultimately persuading communities and individuals to give high priority to mitigation and to make their own investments in mitigation measures—but that it would be extremely difficult to demonstrate a financial benefit commensurate with the cost. Concerns also exist about whether mitigation planning might decrease under the proposed program. According to state and local mitigation officials, Project Impact’s emphasis on planning assisted communities in identifying vulnerabilities, assessing risks, and developing and prioritizing mitigation projects to address their needs. Some state and community officials pointed out that the development of the mitigation plan would not have been done without Project Impact funding. FEMA had permitted Project Impact to be used to develop and update plans; state and local officials are concerned that with the new nationally competitive program, such support may diminish. FEMA may face challenges in designing and implementing the proposed program, particularly in selecting projects on a competitive, nationwide basis. Most significantly, FEMA has not yet established a viable process for comparing the relative costs and benefits of competing mitigation projects. The current benefit-cost analysis model does not fully measure the indirect benefits associated with projects. FEMA has acknowledged that its current benefit-cost analysis model does not capture all the indirect benefits of projects, such as environmental and social benefits, or mitigation activities such as outreach and planning. In this regard, FEMA is funding a study that examines the benefits of mitigation and which will, in part, address the issue of measuring the benefits of outreach, planning, and other activities that have benefits that are hard to quantify. However, FEMA does not expect this study to be completed and possibly used to improve benefit- cost analyses until 2004 at the earliest. State mitigation officials agreed that FEMA would have difficulty in applying benefit-cost analyses to mitigation projects in a competitive program. They said that not only is it difficult to demonstrate the benefits of certain projects and the indirect benefits of others, but that the current analyses are difficult to do and are used primarily for determining project eligibility rather than on determining the full project benefits. In this regard, they said that in doing these analyses under the HMGP, they frequently discontinued additional analysis of the benefit of a project once the ratio of benefits to cost were equal—which meets the minimum program requirements. The officials said that this is the primary reason why the administration views many projects as only minimally cost effective. Further, FEMA faces challenges in staffing and operating a nationally competitive mitigation program. Both FEMA and state officials said that states currently perform most of the analysis and selection of projects, while FEMA provides final approval. However, under a nationally competitive program, they said that FEMA will be required to play a greater role in order to administer a fair and effective competition, and will need additional staffing. FEMA mitigation officials expect that a minimum of 41 permanent employees will be needed to staff a new competitive predisaster mitigation program. Additionally, the officials said that FEMA would need budget authority to fund the new positions because they are prohibited from using the disaster relief fund—currently used to fund temporary employees to conduct the HMGP—for predisaster activities. FEMA officials are aware of concerns about the proposed predisaster mitigation program and plan to address these concerns if legislation creating the new program is enacted. Moreover, FEMA has already provided some indications of how it might implement the program. Regarding concerns about the elimination of the HMGP, the FEMA Director acknowledged, in response to questions raised during appropriation hearings this year, that unique opportunities for mitigation exist in the immediate aftermath of a disaster and agreed that the HMGP has been effective in enabling states and communities to complete critical mitigation work during this period. Consequently, he stated there is a need for both pre- and postdisaster mitigation efforts and that if the Congress adopts the proposal in its current form, FEMA would attempt to design the program with sufficient flexibility to assist communities with postdisaster mitigation activities. FEMA officials also said that they agree with states that a base level of funding for all states will better enable mitigation programs to succeed. They told us that this funding would be essential for states to enable them to participate in the proposed competitive program. However, as discussed earlier, the proposal, as currently written, would appear to prohibit FEMA from providing this guaranteed base. FEMA officials stated that they would attempt to work with states as well as OMB to develop a funding mechanism that would ensure that all states maintain a mitigation program. Regarding the challenges that FEMA might face in implementing a competitive evaluation and selection process, FEMA has emphasized that it will collaborate with its state partners and other stakeholders in defining the competitive grant program and policy. This effort would include developing a fair, reasonable, and appropriate means for competitive review and selection of grant proposals. For example, FEMA officials stated that they would like to base their decisions on more than just cost- effectiveness and that they currently envision criteria that would focus on the quality of the proposed projects and the ability of the projects to address state and community mitigation priorities, as well as cost- effectiveness. FEMA recently asked for input on how to best address these challenges. On August 6, 2002, it issued a notice in the Federal Register soliciting comments and ideas from interested parties on the process for implementing the mitigation grant program on a competitive basis. FEMA requested responses on specific concerns, which among other things included (1) how applications should be evaluated to ensure that the most cost-beneficial projects are funded, (2) the type of activities that should be funded, (3) whether funds should be set aside for states to maintain a level of mitigation capability, and (4) whether funds should be set aside for planning in addition to competitive grants. FEMA expects to begin consideration of the comments it receives in the fall if the proposed predisaster mitigation grant program is included in its fiscal year 2003 appropriations. The events of September 11, 2001, demonstrated the vulnerability of our nation to terrorist attack, and subsequent efforts have been initiated to strengthen the nation’s homeland security. These events, as well as the proposal to establish a Department of Homeland Security, represent a substantially changed environment under which FEMA and its hazard mitigation programs operate now and will operate in the future. As a result, in addition to the proposal to change the multihazard mitigation programs, a number of broader issues face hazard mitigation efforts. These issues include the following: The potential that an emphasis on terrorism efforts may result in a decrease in natural hazard mitigation activities. The proliferation and overlap of plans and programs that address mitigation-related concerns that may cause duplication of effort and confusion. The effective use of HMGP mitigation funds to reduce the risk of terrorism damage and associated hardship, loss, and suffering is not clear. The proposed placement of FEMA within the DHS places functions that have traditionally not been security related, such as hazard mitigation, into a department whose primary mission will be to provide a secure national environment, including actions to prevent and prepare for possible terrorist events. Supporters of FEMA’s transfer in its entirety to DHS argue that dual use of funding for natural and man-made disasters and emergencies is appropriate in an “all hazards” approach to disaster assistance. For example, the Director of FEMA’s Office of National Preparedness said that leaving FEMA intact in DHS would enhance the agency’s preparedness capabilities, not detract from the agency’s natural disaster response and recovery functions. Further, FEMA mitigation officials said that they are currently working to identify terrorism mitigation activities that are also “all hazard” and address natural hazard mitigation priorities. Concerns have been raised that with the emphasis on terrorism preparedness in the aftermath of September 11th, the transfer of FEMA to DHS may result in decreased emphasis on mitigation of natural hazards. Opponents of the FEMA transfer, such as a former FEMA director, said that activities not associated with homeland security would suffer if relocated to a large department dedicated essentially to issues of homeland security. They contend that the agency’s disaster mitigation programs and other efforts integral to FEMA’s current mission that have no bearing on homeland security will be compromised. They argue that agency resources dedicated to those functions have already been, and would continue to be, diverted to the homeland security mission, resulting in diminished federal capabilities for nonnational security activities. As a result of the terrorist attacks, many new initiatives have been undertaken to begin addressing security concerns; however, many of them raise questions regarding the role and relationship of preparedness and mitigation efforts. FEMA requires states and communities to develop mitigation plans to obtain mitigation funding; however, other proposed legislation calls for similar, but more specialized, homeland security preparedness plans that may overlap with the required mitigation plans. For example, proposed legislation directed at increasing port security will require all facilities in port areas, as well as the Department of Transportation, to develop plans for action to deter and minimize damage from catastrophic emergencies. FEMA hazard mitigation officials said that they are aware that there were numerous and related planning requirements being placed on communities, and that they are working toward identifying and minimizing the impact of such requirements. The officials said they are confident that they will become aware of all such requirements due to the plans to consolidate preparedness efforts within FEMA. They said that planning requirements that address mitigation-type efforts will be adequately coordinated and, where appropriate, incorporated by reference into overlapping or related plans to minimize the burden on all stakeholders. The new initiatives may also result in duplication or overlap in programs. Many programs are being initiated that address the predisaster environment, most significantly the $3.5 billion first responder grant program proposed by the administration to fund state and local first responders for terrorist attacks. The first responder grant program would provide funding to states and local governments to prepare for terrorist events, and a portion of this preparedness may involve activities that could be viewed as mitigation. Other programs, such as the Emergency Preparedness Enhancement Pilot Program, which is contained in proposed DHS legislation, may also involve the development of and funding for mitigation related activities, because it will provide funds for improved security measures at private entities. The number and size of these programs could result in duplication of effort and confusion among the state and local governments partnering in mitigation efforts. We found such problems occurred in the past with other assistance being provided to states and localities. We reported in September 2001, for example, that first responder training and assistance programs were being conducted by three federal organizations—FEMA, the Department of Justice, and the Federal Bureau of Investigation—which resulted in overlapping and duplicative activities and caused confusion on the part of state and local officials. As discussed earlier, HMGP funds have been typically made available to states following presidentially declared disasters in amounts totaling as much as 15 percent of the federal grant funds spent on the disaster. HMGP funds have historically been used for natural hazard mitigation, although no restrictions have been made on the types of disasters for which these funds are made available. Consequently, HMGP funds can be, and have been, made available after disasters resulting from terrorist attacks. In fact, according to FEMA officials, after the 1995 explosion at the federal building in Oklahoma City, HMGP funds were made available to Oklahoma. The amount provided was relatively small—$1 million—which FEMA officials said was due to the low amount of disaster assistance funds spent on this disaster. According to these officials, the mitigation funding provided to Oklahoma was used for natural hazard mitigation because FEMA has traditionally interpreted the HMGP authority to limit funding to only natural hazard mitigation projects. As shown by the disaster in New York, the HMGP funding that could be provided in response to terrorist events may be substantial. Currently, FEMA has been authorized to fund disaster assistance to New York approaching $9 billion. Based on this level of assistance funding and the current 15 percent HMGP funding formula, New York could have received about $1.3 billion in HMGP funding for mitigation projects. President Bush, however, has limited the amount of HMGP funds the state can receive. In a September 18, 2001, amendment to his major disaster declaration for New York, the President stated that because of the unique nature and magnitude of this event, federal funds from the HMGP would be limited to 5 percent of the aggregate amount of federal grant assistance. FEMA officials said that at this percentage rate, HMGP funding to New York might total about $417 million. The key objective of the HMGP is to reduce the risk of future damage, hardship, loss, or suffering; however, it is not clear how mitigation funds can be effectively used to reduce the risk of terrorism damage and associated hardship, loss, or suffering. FEMA officials said that it would be difficult to develop a benefit-cost methodology for terrorism mitigation, because there is little data upon which to calculate the likelihood of an event and thereby determine the project’s benefit. FEMA officials said that they are undertaking a pilot program with New York to identify terrorism- related hazard mitigation measures, such as physical protection and security-related projects that can meet cost-effectiveness criteria. FEMA’s current multihazard mitigation programs are viewed positively by the emergency management community, but questions about the programs’ cost-effective projects have lead to a proposal to consolidate and revise them. The focus of the proposed new program on obtaining the most cost- effectiveness projects, in light of current budget concerns, is well intended. However, the issue facing decisionmakers is whether the proposed revision to the program will make the program more effective in achieving disaster mitigation objectives. The structure of the new program may not be able to capitalize on the characteristics of the current programs that have been viewed as successful—such as acting in the postdisaster environment to quickly take advantage of mitigation opportunities and undertaking outreach activities to develop grassroot support for mitigation. A balance that includes these characteristics in the program may need to be struck, and we are encouraged to see that FEMA is obtaining input and consensus on how to best structure the new program if it obtains congressional approval. Furthermore, without careful structuring of the program, FEMA’s hazard mitigation program may not remain consistent with the approach of disaster mitigation legislation passed only 2 years ago by the Congress that emphasized involvement by all states, funding for planning activities, and increased postdisaster mitigation funding for states willing to undertake enhanced mitigation planning efforts. The proposed inclusion of FEMA in DHS and, in the broader context, the heightened concern over terrorism raises more fundamental issues about hazard mitigation efforts, such as (1) how natural hazard mitigation activities would fare in the new department that focuses on terrorism, (2) whether planning and program efforts in the mitigation and preparedness area should remain separate and distinct, and (3) how the HMGP—and the legislation authorizing it—address the role and rationale for mitigation after a terrorism-caused disaster. We provided a draft copy of this report to FEMA for its review. The FEMA Director, in a September 24, 2002, letter commenting on the report, generally agreed with the information presented and noted that the report supports his belief that both pre- and postdisaster mitigation programs are critical to FEMA’s success in reducing disaster losses. Additionally, the Director stated that the expertise the agency has developed in natural hazard mitigation is clearly applicable to the homeland security mission, and FEMA looked forward to addressing the opportunities presented by the proposal to include it in the new Department of Homeland Security. FEMA also provided some technical comments that we considered and have incorporated into this report where appropriate. FEMA comments are contained in appendix IV. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report for 10 days. At that time, we will send copies of this report to the appropriate congressional committees; the Director of the Federal Emergency Management Agency; and the Director of the Office of Management and Budget. We will also make copies available to others upon request. In addition, this report will be available at no charge on the GAO Web site at http://www.gao.gov. If you have any questions regarding this report, please contact me at (202) 512-2834 or at heckerj@gao.gov. Key contributors to this report were Mark Abraham, Colin Fallon, Kirk Kiester, Aisha Cabrer, John McGrail, and Jack Schulze. Debate has emerged in recent years about the effectiveness of the Federal Emergency Management Agency’s (FEMA) multihazard mitigation programs—the Hazard Mitigation Grant Program (HMGP) and the Project Impact program. The administration also has proposed, in FEMA’s fiscal year 2003 budget request, to change the multihazard mitigation programs to improve their effectiveness. Further, the recent proposal to create the Department of Homeland Security includes moving FEMA into that department, which may also impact on the overall conduct and content of these programs. The Chairman of the Subcommittee on International Security, Proliferation, and Federal Services, Senate Committee on Governmental Affairs, asked us to determine the available viewpoints on the effectiveness of these mitigation programs and the possible impacts of the two proposals. We addressed the following objectives: What are the characteristics of FEMA’s current multihazard mitigation programs, and what do states perceive as these programs’ most successful features? How would the proposed program change FEMA’s current approach to mitigation, and what are some of the concerns that have been raised about this proposal? What are the issues resulting from the increased federal focus on homeland security on conducting hazard mitigation efforts? To determine the characteristics of FEMA’s multihazard mitigation programs, we reviewed FEMA’s Hazard Mitigation Grant Program and Project Impact program regulations, policy guidance and handbooks, which identified and described the programs’ purpose, goals, eligibility criteria, cost-effective criteria and funding. We also examined relevant legislation that described the programs’ objectives, funding, and the focus of its activities. We conducted a review of the available literature on the multihazard mitigation programs, including past GAO, FEMA Inspector General, and other reports that provided a perspective on these programs. We also discussed these programs with FEMA officials in Washington, D.C., as well as in its regional offices in Atlanta, Georgia; Denver, Colorado; and Chicago, Illinois. To determine state mitigation officials viewpoints on the successful features of these programs, as well as their overall perspectives on the programs, we interviewed state hazard mitigation officials from 24 states within 4 FEMA regions (IV, V, VII and VIII) to obtain their views about their experiences administering and utilizing these programs. We selected these regions and the states within these regions because they provide a representation of small and large states that contain urban and rural communities that have received both small and larger amounts of multihazard mitigation funding. These states also have varied experience with disasters. We examined and synthesized documents provided by these officials detailing their experiences with these mitigation programs. We also conducted site visits and interviewed local hazard mitigation officials in Georgia, Florida, and North Carolina because these states have a wide variety of pre- and postdisaster mitigation projects and are very active in both the HMGP and Project Impact program. We also reviewed studies available from the Natural Hazards Research and Applications Information Center and from FEMA that addressed the benefits and results of both the HMGP and the Project Impact program. In addition, we met with officials in OMB’s Financial Institutions Branch to obtain their perspectives on the effectiveness of the current programs as well as on the objectives for the proposed new mitigation program. To determine how the current legislative proposals might change FEMA’s mitigation programs, we interviewed FEMA headquarters and regional mitigation officials to gain their perspective about the proposed changes. Specifically, with regard to the proposal to establish a new predisaster mitigation program, we obtained their viewpoints on what challenges they would confront in (1) developing the criterion and processes of selecting mitigation projects; (2) addressing administrative issues, such as staffing; and (3) addressing any statutory issues from replacing the HMGP with a new competitive grant program. We also gained the perspectives of state hazard mitigation officials we interviewed on how they perceived the proposed changes would impact their ability to pursue mitigation activities. We also reviewed available literature that presented the viewpoints of various organizations on either the advantages or disadvantages of the proposed program. To determine the issues related to conducting hazard mitigation efforts as a result of the increased federal focus on homeland security, we drew upon recently completed work that is examining the challenges surrounding the establishment of that department. This work included assessments of the administration’s proposal to establish a Department of Homeland Security, examinations of the relationships between federal, state, and local governments in undertaking terrorism preparedness efforts, a review of legislative proposals related to the Coast Guard and port security, as well as ongoing work that assesses port security efforts. We also discussed the effects of including mitigation activities with FEMA mitigation officials to determine from them the concerns that exist over the movement of mitigation activities into the Department of Homeland Security. We conducted our review from November 2001 through August 2002 in accordance with generally accepted government auditing standards. Sum of federal share – obligated (FY 1996 thru 2001) $3,081,072 (Continued From Previous Page) Mobile County with Town of Dauphin Island & City of Bayou La Batre Clay County/City of Piggott/City of Corning/City of Rector City of West Memphis; South Arkansas Community Development County of Santa Barbara/City of Santa Barbara Las Virgenes Malibu Council of Governments (includes the cities of Agoura Hills, Calabassas, Hidden Hills, Malibu & Westlake Village) Region of San Luis Valley (Counties of Alamosa, Conejos, Costilla, Mineral, Rio Grande, & Saguache) City of Deerfield Beach/Broward County Tampa Bay Region (Counties of Hillsborough, Manatee, Pasco, and Pinellas & 38 incorporated municipalities) Tri-County Council of Southern Maryland: Calvert, Charles, & St. Mary’s Counties Upper Mystic River Basin Watershed (in Middlesex County; includes communities of Arlington, Burlington, Lexington, Medford, Reading, Stoneham, Wilmington, Winchester, & Woburn) Cape Cod Commission (includes 15 Towns that comprise Barnstable County) City of Charlotte & Mecklenburg County City of Wilmington & New Hanover County Buncombe County & all incorporated municipalities Lenoir County & all incorporated municipalities The Eastern Band of Cherokee Indians Research Triangle Region (includes Wake, Durham, & Orange Counties with Research Triangle Park) Luzerne County Flood Control Authority/Mitigation Advisory Board (Includes the counties of Luzerene, Columbia, Montour, Northumberland, and Snyder) Anderson County, including the cities of Clinton, Lake City, Norris, Oak Ridge, & Oliver Springs Harris County to include Bellaire, Webster, & Houston Two River-Ottauquechee Regional Planning Commission (includes most of Orange & Northern Windsor Counties and the Towns of Pittsfield, Hancock, and Granville) North West Regional Planning Commission (includes 23 towns in Franklin & Grand Isle Counties) Addison County Regional Planning Commission (includes Addison County and 21 Towns in the Region) Roanoke Valley District Planning Commission (Roanoke County, City of Roanoke, City of Salem, Town of Vinton) Central Shenandoah Planning District (Augusta, Bath, Highland, Rockbridge & Rockingham Counties; Cities of Buena Vista, Harrisonburg, Lexington, Staunton, & Waynesboro; and 11 towns) The General Accounting Office, the investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability. The fastest and easiest way to obtain copies of GAO documents at no cost is through the Internet. GAO’s Web site (www.gao.gov) contains abstracts and full- text files of current reports and testimony and an expanding archive of older products. The Web site features a search engine to help you locate documents using key words and phrases. You can print these documents in their entirety, including charts and other graphics. Each day, GAO issues a list of newly released reports, testimony, and correspondence. GAO posts this list, known as “Today’s Reports,” on its Web site daily. The list contains links to the full-text document files. To have GAO e-mail this list to you every afternoon, go to www.gao.gov and select “Subscribe to daily E-mail alert for newly released products” under the GAO Reports heading. | Over the past 12 years, federal disaster assistance costs have totaled more than $39 billion (in fiscal year 2001 dollars)--a nearly fivefold increase over the previous 12-year period--as a result of a series of unusually large and frequent disasters and an increasing federal role in assisting communities and individuals affected by disasters. The Federal Emergency Management Agency (FEMA), the lead agency for providing federal disaster relief, has provided the bulk of the assistance to help those in need respond to and recover from disasters. As the costs for disaster assistance have risen, FEMA has made disaster mitigation a primary goal in its efforts to reduce the long-term cost of disasters and has developed mitigation programs designed to minimize risk to property or individuals from natural or man-made hazards. FEMA's multihazard mitigation programs differ substantially in how they have sought to reduce the risks from hazards but each has features that the state emergency management community believes has been successful for mitigation. The Hazard Mitigation Grant Program (HMGP), FEMA's oldest multihazard mitigation programs, is a post disaster program that has provided the bulk of mitigation assistance to states and communities. State mitigation officials view the HMGP as a highly successful means for achieving mitigation because commitment to undertake mitigation efforts is greatest in the aftermath of a disaster, and the HMGP takes advantage of this "window of opportunity." FEMA has used its more recent and smaller predisaster Project Impact program to provide funding directly to communities in every state, regardless of whether the state had recently experienced a disaster. State and local officials said that Project Impact has been successful in increasing awareness of and community support for mitigation efforts due to its funding of these types of activities. The proposed new mitigation program would fundamentally change FEMA's approach by eliminating the postdisaster HMGP and by funding mitigation activities on a nationally competitive basis. The administration believes that the new program will ensure that mitigation funding remains stable from year to year and that the most cost-beneficial projects receive funding. The heightened focus on homeland security has raised several issues related to the conduct of hazard mitigation activities. Foremost among these issues is whether the increased emphasis on preventing and preparing for terrorist events will result in less focus on natural hazard mitigation concerns. |
The Food Stamp Program helps low-income households (individuals and families) obtain a more nutritious diet by supplementing their income with food stamp benefits. In fiscal year 1996, the average monthly food stamp benefit was $73 per person. These benefits are generally provided through coupons or electronically on a debit card (similar to a bank card) that may be used to purchase food at stores authorized to receive food stamps. The Food Stamp Program is a federal-state partnership, in which the federal government pays the full cost of the food stamp benefits and approximately half of the states’ administrative costs. USDA’s Food and Nutrition Service (FNS)—formerly, the Food and Consumer Service—administers the program at the federal level. The states’ responsibilities include certifying eligible households, calculating the amount of benefits, and issuing benefits to participants who meet the requirements set by law. The Welfare Reform Act overhauled the nation’s welfare system and significantly changed the Food Stamp Program. In addition, the Fiscal Year 1997 Supplemental Appropriations Act (P.L. 105-18, June 12, 1997) included new authority allowing the states to purchase federal food stamps to provide state-funded food assistance for legal immigrants and able-bodied adults without dependents who are no longer eligible for federal food stamps under the Welfare Reform Act. Under the supplemental act, states are required to receive approval from FNS to distribute additional food stamps and to fully reimburse the federal government in advance for all costs associated with providing the benefits. In addition, the states’ food stamp programs must be cost-neutral to the federal government. Changes to the Food Stamp Program included imposing time limits on those able-bodied individuals between the ages of 18 and 50 without dependents who were not working at least 80 hours a month or participating in certain kinds of employment and training programs. This work requirement was effective not later than November 22, 1996. States were required to terminate food stamps for these nonworking able-bodied adults without dependents after 3 months within any 36-month period. Disabled individuals, if they meet eligibility requirements, can still receive assistance. The act allows FNS to grant waivers to states for exempting able-bodied adults without dependents from the work requirement if they live in an area where unemployment is over 10 percent or in an area with an insufficient number of jobs. FNS generally grants waivers for a 1-year period. Once approved, these waivers may be renewed if the areas covered continue to have high unemployment or insufficient jobs. Once the waivers are approved, the states or localities can choose to either implement them in whole or in part, or choose not to implement them at all. In addition, The Balanced Budget Act (P.L. 105-33, Aug. 5, 1997) gives states the discretion to exempt certain types of able-bodied adults without dependents from the work requirement—up to 15 percent of those not otherwise waived. The Balanced Budget Act also provided an additional $131 million for each of the next 4 years to the Food Stamp Program—80 percent of it is designated for employment and training opportunities for these adults. According to data from FNS for fiscal year 1995—the latest year for which data were available—in an average month, the Food Stamp Program provided benefits for 27 million people. Of these, 2.5 million were able-bodied adults between the ages of 18 and 50 without dependents. An estimated one-half of these adults, about 1.3 million, are subject to the 3-month time limit. In addition, the Welfare Reform Act and the Supplemental Appropriations Act allowed immigrants with legal status as of August 22, 1996, to retain food stamps up to August 22, 1997. However, if legal immigrants have 40 quarters or more of work history in the United States or are veterans or active duty members of the U.S. military, they may continue to retain food stamps. Spouses and minor children of veterans are also eligible. At the time of our survey of states in the summer of 1997, the states were pursuing a variety of options to address changes in the Food Stamp Program that affected able-bodied adults without dependents and legal immigrants. Some state actions, such as job training assistance, although primarily intended to move individuals toward self-sufficiency, may have the effect of allowing some able-bodied adults without dependents to retain food stamps by meeting the act’s work requirements. Twenty states provided legal immigrants with information on how to become citizens so that they can be eligible for food stamps. Other state actions are intended to replace the food stamp benefits that individuals have lost. According to our survey results, when the states notified able-bodied adults without dependents that they were subject to the work requirements in order to retain food stamps, many states told us that they chose to also notify these adults of job placement and/or training services that were available. Although these programs are intended primarily to move individuals toward self-sufficiency, participants may still receive food stamps if income and other requirements are met. For example, our survey indicated that Texas provided information on jobs and/or employment resources and training. Thirty-two states provided information about jobs and/or employment resources; 29 provided information on training; 19 provided information on workfare. In addition, 20 states helped assess an individual’s employment skills. The states also offered one or more ways to meet the work requirements: 25 states counted volunteer work, 25 counted workfare, and 33 counted employment training that leads to a job. In addition, as allowed under the Welfare Reform Act, our survey indicated that 43 states had applied for, and 42 received, authority to waive the work requirement for able-bodied adults without dependents in areas where unemployment exceeded 10 percent or in areas with insufficient jobs. (See app. III for the waiver status of each state.) FNS estimated that as many as 35 percent of the affected able-bodied adults without dependents would retain their eligibility through a waiver. However, 8 of the 43 states were not planning to implement their waivers—either in their entirety or in part. In seven states—California, Indiana, Nevada, New York, Ohio, Oklahoma, and Virginia—waivers were approved for selected regions but the local governments, which are authorized to implement the waivers, did not plan to do so. Texas planned to implement its waiver only in localities with an unemployment rate of over 10 percent. Two of the eight states or their localities had not fully implemented the waivers because they believed that it was unfair to exempt able-bodied adults without dependents from the work requirement while single mothers receiving federal assistance, like Temporary Assistance for Needy Families (TANF), are required to participate in work activities. At the time of our survey, 20 states provided or planned to provide legal immigrants—who were scheduled to lose their food stamps—with information on how to become U.S. citizens. In May 1997, we reported that it took between 112 and 678 days (with an average of 373 days) to process applications for citizenship at INS between June of 1994 and June of 1996. For example, it took just over 1 year to process a request for citizenship in Los Angeles—a city with one of the largest immigrant populations in the nation—and almost 2 years to process an application in Houston. INS officials told us that among the reasons for the significant increase in the number of applications that INS has received since fiscal year 1989 is that there are incentives to becoming a citizen because of the benefits that can be derived. Because it takes an average of over 1 year to process applications for citizenship and legal immigrants were not eligible to receive food stamps after August 22, 1997, many legal immigrants have lost their federal food stamp benefits. FNS’ most current estimate is that 935,000 legal immigrants lost their federal food stamps under the welfare reform provisions. However, as of December 1997, estimates are that over one-quarter, or about 241,000, of these individuals are receiving food stamps funded by the states. For those individuals who lose federal food stamp benefits, 20 states were taking one or more actions to provide state-funded food assistance. (App. IV identifies the 20 states with food assistance programs that serve able-bodied adults without dependents and/or legal immigrants.) Ten states decided to purchase federal food stamps with their own funds for certain legal immigrants—primarily children and the elderly. According to FNS and the states, 9 of the 10 states have estimated that about 241,000 legal immigrants are now receiving state-funded food stamps. Among these states are California, Florida, New York, and Texas, which, according to an FNS report, had about 70 percent of the legal immigrants receiving food stamps in fiscal year 1995—the latest year for which data were available. These states will generally use the Food Stamp Program’s infrastructure and benefit structure to deliver food assistance, according to FNS. For example, Washington State appropriated just over $60 million for fiscal years 1998-99 to fully restore benefits to an estimated 38,000 legal immigrants—all of whom were slated to become ineligible for federal food stamps. Households eligible for participation receive the same benefits that they did under the federal program. However, FNS also told us that, unlike Washington State’s, most states’ eligibility standards are likely to apply to only certain categories of legal immigrants. California, for example, recently appropriated $34.6 million to provide food stamps for legal immigrants who are children or elderly. Thirteen of the 20 states reported that they were using their own state-funded food assistance programs, and 2 of the 13 states created programs in response to welfare reform. These two states—Colorado and Minnesota—developed state-funded food assistance programs to aid those legal immigrants losing their federal food stamp eligibility as a result of welfare reform. Colorado, for example, has appropriated $2 million to provide emergency assistance, including food, for legal immigrants. Minnesota has allocated just over $4.7 million for two programs to provide food assistance for legal immigrant families in that state. The remaining 11 states had food assistance programs—created before the Welfare Reform Act was passed—that ranged from those that provide individuals with cash directly to those that provide funds for local food banks and food pantries that serve, among others, both able-bodied adults without dependents and legal immigrants. A program with significant funding is Pennsylvania’s State Food Purchase Program, which provided about $13 million in fiscal year 1997 and $13.6 million in fiscal 1998 to counties for the purchase of food. This program is intended to supplement the efforts of food pantries, shelters for the homeless, soup kitchens, food banks, and similar organizations to reduce hunger. Two states with state-funded programs are also providing existing state or local programs with additional funding to assist able-bodied adults without dependents and legal immigrants. Rhode Island appropriated $250,000 in fiscal year 1998 for a community-run food bank. Massachusetts increased the funding it provides for local food banks and food pantries from just under $1 million to $3 million in fiscal year 1998 in anticipation of an increased need by both groups. Seven of the 20 states reported that they had allocated additional money to federally funded programs that assist groups of individuals, which may include those losing food stamp benefits. Programs identified by the states in our survey include The Emergency Food Assistance Program (TEFAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). In the five localities we visited, government officials reported that their assistance largely consists of implementing state programs. Most nonprofit organizations that we contacted said that although it is too soon to assess the impact of welfare reform, they anticipate an increased need for their services. Given their limited resources, however, these organizations are concerned that the supplemental assistance they provide will not compensate for the basic food assistance provided by the federal program. (See apps. V-IX for information about food assistance programs in these localities.) In the five localities we visited—Denver, Detroit, Hartford, Houston, and Los Angeles—employment and training programs were offered to able-bodied adults without dependents through existing or expanded programs. Although these programs are intended to promote self-sufficiency, they may also help participants to retain food stamps if they meet income and work requirements. For example, in Hartford, Connecticut, able-bodied adults without dependents can participate in the statewide Connecticut Works System. This program’s objective is to enhance the state’s economy by helping to match the needs of businesses with workers’ skills. The Connecticut Works System brings together state, regional, and local organizations to provide job listings, job search assistance, access to training and education programs, resume assistance, interviewing, and networking assistance. In Detroit, Michigan, able-bodied adults without dependents can participate in a new state employment and training assistance program that specifically targets this population; about one-half of these adults live in the greater Detroit area. For fiscal year 1998, the state will receive $13.4 million from FNS to expand work programs for this population. For legal immigrants, three out of the five localities—Denver, Houston, and Los Angeles—had plans to offer limited food assistance through state-funded programs. California passed legislation to provide food stamps for legal immigrant children and the elderly by purchasing federal food stamps. Similarly, Colorado passed legislation that provides legal immigrant families with special emergency assistance. As a result, families in Denver can receive, among other things, food coupons redeemable at designated food pantries. Finally, Texas is planning to offer food assistance to elderly and disabled legal immigrants. In addition, Los Angeles County launched two special efforts on behalf of legal immigrants after the passage of welfare reform. First, Los Angeles initiated a countywide citizenship campaign that brought together 200 public and nonprofit organizations whose goal was to assist legal immigrants in obtaining citizenship. Los Angeles County coordinated the efforts of these organizations, worked with the local district office of the INS, and directly contacted 400,000 potentially affected legal immigrants. However, Los Angeles officials told us that (1) because of the time it takes to process applications for citizenship—including the fact that criminal background checks are required on all applicants—and (2) because many of the applications remain unprocessed owing to the volume of applications received by the INS, they estimate that about 91,000 legal immigrants lost their food stamps. Los Angeles County officials said they continue to encourage legal immigrants to become U.S. citizens and, for those who do become citizens, hope to restore benefits to those who meet the Food Stamp Program’s requirements. Second, Los Angeles County and the local United Way jointly sponsored the efforts of a local referral service to provide information on food assistance. Information on how to contact this referral service was enclosed with termination notices to legal immigrants. In every locality that we visited, nonprofit organizations—including food banks, food pantries, soup kitchens, and religious organizations—generally serve anyone who needs their services, including able-bodied adults without dependents and legal immigrants. Historically, these organizations provide supplemental food assistance on an emergency basis, perhaps once or twice a month. According to these nonprofit organizations, food stamp recipients—even before welfare reform—had turned to them for assistance. These organizations generally expect an increase in the need for their services—both in terms of the numbers of people and frequency of visits—as a result of welfare reform. For example, in Denver, one organization was getting 40 to 50 more applicants for food assistance per week in August 1997. Furthermore, the organizations were generally concerned that they could not replace the long-term, sustained assistance that food stamps provided. At the time of our visits in the late summer of 1997, however, most organizations had not experienced this anticipated increase. Our visits occurred before benefits were cut off for legal immigrants and before the usual increase in the need for food assistance in the winter months. The organizations are unsure how they will meet the expected increase because they have limited resources. Furthermore, these organizations are competing for these limited resources, and officials told us that they do not anticipate larger contributions as a result of welfare reform. While most organizations were waiting to see the full impact of welfare reform, some were developing contingency plans to handle the expected increase. For example, in Detroit, a kosher food pantry surveyed its existing clientele to determine which individuals would lose their benefits. The pantry learned that it would need about $100,000 the first year to serve its existing population. According to officials from the food pantry, this effort is not likely to be duplicated by other organizations because, unlike most other organizations, the kosher food pantry serves a known group of legal immigrants. More typically, most organizations are unsure how they will sustain a long-term increase in the number of people needing their services because they typically provide assistance on an emergency basis for anyone in need, and their resources are already limited. These organizations are considering strategies that would restrict eligibility, such as limiting eligibility to serve children or the elderly, in order to accommodate the anticipated increase and/or reduce their existing levels of service in order to accommodate the needs of more individuals. It is too soon to assess how able-bodied adults without dependents and legal immigrants will fare in the long term under welfare reform. However, many states have taken actions that could result in continuing food assistance, under certain conditions, for some of these individuals. For able-bodied adults, some of these actions—employment assistance and training—may help move these individuals towards self-sufficiency. For legal immigrants, citizenship could restore federal food stamps to those who meet income and work eligibility requirements. However, because of the amount of time it takes to process citizenship applications, many individuals have likely lost their food stamps. We provided USDA with a copy of a draft of this report for review and comment. We met with FNS officials, including the Acting Deputy Administrator for the Food Stamp Program. USDA concurred with the accuracy of the report but stated that while some states are providing or will provide food assistance for legal immigrants with state funds, in many cases, the assistance will not replace federal benefits because it generally targets only certain portions of the legal immigrant population, such as the elderly or children. USDA officials indicated that about one-quarter of the 935,000 legal immigrants that they estimated would lose food stamp benefits are now being covered under state funded programs. The USDA officials also pointed out that while many states are offering employment and training services for able-bodied adults without dependents, often, the services offered are job search activities, which do not satisfy the work requirements under the Welfare Reform Act and, thus, do not qualify these individuals for food stamps. We expanded our discussion of these points where appropriate and made some additional minor clarifications to the report on the basis of USDA’s comments. As arranged with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copies of this report to the Senate Committee on Agriculture, Nutrition, and Forestry; the House Committee on Agriculture; other interested congressional committees, and the Secretary of Agriculture. We will also make copies available upon request. If you have any questions about this report, I can be reached at (202) 512-5138. Major contributors to this report are listed in appendix X. In October 1996, the Ranking Minority Member, Subcommittee on Children and Families, Senate Committee on Labor and Human Resources, asked us to study several issues concerning the impact of welfare reform on the Food Stamp Program. This report focuses on the two groups of individuals most likely to lose their food stamp benefits—able-bodied adults without dependents and legal immigrants. Specifically, we describe the (1) actions, if any, that states have taken to assist those individuals who lose eligibility for the Food Stamp Program and (2) related actions, if any, taken by other organizations in selected localities—local governments and nonprofit organizations—to assist those individuals who lose their eligibility for the Food Stamp Program. To address the first objective, we surveyed and received responses from the 50 states and the District of Columbia. We also updated our results as appropriate. The tabulated results of the survey are included as appendix I. To address the second objective, we visited five localities. These localities were selected using the following criteria regarding the states in which they are located: (1) whether the states offered general relief to able-bodied adults without dependents and (2) whether the states had filed waivers precluding able-bodied adults without dependents from meeting the work requirement because of high unemployment or an insufficient number of jobs. We then selected states within these categories by (1) those with the highest food stamp participation of able-bodied adults without dependents and legal immigrants and (2) geographic diversity. Within the states, we chose the locality, usually a county, with the highest participation in the Food Stamp Program. We visited these localities in the late summer of 1997. We contacted several organizations that were significantly involved in providing the localities with food assistance. We also met with government officials responsible for food stamps and other officials involved in welfare reform. In several localities, we also met with officials affiliated with the Federal Emergency Management Agency (FEMA) because of their expertise in providing emergency food assistance after natural disasters. We also visited nonprofit organizations, such as community action agencies; food banks; church-affiliated food assistance providers, such as soup kitchens; local advocacy groups; local United Way affiliates; and food pantries. (See apps. V-IX for individual reports on the food assistance provided in these localities.) In addition, we contacted several national organizations that provide local communities with food assistance, including Catholic Charities USA; Lutheran Social Services; Second Harvest; World Share, Inc.; and the United Way of America. We also attended a conference sponsored by Second Harvest on the implications of welfare reform on food assistance. Additionally, we attended the American Public Welfare Association’s National Conference for Food Stamp Directors to obtain information on current state and local food assistance programs. Finally, we met with officials from the U.S. Department of Agriculture’s (USDA) Food and Nutrition Service (FNS) to obtain program information and statistics. We performed our work in accordance with generally accepted government auditing standards from March through December 1997. Los Angeles County had a population of about 9.1 million in 1995. In 1996, its unemployment rate was 8.3 percent, and its poverty rate for 1995-96 averaged 18.7 percent. In comparison, the state’s unemployment rate was 7 percent, and the poverty rate for 1995-96 averaged 16.8 percent. Nationwide, unemployment was 5.4 percent and the poverty rate for 1995-96 averaged 13.8 percent. In January 1997, as states were beginning to implement the Welfare Reform Act, over 1 million individuals participated in the Food Stamp Program in Los Angeles County. Of this total, over 189,000 were legal immigrants, and an estimated 56,400 were able-bodied adults without dependents. As of September 1997, after many changes to the Food Stamp Program were implemented, the county had about 870,000 food stamp participants of which about 31,000 were able-bodied adults without dependents and about 24,000 were legal immigrants. In addition, 29,000 legal immigrant children and elderly were receiving state-funded food stamps. The Los Angeles County Department of Public Social Services (DPSS) administers the Food Stamp Program, with guidance from the California Health and Welfare Agency. DPSS officials told us that they assist both able-bodied adults without dependents and legal immigrants in retaining food stamp benefits to the extent possible. For able-bodied adults without dependents, local officials were providing employment and training experiences through workfare. The county has expanded its workfare program to include from 80 percent of these adults prior to welfare reform to 100 percent. Officials were concerned that if these adults were not offered workfare to meet the work requirement, they would lose their food stamp benefits. For legal immigrants losing food stamp benefits, DPSS had an extensive notification process to advise them of their impending change in status for the Food Stamp Program as a result of welfare reform. DPSS sent out notification flyers entitled “You May Lose Your Food Stamp Benefits” to legal immigrants on five occasions and in several languages. The flyers described the process for obtaining citizenship. DPSS is providing assistance through a countywide effort in partnership with 200 public and nonprofit organizations. Activities have included providing assistance with applications for U.S. citizenship, including completing forms, and offering classes in English as a second language and in American government. However, because of the time needed to process citizenship applications by the Immigration and Naturalization Service (INS), including the fact that the INS has to do criminal background checks on all applicants for U.S. citizenship, Los Angeles officials indicated that about 91,000 legal immigrants lost their federal food stamp benefits. These officials indicated, however, that the citizenship campaign continues and they hope to be able to restore food stamps to those who qualify once they become U.S. citizens. At the time of our visit, DPSS was also considering what state and federal assistance could be provided. DPSS officials were awaiting the outcome of pending state legislation that would assist legal immigrants who were losing food stamps. In August 1997, the state legislature restored food stamps by purchasing federal food stamps for legal immigrants who are elderly or are children. County officials believed it was important to restore food stamps and other benefits to legal immigrants—particularly because they represent 15-20 percent of the population in Los Angeles County. Nonprofit organizations in Los Angeles County, some of which are affiliated with national groups, provide direct and indirect food assistance through a well-established network. These organizations are also connected with federal, state, and local government agencies to provide services. Officials in different organizations told us that this locality’s food assistance providers are effective in their efforts because of their experience in providing assistance following natural disasters, such as earthquakes, brush fires, and land slides, and because of experiences with rioting. These organizations generally expected to see an increase in the number of people needing their services as a result of welfare reform. Officials expressed concern that they would not be able to provide more services if their current level of resources remained the same. Additionally, several officials told us that resources for food and funding were diminishing. Accordingly, the organizations had developed the following approaches for handling the anticipated increase in needed services: (1) seeking additional donations for funds and food, (2) considering decreasing the amount of services that each recipient receives, and (3) targeting certain populations, such as the elderly, for services. Table V.1 describes the nonprofit organizations that we contacted. Religious-based social service nonprofit (local affiliate of Catholic Charities) Distributes food through its community centers to about 90,000 individuals twice a week. Food bank (local affiliate of Second Harvest) Distributes food to 750 charitable organizations at 14¢ per pound. These organizations distribute food to an estimated 200,000 individuals per week; also distributes federal agricultural commodities. Religious organization (a local affiliate of Lutheran Social Services) Distributes bags of groceries and hot meals to more than 2,500 families. Interfaith/Religious social service nonprofit (represents over 300 denominations and temples) Distributes food through programs such as food pantries, meals-on-wheels, “homebound meals,” and nutrition sites. Distributes funding to and purchases food for food pantries, soup kitchens, food banks, and homeless shelters. Community social service agency (local affiliate of the United Way) Donates funding for food assistance to 15 food service providers. Estimates serving 408,000 clients with food and meals service. Provides the needy with information about food pantries and soup kitchens throughout the Los Angeles area. Handles about 150 food assistance inquiries per day. Provides advocacy assistance for the poor in representing their views to local political officials on a number of issues, including food assistance. Most organizations did not have specific eligibility requirements for recipients of their food assistance services and did not keep demographic information on those they served. Generally, they serve anyone in need, including able-bodied adults without dependents and legal immigrants. Officials told us that their organizations serve the working poor, including single mothers with children and grandparents raising young children. The resources available to these organizations included federal, state, and local government grants, philanthropic grants, private donations, and in-kind donations, such as voluntary services and housing. For example, the city of Los Angeles provides some of these organizations with funding from its federal Community Development Block Grant. The population of Denver County in 1995 was approximately 500,000. In May 1996, the unemployment rate for the Denver metropolitan area was 3.9 percent. At the state level, the unemployment rate was 4.3 percent in May 1996, and the poverty rate for 1995-96 averaged 9.7 percent.Nationally, in May 1996, the unemployment rate was 5.4 percent, and the poverty rate for 1995-96 averaged 13.8 percent. In January 1997, as states were beginning to implement the Welfare Reform Act, about 56,000 individuals participated in the Food Stamp Program. By September 1997, after many changes to the Food Stamp Program were implemented, participation had declined to approximately 47,000. Between January and September, the number of able-bodied adults without dependents with food stamps decreased from about 1,600 to about 300. According to an official with the Denver Department of Social Services (DDSS), most of these adults lost food stamp benefits because they did not attend a required orientation session informing them of their work requirements under welfare reform. The information on this session was publicized through fliers at food pantries and soup kitchens as well as in the food stamp office. Although the number of legal immigrants on food stamps is unknown, a 1996 study by the Colorado Department of Human Services estimated that, statewide, approximately 5,700 immigrants would lose their benefits as a result of welfare reform. DDSS administers the Food Stamp Program in Denver County with supervision from the Colorado Department of Human Services. To assist able-bodied adults without dependents in meeting the Food Stamp Program’s work requirements, DDSS provides employment and training assistance through Denver Employment First. This program helps these adults prepare for jobs by teaching them resume writing, interviewing techniques, and appropriate dress. The program also offers General Educational Developmental (GED) self-study courses to move adults without a high school education toward earning a high school equivalency diploma. The program also operates the county workfare program for able-bodied adults without dependents and maintains a list of approved nonprofit agencies at which participants can meet their work requirements. In addition, DDSS is administering an emergency assistance program for legal immigrants in Denver County who lost federal food stamps. Colorado appropriated $2 million for emergency assistance to legal immigrants from July 1997 to June 1998. Under this program, legal immigrants can receive assistance, including food vouchers, that can be redeemed at designated food pantries. In order to receive this special emergency food assistance, the legal immigrants’ participation must be approved by DDSS. With DDSS’ approval, legal immigrants can continue to receive this emergency assistance on a monthly basis as long as they continue to be in an emergency situation. Nonprofit organizations in Denver County provide direct and indirect food assistance. Most of the organizations we visited were affiliated with national groups; others were state or local. In the last several years, these nonprofit organizations and some government agencies have established a network to discuss food assistance problems. Several organizations expected the number of individuals requesting services to increase as a result of welfare reform. Most organizations reported that they were already experiencing an increased demand, with one organization reporting 40 to 50 more applicants per week. Although many of the nonprofit organizations we contacted expect more individuals to request services, none are sure how they will deal with the expected increase. They are also concerned about their ability to meet an increased need for their services because of their limited resources. A few of them also reported that they would try to raise additional money through fund-raising activities and grants. Two officials also voiced concern about their ability to meet the demand for emergency food assistance in an economic downturn. Table VI.1 describes the nonprofit organizations that we contacted. Provides bags of food to clients. Serves approximately 26,000 individuals per year. Provides funding for local food assistance programs. Provides advocacy on issues, including food assistance, in Colorado. Foodbank serving metropolitan Denver, northern Colorado, and Wyoming (local affiliate of Second Harvest) Serves approximately 750 hunger- relief programs, including, for example, a program to pick up surplus prepared foods and a “kid’s cafe” providing food for children in Denver’s inner city. (continued) Manages a kosher food pantry that provides food for those who meet income requirements. Serves approximately 250 people per month. Provides a variety of services, including funding for approximately 13 emergency food assistance programs. Provides advocacy on food assistance in Colorado. A religious organization (local member of Catholic Charities) Provides food through a network of emergency assistance centers in the Denver metropolitan area; a food bank, which pools together the resources of 22 food banks to buy food in bulk at lower cost; the SHARE program; and meals at a temporary shelter for the homeless. The nonprofit organizations we contacted generally required their clients to meet some type of eligibility requirement in order to receive services. The organizations said that they serve many different groups of people besides legal immigrants and able-bodied adults without dependents, including the working poor, single mothers with children, and the elderly. The organizations use various resources to fund their operations, including federal government grants, foundation grants, individual contributions, and volunteer services. For example, one organization received approximately $264,000 in volunteer services and $1.4 million in in-kind food pantry donations during the last year. The greater Hartford area had a population of approximately 835,000 in 1995. The area consists of three jurisdictions—the city of Hartford and the towns of East and West Hartford. In May 1996, the unemployment rate for the Hartford area was 5.9 percent. By comparison, the state’s unemployment rate was 5.5 percent in May 1996, and the poverty rate for 1995-96 averaged 10.7 percent. Nationally, in May 1996, the unemployment rate was 5.4 percent, and the poverty rate for 1995-96 averaged 13.8 percent. In August 1996, before the Welfare Reform Act was implemented, over 219,000 individuals were participating in the federal Food Stamp Program statewide, according to Connecticut officials. (Statistics were not available for the greater Hartford area.) Of this total, about 5,800 were able-bodied adults without dependents. Furthermore, as of August 1996, an estimated 9,700 food stamp participants were legal immigrants. By September 1997, after many changes to the Food Stamp Program were implemented, participation had declined to about 202,000. Of this total, about 5,400 were able-bodied adults without dependents, and about 7,100 were legal immigrants. Connecticut’s Department of Social Services (DSS) administers the Food Stamp Program throughout the state. At the time of our visit, DSS officials told us that they did not have and did not plan to develop outreach services to help individuals retain their food stamps. In August 1997, however, the state received approval for a waiver of the work requirement for areas with limited employment opportunities and began to notify able-bodied adults without dependents who lost benefits because of welfare reform that their benefits could be restored. DSS’ goal is to provide access to information and services for employment and training. However, if participants in these programs meet income and work requirements, they may still qualify for food stamps. In addition, the state has developed a referral system to provide individuals with information on available food assistance. Able-bodied adults without dependents can participate in employment and training in a number of ways. For example, they can obtain training through the Connecticut Works System, which offers a “one-stop” approach to employment services and unemployment benefits. In addition, able-bodied adults without dependents can participate in the Self-Initiated Food Stamp Community Service Program/Working for In-Kind Income. In this state program, an able-bodied adult without dependents can meet workfare requirements by participating in a community service activity. The state will provide these adults with information on potential community service opportunities. Individuals accepting these community service positions will be able to maintain their eligibility for food stamps. State officials told us they had not made plans to provide outreach programs/services for legal immigrants losing food stamps because they were uncertain how many legal immigrants would become ineligible for the food stamp program. However, the state provided legal immigrants with information about obtaining U.S. citizenship when they were notified about changes in their eligibility for food stamps. The nonprofit organizations at work in the greater Hartford area provide food assistance directly and indirectly through food banks, food pantries, shelters, and soup kitchens. These organizations are affiliated with a network overseen by the local board of FEMA. The local board provides an opportunity for nonprofit organizations to communicate and coordinate the efforts or services they provide. Several of the organizations noted that it was too soon to clearly determine the effects of welfare reform. Nevertheless, they expected an increased need for food assistance because of the loss of eligibility for food stamps and were concerned about their ability to respond to that increase with little or no additional funding. Organizations told us that they plan to (1) seek additional funding and food donations and (2) make adjustments with the amounts and/or types of services they normally provide. Table VII.1 lists the nonprofit organizations we contacted. Community Renewal Team of Greater Hartford, Inc. Distributes funding to five agencies to provide food assistance. Food bank (local affiliate of Second Harvest) Distributes donated food to over 200 private, nonprofit programs that feed the hungry (e.g., food pantries, soup kitchens, shelters). Center City Churches (Center for Hope) Serves meals to approximately 1,200 to 1,400 individuals and provides referrals to other food assistance programs. Conducts research, outreach, training, advocacy, and provides referrals to other food assistance. Serves meals or provides bags of food. Food bank (local affiliate of the Second Harvest) Provides donated food to 450 private, nonprofit feeding agencies. Religious organization (local affiliate of Jewish Federation) Provides kosher lunches for approximately 300 to 350 individuals. Distributes funding to the local food bank to service shelters, food pantries, and soup kitchens. Provides food vouchers, Senior meal programs, hot meals, home-meal delivery for the elderly, seasonal meal programs, and part-time soup kitchens. Information referral service Makes referrals to food assistance programs. Religious organization (local affiliate of Lutheran Social Services) Provides referrals to food assistance for refugees and immigrants. These nonprofit organizations have no or minimal eligibility requirements for participation, such as picture identification and documentation of income. Currently, the nonprofit organizations receive funding from federal, state, and local government grants; individual and corporate contributions; and volunteer services. Two municipalities—the town of West Hartford and the town of East Hartford—maintain town food pantries where the needy can obtain either bags of groceries or food vouchers redeemable at local grocery stores. In East Hartford, participants must show a photo identification, which includes a social security number and date of birth; provide verification of income for all family members; and sign a “Client Information Form” that provides proof of dependents. The Detroit Tri-County area—Macomb, Oakland, and Wayne counties—had a population of about 3.9 million people in 1995—about 41 percent of Michigan’s population. In May 1996, the Detroit metropolitan area had an unemployment rate of 4.3 percent. In comparison, the state unemployment rate in May 1996 was 4.6 percent, and the poverty rate for 1995-96 averaged 11.7 percent. Nationally, in May 1996, the unemployment rate was 5.4 percent and the poverty rate for 1995-96 averaged 13.8 percent. In November 1996, as states were beginning to implement the Welfare Reform Act, about 427,600 persons received food stamps in the tri-county area. Of this total, about 29,500 were able-bodied adults without dependents. According to a Michigan state official, the agency does not track the number of legal immigrants receiving food stamps. As of September 1997, after many changes to the Food Stamp Program were implemented, about 381,500 individuals were receiving food stamps. Michigan’s Family Independence Agency (FIA) administers the Food Stamp Program in all counties, including Macomb, Oakland, and Wayne. An FIA official told us that FIA was assisting able-bodied adults without dependents with employment and training so that they can become self-sufficient while meeting work requirements that allow them to continue receiving food stamps. According to an FIA official, the state is not planning to create a new food assistance program to assist legal immigrants who lost food stamp benefits. Able-bodied adults without dependents have several opportunities to participate in employment and training and meet the Food Stamp Program’s work requirements. They can participate in a state-approved employment training program, work 20 hours a week, or perform 25 hours of public service at a nonprofit agency. Effective October 1, 1997, the number of community service hours must equal the benefit divided by the minimum wage ($5.15 per hour). In addition, in 1996, Michigan established the Food Stamp Community Service Program, which focuses on able-bodied adults without dependents. In fiscal year 1998, the state will receive $13.4 million from USDA’s FNS to expand work programs for this population. Nonprofit organizations in greater Detroit, Michigan, some of which are affiliated with national groups, provide direct and indirect food assistance through an established network that includes soup kitchens, food pantries, and food banks. According to officials of 10 nonprofit agencies, able-bodied adults without dependents and legal immigrants who lose their food stamps as a result of welfare reform will look for food assistance from these nonprofit organizations. Several of these officials told us that they had already experienced an increased need for their services as a result of welfare reform. They expressed concern about their ability to provide these additional services because of limited funding. However, several organizations we visited have developed strategies to increase the supply of food. These strategies include (1) raising additional money through fund-raising activities, (2) seeking more government and corporate grants, (3) encouraging Michigan to apply for federal food stamp waivers, (4) raising funds for target groups of legal immigrants, and (5) improving the emergency provider infrastructure. Table VIII.1 describes the nonprofit organizations that we contacted. Food bank (a Second Harvest affiliate) Serves about 300 emergency food providers, including soup kitchens and food pantries. Refers clients to 124 soup kitchens and food pantries in Wayne County and provides technical assistance for emergency food providers. Retrieves perishable food from restaurants and other food service organizations. Each month transports 60,000 pounds of food to tri-county soup kitchens and shelters. Provides about 500,000 pounds of food per year to about 1,100 Jewish families in the metropolitan area. Establishes about 20 new emergency food assistance providers each year. Since 1991, the coalition has received about $1.8 million in grants which it distributed to about 270 emergency food providers. Provides 7 days of food for needy families. Also serves two meals daily for the homeless in downtown Detroit. Commodity Supplemental Food Program, providing groceries for mothers, infants, preschool children, and seniors over the age of 60 meeting certain income guidelines. Many nonprofit organizations had eligibility requirements for individuals to receive their food assistance services. One required a certain qualifying income level but frequently made exceptions. Some providers serve only people within certain geographical boundaries. Others will provide food for anyone who asks. A few groups provide groceries for people on special diets; for example, an Oakland County food pantry provides groceries for those who keep kosher kitchens. Other providers primarily serve specific groups such as Hmong, Vietnamese, and migrant farm workers. Detroit’s emergency food assistance providers depend upon a variety of resources to fund their operations: grants from large corporate foundations; federal, state, and local governments; and community fund-raising activities that donate food and money. Community-based organizations also depend upon volunteers to manage and staff their food pantries and soup kitchens. According to the emergency food providers we interviewed, food is generally available for soup kitchens and food pantries, but additional funding for infrastructure is needed. The supply of food available for emergency food services does not depend only on the number of people needing emergency food services or the amount of food available for donation. The availability of funding for infrastructure—transportation and storage space, including refrigeration, and staff—is key to a successful food assistance operation. For example, many smaller soup kitchens and food pantries lack refrigeration and storage space, which prevents them from obtaining and keeping donated meats, vegetables, fruits, or dairy products. Furthermore, these organizations anticipate an increase in individuals and families needing their services. For example, early in 1997, one food pantry—Yad Ezra—realized that welfare reform would affect a number of Russian Jewish immigrants and that some means had to be found to replace the food stamps that would no longer be available. A Yad Ezra survey indicated that 212 of the 1,006 families currently being assisted would be affected by welfare reform and that many of the families are elderly and sick. Therefore, their ability or desire to learn English and gain citizenship is doubtful. Only 39 percent of the surveyed immigrants are taking citizenship classes. Specifically, to assist the 212 families, Yad Ezra will need $100,000 to augment the food pantry’s current year’s budget of $680,000. Each subsequent year, the organization will need to raise additional money to assist needy legal immigrants. Any additional money could exceed $100,000 each year, since Yad Ezra did not attempt to identify any new families or individuals whom it was not currently serving and who could be affected by welfare reform. Officials from Yad Ezra believe that Yad Ezra’s effort to replace food stamps for its legal immigrants is not likely to be duplicated by other food pantries because, unlike most other organizations, Yad Ezra serves a specific group of legal immigrants and was able to obtain the necessary resources to meet its needs. Houston, Texas (Harris County) had a population of 3.1 million in 1995. In May 1996, the unemployment rate was 5.1 percent in Houston, while the state unemployment rate was 5.4 percent. The poverty rate in 1995-96 averaged 17 percent. Nationally, in May 1996, the unemployment rate was 5.4 percent, and the poverty rate for 1995-96 averaged 13.8 percent. In February 1997, as states were beginning to implement the Welfare Reform Act, over 333,000 individuals in Houston received food stamps. Of this population, about 14,000 were able-bodied adults without dependents, and about 17,000 were legal immigrants. As of September 1997, after many changes to the Food Stamp Program were implemented, the number of able-bodied adults without dependents receiving food stamps had decreased to about 4,900, and the number of legal immigrants had decreased to about 2,700. The Texas Department of Human Services (TDHS) administers USDA’s Food Stamp Program. Texas had obtained a waiver of the work requirement for selected counties; however, it decided not to implement the waiver in Houston because the unemployment rate was less than 10 percent. Therefore, all able-bodied adults in Houston are required to participate in employment and training activities in order to continue receiving food stamps. According to TDHS officials, activities meeting these work requirements include regular employment; self-employment; volunteer work with a business, government entity, or nonprofit organization; and/or participation in the Job Training Partnership Act or the Trade Adjustment Assistance Act Program. In addition, these adults can obtain assistance from the Texas Workforce Commission’s Food Stamp Employment and Training Program. The purpose of this program is to move welfare recipients to work as quickly as possible. However, participation in the job search and the job search training component of this program does not satisfy the work requirement. For legal immigrants, Texas is providing, effective February 1998, targeted food assistance to elderly and disabled legal immigrants who were receiving food stamp benefits as of August 22, 1996, and lost benefits because of welfare reform. Texas is providing about $18 million for this effort. Benefits will range from $10 to $122 per month per individual. The nonprofit organizations that provide direct and indirect food assistance—including food banks and food pantries—in Houston generally operate independently of each other. At the time of our visit, most of the organizations reported that their ability to provide food assistance for those needing it had not yet been affected by welfare reform. However, most organizations told us that they expected the amount of food assistance provided by them to increase within the next 2 to 3 years because of welfare reform. In addition, many organizations’ officials expressed concern that they may have difficulty in providing food assistance in the future. One organization attributed this difficulty to the fact that so many organizations were competing for the same monetary and food donation resources. Most organizations did not have planned approaches for dealing with the expected increase in the need for their services. However, one organization is considering a reduction in the number of items that it distributes in bags of groceries in order to meet the expected increased need for its services. Table IX.1 list the nonprofit organizations that we contacted. Community service agency Provides supplemental food for low-income families in emergency situations. Food bank (local affiliate of Second Harvest) Distributes food to local charities that care for the needy. Community social service agency (local affiliate of United Way) Distributes funds to community groups that target hunger. Distributes food in low-income neighborhoods through volunteers who operate food pantries and community gardens. Distributes rice, instant noodles, and canned food within the Vietnamese community. Conducts fund-raising activities for member coalitions that provide food assistance. Provides funding that supplements local food assistance programs. Provides food support for 30 agencies and 110 food pantries. Associated Catholic Charities of the Diocese of Galveston-Houston (Guadalupe Social Services) Provides emergency food assistance and bags of groceries once a month for senior citizens and disabled persons. Most organizations had income eligibility requirements for their food assistance services, and several limited their assistance to individuals residing in certain areas. In addition, one organization focused its efforts in the Vietnamese community. Funding sources for the nonprofit organizations we visited varied. Most received funding from federal, state, and local government grants and donations from religious organizations and individuals. Robert E. Robertson, Associate Director Patricia Gleason, Assistant Director Tracy Kelly Solheim, Project Leader Carolyn Boyce, Senior Social Science Analyst Carol H. Shulman, Communications Analyst Kathy Alexander Renee McGhee-Lenart Janice Turner Sheldon Wood, Jr. Patricia A. Yorkman The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO reviewed the impact of welfare reform on the Food Stamp Program, focusing on: (1) the actions, if any, that states have taken to assist those individuals who lose eligibility for the Food Stamp Program; and (2) related actions, if any, taken by other organizations in selected localities--local governments and nonprofit organizations--to assist those individuals who lose their eligibility for the Food Stamp Program. GAO noted that: (1) most states are taking a variety of measures to address the changes in the Food Stamp Program as a result of welfare reform; (2) for able-bodied adults without dependents, many states are providing employment and training assistance; (3) this assistance, although primarily intended to move these individuals toward self-sufficiency, may still allow them to qualify for food stamp benefits if they meet both income and work requirements; (4) most states have obtained the authority from the Department of Agriculture, if they choose to exercise it, to continue providing food stamp benefits for individuals in areas with high unemployment or in areas with insufficient jobs; (5) 20 states are providing or plan to provide legal immigrants with information on how to become U.S. citizens; (6) because it takes over 1 year on average to process citizenship applications, many legal immigrants lost their food stamp benefits as of August 22, 1997; (7) the Food and Nutrition Service estimated that 935,000 legal immigrants had lost their federal food stamp benefits; (8) some states have existing programs that provide food assistance for the needy--such as food pantries--that able-bodied adults without dependents and legal immigrants who have lost their food stamps already had access to; (9) some states have developed new programs to specifically meet the needs of individuals who lose their food stamps; (10) 10 states--including 4 states estimated to have about 70 percent of the legal immigrants who receive food stamps in the U.S.--are purchasing or planning to purchase federal food stamps with their own funds--primarily for legal immigrant children and the elderly; (11) in December 1997, the states involved indicated that about 241,000 of these individuals are now receiving food stamp benefits funded by the states; (12) the extent to which any of these actions will meet the food assistance needs of those affected remains unknown; (13) in the five localities GAO visited, government officials are implementing their state's efforts to address changes in the Food Stamp Program and, in some cases, are working with local nonprofit organizations to plan for an expected increase in the need for food assistance; (14) most of the nonprofit organizations GAO visited said that it is too early to assess the impact of welfare reform on their food assistance programs; and (15) however, the organizations fear that their limited resources may be insufficient to meet the needs of the individuals who have lost their food stamps, which included the basic foods that the program provided. |
The four major federal land management agencies administer approximately 628 million acres, or about 28 percent of the land area in the United States. These public lands are mostly in Alaska and the 11 western states: Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming. Alaska is not currently participating in the FLTFA program because of its priority to settle Alaska Native land claims. BLM is authorized to sell or exchange land identified in its land use plans; the other three land management agencies have limited or no sales authority. Once BLM has sold land, FLTFA directs BLM to deposit the revenue generated from these sales into a special U.S. Treasury account created by FLTFA. However, the act limits the revenue deposited into this account to that generated from sales or exchanges of public lands identified for disposal in a land use plan in effect as of July 25, 2000—the date of FLTFA’s enactment. Money in the new account is available to BLM and the other three agencies to purchase inholdings, and in some cases, land adjacent to federally designated areas that contain exceptional resources. The federal land management agencies have two methods for identifying land to acquire under FLTFA. First, the agencies can nominate parcels through a process laid out in state-level implementation agreements that were developed under the direction of a national memorandum of understanding (MOU). Second, the Secretaries can directly use a portion of FLTFA revenue to acquire specific parcels of land at their own discretion. The national MOU laid out the expectation that most acquisitions would occur through the state-level process. FLTFA places several restrictions on using funds from the new U.S. Treasury account. Among other things, FLTFA requires that (1) no more than 20 percent of the revenue can be used for BLM’s administrative and other activities necessary to carry out the land disposal program; (2) of the amount not spent on administrative expenses, at least 80 percent must be expended in the state in which the funds were generated; and (3) at least 80 percent of FLTFA revenue required to be spent on land acquisitions within a state must be used to acquire inholdings (as opposed to adjacent land) within that state. In addition, the national MOU sets the allocation of funds from the FLTFA account for each agency—60 percent for BLM, 20 percent for the Forest Service, and 10 percent each for the Fish and Wildlife Service and the Park Service, but the Secretaries may vary from these allocations by mutual agreement. At the time of our review, BLM had raised $95.7 million in revenue, mostly from selling 16,659 acres. As of May 2007, about 92 percent of the revenue raised, or $88 million, came from land sales in Nevada. Revenue grew slowly during the first years of the program and peaked in fiscal year 2006, when a total of $71.1 million was generated. BLM’s Nevada offices accounted for the lion’s share of the sales because (1) demand for land to develop had been high in rapidly expanding population centers such as Las Vegas, (2) BLM had a high percentage of land in proximity to these centers, and (3) BLM had experience selling land under another federal land sales program authorized for southern Nevada. During the period we reviewed, BLM offices covering three other states—New Mexico, Oregon, and Washington—had raised over $1 million each, and the remaining seven BLM state offices—Arizona, California, Colorado, Idaho, Montana, Utah, and Wyoming—had each raised less than $1 million. Most BLM field offices had not generated revenue under FLTFA. As of August 2009, BLM reported raising a total of $113.4 million in revenue for the FLTFA account from the sale of about 29,400 acres. According to these revised BLM data, Nevada still accounted for the majority of FLTFA sales revenues—about $88 million, or 78 percent of the total revenue. BLM faces several challenges to raising revenue through future FLTFA sales, according to officials in the 10 BLM state offices and 18 BLM field offices we interviewed for our 2008 report. Many of these challenges are likely to continue if FLTFA is reauthorized. The following lists, in order of most frequently cited, the challenges officials identified and provides examples: The availability of knowledgeable realty staff to conduct the sales. BLM staff said realty staff must address higher priority work before land sales. For example, Colorado BLM staff said that processing rights-of-way for energy pipelines takes a huge amount of realty staff time, 100 percent in some field offices, and poses one of the top challenges to carrying out FLTFA sales in Colorado. In Idaho, staff also cited the lack of realty staffing, which was down 40 percent from 10 years ago. Time, cost, and complexity of the sales process. Much preparation must be completed before a property can be sold. For example, several offices cited the cost and length of the process to ensure that a sale complies with environmental laws and regulations. In addition, obtaining clearances from experts on cultural and natural resources on a proposed sale can be time-consuming. External factors. BLM officials cited such factors such as public opposition to a sale, market conditions, or lack of political support as challenges. For example, Colorado BLM officials said that they have faced strong local opposition to sales, and the El Centro Field Office staff in California cited the lack of demand for the land from buyers as a challenge. Program and legal restrictions. The Arizona State Office staff and the Elko, Nevada Field Office staff cited the sunset date of FLTFA, less than 3 years away at the time of our review, as a challenge to the disposal of land under FLTFA because the sunset date might not allow enough time to complete many more sales. Other offices said the MOU provision requiring a portion of the land sale proceeds to be used by the three other agencies reduces BLM’s incentive to conduct land sales because BLM keeps only 60 percent of the revenue. Another challenge, especially in Nevada, has been the enactment of land bills for Lincoln and White Pine counties. In total, BLM staff estimated that, once mandated land use plan amendments were completed, these two acts would result in the removal of about 148,000 acres from FLTFA eligibility. Land use planning. Some offices cited problems with the land use plans. For example, the Idaho Falls District Office staff said that specific land for sale is hard to identify in old land use plans. Nevada’s Elko Field Office staff said that some lands that could be offered for sale were not available because they were not designated in the land use plan at the time of FLTFA’s enactment. We identified two additional issues hampering land sales activity under FLTFA. First, while BLM had identified land for sale in its land use plans, it had not made the sale of this land a priority during the first 7 years of the program. Furthermore, BLM had not set goals for sales or developed a sales implementation strategy. Second, some of the additional land BLM had identified for sale since FLTFA was enacted would not generate revenue for acquisitions because the act only allows the deposit of revenue from the sale of lands identified for disposal on or before the date of the act. At the time of our review, BLM had reported that the four land management agencies had spent $13.3 million of the $95.7 million in the FLTFA account. More specifically: The four agencies spent $10.1 million to acquire nine parcels totaling 3,381 acres in seven states—Arizona, California, Idaho, Montana, New Mexico, Oregon, and Wyoming. BLM spent $3.2 million for administrative expenses between 2000 and 2007 to conduct FLTFA-eligible sales, primarily in Nevada. The agencies acquired these lands between August 2007 and January 2008—more than 7 years after FLTFA was enacted. These acquisitions were initiated using the Secretaries’ discretion, and most had been identified but not funded for purchase under another land acquisition program. As of October 2007, no land had been purchased through the state-level interagency nomination process that was established by the national MOU and state agreements. Acquisitions had not yet occurred under the state-level process because it took 6 years to complete the interagency agreements needed to implement the program and because relatively little revenue was available for acquisitions outside of Nevada, owing to FLTFA requirements. As of November 2009, BLM reported the following: The Secretaries had approved $66.8 million for the acquisition of 39 parcels since FLTFA’s enactment in 2000. Of the $66.8 million, agencies spent a total of about $43.8 million to acquire 28 parcels totaling 16,738 acres and the remainder of the approved acquisitions was being processed. $48.6 million of the $66.8 million in acquisitions for 22 parcels had been nominated through the state-level interagency process rather than through Secretarial discretion. Of the $48.6 million nominated through the state- level process, the agencies have acquired 12 parcels with $24.6 million in FLTFA funding. $5.1 million has been spent on FLTFA administrative expenses to conduct land sales overall. BLM state and field officials we interviewed for our 2008 report cited several challenges to completing additional acquisitions under FLTFA. Many of these challenges are likely to continue if FLTFA is reauthorized. The following lists, in order of most frequently cited, the challenges officials identified, and provides examples of these challenges. Time, cost, and complexity of the land acquisition process. To complete an acquisition under FLTFA, four agencies must work together to identify, nominate, and rank proposed acquisitions, which must then be approved by the two Secretaries. Officials at two field offices estimated the acquisition process took about 2-1/2 to 3 years. BLM officials from the Wyoming State Office and the Las Cruces, New Mexico, Field Office said that, with this length of time, BLM must either identify a very committed seller willing to wait to complete an acquisition or obtain the assistance of a third party in completing an acquisition. In terms of cost, some offices noted that they did not have the funding required to complete all of the work involved to prepare land acquisitions. In terms of complexity, a Utah State Office official said BLM has more control over the process for submitting land acquisitions under the Land and Water Conservation Fund than FLTFA because FLTFA requires four agencies in two departments to coordinate their efforts. Identifying a willing seller. Identification of a willing seller can be problematic because, among other things, the seller might have higher expectations of the property’s value. For example, an Ely, Nevada, Field Office official explained that, because of the then-high real estate values, sellers believed they could obtain higher prices from developers than from the federal government. Furthermore, an Idaho State Office official said that it is difficult to find a seller willing to accept the appraised price and wait for the government to complete the purchase. Even when land acquisition nominations are approved, they may not result in a purchase. For example, in 2004, under FLTFA, two approved acquisitions for inholdings within a national forest in Nevada were terminated. In one case, property values rose sharply during the nomination process and, in an effort to retain some of his land, the seller decided to reduce the acres for sale but maintain the price expectation. Furthermore, the seller decided not to grant the Forest Service access through the parcel he was retaining, thus eliminating the opportunity to secure access to an inaccessible area of the national forest. In the other case, during the course of the secretarial approval process, the landowner sold portions of the land included in the original transaction to another party, reducing the land available for the Forest Service to purchase. According to Forest Service officials, in both cases the purchase of the remaining parcels would not fulfill the original purpose of the acquisitions owing to reductions in resource benefits. Therefore, the Forest Service terminated both projects. Availability of knowledgeable staff to conduct acquisitions. BLM officials reported that they lacked knowledgeable realty staff to conduct land acquisitions, as well as other BLM or department staff to conduct appraisals, surveys, and resource studies. Staff were occupied working on higher priority activities, particularly in the energy area. Lack of funding to purchase land. BLM officials in some states said they lack adequate funds to acquire land under FLTFA. For example, according to a field office official in Burns, Oregon, just one acquisition in a nearby conservation area would have nearly drained that state’s FLTFA account. Restrictions imposed by laws and regulations. BLM officials said that legal and other restrictions pose a challenge to acquiring land. For example, officials in the BLM Arizona State Office and the Grand Junction, Colorado, Field Office said that some federally designated areas in their jurisdictions were established after the date of FLTFA’s enactment, making the land within them ineligible for acquisition under the act. In terms of regulations, BLM Carson City, Nevada, Field Office officials told us that the requirements they must follow regarding the processing of title, survey, and hazardous materials issues posed a challenge to conducting acquisitions. Public opposition to land acquisitions. According to BLM officials from the Elko and Ely Field Offices in Nevada, the public did not support the federal government’s acquisition of federal land in their areas, arguing that the government already owned a high percentage of land and that such acquisitions resulted in the removal of land from the local tax base. We also found that the act’s restriction on the use of revenues outside of the state in which they were raised continues to limit acquisitions. Specifically, little revenue was, and still is available for acquisitions outside of Nevada. Furthermore, progress in acquiring priority land had been hampered by the agencies’ weak performance in identifying inholdings and setting priorities for acquiring them, as required by the act. Finally, the agencies had yet to develop effective procedures to fully comply with the act and national MOU. Specifically, the agencies—and primarily BLM, as the manager of the FLTFA account—had not established a procedure to track the act’s requirement that at least 80 percent of funds allocated toward the purchase of land within each state must be used to purchase inholdings and that up to 20 percent may be used to purchase adjacent land. And with respect to the national MOU, BLM had not established a procedure to track agreed-upon fund allocations—60 percent for BLM, 20 percent for the Forest Service, and 10 percent each for the Fish and Wildlife Service and the Park Service. In 2008, we concluded that 7 years after FLTFA had been enacted, BLM had not taken full advantage of the opportunity the act offered. We recognized that a number of challenges prevented BLM from completing many sales in most states, which limited the number of possible acquisitions. Many of the challenges that BLM cited are likely faced in many public land sales because FLTFA did not change the land sales process. However, we believed that BLM’s failure to set goals for FLTFA sales and develop a sales implementation strategy limited the agency’s ability to raise revenue for acquisitions. Without goals and a strategy to achieve them, BLM field offices did not have direction for FLTFA sales. Moreover, the lack of goals made it difficult to determine the extent of BLM’s progress in disposing of unneeded lands to raise funds for acquisitions. As with sales, progress in acquiring priority land had been hampered by weak agency performance in developing an effective mechanism to identify potential land acquisitions and set priorities for inholdings and adjacent land with exceptional resources, which FLTFA requires. Moreover, because the agencies had not tracked the amounts spent on inholdings and agency allocations, they could not ensure compliance with the act or full implementation of the MOU. Our report contained two matters for congressional consideration and five recommendations for executive action. We said that if Congress decided to reauthorize FLTFA in 2010, it might wish to consider revising the following two provisions to better achieve the goals of the act: FLTFA’s limitation of eligible land sales to those lands identified in land use plans in effect as of July 25, 2000. This provision excludes more recently identified land available for disposal, thereby reducing opportunities for raising additional revenue for land acquisition. The requirement that agencies spend the majority of funds raised from eligible sales for acquisitions in the same state. This provision makes it difficult for agencies to acquire more desirable land in states that have generated little revenue. Our report also contained five recommendations for executive action to improve FLTFA implementation. BLM has taken several actions to implement our recommendations. Table 1 shows the recommendations from our 2008 report and the actions the agencies reported as of November 2009. Table 1. GAO Recommendations to Improve FLTFA Implementation and Agency Actions, as of November 2009 August 2008. BLM established FLTFA land sale goals for fiscal years 2009 and 2010 of $25 million each, according to agency officials. To set these goals, a BLM headquarters official contacted each of the BLM state offices to determine the amount of eligible land sales that could be conducted in the final 2 years of FLTFA. Fall 2009. BLM revised its land sales goal for fiscal year 2010 to $20 million. August 2008. BLM developed a sales incentive program that provides seed money for planning and carrying out FLTFA-eligible land sales. Specifically, the program makes available up to $300,000 to eligible state and field offices for activities necessary to identify and pre-screen properties for possible sale under FLTFA. At a minimum, offices are to prepare a list of specific tracts for sale, with legal descriptions and a copy of the respective land use plan that supports the potential sale. As of November 2009, six states—Arizona, California, Colorado, Idaho, Montana, and New Mexico—had agreed to participate in the program, according to BLM officials. May 2008. USDA stated that its Land Acquisition Prioritization System, generally used for land acquisitions under the Land and Water Conservation Fund, also satisfies the land acquisition prioritization requirements under FLTFA. USDA further stated that Forest Service would continue working with BLM to identify and set priorities for acquiring inholdings and that the Forest Service would coordinate with BLM to formalize the use of a single process to set priorities for land acquisitions. November 2009. The Forest Service FLTFA program lead said that Forest Service has coordinated with BLM to formalize the use of a single process to set priorities for land acquisitions. She said that the agencies meet regularly to discuss FLTFA nominations. April 2008. Interior agreed to continue to improve the procedures to identify and set priorities for acquiring inholdings. November 2009. BLM officials said that the current Land and Water Conservation Fund system works well for FLTFA acquisitions and no changes have been made to this system. BLM has, however, intensified its efforts to educate state-level FLTFA implementation teams on the FLTFA land acquisition process. For example, the FLTFA lead said he has attended numerous state-level interagency team meetings to educate team members about the availability and use of FLTFA funds. November 2009. BLM officials reported that BLM gathers and maintains data on each transaction and tracks whether the parcel is an inholding or adjacent land. Officials also reported that BLM is directing field staff to note in BLM’s automated land status tracking system (LR2000) whether a parcel is an inholding or adjacent land. May 2008. USDA stated that BLM is responsible under FLTFA for tracking the sales, proceeds, and disbursement of funds and that USDA will continue to assist BLM in tracking these funds. November 2009. The Forest Service FLTFA program lead reiterated USDA’s May 2008 statement that BLM is responsible for tracking the use of FLTFA funding. She said that the Forest Service is merely a recipient of FLTFA funding. She added that the national MOU allocations are only targets and that they do not necessarily represent a limit on how much funding an agency can receive. November 2009. The BLM FLTFA program lead reported that BLM is gathering data on each FLTFA transaction by agency and will prepare a final report in compliance with the MOU at FLTFA’s sunset if not reauthorized. He added that the allocations established in the MOU are goals only, and that, while the agencies will try to adhere to them, they ultimately will not be held to those allocations. As of November 2009, BLM reports that of the $66.8 million approved by the Secretaries, 60 percent is for BLM, 30 percent is for the Forest Service, 5.5 percent is for the Park Service, and 4.5 percent is for the Fish and Wildlife Service. Mr. Chairman, this concludes my prepared statement. I would be happy to respond to any questions that you or Members of the Subcommittee may have. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this testimony. For further information about this testimony, please contact Robin M. Nazzaro at (202) 512-3841. Individuals making key contributions to this testimony were Andrea Wamstad Brown, Assistant Director; Rich Johnson; Mark Keenan; Paul Kinney; Emily Larson; John Scott; Rebecca Shea and Carol Herrnstadt Shulman. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | The U.S. Department of the Interior's Bureau of Land Management (BLM), Fish and Wildlife Service, and National Park Service, and the U.S. Department of Agriculture's Forest Service manage about 628 million acres of public land, mostly in 11 western states and Alaska. Under the Federal Land Transaction Facilitation Act (FLTFA) of 2000, revenue raised from selling BLM lands is available to the agencies, primarily to acquire nonfederal land within the boundaries of land they already own--known as inholdings. These inholdings can create significant land management problems. To acquire land, the agencies can nominate parcels under state-level interagency agreements or the Secretaries can use their discretion to initiate acquisitions. FLTFA expires in July 2010. This testimony discusses GAO's 2008 report: Federal Land Management: Federal Land Transaction Facilitation Act Restrictions and Management Weaknesses Limit Future Sales and Acquisitions (GAO-08-196). Specifically, the testimony discusses (1) FLTFA revenue generated, (2) challenges to future sales, (3) FLTFA expenditures, (4) challenges to future acquisitions, and (5) agencies' implementation of GAO's recommendations. Among other things, GAO examined the act, agency guidance, and FLTFA sale and acquisition data, interviewed agency officials, and obtained some updated information. (1) BLM raised most FLTFA revenue from land sales in Nevada. As of August 2009, BLM reported raising a total of $113.4 million from sale of about 29,400 acres. Since FLTFA was enacted in 2000 through August 2009, about 78 percent of the revenue raised, or about $88 million, has come from land transactions in Nevada. (2) BLM faces challenges to future sales under FLTFA. In particular, BLM state and field officials most frequently cited the limited availability of knowledgeable realty staff to conduct sales. We identified two additional issues hampering land sales activity under FLTFA. First, while BLM had identified land for sale in its land use plans, it had not made these sales a priority during the first 7 years of the FLTFA program. Furthermore, BLM had not set goals for sales or developed a sales implementation strategy. Second, some of the additional land BLM had identified for sale since the act would not generate revenue for acquisitions because the act only allows the deposit of revenue from the sale of lands identified for disposal on or before the date of the act. (3) Agencies had purchased few parcels with FLTFA revenue. In 2008, we reported that between August 2007--7 years after FLTFA was enacted--and January 2008, the four land management agencies had spent $13.3 million of the $95.7 million in revenue raised under FLTFA: $10.1 million using the Secretaries' discretion to acquire nine parcels of land and $3.2 million for administrative expenses to prepare land for FLTFA sales. More recently, as of November 2009, BLM reported spending a total of $43.8 million to acquire 28 parcels, including $24.6 million for 12 parcels through the state-level interagency process. (4) Agencies face challenges to completing additional acquisitions. BLM state and field officials GAO interviewed most commonly cited the time, cost, and complexity of the land acquisition process as a challenge to completing land acquisitions. Furthermore, the act's requirement to spend the majority of funds in the state in which revenue was generated has had the effect of making little revenue available for acquisitions outside of Nevada. The agencies also had not established procedures to track the implementation of the act's requirement that at least 80 percent of FLTFA revenue raised in each state be used to acquire inholdings in that state or to track the extent to which BLM is complying with agreed-upon fund allocations among the four participating agencies. (5) BLM has taken steps to implement GAO's recommendations. Specifically, BLM established FLTFA sale goals for fiscal years 2009 and 2010 and established a sales incentive program providing seed funds to state and field offices to identify and pre-screen properties for possible sale under FLTFA. As of November 2009, six states have agreed to participate in the program. |
Over the past decade, NASA has experienced significant problems with several of its projects, which GAO and others have reported on (see table 1 for some examples of such problems). In addition, in 2002 we reported on several sources of failures in NASA programs, including underestimating complexity and technology maturity, and inadequate review and systems engineering processes. Further, we reported that the sources of these problems were not new and that NASA failed to consistently apply lessons previously learned. The failures identified in our 2002 report were in part the result of the “faster, better, cheaper” approach to managing its major acquisitions, which NASA adopted in the 1990s. These problems, along with others, highlighted the need for the agency to reevaluate its approach to cost and schedule estimating, risk assessments, technology development, project reviews, and systems engineering. In 1998, NASA adopted a new program and project management policy, which was revised in 2002. The policy provided significant flexibility, including allowing tailoring and projects to opt out of requirements at the discretion of the project manager. In March 2005—following a series of internal and external assessments of NASA that showed that the agency faced significant problems with project management—NASA again revised its program and project management policy, which includes policy requirements for the development of flight systems and ground support projects. According to NASA officials, further changes to the policy are anticipated in light of new agency leadership. The March 2005 policy document, NASA Procedural Requirement (NPR) 7120.5C, differs from previous versions of the policy in that it delineates requirements based on four NASA investment areas: Basic and Applied Research, Advanced Technology Development, Flight Systems and Ground Support, and Institutional Infrastructure. The policy also reinstitutes a life cycle phased approach to product development and institutes a project categorization scheme, based on project cost and priority, which denotes the oversight authorities and the level of detail that is needed to support project planning documents. See figure 2 below. NASA has also attempted to address some of its cost-estimating weaknesses by instituting a cost analysis data requirement (CADRe) and establishing thresholds for the use of earned value management (EVM). Another major change to the policy is the establishment of the Independent Technical Authority (ITA). According to the policy, the purpose of ITA is to establish sound technical requirements and decisions for safe and reliable system operations separate from the project management and reporting chain. Finally, agency officials told us that while the requirements in previous versions of the policy were easy to tailor, projects now must document compliance with the requirements in a “compliance matrix” and must request and have approved any deviations and/or waivers to requirements. Although NPR 7120.5C contains some mandatory requirements with regard to systems engineering, according to agency officials, NASA has never had an agencywide systems engineering policy to inform the development of flight systems and ground support projects. Since 1995, in the absence of a policy on systems engineering, project managers and systems engineers have relied on information contained in NASA’s Systems Engineering Handbook to guide their systems engineering approach on projects. Project managers and systems engineers, however, are not required to follow the handbook. Recognizing the need for a more structured and rigorous approach to systems engineering agencywide, NASA’s Office of the Chief Engineer is currently leading the effort to develop a systems engineering policy. The policy is in draft form. Over the last several years, we have undertaken a body of work on how leading developers in industry and government use a knowledge-based approach to deliver high quality products on time and within budget. A knowledge-based approach to product development efforts enables developers to be reasonably certain, at critical junctures or “knowledge points” in the acquisition life cycle, that their products are more likely to meet established cost, schedule, and performance baselines and, therefore provides them with information needed to make sound investment decisions. See figure 3 for a depiction of a knowledge-based acquisition life cycle. Knowledge point 1 (KP1): Resources and needs match. Knowledge point 1 occurs when a sound business case is made for the product— that is, a match is made between the customer’s requirements and the product developer’s available resources in terms of knowledge, time, workforce, and money. To determine available resources, successful developers rely on current and valid information from predecessor projects, new technologies that have demonstrated a high level of maturity, system engineering data, and experienced people. Successful developers also communicate extensively with customers to match their wants and needs with available resources and with the developers ability to manufacture an appropriate product. Knowledge point 2 (KP2): Product design is stable. Knowledge point 2 occurs when a developer determines that a product’s design is stable—that is, it will meet customer requirements and cost and schedule targets. A best practice is to achieve design stability at the product’s critical design review (CDR), usually held midway through development. Completion of at least 90 percent of engineering drawings at the CDR provides tangible evidence to decision makers that the design is stable. Knowledge point 3 (KP3): Production processes are mature. This level of knowledge is achieved when it has been demonstrated that the product can be manufactured within cost, schedule, and quality targets. A best practice is to ensure that all key manufacturing processes are in statistical control— that is, they are repeatable, sustainable, and capable of consistently producing parts within the product’s quality tolerances and standards—at the start of production. It is important that the product’s reliability be demonstrated before production begins, as investments can increase significantly if defective parts need to be repaired or reworked. If the knowledge attained at each juncture does not confirm the business case on which the initial investment was originally justified, the project should not go forward and additional resources should not be committed. Product development efforts that do not follow a knowledge-based approach can be frequently characterized by poor cost, schedule, and performance outcomes. NASA’s revised acquisition policy for developing flight and ground support systems incorporates some of the elements of a knowledge-based acquisition approach. However, it lacks specific key criteria and decision reviews necessary to fully support such an approach. NASA’s policy defines a phased life cycle approach and requires a major decision review to move from project formulation to implementation. The policy requirements for this review address many of the key elements necessary to match needs to resources, such as requirements to establish project baselines. However, the policy does not require that projects demonstrate technologies at high levels of maturity before launching a project and investing a large amount of resources. In addition, NASA’s policy does not require any further major decision reviews following the formulation phase of the project. Major decision reviews in the implementation phase based on specific evaluation criteria at the final design and fabrication, assembly, and testing milestones—critical decision points in any product development—could help NASA ensure that sufficient knowledge has been gained to warrant moving forward in the development process. The life cycle for all NASA projects is divided into two major phases— formulation and implementation. Because flight systems and ground support projects are particularly complex and have long life cycles, NASA has further divided the formulation and implementation phases for these projects to allow managers to assess management and engineering progress (see fig. 4). Flight systems and ground support projects must successfully complete two major decision reviews: a Preliminary-Non Advocate Review (Pre-NAR) between Phases A and B and a Non-Advocate Review (NAR) between Phases B and C. At these reviews, the Governing Program Management Committee (GPMC) evaluates the cost, schedule, safety, and technical content of the project to ensure that the project is meeting commitments specified in key management documents. Following each of these reviews, the GPMC recommends to the appropriate decision authority whether the project should be authorized to proceed. As with KP1 in a knowledge-based acquisition life cycle, the NAR marks the official project approval point in the life cycle. After approval at the NAR, projects are included as part of implementation reviews of their parent program. These implementation reviews are conducted biennially and are not tied to any design or production milestones. After entering the implementation phase, the GPMC is notified if cost or schedule performance exceeds the baselines established at the NAR by 10 percent or when key performance criteria are not met. Exceeding the NAR baselines can result in the GPMC conducting a termination review to determine whether or not to continue the project. To help ensure project requirements do not outstrip resources, leading developers obtain the right knowledge about a new product’s technology, design, and production at the right time. NASA’s policy emphasizes many elements needed at the NAR (or KP1) to match needs to resources, such as validating requirements, developing realistic cost and schedule estimates and human capital plans, and establishing a preliminary design. The policy does not, however, require projects to demonstrate technologies at high levels of maturity before launching a project. Table 2 compares KP1 criteria and NASA’s policy criteria. While NASA requires projects to develop plans that describe how technologies will be matured and to provide alternative development strategies for technologies that do not mature as expected, it does not establish a minimum threshold for technology maturity. Consequently, projects can enter the implementation phase with immature technologies and embark on a risky path of having to build technology, design, and production knowledge concurrently. Our best practices work has shown that maturing technologies during the preliminary design phase and before entering product development is a key element of matching needs to resources and that there is a direct relationship between the maturity of technologies and the risk of cost and schedule growth. Allowing technology development to carry over into the product development phase increases the risk that significant problems will be discovered late in development. Addressing such problems at this stage may require extensive retrofitting and redesign as well as retesting, which can jeopardize performance and result in more time and money to fix. This approach also makes it more difficult for projects to demonstrate the same level of design stability in later phases of implementation since technology and design activities will be done concurrently. Technology readiness levels (TRL)—a concept developed by NASA—can be used to gauge the maturity of individual technologies. The higher the TRL, the more the technology has been proven and the lower the risk of performance problems and cost and schedule overruns (see fig. 5). TRL 7—demonstrating a technology as a fully integrated prototype in an operational environment—is the level of maturity preferred by product developers to minimize risks when entering product development. Successful developers will not commit to undertaking product development, and more importantly investing resources, unless they have high confidence that they have achieved a match between what the customer wants and what the project can deliver. Technologies that are not mature continue to be developed in an environment that is focused solely on technology development. Once matured, these technologies can be transitioned to projects. This puts developers in a better position to succeed because they can focus on integrating the technologies and testing and proving the product design. Our prior work has shown that successful developers establish specific criteria to ensure that requisite knowledge has been attained before moving forward from final design into the latter stages of development. Before making significant increases in investments to fabricate, assemble, and test the product, these developers conduct a decision review to determine if the design is stable and performs as expected and the project is ready to enter the next phase. To make this determination and reach KP2, successful developers use specific, knowledge-based standards and criteria. (See app. II for more information on the specific knowledge-based standards and criteria successful developers use to judge readiness to proceed beyond detailed design activities.) Successful developers also demand proof that manufacturing processes are in control and product reliability goals are attained before committing to production. To determine whether they have achieved this knowledge point, KP3, successful developers conduct another mandatory decision review in which they use specific, knowledge-based standards and criteria to determine if the product can be produced within cost, schedule, and quality targets. (See app. II for more information on specific knowledge- based standards and criteria used by successful developers to judge readiness to enter into production.) Contrary to these best practices, the NAR at the end of the preliminary design phase—KP1—is the last major decision review in the NASA project life cycle. (See fig. 6.) Although NPR 7120.5C requires that projects document in the project plan a continuum of technical and management reviews, such as a PDR and CDR, it does not require any specific reviews. In addition, NASA’s March 2005 policy does not require a NAR-type decision review to ensure a project has obtained the knowledge needed to proceed beyond the final design phase into the fabrication, assembly, and test phase, which serves as both the demonstration and production phase of the NASA life cycle. According to NASA officials, projects conduct a CDR at the end of the final design phase to ensure adequate information is available about product design and producibility before entering the fabrication, assembly, and test phase. The CDR, however, is a technical review—not a major decision review like the NAR. Furthermore, the policy does not establish criteria as to what constitutes successful completion of a CDR. NASA’s policy also does not require a major decision review before beginning manufacturing. (See fig. 6 above.) Therefore, the transition from final design to fabrication, assembly, and test often marks a de facto production decision. According to NASA officials, the agency rarely enters a formal production phase due to the small quantities of space systems that they build. However, due to the high cost of failure associated with NASA projects and the costs and risks involved in repairing a system in-orbit, a major decision review at production that assesses product reliability is essential even for these limited production systems. In addition, although NASA’s production quantities are typically low, in some instances NASA does produce larger quantities of a system or subsystem, such as the external tanks for the Space Shuttle. Furthermore, NASA’s plans indicate that the agency may be increasing production for elements of their future systems. For example, NASA’s Exploration Systems Architecture Study indicates that NASA plans to build several new crew exploration vehicles, with disposable elements, such as the lunar lander, solid rocket boosters, and space shuttle main engines that will require higher numbers of production runs. Rather than establish specific criteria by which all projects are judged, NASA’s policy requires that projects manage to baselines and plans established in key management documents and approved at the project NAR. The baselines and plans serve as their primary tools for measuring project progress and as the primary basis for judgment at project reviews. While the plans may include some information that addresses knowledge-based criteria for design and production, the instructions for preparing them leaves the establishment of thresholds and success criteria to the discretion of the project manager. For example, NASA policy requires that projects include, as part of the project plan, a Verification and Validation Sub Plan that describes the project’s approach to verifying and validating hardware and software as part of the project plan. The policy, however, includes no instruction as to what constitutes a sufficient approach to testing. In other words, there are no requirements concerning the fidelity of test articles or the realism of the test environment. Similarly, NASA’s policy requires that projects include, as part of the project plan, a Systems Engineering Sub Plan that describes the project’s approach to systems engineering and the technical standards that are applicable, including metrics that verify the processes. The policy, however, does not identify the types of metrics appropriate to verify the process or establish any threshold criteria. The absence of major decision reviews, along with specific criteria in the fabrication, assembly, and test phase, could result in concurrent design and manufacturing activities, a practice our past work has found increases risk in acquisition programs. Furthermore, lacking a major decision review to ensure that projects have gained the appropriate levels of knowledge at KP2 and KP3, NASA decision makers cannot be provided a high level of certainty that the project will meet cost, schedule, and performance requirements and have no assurance that the provisions of the key management documents required by NPR 7120.5C are being executed after the NAR. NASA centers have varying approaches to implementing project management policies and systems engineering guidance for flight systems and ground support projects. Some centers use criteria at key decision points that are similar to the criteria required to ensure a knowledge-based approach is followed, while others lack such criteria. As a result, each center reports a different level and type of knowledge about a project at key decision points. Centers also rely on project managers and systems engineers to employ good project management and systems engineering practices. However, given the loss of experienced project managers and the decline of in-house systems engineering and technical capabilities agencywide, that reliance could be problematic. These situations make it difficult for NASA decision makers to evaluate center projects on a common foundation of knowledge and make sound investment decisions and tradeoffs based on those evaluations. A standardized, knowledge- based approach would prepare NASA to face competing budgetary priorities and make difficult decisions regarding the investment in and termination of projects. While NASA centers are given discretion about how they implement agencywide policies, they are expected to have procedures and guidelines in place for implementing those policies. Some centers have developed center-specific policies and criteria for implementing NASA’s project management policies and system engineering guidance, while others have not. Centers also rely on project managers and systems engineers to implement the requirements of NPR 7120.5C and to use NASA’s Systems Engineering Handbook as guidance for good systems engineering practices. Some NASA centers have also developed criteria in their policies that are similar to the criteria used to ensure a knowledge-based approach is followed; other centers lack such criteria. Because of their varying policies and criteria, each center requires a different level of knowledge at the same point in a project’s development cycle. For example, GSFC requires its projects to mature technologies to TRL 6 by the preliminary design review—before entering the implementation phase. On the other hand, JPL, MSFC, and JSC policies do not require projects to mature technology to a particular level before entering implementation, leaving the determination of needed technology maturity up to the project manager. Requirements for assessing design maturity also vary across the centers. While most of the center policies require a CDR to enter “Phase D— Fabrication, Assembly, and Test”—the criteria used to assess projects at this point vary. For example, both GSFC and MSFC require projects to have completed a percentage of design drawings at this review. MSFC requires that 90 percent of design drawings to be complete by CDR, which is consistent with best practices. While GSFC establishes a minimum threshold of drawings to be complete by CDR—greater than 80 percent— neither JSC nor JPL establish a minimum percentage drawing requirement. Instead, JSC requires the design be complete and drawings ready to begin production, and JPL requires that the design be mature and provide confidence in the integrity of the flight system design. Almost none of the center policies include a requirement to assess the maturity of production processes. According to NASA officials, due to the low quantities of systems generally produced by the agency, most of the center policies do not require a review before beginning manufacturing. Only JSC requires a Production Readiness Review to ensure that production plans, facilities, and personnel are in place and ready to begin production. However, the criteria do not specify quantifiable thresholds to measure production readiness at the review. JPL, MSFC, and GSFC do not have policies that outline requirements for such a review. In addition to individual policy requirements, centers may rely on project managers and systems engineers to employ good project management and systems engineering practices. For example, an experienced project manager at JPL told us that although JPL policy does not require a particular TRL at PDR to enter implementation, he required that all technologies for his project be around a TRL 6 in order to separate technology development from systems development. Reliance on project managers to implement good practices, however, could be problematic given the diminishing number of experienced project managers available to lead projects and the decline of in-house systems engineering and technical capabilities agencywide caused by increasing retirements and outsourcing. In a knowledge-based process, the achievement of each successive knowledge point builds on the preceding one, giving decision makers the information they need, when they need it, to make decisions about whether to invest significant additional funds to move forward with product development. Our work has shown that successful product development efforts are marked by adherence to a disciplined process that establishes and uses common and consistent criteria for decision making at these key points. With varying project management and systems engineering criteria, NASA centers’ technical reviews, such as PDR, provide different levels of knowledge to support NASA’s major decision reviews, such as the NAR, which NASA uses to support its major investment decisions for flight systems and ground support projects. In the near future, NASA will need to determine the resources necessary to develop the systems and supporting technologies to achieve the President’s Vision for Space Exploration and structure its investment strategy accordingly. Initial implementation of the Vision as explained in NASA’s Exploration Systems Architecture Study calls for completing the International Space Station, developing a new crew exploration vehicle, and returning to the moon no later than 2020. NASA estimates that it will cost approximately $104 billion over the next 13 years to accomplish these initial goals. These priorities, along with NASA’s other missions, will be competing within NASA for funding. It will likely be difficult for NASA managers to agree on which projects to invest in and which projects to terminate. The NASA Administrator has acknowledged that NASA faces difficult choices about its missions in the future—for example, between human space flight, science, and aeronautics missions. Using consistent criteria to evaluate all NASA projects would help ensure that the same level and type of knowledge is available about individual projects at key decision points. Analogous information about all flight systems and ground support projects would allow decision makers to make apples-to-apples comparisons across projects and make investment decisions, and trade-offs, based upon these comparisons. Further, policies with consistent criteria can provide inexperienced project managers and systems engineers with the necessary guidance to implement good project management and systems engineering practices and ensure that the right knowledge is available for decision makers. NASA officials within the Chief Engineer’s Office acknowledged the need for more consistency in criteria across the various NASA centers in order to successfully achieve NASA’s Vision for Space Exploration. Some NASA centers, however, are resistant to standardized criteria because they feel it could be overly prescriptive. These centers indicated that because of the unique nature of the work done at each of the 10 NASA centers, it would be unrealistic to hold every project to the same criteria. Nonetheless, our work has shown that the process used by successful product developers to develop leading-edge technology and products does not differ based upon the type of product being developed. Further, these developers adhere to a disciplined process that establishes and uses common and consistent criteria for decision making—regardless of the type of product or technology being developed. Using consistent criteria can allow NASA decision makers to assess the likely return on competing investment priorities and to reevaluate alternatives and make investment decisions across projects to increase the likelihood of attaining the strategic goals of the agency. Several of NASA’s major acquisitions have been marked by cost, schedule, and performance problems. Yet the challenges NASA faces in the future are likely to far exceed those it has faced in the past. The complex technical requirements associated with fulfilling the President’s Vision, the fiscal constraints under which NASA will be required to operate, the diminished number of experienced project managers and systems engineers, and the potential for increased production make following a knowledge-based approach for flight systems and ground support projects all the more critical. While NASA has made improvements to its policies governing project management, the lack of major decision reviews beyond the initial project approval gate leaves decision makers with little knowledge about the progress of the agency’s projects. Further, without a standard set of criteria to measure projects at crucial phases in the development life cycle, NASA cannot be assured that its decisions will result in the best possible return on its investments. Since NASA is currently in the process of revising its program and project management policies and developing an agencywide systems engineering policy, we believe this presents a unique opportunity for the agency to correct some of the problems identified during our review. In order to close the gaps between NASA’s current acquisition environment and best practices on knowledge-based acquisition, we recommend that NASA take steps to ensure NASA projects follow a knowledge-based approach for product development. Specifically, we recommend that the NASA Administrator direct the Office of the Chief Engineer to take the following two actions: In drafting its systems engineering policy, incorporate requirements in the policy for flight systems and ground support projects to capture specific product knowledge by key junctures in project development. The demonstration of this knowledge should be used as exit criteria for decision making at the following key junctures: Before projects are approved to transition from formulation to implementation, the policy should require that projects demonstrate that key technologies have reached a high maturity level. Before projects are approved to transition from final design to fabrication, assembly, and test, the policy should require that projects demonstrate that the design is stable. Before projects are approved to transition into production, the policy should require projects to demonstrate that the design can be manufactured within cost, schedule, and quality targets. Revise NPR 7120.5C to institute additional major decision reviews following the NAR for flight systems and ground support projects, which result in recommendations to the appropriate decision authority. These reviews should be tied to the key junctures during project development mentioned above in order to increase the likelihood that cost, schedule, and performance requirements of the project will be met. We provided a draft of this report to NASA for review and comment. In written comments, NASA indicated that it agreed with our recommendations and outlined specific actions that the agency plans to take to address them. The actions that NASA plans to take to address our recommendations are a positive step toward achieving successful project outcomes and ensuring that decision makers are appropriately investing the agency’s resources. We are pleased to hear that many knowledge-based practices identified in our report are currently being practiced agencywide in the management and development NASA systems. The addition of such practices to NASA’s policies will only strengthen their use agencywide and ensure that these practices continue to be utilized by less experienced project managers and systems engineers as the more experienced workforce retires. The effectiveness of such practices, however, will be limited if project officials are not held accountable for demonstrating a high level of knowledge, consistent with the success criteria that NASA plans to require in its policies, at key junctures in development. It is critical that project officials not only have a high level of knowledge about a project at key junctures, but also that this information is used by decision makers to make decisions on whether to invest additional resources and allow a project to proceed through the development life cycle. NASA’s comments are reprinted in appendix III. NASA also provided technical comments, which we addressed throughout the report as appropriate. As agreed with your offices, unless you announce its contents earlier, we will not distribute this report further until 30 days from its date. At that time, we will send copies to NASA’s Administrator and interested congressional committees. We will make copies available to others upon request. In addition, the report will be available at no charge on the GAO Web site at http://www.gao.gov. If your or your staff have any questions concerning this report, please contact me at (202) 512-4841 or lia@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are acknowledged in appendix IV. To determine the extent to which NASA’s current policies support an acquisition approach consistent with best practices identified in GAO’s work on system acquisitions, we reviewed and analyzed NASA-wide program and project management policies and systems engineering guidance. Our review and analysis of the policy focused on requirements for flight systems and ground support projects. We interviewed NASA Headquarters officials from the Office of the Chief Engineer who are responsible for the policy and guidance. We compared NASA’s policy on program and project management with criteria contained in GAO best practices work on systems acquisition and space system acquisitions. We concentrated on whether the policy provides a framework for a knowledge-based process and the criteria necessary to carry out this intent. To determine how NASA-wide acquisition policies are implemented across the various NASA centers, we reviewed NASA center-specific program and project management policies and systems engineering policies and interviewed officials responsible for implementing those policies. While we interviewed officials from all centers, we focused on the centers that manage the majority of NASA’s flight systems and ground support projects—GSFC, JPL, JSC, and MSFC. This approach included site visits with GSFC, JPL, JSC, and MSFC and teleconferences with the remaining centers. We compared examples of the centers’ implementation of the policies and specific criteria included in these policies with our best practices work on systems acquisition. We performed our work from May 2005 to November 2005 in accordance with generally accepted government auditing standards. In addition to the individual named above, James L. Morrison, Assistant Director; Valerie Colaiaco, Tom Gordon, Alison Heafitz, Shelby S. Oakley, Ron Schwenn, Karen Sloan, and John S. Warren, Jr., made key contributions to this report. Space Acquisitions: Stronger Development Practices and Investment Planning Need to Address Continuing Problems. GAO-05-891T. Washington, D.C.: July 12, 2005. Defense Acquisitions: Incentives and Pressures That Drive Problems Affecting Satellite and Related Acquisitions. GAO-05-570R. Washington, D.C.: June 23, 2005. Defense Acquisitions: Space-Based Radar Effort Needs Additional Knowledge before Starting Development. GAO-04-759. Washington, D.C.: July 23, 2004. Defense Acquisitions: Risks Posed by DOD’s New Space Systems Acquisition Policy. GAO-04-379R. Washington, D.C.: January 29, 2004. Space Acquisitions: Committing Prematurely to the Transformational Satellite Program Elevates Risks for Poor Cost, Schedule, and Performance Outcomes. GAO-04-71R. Washington, D.C.: December 4, 2003. Defense Acquisitions: Improvements Needed in Space Systems Acquisition Policy to Optimize Growing Investment in Space. GAO-04-253T. Washington, D.C.: November 18, 2003. Defense Acquisitions: Despite Restructuring, SBIRS High Program Remains at Risk of Cost and Schedule Overruns. GAO-04-48. Washington, D.C.: October 31, 2003. Defense Acquisitions: Improvements Needed in Space Systems Acquisition Management Policy. GAO-03-1073. Washington, D.C.: September 15, 2003. Military Space Operations: Common Problems and Their Effects on Satellite and Related Acquisitions. GAO-03-825R. Washington, D.C.: June 2, 2003. Military Space Operations: Planning, Funding, and Acquisition Challenges Facing Efforts to Strengthen Space Control. GAO-02-738. Washington, D.C.: September 23, 2002. Polar-Orbiting Environmental Satellites: Status, Plans, and Future Data Management Challenges. GAO-02-684T. Washington, D.C.: July 24, 2002. Defense Acquisitions: Space-Based Infrared System-Low at Risk of Missing Initial Deployment Date. GAO-01-6. Washington, D.C.: February 28, 2001. Defense Acquisitions: Assessments of Selected Major Weapon Programs. GAO-05-301. Washington, D.C.: March 31, 2005. Defense Acquisitions: Stronger Management Practices Are Needed to Improve DOD’s Software-Intensive Weapon Acquisitions. GAO-04-393. Washington, D.C.: March 1, 2004. Defense Acquisitions: Assessments of Selected Major Weapon Programs. GAO-04-248. Washington, D.C.: March 31, 2004. Defense Acquisitions: DOD’s Revised Policy Emphasizes Best Practices, but More Controls Are Needed. GAO-04-53. Washington, D.C.: November 10, 2003. Defense Acquisitions: Assessments of Selected Major Weapon Programs. GAO-03-476. Washington, D.C.: May 15, 2003. Best Practices: Setting Requirements Differently Could Reduce Weapon Systems’ Total Ownership Costs. GAO-03-57. Washington, D.C.: February 11, 2003. Best Practices: Capturing Design and Manufacturing Knowledge Early Improves Acquisition Outcomes. GAO-02-701. Washington, D.C.: July 15, 2002. Defense Acquisitions: DOD Faces Challenges in Implementing Best Practices. GAO-02-469T. Washington, D.C.: February 27, 2002. Best Practices: Better Matching of Needs and Resources Will Lead to Better Weapon System Outcomes. GAO-01-288. Washington, D.C.: March 8, 2001. Best Practices: A More Constructive Test Approach Is Key to Better Weapon System Outcomes. GAO/NSIAD-00-199. Washington, D.C.: July 31, 2000. Defense Acquisition: Employing Best Practices Can Shape Better Weapon System Decisions. GAO/T-NSIAD-00-137. Washington, D.C.: April 26, 2000. Best Practices: DOD Training Can Do More to Help Weapon System Program Implement Best Practices. GAO/NSIAD-99-206. Washington, D.C.: August 16, 1999. Best Practices: Better Management of Technology Development Can Improve Weapon System Outcomes. GAO/NSIAD-99-162. Washington, D.C.: July 30, 1999. Defense Acquisitions: Best Commercial Practices Can Improve Program Outcomes. GAO/T-NSIAD-99-116. Washington, D.C.: March 17, 1999. Defense Acquisition: Improved Program Outcomes Are Possible. GAO/T-NSIAD-98-123. Washington, D.C.: March 18, 1998. Best Practices: Successful Application to Weapon Acquisition Requires Changes in DOD’s Environment. GAO/NSIAD-98-56. Washington, D.C.: February 24, 1998. Major Acquisitions: Significant Changes Underway in DOD’s Earned Value Management Process. GAO/NSIAD-97-108. Washington, D.C.: May 5, 1997. Best Practices: Commercial Quality Assurance Practices Offer Improvements for DOD. GAO/NSIAD-96-162. Washington, D.C.: August 26, 1996. | The National Aeronautics and Space Administration (NASA) plans to spend over $100 billion on capabilities and technologies to achieve the initial goals of the President's 2004 Vision for Space Exploration. In the past, NASA has had difficulty meeting cost, schedule, and performance objectives for some of its projects because it failed to adequately define project requirements and quantify resources. NASA will be further challenged by a constrained federal budget and a shrinking experienced NASA workforce. To help face these challenges and manage projects with greater efficiency and accountability, NASA recently updated its program and project management policy and is developing an agencywide systems engineering policy. GAO has issued a series of reports on the importance of obtaining critical information and knowledge at key junctures in major system acquisitions to help meet cost and schedule objectives. This report (1) evaluates whether NASA's policy supports a knowledge-based acquisition approach and (2) describes how NASA centers are implementing the agency's acquisition policies and guidance. While NASA's revised policy for developing flight systems and ground support projects incorporates some of the best practices used by successful developers, it lacks certain key criteria and major decision reviews that support a knowledge-based acquisition framework. For example, NASA's policy requires projects to conduct a major decision review before moving from formulation to implementation. Further, before moving from formulation to implementation, projects must validate requirements and develop realistic cost and schedule estimates, human capital plans, a preliminary design, and a technology plan--key elements for matching needs to resources. However, NASA's policies do not require projects to demonstrate technologies at high levels of maturity before program start. By not establishing a minimum threshold for technology maturity, NASA increases the risk that design changes will be required later in development, when such changes are typically more costly to make. In addition, although NASA's policy does require project managers to establish a continuum of technical and management reviews, it does not specify what these reviews should be, nor does it require major decision reviews at other key points in a product's development. Acquiring knowledge at key junctures will become increasingly important as NASA proceeds to implement elements of the Vision. Without a major decision review at key milestones to ensure that the appropriate level of knowledge has been achieved to proceed to the next phase, the risk of cost and schedule overruns, as well as performance shortfalls, increases. NASA centers have varying approaches for implementing the agency's policies and guidance. Some centers have established product development criteria that are similar to the criteria used in a knowledge-based acquisition, while other centers have not. As a result, each center reports a different level and type of knowledge about a project at key decision points. Centers also rely on project managers and systems engineers to employ good project management and systems engineering practices. However, given the loss of experienced project managers and the decline of in-house systems engineering and technical capabilities, that reliance could be problematic. These situations make it difficult for decision makers to evaluate projects on the same basis and make sound investment decisions and tradeoffs based on those evaluations. A standardized, knowledge-based approach would prepare NASA to face competing budgetary priorities and better position the agency to make difficult decisions regarding the investment in and termination of projects. |
Although the current focus of concern is largely on the potential for several years of declining physician fees, the historic challenge for Medicare has been to find ways to moderate the rapid growth in spending for physician services. Before 1992, the fees that Medicare paid for those services were largely based on physicians’ historical charges. Spending for physician services grew rapidly in the 1980s, at a rate that the Secretary of Health and Human Services (HHS) characterized as out of control. Although Congress froze fees or limited fee increases, spending continued to rise because of increases in the volume and intensity of physician services. From 1980 through 1991, for example, Medicare spending per beneficiary for physician services grew at an average annual rate of 11.6 percent. The ineffectiveness of fee controls alone led Congress to reform the way that Medicare set physician fees. The Omnibus Budget Reconciliation Act of 1989 (OBRA 1989) established both a national fee schedule and a system of spending targets, which first affected physician fees in 1992. From 1992 through 1997, annual spending growth for physician services was far lower than the previous decade. The decline in spending growth was the result in large part of slower volume and intensity growth. (See fig. 1.) Over time, Medicare’s spending target system has been revised and renamed. The SGR system, Medicare’s current system for updating physician fees, was established in the Balanced Budget Act of 1997 (BBA) and was first used to adjust fees in 1999. Following the implementation of the fee schedule and spending targets in 1992, through 1999, average annual growth in volume and intensity of service use per beneficiary fell to 1.1 percent. More recently volume and intensity growth has trended upward, rising at an average annual rate of about 5 percent from 2000 through 2003. Although this average annual rate of growth remains substantially below that experienced before spending targets were introduced, the recent increases in volume and intensity growth are a reminder that inflationary pressures continue to challenge efforts to moderate growth in physician expenditures. The SGR system establishes spending targets to moderate physician services spending increases caused by excess growth in volume and intensity. SGR’s spending targets do not cap expenditures for physician services. Instead, spending in excess of the target triggers a reduced fee update or a fee cut. In this way, the SGR system applies financial brakes to physician services spending and thus serves as an automatic budgetary control device. In addition, reduced fee updates signal physicians collectively and Congress that spending due to volume and intensity has increased more than allowed. To apply the SGR system, every year the Centers for Medicare & Medicaid Services (CMS) follows a statutory formula to estimate the allowed rate of increase in spending for physician services and uses that rate to construct the spending target for the following calendar year. The sustainable growth rate is the product of the estimated percentage change in (1) input prices for physician services;9, 10 (2) the average number of Medicare beneficiaries in the traditional fee-for-service (FFS) program; (3) national economic output, as measured by real (inflation-adjusted) GDP per capita; and (4) expected expenditures for physician services resulting from changes in laws or regulations. SGR spending targets are cumulative. That is, the sum of all physician services spending since 1996 is compared to the sum of all annual targets since the same year to determine whether spending has fallen short of, equaled, or exceeded the SGR targets. The use of cumulative targets means, for example, that if actual spending has exceeded the SGR system targets, fee updates in future years must be lowered sufficiently both to offset the accumulated excess spending and to slow expected spending for the coming year. CMS calculates changes in physician input prices based on the growth in the costs of providing physician services as measured by the Medicare Economic Index, growth in the costs of providing laboratory tests as measured by the consumer price index for urban consumers, and growth in the cost of Medicare Part B prescription drugs included in SGR spending. Under the SGR and MVPS systems, the Secretary of Health and Human Services defined physician services to include “services and supplies incident to physicians’ services,” such as laboratory tests and most Part B prescription drugs. Under SGR, spending per beneficiary adjusted for the estimated underlying cost of providing physician services is allowed to grow at the same rate that the national economy grows over time on a per-capita basis—currently projected to be slightly more than 2 percent annually. If volume and intensity grow faster, the annual increase in physician fees will be less than the estimated increase in the cost of providing services. Conversely, if volume and intensity grow more slowly than 2 percent annually, the SGR system permits physicians to benefit from fee increases that exceed the increased cost of providing services. To reduce the effect of business cycles on physician fees, MMA modified the SGR system to require that economic growth be measured as the 10-year moving average change in real per capita GDP. This measure is projected to range from 2.1 percent to 2.5 percent during the 2005 through 2014 period. When the SGR system was established, GDP growth was seen as a benchmark that would allow for affordable increases in volume and intensity. In its 1995 annual report to Congress, the Physician Payment Review Commission stated that limiting real expenditure growth to 1 or 2 percentage points above GDP would be a “realistic and affordable goal.” Ultimately, BBA specified the growth rate of GDP alone. This limit was an indicator of what the 105th Congress thought the nation could afford to spend on volume and intensity increases. If cumulative spending on physician services is in line with SGR’s target, the physician fee schedule update for the next calendar year is set equal to the estimated increase in the average cost of providing physician services as measured by the Medicare Economic Index (MEI). If cumulative spending exceeds the target, the fee update will be less than the change in MEI or may even be negative. If cumulative spending falls short of the target, the update will exceed the change in MEI. The SGR system places bounds on the extent to which fee updates can deviate from MEI. In general, with an MEI of about 2 percent, the largest allowable fee decrease would be about 5 percent and the largest fee increase would be about 5 percent. The 2004 Medicare Trustees Report announced that the projected physician fee update would be about negative 5 percent for 7 consecutive years beginning in 2006; the result is a cumulative reduction in physician fees of more than 31 percent from 2005 to 2012, while physicians’ costs of providing services, as measured by MEI, are projected to rise by 19 percent. According to projections made by CMS Office of the Actuary (OACT) in July 2004, maximum fee reductions will be in effect from 2006 through 2012, while fee updates will be positive in 2014. (See fig. 2.) There are two principal reasons for the projected fee declines: increases in volume and intensity that exceed the SGR’s allowance—partly as a result of spending for Part B prescription drugs—and the minimum fee updates for 2004 and 2005 specified by MMA. Recent growth in spending due to volume and intensity increases has been larger than SGR targets allow, resulting in excess spending that must be recouped through reduced fee updates. In general, the SGR system allows physician fee updates to equal or exceed the MEI as long as spending growth due to volume and intensity increases is no higher than the average growth in real GDP per capita—about 2.3 percent annually. However, in July 2004, CMS OACT projected that the volume and intensity of physician services paid for under the physician fee schedule would grow by 3 percent per year. To offset the resulting excess spending, the SGR system will have to reduce future physician fee updates. Additional downward pressure on physician fees arises from the growth in spending for other Medicare services that are included in the SGR system, but that are not paid for under the physician fee schedule. Such services include laboratory tests and many Part B outpatient prescription drugs that physicians provide to patients. Because physicians influence the volume of services they provide directly—that is, fee schedule services— as well as these other services, defined by the Secretary of HHS as “incident to” physician services, expenditures for both types of services were included when spending targets were introduced. In July 2004, CMS OACT projected that SGR-covered Part B drug expenditures would grow more rapidly than other physician service expenditures, thus increasing the likelihood that future spending would exceed SGR system targets. To the extent that spending for SGR Part B drugs and other “incident to” services grows larger as a share of overall SGR spending, additional pressure is put on fee adjustments to offset excess spending and bring overall SGR spending in line with the system’s targets. This occurs because the SGR system attempts to moderate spending only through the fee schedule, even when the excess spending is caused by expenditures for “incident to” services, such as Part B drugs, which are not paid for under the fee schedule. The MMA averted fee reductions projected for 2004 and 2005 by specifying an update to physician fees of no less than 1.5 percent for those 2 years. The MMA increases replaced SGR system fee reductions of 4.5 percent in 2004 and 3.3 percent in 2005 and thus will result in additional aggregate spending. Because MMA did not make corresponding revisions to the SGR system’s spending targets, the SGR system must offset the additional spending by reducing fees beginning in 2006. An examination of the SGR fee update that would have gone into effect in 2005, absent the MMA minimum updates, illustrates the impact of the system’s cumulative spending targets. To begin with, actual expenditures under the SGR system in 2004 are estimated to be $84.9 billion, whereas target expenditures for 2004 were $77.1 billion. As a result, SGR’s 2005 fee updates would have needed to offset the $7.8 billion deficit from excess spending in 2004 plus the accumulated excess spending of $5.9 billion from previous years to realign expected spending with target spending. Because the SGR system is designed to offset accumulated excess spending over a period of years, the deficit for 2004 and preceding years reduces fee updates for multiple years. The projected sustained period of declining physician fees and the potential for beneficiaries’ access to physician services to be disrupted have heightened interest in alternatives for the current SGR system. In general, potential alternatives cluster around two approaches. One approach would end the use of spending targets as a method for updating physician fees and encouraging fiscal discipline. The other approach would retain spending targets but modify the current SGR system to address perceived shortcomings. These modifications include such options as removing the prescription drug expenditures that are currently counted in the SGR system; resetting the targets and not requiring the system to recoup previous excess spending; and raising the allowance for increased spending due to volume and intensity growth. The Part B premium amount is adjusted each year so that expected premium revenues equal 25 percent of expected Part B spending. Beneficiaries must pay coinsurance—usually 20 percent—for most Part B services. physicians appropriately, it is important to consider how modifications or alterations to the SGR system would affect the long-term sustainability and affordability of the Medicare program. See GAO-05-85 for more information about these alternatives. See Medicare Payment Advisory Commission, Report to the Congress: Medicare Payment Policy (Washington, D.C.: March 2001, 2002, 2003, and 2004). MedPAC suggested that other adjustments to the update might be necessary, for example, to ensure overall payment adequacy, correct for previous MEI forecast errors, and to address other factors. likely produce fee updates that ranged from 2.1 percent to 2.4 percent over the period from 2006 through 2014. (See table 1.) However, Medicare spending for physician services would rise, resulting in cumulative expenditures that are 22 percent greater over a 10-year period than under current law, based on CMS OACT estimates. Although MedPAC’s recommended update approach would limit annual increases in the price Medicare pays for each service, the approach does not contain an explicit mechanism for constraining aggregate spending resulting from increases in the volume and intensity of services physicians provide. In 2004 testimony, MedPAC stated that fee updates for physician services should not be automatic, but should be informed by changes in beneficiaries’ access to services, the quality of services provided, the appropriateness of cost increases, and other factors, similar to those that MedPAC takes into consideration when considering updates for other providers. Another approach for addressing the perceived shortcoming of the current SGR system would retain spending targets but modify one or more elements of the system. The key distinction of this approach, in contrast to basing updates on MEI, is that fiscal controls designed to moderate spending would continue to be integral to the system used to update fees. Although spending for physician services would likely also rise under this approach, the advantage of retaining spending targets is that the fee update system would automatically work to moderate spending if volume and intensity growth began to increase above allowable rates. The SGR system could be modified in a number of ways: for example, by raising the allowance for increased spending due to volume and intensity growth; resetting the base for the spending targets and not requiring the system to recoup previous excess spending; or removing the prescription drug expenditures that are currently counted in the SGR system. The current SGR system’s allowance for volume and intensity growth could be increased, through congressional action, by some factor above the percentage change in real GDP per capita. As stated earlier, the current SGR system’s allowance for volume and intensity growth is approximately 2.3 percent per year—the 10-year moving average in real GDP per capita— while CMS OACT projected that volume and intensity growth would be more than 3 percent per year. To offset the increased spending associated with the higher volume and intensity growth, the SGR system will reduce updates below the increase in MEI. According to CMS OACT simulations, increasing the allowance for volume and intensity growth to GDP plus 1 percentage point would likely produce positive fee updates beginning in 2012—2 years earlier than is projected under current law. Because fee updates would be on average greater than under current law during the 10- year period from 2005 through 2014, Medicare spending for physician services would rise. CMS OACT estimated that cumulative expenditures over the 10-year period would increase by 4 percent more than under current law. (See table 1.) In 2002, we testified that physician spending targets and fees may need to be adjusted periodically as health needs change, technology improves, or health care markets evolve. Such adjustments could involve specifying a new base year from which to set future targets. Currently, the SGR system uses spending from 1996, trended forward by the sustainable growth rate computed for each year, to determine allowable spending. MMA avoided fee declines in 2004 and in 2005 by stipulating a minimum update of 1.5 percent in each of those 2 years, but the law did not similarly adjust the spending targets to account for the additional spending that would result from the minimum update. Consequently, under the SGR system the additional MMA spending and other accumulated excess spending will have to be recouped through fee reductions beginning in 2006. If the resulting negative fee updates are considered inappropriately low, one solution would be, through congressional action, to use actual spending from a recent year as a basis for setting future SGR system targets and forgiving the accumulated excess spending attributable to MMA and other factors. The effect of this action would be to increase future updates and, as with other alternatives presented here, overall spending. According to CMS OACT simulations, forgiving the accumulated excess spending as of 2005—that is, resetting the cumulative spending target so that it equals cumulative actual spending—would raise fees in 2006. However, because volume and intensity growth is projected to exceed the SGR system’s allowance for such growth, negative updates would return beginning in 2008 and continue through 2013. Resulting cumulative spending over the 10-year period from 2005 through 2014 would be 13 percent higher than is projected under current law. (See table 1.) The Secretary of HHS could, under current authority, consider excluding Part B drugs from the definition of services furnished incident to physician services for purposes of the SGR system. Expenditures for these drugs have been growing rapidly, which, in turn, has put downward pressure on the fees paid to Medicare physicians. However, according to CMS OACT simulations, removing Part B drugs from the SGR system beginning in 2005 would not prevent several years of fee declines and would not decrease the volatility in the updates. Fees would decline by about 5 percent per year from 2006 through 2010. There would be positive updates beginning in 2011—3 years earlier than is projected under current law. (See table 1.) CMS OACT estimated that removing Part B drugs from the SGR system would result in cumulative spending over the 10-year period from 2005 through 2014 that is 5 percent higher than is projected under current law. Together Congress and CMS could implement several modifications to the SGR system, for example, by increasing the allowance for volume and intensity growth to GDP plus 1 percentage point, resetting the spending base for future SGR targets, and removing prescription drugs. According to CMS OACT simulations, this combination of options would result in positive updates ranging from 2.2 percent to 2.8 percent for the 2006–2014 period. CMS OACT projected that the combined options would increase aggregate spending by 23 percent over the 10-year period. (See table 1.) Medicare faces the challenge of moderating the growth in spending for physician services while ensuring that physicians are paid fairly so that beneficiaries have appropriate access to their services. Concerns have been raised that access to physician services could eventually be compromised if the SGR system is left unchanged and the projected fee cuts become a reality. These concerns have prompted policymakers to consider two broad approaches for updating physician fees. The first approach—eliminating targets—emphasizes fee stability while the second approach—retaining and modifying targets—includes an automatic fiscal brake. Either of the two approaches could be implemented in a way that would likely generate positive fee updates and each could be accompanied by separate, focused efforts to moderate volume and intensity growth. Because multiple years of projected 5 percent fee cuts are incorporated in Medicare’s budgeting baseline, almost any change to the SGR system is likely to increase program spending above the baseline. As policymakers consider options for updating physician fees, it is important to be mindful of the serious financial challenges facing Medicare and the need to design policies that help ensure the long-term sustainability and affordability of the program. We look forward to working with the Subcommittee and others in Congress as policymakers seek to moderate program spending growth while ensuring appropriate physician payments. Madam Chairman, this concludes my prepared statement. I will be happy to answer questions you or the other Subcommittee Members may have. For further information regarding this testimony, please contact A. Bruce Steinwald at (202) 512-7101. James Cosgrove, Jessica Farb, Hannah Fein, and Jennifer Podulka contributed to this statement. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | Concerns were raised about the system Medicare uses to determine annual changes to physician fees--the sustainable growth rate (SGR) system--when it reduced physician fees by almost 5 percent in 2002. Subsequent administrative and legislative actions modified or overrode the SGR system to avert fee declines in 2003, 2004, and 2005. However, projected fee reductions for 2006 to 2012 have raised new concerns about the SGR system. Policymakers question the appropriateness of the SGR system for updating physician fees and its effect on physicians' continued participation in the Medicare program if fees are permitted to decline. At the same time, there are concerns about the impact of increased spending on the long-term fiscal sustainability of Medicare. GAO was asked to discuss the SGR system. Specifically, this statement addresses the following: (1) how the SGR system is designed to moderate the growth in spending for physician services, (2) why physician fees are projected to decline under the SGR system, and (3) options for revising or replacing the SGR system and their implications for physician fee updates and Medicare spending. This statement is based on GAO's most recent report on the SGR system, Medicare Physician Payments: Concerns about Spending Target System Prompt Interest in Considering Reforms (GAO-05-85). To moderate Medicare spending for physician services, the SGR system sets spending targets and adjusts physician fees based on the extent to which actual spending aligns with specified targets. If growth in the number of services provided to each beneficiary--referred to as volume--and in the average complexity and costliness of services--referred to as intensity--is high enough, spending will exceed the SGR target. While the SGR system allows for some volume and intensity spending growth, this allowance is limited. If such growth exceeds the average growth in the national economy, as measured by the gross domestic product per capita, fee updates are set lower than inflation in the cost of operating a medical practice. A large gap between spending and the target may result in fee reductions. There are two principal reasons why physician fees are projected to decline under the SGR system beginning in 2006. One problem is that projected volume and intensity spending growth exceeds the SGR allowance for such growth. Second, the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (MMA) increased the update for 2004 and 2005--thus increasing spending--but did not raise the spending targets for those years. The SGR system, which is designed to keep spending in line with its targets, must reduce fees beginning in 2006 to offset excess spending attributable to both volume and intensity growth and the MMA provision. In general, proposals to reform Medicare's method for updating physician fees would either (1) eliminate spending targets and establish new considerations for the annual fee updates or (2) retain spending targets, but modify certain aspects of the current system. The first approach emphasizes stable and positive fee updates, while the second approach automatically applies financial brakes whenever spending for physician services exceeds predefined spending targets. Either approach could be complemented by focused efforts to moderate volume and intensity growth directly. As policymakers consider options for updating physician fees, it is important to be mindful of the serious financial challenges facing Medicare and the need to design policies that help ensure the long-term sustainability and affordability of the program. |
Survey questions about sensitive topics carry a “threat” for some respondents, because they fear that a truthful answer could result in some degree of negative consequence (at a minimum, social disapproval). The grouped answers approach is designed to reduce this threat when asking about immigration status. Three key points about the grouped answers approach are that 1. no respondent is ever asked whether he or she, or anyone else, is 2. two pieces of information are separately provided by two subsamples of respondents (completely different people—no one is shown both immigration status cards); and 3. taking the two pieces of information together—like two different pieces of a puzzle—allows indirect estimation of the undocumented population, but no individual respondent (and no piece of data on an individual respondent) is ever categorized as undocumented. We discuss each point in some detail. 1. No respondent is ever asked whether he or she is in the undocumented category. Unlike questions that ask respondents to choose among specific answer categories, the grouped answers approach combines answer categories in sets or “boxes,” as shown in figure 1. Box B includes the sensitive answer category-—currently “undocumented”—along with other categories that are nonsensitive. Each respondent is asked to “pick the Box”—Box A, Box B, or Box C— that contains the specific answer category that applies to him or her. Respondents are told, in effect: If the specific category that applies to you is in Box B, we don’t want to know which one it is, because right now we are focusing on Box A categories. By using the boxes, the interview avoids “zeroing in” on the sensitive answer. The specific categories shown in the boxes in figure 1 are grouped so that one would expect many respondents who are here legally, as well as those who are undocumented, to choose Box B, and there is virtually no possibility of anyone deducing which specific category within Box B applies to any individual respondent. 2. Two pieces of information are provided separately by two subsamples of respondents (no one is shown both immigration status cards). Respondents are divided into two subsamples, based on randomization procedures or rotation (alternation) procedures conducted outside the interview process. (For example, a rotation procedure might specify that within an interviewing area, every other household will be designated as subsample 1 or subsample 2.) This “split sample” procedure has been used routinely for many surveys over the years. As applied to the grouped answers approach, the two subsamples are shown alternative flash cards. Immigration Status Card 1, described above, represents one way to group immigration statuses in three boxes. A second immigration status flash card (Immigration Status Card 2, shown in figure 2) groups the same statuses differently. The alternative immigration-status cards can be thought of as “mirror images” in that the two nonsensitive legal statuses in Box A of Card 1 appear in Box B of Card 2 and the two nonsensitive legal statuses in Box B of Card 1 appear in Box A of Card 2. However, the undocumented status always appears in Box B. Interviewers ask survey respondents in subsample 1 about immigration status with respect to Card 1. They ask survey respondents in subsample 2 (completely different persons) about immigration status with respect to Card 2. Each respondent is shown one and only one immigration-status flash card. There are no highly unusual or complicated interviewing procedures. Because the two subsamples of respondents are drawn randomly or by rotation, each subsample represents the foreign-born population and, if sufficiently large, can provide “reasonably precise” estimates of the percentages of the foreign-born population in the boxes on one of the alternative cards. Incidentally, a respondent picking a box that does not include the sensitive answer—for example, a respondent picking Box A or Box C in figure 1— can be asked follow-up questions that pinpoint the specific answer category that applies to him or her. Thus, direct information is obtained on all legal immigration statuses. The data on some of the legal categories can be compared to administrative data to check the reasonableness of responses. Additionally, these data provide estimates of legal statuses, which are useful when, for example, policymakers review legislation on the numbers of foreign-born persons who may be admitted to this country under specific legal status programs. 3. No individual respondent is ever categorized as undocumented, but indirect estimates of the undocumented population can be made. Using two slightly different pieces of information provided by the two different subsamples allows indirect estimation of the size of the currently undocumented population—by simple subtraction. The only difference between Box B of Card 1 and Box A of Card 2 is the inclusion of the currently “undocumented” category in Box B of Card 1. Figure 3 shows both cards together for easy comparison. Thus, the percentage of the foreign-born population who are currently undocumented can be estimated as follows: Start with the percentage of subsample 1 respondents who report that they are in Box B of Card 1 (hypothetical figure: 62 percent of subsample 1). Subtract from this the percentage of subsample 2 who say they are in Box A on Card 2 (hypothetical figure: 33 percent of subsample 2). Observe the difference (29 percent, based on the hypothetical figures); this represents an estimate of the percentage of the foreign-born population who are undocumented. Alternatively, a “mirror-image” estimate could be calculated, using Box B of Card 2 and Box A of Card 1. To estimate the numerical size of the undocumented population, a grouped answers estimate of the percentage of the foreign-born who are undocumented would be combined with a census figure. For example, the 2000 census counted 31 million foreign-born, and the Census Bureau issued an updated estimate of 35.7 million for 2005. The procedure would be to simply multiply the percent undocumented (based on the grouped answers data and the subtraction procedure) by a census count or an updated estimate for the year in question. These procedures ensure that no respondents—and no data on any specific respondent—are ever separated out or categorized as undocumented, not even during the analytic process of making indirect, group-level estimates. To further ensure reduction of “question threat,” the grouped answers question series begins with flash cards that ask about nonsensitive topics and familiarize respondents with the 3-box approach. For each nonsensitive-topic card, interviewers ask the respondent which box applies to him or her, saying: If it’s Box B, we do not want to know which specific category applies to you. In this way, most respondents should understand the grouped answers approach before seeing the immigration-status card. To help ensure accurate responses, respondents who choose Box A can be asked a series of clarifying questions. (No follow-up questions are addressed to anyone choosing Box B.) The questions for Box A respondents are designed to prompt them to, essentially, reclassify themselves in Box B, if that is appropriate. The grouped answers question series can potentially be applied in a large- scale general population survey, where the questions on immigration status would be added for the foreign-born respondents—provided that an appropriate survey can be identified. If a new survey of the general foreign-born population were planned, it would involve selecting a general sample of households and then screening out the households that do not include one or more foreign-born persons. Finally, we note that while the initial version of the grouped answers approach involved three alternative flash cards (and was termed the “three-card method”), we recently devised the version described here, which uses two cards rather than three. The two-card method is simpler, is easier to understand, and provides more precise estimates. All cards are alike in that they feature three boxes in which specific answer categories are grouped. Generally, grouped answers questions on immigration status would be asked as part of a larger survey that includes direct questions on demographic characteristics and employment and might include questions on school attendance, use of medical facilities, and so forth; some surveys also ask specific questions that can help estimate taxes paid. Potentially, combining the answers to such questions with grouped answers data can be used to provide further information on the characteristics, costs, and contributions of the undocumented population. For example, the numbers of undocumented persons in major subgroups —such as demographic or employment status subgroups—can be estimated, provided that the sample of foreign-born persons interviewed is sufficiently large. Grouped answers data collected from adult respondents can also be used to estimate the number of children in various immigration statuses, including undocumented—provided that an additional question is asked. Additionally, when combined with separate quantitative data (for example, data on program costs per individual), grouped answers data can be used to estimate quantitative information (such as program costs) for the undocumented population as a whole—or, again, depending on sample size, for specific subgroups. The procedures for deriving these more complex indirect estimates are described in appendix II. No grouped answers respondent is ever categorized as undocumented. The foreign-born population of the United States is large and growing— as is the undocumented population within it. Congressional policymakers, the U.S. Commission on Immigration Reform, and the National Research Council’s (NRC) Committee on National Statistics have indicated a need for statistical information on the undocumented population, including its size, characteristics, costs, and contributions. The Census Bureau estimates that as of 2005, foreign-born residents (both legally present and undocumented) numbered 35.7 million and accounted for at least one-tenth of all persons residing in each of 15 states and the District of Columbia. These figures represent substantial increases over the prior 15 years. For example, in 1990 the foreign-born population totaled fewer than 20 million; only 3 states had a population more than one-tenth foreign-born. One result is that as the Department of Labor has testified, foreign-born workers now constitute almost 15 percent of the U.S. labor force, and the numbers of such workers are growing. A new paper from the Department of Homeland Security (DHS) puts the “unauthorized” immigrant population at 10.5 million as of January 2005 and indicates that if recent trends continued, the figure for January 2006 would be 11 million. The Pew Hispanic Center’s indirect estimate of the undocumented population as of 2006 is 11.5 million to 12 million. These estimates represent roughly one-third of the entire foreign-born population. DHS has variously estimated the size of the undocumented population as of January 2000 as 7 million and 8.5 million . Government and other estimates for 1990 numbered only 3.5 million. These various indirect estimates of the undocumented population are based on the “residual method.” Residual estimation (1) starts with a census count or survey estimate of the number of foreign-born residents who have not become U.S. citizens and (2) subtracts out estimated numbers of legally present individuals in various categories, based on administrative data and assumptions (because censuses and surveys do not ask about legal status). The remainder, or residual, represents an indirect estimate of the size of the undocumented population. To illustrate the role of administrative data and assumptions, residual estimates draw on counts of the number of new green cards issued each year. But they also require assumptions to account for emigration and deaths among those who received green cards in earlier years. A recent DHS paper providing residual estimates of the undocumented population includes ranges of estimates based on alternative assumptions made for two key components. For example, “by lowering or raising the emigration rates 20 percent . . . the estimated unauthorized immigrant population would range from 10.0 million to 11.0 million.” The DHS paper also lists assumptions that were not subjected to alternative specifications. We believe the DHS paper represents an advance because, up to now, analysts producing residual estimates have generally not made public statements regarding the precision of the estimates. (Some critics have, however, indicated that residual estimates are likely to lack precision.) While the residual approach has been used to profile the undocumented population on two characteristics—age and country of birth—it is limited with respect to estimating (1) current geographic location and (2) current employment and benefit use. The reason is that current characteristics of legally present persons are not maintained in administrative records; analysts must therefore rely largely on assumptions . In contrast, the grouped answers method does allow for the possibility of estimating current characteristics based on current self-reports. During the mid-1990s, the U.S. Commission on Immigration Reform determined that better statistical “information on legal status and type of immigrant crucial” to assessing immigration policy. Indeed, the Commission called for a variety of improvements in estimates of the costs and benefits associated with undocumented immigration. NRC’s Committee on National Statistics further emphasized the need for better information on costs, especially state and local costs. (If successfully fielded, the grouped answers method might help provide general information on such costs—and, potentially, specific information for large states such as California. Sample size limitations would be likely to prohibit separate analyses for specific local areas, small states, and states with low percentages of foreign-born or undocumented.) Over the years, we have received numerous congressional requests related to estimating costs associated with the undocumented population. Recent Census Bureau research and conferences reflect the realization that undocumented immigration is a key component of current population growth and that there is a resultant need for information on this group. Additionally, some of the immigrant advocates we interviewed expressed an interest in being able to better describe the contributions of the undocumented population. Various national surveys ask foreign-born respondents to provide information about themselves and, in some cases, other persons in their households. While such surveys provide a wealth of information on a wide variety of areas, including some sensitive topics, national surveys generally do not ask about current immigration status—with the exception of a question on U.S. citizenship, which is included in several surveys. As we reported earlier, it is believed that direct questions on immigration status “are very sensitive, and negative reactions to them could affect the accuracy of responses to other questions on survey.” Two surveys that have asked respondents directly about immigration status for several years are the National Agricultural Workers Survey (NAWS), an ongoing annual cross-sectional self-report survey of farmworkers, fielded by Aguirre International, a private sector firm under contract to the Department of Labor, since 1988, and the Survey of Income and Program Participation (SIPP), a longitudinal panel survey of the general population, conducted by the Census Bureau, which has asked immigration status questions since 1996. Of the two, SIPP is the more relevant, because its immigration status questions have been administered to a sample of the general foreign-born population. SIPP has asked an adult respondent-informant from each household to provide information about himself or herself and about others in his or her household, including which immigration-status category applied to each person when he or she came to this country. Answers are facilitated by a flash card that lists major legal immigration statuses (see fig. 4). A further question asks whether each person obtained a green card after arriving in this country. The SIPP questions come close to asking about—but do not actually allow an estimate of—the number of foreign-born U.S. residents who are currently undocumented. According to the Census Bureau, SIPP is now scheduled to be “reengineered,” but the full outlines of the revised effort have not been set. In the middle to late 1990s, the grouped answers question series was subjected to preliminary development and testing with Hispanic respondents, including interviews with farmworkers conducted by Aguirre International, under contract to GAO. In these tests, every respondent picked a box. However, these interviews were not conducted under conditions of a typical large-scale survey in which interviewers initiate contact with respondents in their homes. To further test respondents’ acceptance of the grouped answers approach, the Census Bureau created a question module with 3-box flash cards and contracted for it to be added to the 2004 GSS. When presenting the survey to respondents, interviewers explained that NORC of the University of Chicago fielded the GSS survey, with “core funding” from an NSF grant. The Census Bureau’s question module included cards from the three-card version of the grouped answers approach—which features only one immigration status category in Box A. The cards used were the two training cards shown in figures 5 and 6 and the immigration status card shown in figure 7. the “overwhelming majority of foreign-born respondents” picked a box on the immigration status card without—according to interviewers—any objection, hesitation, or periods of silence; while some interviewers did not give a judgment or were confused about rating respondents’ understanding, about 80 percent of respondents were coded as understanding and about 10 percent as not; and some respondents’ comments, written in by interviewers, indicated that although the GSS is a “personal interview” survey, telephone interviews had been substituted, in some cases, and this meant that respondents could not see the cards—making the use of the 3-box format difficult. The Census Bureau’s paper highlighted various limitations of the 2004 GSS test, including (1) testing only one immigration status card, (2) underrepresenting Hispanics, and (3) in some instances interviewing over the telephone (instead of in person), so that respondents did not see the flash cards. The acceptability of the grouped answers approach appears to be high, when implemented in surveys fielded by a university or private sector organization. Many immigration experts, including advocates, accepted the grouped answers approach, although some conditioned their acceptance on a quality implementation in a survey fielded by a university or other private sector organization. An independent statistical expert believed that the grouped answers approach would be generally usable with survey respondents. Some of the researchers and advocates we contacted were extremely enthusiastic about the potential for new data. No one objected to statistical, policy-relevant information being developed on the size, characteristics, costs, and contributions of the undocumented population. Overall, the immigration experts we contacted (listed in appendix I, table 5) accepted the grouped-answers question approach—although advocates sometimes conditioned their acceptance on, for example, the questions being asked in a survey fielded by a university or private sector organization—with data protections built in. Many also offered suggestions for maximizing cooperation by foreign-born respondents or ideas about how advocacy organizations might help. Some advocates indicated that a key condition of their support would be that (1) the grouped answers question on immigration status be asked by a university or private sector organization and (2) identifiable data (that is, respondents’ answers linked to personal identifiers) be maintained by that organization. Two advocate organizations specifically stated that they “could not endorse,” or implied they would not support, the grouped answers approach, assuming the data were collected and maintained by, in one case, the Census Bureau and, in the other case, the government. Many other immigration experts and advocates preferred that grouped answers data on immigration status be collected by a university or other reputable private sector organization pledged to protect the data. The immigration advocates said that private sector fielding of a grouped answers survey and protection of such data from nonstatistical uses that might harm immigrants were key issues because Some foreign-born persons are from countries with repressive regimes and thus have more fear of (less trust in) government than the typical U.S.-born person. Despite current law protecting individual data from disclosure, some persons believe that information collected by a government agency such as the Census Bureau is routinely shared (or that in some circumstances it might be shared) across government agencies. Further, one advocate pointed out that the Congress could change the current law, eliminating that protection. (Although the grouped answers approach does not identify anyone as undocumented, it does provide some information regarding each respondent’s immigration status.) Extremely large-scale data collections—notably, the American Community Survey (ACS)—can yield estimates for areas small enough that if the data were publicly available, they could be used for nonstatistical, nonpolicy purposes. Some advocates referred to the World War II use of census data to identify the areas where specific numbers of persons of Japanese origin or descent resided. They also pointed out that Census Bureau data on ethnicity— including counts of Arab Americans—are publicly available by zip code. (The Census Bureau, unlike other government agencies and private sector survey organizations, is associated with extremely large-scale data collections, and some persons may not fully differentiate Census Bureau data collection efforts of different sizes.) Hostility to or lack of trust in the Census Bureau might result in potentially lower response rates for foreign-born persons, based on the World War II experience of the Japanese or a more recent incident in which Census Bureau staff helped a DHS enforcement unit access publicly available data on ethnicity by zip code. 51, DHS stated that it did not use these data and had not requested the information by zip code. The Census Bureau clarified its position on providing help to others requesting publicly available data. Various advocates saw the issues listed above as linked to their own acceptance, as well as to respondent acceptance, of a survey. Linking these issues to respondent acceptance of a survey was, in some cases, echoed by other immigration experts we consulted. Some immigrant advocates and other immigration experts counseled us that if there were an increase in enforcement efforts in the interior of the United States (as opposed to border-crossing areas), foreign-born respondents’ acceptance of the grouped answers questions would be likely to decrease—at least, if the questions were asked in a survey fielded by the government. See Samia El-Badry and David A. Swanson, “Providing Special Census Tabulations to Government Security Agencies in the United States: The Case of Arab-Americans,” paper presented at the 25th International Population Conference of the International Union for the Scientific Study of Population, Tours, France, July 18–23, 2005. One advocate was particularly concerned about the possibility that lower respondent cooperation might have resulted from these incidents and, if so, might have led to underrepresentation of these communities in Census Bureau data. Additionally, one advocate questioned whether local estimates of the undocumented might, in future, facilitate possible efforts to base apportionment on population counts that do not include undocumented residents. We note that most large-scale personal-interview surveys do not include sufficient numbers of foreign-born respondents to allow indirect grouped answers estimates of undocumented persons for small geographic areas, such as zip codes. sector organization with government funding. In some cases, we specifically referred to one or both of the following surveys, which (1) have been conducted for many years without inappropriate data disclosures and (2) ask direct sensitive questions: the National Survey on Drug Use and Health (NSDUH), fielded by RTI International under a contract from HHS’s Substance Abuse and Mental Health Services Administration (SAMHSA), and the National Agricultural Workers Survey (NAWS), fielded by Aguirre International, under a contract from the Department of Labor. The advocates’ response was generally to accept the concept of government funding of a university’s or private sector survey organization’s field work, provided that appropriate protections of the data were built into the funding agreement. GAO’s contract with Aguirre International for early testing of the grouped answers approach with farmworker respondents specified that data on respondents’ answers would be “stripped of person-identifiers and related information.” Additionally, the GSS “core funding” grant with NSF and its contractual arrangements with sponsors of question modules—such as the grouped-answers question insert contracted for by the Census Bureau— do not involve the transfer of any data other than publicly available data, stripped of identifiers, and limited so as to avoid the possibility of “deductive disclosure” with respect to respondent identities or local areas. Various advocates said that their acceptance was also contingent on factors such as 1. high-quality data, including coverage of persons who have limited English proficiency, with special attempts to reach those who are linguistically isolated (that is, members of households in which no one 14 or older speaks English “very well”) and to overcome other potential barriers (such as cultural differences); 2. appropriate presentation of the survey, including an appropriate explanation of its purpose and how respondents were selected for interview; and 3. transparency—that is, keeping the immigrant community informed about or involved in the development and progress of the survey. One advocate specifically said that her organization’s support would be contingent on both (1) the development of more information on respondent acceptance within the Asian community—particularly among Asians who have limited English proficiency or are linguistically isolated— and (2) a survey implementation that is planned to adequately communicate with Asian respondents, including those who are linguistically isolated or have little education. Although one-fourth of the 2004 GSS test respondents were Asian, the test was conducted in English (allowing help from bilingual household members), and no other tests have included linguistically isolated Asians. Advocates and other experts made several suggestions for maximizing respondent cooperation with a survey using the grouped answers question series—that is, maximizing response rates for such a survey as well as maximizing authentic participation. Advocates suggested that the survey (1) avoid taking names or Social Security numbers, (2) hire interviewers who speak the respondents’ home-country language, (3) let respondents know why the questions are being asked and how their households came to be selected, (4) conduct public relations efforts, (5) obtain the support of opinion leaders, (6) select a survey group from a well-known and trusted university to collect the data, and (7) ask respondents about their contributions to the American economy through, for example, working and paying taxes. Additionally, survey experts suggested using audio–Computer Assisted Self Interview (audio-CASI), carefully explaining to respondents how anonymity of response is paying respondents $25 or $30 for participating in the interview. Survey experts viewed these elements as key ways of boosting response rates or encouraging authentic responses to sensitive questions. For example, NAWS, which uses respondent incentives, achieves extremely high response rates within cooperating farms—97 percent in 2002, with a $20 payment to farmworkers selected. Some immigrant advocates also offered suggestions for how their organizations or other advocates might help the effort to develop and field the grouped answers approach, including 1. providing contacts at local organizations to help with arrangements for 2. developing or reviewing Box A follow-up questions, and 3. serving on an advisory board with other representatives from immigrant communities. As we report above, the Census Bureau’s recent analysis of the 2004 GSS grouped answers data concluded that the “overwhelming majority of foreign-born respondents” picked a box without objection, hesitation, or silence. The Census Bureau reported, more specifically, that roughly 90 percent (216 of 237 respondents) chose a box, 4 gave other answers, and 17 refused to answer or said “don’t know.” Our subsequent analysis excluded 19 of the 237 respondents in the Census Bureau analysis because 4 were not foreign-born (for example, 1 had been born abroad to parents who had, by the time he was born, become naturalized U.S. citizens); 1 was not classifiable as either foreign-born or not foreign-born (because he did not know whether his parents were born in the United States); 4 others were known to have been interviewed on the telephone, based on written-in interviewers’ comments recorded in the computer file (for example, one wrote that the respondent could not see the cards because the interview was on the telephone); and 10 others were subsequently found to have been interviewed on the telephone, based on a special GSS hand check of the interview forms for respondents who had refused or said “don’t know,” which was carried out in response to our request. As a result, in our analysis we found that only 6 personally interviewed foreign-born GSS respondents refused or said “don’t know.” One of the 6 was an 18-year-old Mexican who told the interviewer that he did not know whether or not he was a legal immigrant. Additionally, we found that the 4 respondents who gave “other answers” had provided usable information (for example, one called out that he had a student visa) and thus could be recoded into an appropriate box. The test confirms the general usability of the with subjects similar to the target population for its potential large-scale use—that is, foreign-born members of the general population. Out of about 218 respondents meeting eligibility criteria and who were most likely administered the cards in person (possibly including a few who had telephone interviews but responded without problems), only 9 did not respond by checking one of the 3 boxes. Of these, 3 provided verbal information that allowed coding of a box, and 6 declined to answer the question altogether. Furthermore, several of these raised similar difficulties with other 3-box questions on nonsensitive topics (type of house where born, mode of transportation to enter United States), suggesting that the difficulties with the question format were at least in part related to the format and not to the particular content of the answers. Thus, indications were that there would not be a systematic bias due to respondents whose immigration status is more sensitive being unwilling to address the 3-box format. Dr. Zaslavsky emphasized the importance of minimizing or completely avoiding telephone interviews when using the grouped answers approach—or, alternatively, providing advance copies of the cards to respondents before interviewing over the telephone. (Dr. Zaslavsky’s written review is presented in full in appendix III.) The findings on respondent acceptance—that is, the GSS test—raised some unanswered questions about acceptance that experts said should be addressed. Additionally, the experts said that one or more tests of response validity are needed to determine whether respondents “pick the correct box” versus systematically avoiding Box B. four issues should be addressed in future field tests: (a) Equivalent acceptability of all forms of the response card, (b) Usability with special populations including those with low literacy, the linguistically isolated, and concentrated immigrant populations, (c) Methods that avoid telephone interviews, or reduce bias and nonresponse due to use of the telephone, (d) Use of follow-up questions to improve the accuracy of box choices. As the independent expert explained with respect to point (b), GSS undercoverage of the foreign-born population occurred at least in part because interviews were conducted only in English, although household members could help respondents with limited English. Various colleagues and experts we talked with supported points (a) through (d). We further note that points (a) and (c) were covered or touched on in the Census Bureau’s paper reporting its analysis of the 2004 GSS data. In our discussions with Census Bureau staff, they also mentioned that further tests of acceptance should include (d) follow-up questions for Box A respondents. Additionally, some advocates and an immigration researcher suggested improving the cards, which might minimize the potential for “don’t know” or inaccurate answers. A survey expert suggested using focus groups to further explore respondent perceptions of the cards—and to potentially improve them. Earlier testing covered a key portion of the populations (Hispanic farmworkers) cited in (b) above, was conducted in Spanish, and included Box A follow-up questions as recommended in (d) above. In those interviews, every respondent picked a box. However, 1. No language other than Spanish or English has been used in testing; thus, as one immigrant advocate pointed out, no testing has focused on linguistically isolated Asians (those living in households in which no adult member speaks English). 2. The interviews with Hispanic farmworkers were not conducted under typical conditions of a household survey. 3. Only one immigration status card was tested with Hispanic farmworkers and in the GSS. Therefore, we agree that the acceptance-testing issues the experts raised should be considered in assessing the grouped answers approach. Several experts told us that tests of respondent accuracy—or at least respondents’ intent to respond accurately—should be conducted. These experts emphasized that grouped answers data would not be useful if substantial numbers of respondents were to systematically avoid picking Box B (that is, to not pick the box with the undocumented category). However, one immigration study expert believed that if a response validity study involved lengthy delays, fielding a grouped answers survey should proceed in advance of a validity study. We agree with the experts’ position that tests are needed to determine whether respondents systematically avoid Box B (even after Box A follow- up check questions). Tests of response validity would ideally be conducted with the methods of encouraging truthful answers that experts mentioned, such as (1) explaining why the survey is being conducted, how the respondent was selected, and how the anonymity of answers is ensured, and (2) using audio-CASI and, if appropriate, paying respondents for participating in the interview. And, as the Census Bureau pointed out, such a study should include the full grouped answers question series, including follow-up questions, and it should test both Card 1 and Card 2. Even if small numbers of respondents were to respond inaccurately, it would be helpful to estimate this and adjust for any resulting bias. We discussed various approaches to conducting validity studies with immigration experts, including immigrant advocates, and with agencies conducting surveys. In reviewing these approaches, we found that response validity tests vary according to whether they are conducted before, during, or after a survey is fielded. Before a large-scale survey is conducted. The grouped answers question series could be asked of a special sample of respondents for whom the answers are known, in advance, by study investigators on an individual-respondent basis. Such knowledge might be based, for example, on information that recent applicants for green cards have submitted to DHS. “Firewalls” could be used to prevent survey information from being given to DHS. We discussed this approach with DHS; however, experts criticized a DHS-based validity study on both methodological and public relations grounds. An alternative source of data on individuals’ immigration statuses might avoid these problems, but no alternative source has yet been identified. Before or as part of a large-scale survey. In either situation (that is, in a presurvey study or as part of a survey), respondents could be asked if they would be willing to participate in special validity-test activities in return for a payment of, say, $25 or $30 for each activity. Later, after interviewing had been completed in a given location—not as part of the interview process—a sample of respondents who chose Box A (that is, those who claimed to be here legally) could be asked to participate in a focus group in which respondents would discuss how they felt answering the grouped answers questions when the interviewer came to their house and, also, could possibly be asked to fill out a “secret ballot” indicating whether they had answered authentically in the earlier home interview; give permission for a record check and provide information that could subsequently be used in a record check (for example, their name, date of birth, and Social Security number) and permission to check these data with the Social Security Administration; or show his or her documentation (for example, green card) to a documents expert. These checks would logically be focused on Box A respondents, for most of whom such checks would be less threatening. We believe that it is reasonable to assume that most respondents who chose Box B picked the correct box. Further, because the survey interview states that there are no more questions on immigration if the respondent picks Box B, pursuing follow-up validity checks might be deemed inappropriate for Box B respondents. After data are collected. With a large-scale survey, it would be possible to conduct comparative analyses after the data were collected. We provide three examples.1. Grouped answers estimates of the percentage undocumented could be compared for (a) all foreign-born versus (b) high-risk groups, such as those who arrived in the United States within the past 5 or 10 years. The expectation would be that with valid responses, a higher estimate of the percentage undocumented would be obtained for those who arrived more recently—because, for example, persons who had arrived recently were not here during the amnesty in the late 1980s. 2. Comparisons could be made of (a) Box A estimates of specific legal statuses and the approximate dates received—notably, the numbers of persons claiming to have received valid green cards in 1990 or more recently—with (b) publicly available DHS reports of the numbers of green cards issued from 1990 to the survey date. 3. Analysts could compare (a) grouped answers estimates of the number undocumented overall to (b) estimates of total undocumented obtained by the residual method. Wherever possible, Card 1 and Card 2 should be tested separately for accuracy of response. The advantage of conducting a validity study in advance of a survey is that if significant problems surface, adjustments in the approach can be made. Or if the problems are substantial and cannot be easily corrected—and if the anticipated survey were to be fielded mostly or only to collect grouped answers data—then that survey could be postponed or canceled. However, the results of validity tests conducted during or after a survey can be used to interpret the data and, potentially, to adjust estimates if it appears that, for example, 5 to 10 percent of undocumented respondents had erroneously claimed to be in Box A of Card 1. As one expert noted, conducting an advance study does not preclude conducting a subsequent study during or after the survey. Although several factors are involved, and it is not possible to guarantee a specific level of precision in advance, we estimate that roughly 6,000 foreign-born respondents, or more, would be needed for a grouped answers survey. As we explain below, this is based on (1) a precision requirement (that is, a 95 percent confidence interval consisting of plus or minus 3 percentage points), (2) assumptions about the sampling design of the survey in which the questions are asked, and (3) the assumption that approximately 30 percent of the foreign-born population is currently undocumented. An indirect grouped answers estimate of the undocumented population generally requires interviews with more foreign-born respondents than a corresponding hypothetical direct estimate would—assuming it were possible to ask such questions directly in a major national survey. One key reason is that the main sample of foreign-born respondents must be divided into two subsamples. Half the respondents answer each immigration status card. On this basis alone, one would have to double the sample size required for a direct estimate based on a question asked of all respondents. Further, the estimate of undocumented, which is achieved by subtraction, combines two separate estimates, each characterized by some degree of uncertainty. Determining the number of respondents required for a “reasonably precise” estimate of the percentage of the foreign-born population who are undocumented involves three key factors: 1. specification of a precision level—that is, choice of a 90 percent or 95 percent confidence level and an interval defined by plus or minus 2, 3, or 4 percentage points; 2. information on (or assumptions about) the sampling design for the main survey and for subsamples 1 and 2; and 3. to the extent possible, consideration of the likely distribution of the foreign-born population across immigration status categories, including the various legal categories and the undocumented category. With respect to the first factor involved in determining sample size, some agencies—for example, the Census Bureau and the Bureau of Labor Statistics (BLS)—use the 90 percent confidence level. Other agencies use the 95 percent level. With respect to the second factor, the sampling design of a large-scale, nationally representative, personal-interview survey is based on probabilistic area sampling rather than simple random sampling of individuals. This often reduces the precision of estimates (relative to simple random sampling). The reason is that persons selected for interview are clustered in a limited number of areas or neighborhoods (and residents of a particular neighborhood may tend to be similar). It is possible that the design for selecting subsamples 1 and 2 could increase precision; however, it is not possible to predict by how much. With respect to the third factor, existing residual estimates point to a fairly even 3-way split between three main categories—undocumented, U.S. citizen, and legal permanent resident. However, there is some uncertainty associated with these estimates, the distribution may vary across subgroups, and the percentages may change in future. Therefore, a range of distributions is relevant. Taking each of these factors into account (to the extent possible) and using conservative assumptions, we estimated the approximate numbers of respondents required for indirect estimates of the undocumented population that are “reasonably precise.” Table 1 shows required sample sizes for the 90 percent confidence level, table 2 for the 95 percent level, with precision at plus or minus 2, 3, and 4 percentage points. In estimating these required sample sizes, we made conservative assumptions and specified a range of possibilities for the distribution with respect to the undocumented category. To identify a single, rough figure for the sample size needed for reasonably precise estimates, we focused on 1. the 95 percent level, which is more certain and, we believe, preferable; 2. the 30 percent column, because a current residual estimate of the undocumented population is in this range; and 3. the middle row (for plus or minus 3 percentage points), which is a midpoint within the area of “reasonable precision” as defined above. With this focus, we estimate that roughly 6,000 or more respondents would be required. High-risk subgroups—subgroups with higher percentages of undocumented (such as adults 18 to 44 and persons who arrived in the United States within the past 10 years)—would require fewer respondents for the same level of precision, as illustrated in the tables’ middle and right columns. For example, if about 70 percent of a subgroup were undocumented, a survey with about 3,500 respondents in that subgroup would produce an estimate of the percentage of the subgroup that is undocumented, correct to within approximately plus or minus 3 percentage points at the 95 percent confidence level. Low precision could obtain for smaller subgroups in which there are relatively few undocumented persons (for example, 10 percent or less), particularly if—as assumed in tables 1 and 2—there is an even split of legally present foreign-born persons across the Box A categories of immigration status cards 1 and 2. The independent statistician we consulted indicated that if more than one grouped answers survey is conducted, combining data across two or more surveys could help provide larger numbers of respondents for subgroup analysis. For example, if a large-scale survey were conducted annually, analysts could combine 2 or 3 years of data to obtain more precise estimates. (One caveat is that combining data from multiple survey years reduces the time-specificity associated with the resulting estimate.) Finally, we note that to estimate the numerical size of the undocumented population, A grouped answers estimate of the percentage of the foreign-born who are undocumented would be combined with a census count of the foreign-born or an updated estimate. For example, the 2000 census counted 31 million foreign-born persons, and the Census Bureau later issued an updated estimate of 35.7 million for 2005. The specific procedure would be to multiply the percentage undocumented (based on the grouped answers data and the subtraction procedure) by a census count or an updated estimate of the foreign-born population for the year in question. The precision of the resulting estimate of the numerical size of the undocumented population would be affected by (1) the precision of the grouped answers percentage estimate, which is closely related to sample size, as described above, and (2) any bias in the census count or updated estimate of the foreign-born population. The precision of the grouped answers percentage is taken into account by using a percentage range (for example, the estimate plus or minus 3 percentage points) when multiplying. Although the amount of bias in a census count or updated estimate is unknown, we believe that any such bias would have a proportional impact on the calculated numerical estimate of the undocumented population. To illustrate the proportional impact, we assume that a census count for total foreign-born is 5 percent too low. Using that count in the multiplication process would cause the resulting estimate of the size of the undocumented population to be 5 percent lower than it should be. The situation is analogous for subgroups. Overall, it seems clear that reasonably precise grouped answers estimates of the undocumented population and its characteristics require large-scale data collection efforts but not impossibly large ones. A low-cost field strategy would be to insert the new question series in an existing, nationally representative, large-scale survey—that is, to pose the grouped answers questions to the foreign-born respondents already being interviewed. However, based on our review of on-going large-scale surveys, the insertion strategy does not seem feasible. Specifically, we identified four potentially relevant surveys but none met criteria based on the grouped answers design and other criteria based on immigrant advocates’ concerns. The dollar costs associated with inserting a grouped answers module are difficult to calculate in advance because many factors are involved. However, to suggest the “ball park” within which the cost of a grouped answers insert might be categorized, if an insertion were possible, we present the following two examples. The GSS test, in which a grouped answers question module was inserted, cost approximately $100 per interview (more than 200 interviews were conducted). On average, the question series took 3.25 minutes. Logically, per-interview costs are likely to be higher in relatively small surveys than in larger surveys with thousands of foreign-born respondents. For the much larger Current Population Survey (CPS), with interviews covering native-born and foreign-born persons in more than 50,000 households, the Census Bureau and BLS told us that “an average 10-minute supplement cost $500,000 in 2005.” This implies $10 per interview at the 50,000 level, but per-interview costs might be higher when the question series applied to only a portion of the respondents. Additional costs might apply for flash cards and foreign-language interviews. BLS noted that still other costs would apply for advance testing and subsequent analyses requested by the customer. A more costly option would be to ask the grouped answers question series in a follow-back survey of foreign-born respondents identified in interviewing for an existing survey. (In-person self-report interviews can cost $400 to $600 each.) More costly still would be the development of a new, personal-interview survey of a representative sample of the foreign- born population devoted to migration issues; the main reason is that there would be additional costs in “screening out” households without foreign- born persons. We identified four potentially relevant ongoing large-scale surveys. All have prerequisites and processes for accepting (or not accepting) new questions. We also developed six criteria for assessing the appropriateness of each survey as a potential vehicle for fielding the grouped answers approach. Three criteria are based on design requirements, and three are based on the views of immigrant advocates. We found that no ongoing large-scale survey met all criteria. We identified four nationally representative, ongoing large-scale surveys in which respondents are or could be personally interviewed. Three of these conduct most or all interviews in person: 1. the Current Population Survey (CPS), sponsored by BLS and the Census Bureau and fielded by Census; 2. the National Health Interview Survey (NHIS), sponsored by the National Center for Health Statistics (NCHS) and fielded by the Census Bureau; and 3. the National Survey on Drug Use and Health (NSDUH), sponsored by SAMHSA and fielded by RTI International, a private sector contractor. The fourth survey is the American Community Survey (ACS), a much larger survey fielded by the Census Bureau and using “mixed mode” data collection. The majority of the data are based on mailed questionnaires or telephone interviews, with the remaining data based on personal interviews. In addition, there is one personal-interview follow-back survey that uses the ACS frame and data to draw its sample. Other follow-back surveys might eventually be possible. For any of these four surveys, inserting a new question or set of questions (or fielding a “follow-back” survey based on respondents’ answers in the main survey) requires approvals by the Office of Management and Budget (OMB), the agencies that sponsor or field the surveys, and in cases in which data are collected by a private sector organization, the organization’s institutional review board. The prerequisites for an ongoing survey’s accepting new questions typically include low anticipated item nonresponse, pretesting and pilot testing (including debriefing of respondents and interviewers) that indicate a minimum of problems, review by stakeholders to determine acceptability, and tests that indicate no effect on either survey response rates or answers to the main survey’s existing questions. Another prerequisite would be the expectation of response validity. Additionally, multiple agencies mentioned a need for prior “cognitive interviewing,” compatibility with existing items (so that there is no need to change existing items), and no significant increase in “respondent burden” (by, for example, substantially lengthening the interview). Agencies sponsoring or conducting large-scale surveys varied on the perceived relevance of immigration to the main topic of their survey. For example, BLS noted that some of its customers would be interested in data on immigration status by employment status (among the foreign-born), and the Census Bureau has indicated the relevance of undocumented immigration to population estimation. But some other agencies saw little relevance to the large-scale surveys they sponsored or conducted. Resistance to including a grouped answers question series might occur where an agency perceives little or no benefit to its survey or its customers. Additionally, one agency raised the issue of informed consent, which we discuss in appendix V. Based on the design of the grouped answers approach, as tested to date, two criteria for an appropriate survey are (1) personal interviews in which respondents can view the 3-box cards and (2) a self-report format in which questions ask the respondents about their own status (rather than asking one adult member of a household to report information on others). A third criterion is that the host survey not include highly sensitive direct questions that could affect foreign-born respondents’ acceptance of the grouped answers questions. We based these criteria on the results of the GSS test, our knowledge of the grouped answers approach, and general logic. As shown in table 3, one of the surveys we reviewed (the CPS) does not meet the self-report criterion; that is, it accepts proxy responses. Two other surveys (the NHIS and NSDUH) do not meet the criterion of an absence of highly sensitive questions, since they include questions on HIV status (NHIS) and the use of illegal drugs (NSDUH). Conducting a follow- back survey based on ACS would meet all three criteria.1. Are the data gathered in personal interviews? YES. Mostly, for in-person waves; 16% of foreign-born interviewed by telephone, in the in-person waves. 3. Are direct questions not highly sensitive? YES, not highly sensitive. YES. Mostly; 17% of foreign-born sample adults interviewed by telephone. No. There are direct questions on HIV, other STDs. National Survey of Drug Use and Health (NSDUH) YES. All interviewed in person. No. An adult respondent reports on self and provides proxy responses for others in his or her household. In- person data for 6,744 households with 1 or more foreign-born members (2006). YES. For some questions, but not all, 4,829 foreign- born adults self-report (2004). YES. 7,364 foreign-born age 12 and older and 4,934 foreign-born age 18+ self- report (2004). YES. A follow-back could specify self-report only. (ACS data include both self- report data and proxy data in which one member of a household provides responses for others.) No. There are direct questions on respondent’s use and sale of drugs like marijuana and cocaine. YES, not highly sensitive. YES. A follow-back could specify personal interviews only. (ACS is mixed mode, mostly mail.) The views of immigrant advocates, which were echoed by some other experts, suggested three additional criteria for a candidate “host” survey: 1. data collection by a university or private sector organization, 2. no request for the respondent’s name or Social Security number, and 3. protection from possible release of grouped answers survey data for small geographic areas (to guard against estimates of the undocumented for such areas). The experts based their views on (1) methodological grounds (foreign- born respondents would be more likely to cooperate, and to respond truthfully, if all or some of these criteria were met) and (2) concerns about privacy protections at the individual or group levels. These criteria are potentially important, in part because the success of a self-report approach hinges on the cooperation of individual immigrants and, most likely, also on the support of opinion leaders in immigrant communities. With respect to the first criterion above, we note that with the exception of initial GAO pretests, all tests of the grouped answers approach have involved data collection by a university or private sector organization. Without further tests, we do not know whether acceptance would be equally high in a government-fielded survey. As shown in table 4, an ACS follow-back would potentially not meet any of the three criteria based on immigrant advocates’ views. Only one survey (NSDUH) met all three criteria based on immigrant advocates’ views—and because of its sensitive questions on drug use, that survey did not meet the design-based table 3 criteria. 1. Does a nongovernment organization conduct field work? 2. Are interviews anonymous (that is, no names or Social Security numbers are taken)? 3. Is sample too small for reliable small-area estimates of undocumented?YES. No. No. The Census Bureau conducts field work.No. The Census Bureau conducts field work.YES. No. Takes names. No. Takes both names and Social Security numbers. YES. YES. YES. YES. No. Only the Census Bureau can conduct field work. No. Takes names in the initial survey, and a follow-back would be based on knowing each person’s identity. Potentially, no. A follow- back might be extremely large. (Also, small-area releases are not prohibited by law or policy.) In conclusion, we did not find a large-scale survey that would be an appropriate vehicle for “piggybacking” the grouped answers question series. For more than a decade, the Congress has recognized the need to obtain reliable information on the immigration status of foreign-born persons living in the United States—particularly, information on the undocumented population—to inform decisions about changing immigration law and policy, evaluate such changes and their effects, and administer relevant federal programs. Until now, reliable data on the undocumented population have seemed impossible to collect. Because of the “question threat” associated with directly asking about immigration status, the conventional wisdom was that foreign-born respondents in a large-scale national survey would not accept such questions—or would not answer them authentically. Using the grouped answers approach to ask about immigration status seems promising because it reduces question threat and is statistically logical. Additionally, this report has established that The grouped answers approach is acceptable to most foreign-born respondents tested (thus far) in surveys fielded by private sector organizations; it is also acceptable—with some conditions, such as private sector fielding of the survey—to the immigrant advocates and other experts we consulted. A variety of research designs are available to help check whether respondents choose (or intend to choose) the correct box. The grouped answers approach requires a fairly large number of personal interviews with foreign-born persons (we estimate 6,000) to achieve reasonably precise indirect estimates of the undocumented population overall and within high-risk subgroups. However, the most cost-efficient method of fielding a grouped answers question series—piggybacking on an existing survey—does not seem feasible. Rather, fielding the grouped answers approach would require a new survey focused on the foreign-born. This raises two new questions about “next steps”—and the answers depend, in large part, on policymaker judgments, as described below. Question 1: Are the costs of a new survey justified by information needs? DHS stated (in its comments on a draft of this report) that the “information on immigration status and the characteristics of those immigrants potentially available through this method would be useful for evaluating immigration programs and policies.” The Census Bureau has indicated that information on the undocumented would help estimate the total population in intercensal years. And an expert reviewer emphasized that a new survey of the foreign-born would be likely to help estimate the total population. Additionally, policymakers might deem a new survey of the foreign-born to be desirable for other reasons than obtaining grouped answers data. Notably, an immigration expert who reviewed a draft of this report pointed out that a survey focused on the foreign-born might provide more in-depth, higher-quality data on that population than existing surveys that cover both the U.S.-born and foreign born populations. For example, more general surveys, such as the ACS and CPS (1) ask a more limited set of migration questions than is possible in a survey focused on the foreign- born, (2) are not designed with a primary goal of maximizing participation by the foreign-born (for example, are not conducted by private sector organizations), and (3) as DHS pointed out in comments on a draft of this report, may not be designed to cover persons who are only temporarily linked to sampled households, because such persons may have arrived only recently in the United States and are temporarily staying with relatives. A new survey aimed at obtaining grouped answers data on immigration status would require roughly 6,000 (or more) personal, self-report interviews with foreign-born adults. Other in-person, self-report interviews in large-scale surveys have cost $400 to $600 each. A major additional cost would be obtaining a representative sample of foreign-born persons; this would likely require a much larger survey of the general population in which “mini-interviews” would screen for households with one or more foreign-born individuals. We did not study the likely costs of such a data collection or options for reducing costs. However, survey costs can be estimated (based on, for example, the experience of survey organizations), and policymakers can, in future, weigh those costs against the information need—keeping in mind the results of research on the grouped answers approach, to date, and experts’ opinions on research needed. Question 2: What further tests of the grouped answers method, if any, should be conducted before planning and fielding a new survey? On one hand, advance testing could assess response validity (that is, whether respondents pick—or intend to pick—the correct box) before committing funds for a survey and in time to allow adjustments to the question series; further delineate respondent acceptance and explore the impact on acceptance of factors such as government funding—or funding by a particular agency—in order to inform decisions about whether or how to conduct a survey; and as suggested in DHS’s comments on a draft of this report, help determine the cost of a full-scale survey. On the other hand, extensive advance testing would likely delay the survey--and may not be needed because response validity could be assessed—and respondent acceptance could be further delineated—concurrently with or subsequent to the survey rather than in advance, the need for advance testing of response validity would be lessened if policymakers see a need for more or better survey data on the foreign- born additional to the need for grouped answers data on immigration status (see discussion in question 1, above); the value of advance testing would be lessened if changes in immigration law and policy occurred between the time of an advance test and the main survey, because such changes could affect the context in which the survey questions are asked and, hence, change the operant levels of acceptance and validity; and survey costs can be estimated—albeit more roughly—on the basis of the experience of survey organizations. Given the arguments for and against advance testing, it seems appropriate for these to be weighed by policymakers. We provided a draft of this report to and received comments from the Department of Commerce, the Department of Homeland Security, and the Department of Health and Human Services (see appendices VII, VIII, and IX, respectively). The Office of Management and Budget provided only technical comments, and the Department of Labor did not comment. The Census Bureau agreed with the report’s discussion of the grouped answers method, including its strengths and limitations; the Census Bureau-GSS evaluation, including the conclusions of the independent consultant (Alan Zaslavsky); and the need for a “validity study” to determine whether the grouped answers method can “generate accurate estimates” of the undocumented population. The Census Bureau also provided technical comments, which we used to clarify the report, as appropriate. The Department of Homeland Security stated that the kinds of information that the grouped answers approach would provide, if successfully implemented, would be useful for evaluating immigration programs and policies. DHS further called for pilot testing by GAO to assess the reliability of data collection and to help estimate the costs of an eventual survey. As we indicate in the “observations” section of this report, two key decisions for policymakers concern whether to invest in a new survey and whether substantial testing is required in advance of planning and fielding a survey. We believe that depending on the answers to these questions, another issue—one we cannot address in this report—would concern identifying the most appropriate agency for conducting or overseeing (1) tests of the grouped answers and (2) an eventual survey of the foreign-born population. However, we believe that conducting or overseeing such tests or surveys is a management responsibility and, accordingly, is not consistent with GAO’s role or authorities. DHS made other technical comments which we incorporated in the report where appropriate. The Department of Health and Human Services (HHS) agreed that the NSDUH would not be an appropriate vehicle for a grouped answers question series. Commenting on a draft of this report, HHS said that the report should include more information on variance calculations and on “mirror-image” estimates. Therefore, we (1) added a footnote illustrating the variance costs of a grouped answers estimate relative to a corresponding direct estimate and (2) developed appendix VI, which gives the formula for calculating the variance of a grouped answers estimate and discusses “mirror image” estimates. Additionally, HHS said that interviewers should more accurately communicate with respondents when presenting the three-box cards. We believe that the text of appendix V on informed consent, based on our earlier discussions with privacy experts at the Census Bureau, deals with this issue appropriately. As we state in appendix V, it would be possible to explain to respondents that “there will be other interviews in which other respondents will be asked about some of the Box B categories or statuses.” Finally, HHS made other, technical comments, which we incorporated in the report, as appropriate. The Office of Management and Budget provided technical comments. In addition, our discussions with OMB prompted us to re-order some of the points in the “observations” section of the report. The Department of Labor informed us that it had no substantive or technical comments on the draft of the report. We are sending copies of this report to the Director of the Census Bureau, Secretary of Homeland Security, Secretary of Health and Human Services, Secretary of Labor, Director of the Office of Management and Budget, and to others who are interested. We will also provide copies to others on request. In addition, the report will be available at no charge on GAO’s Web site at http://www.gao.gov. If you or your staff have any questions regarding this report, please call me at (202) 512-2700. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Other key contributors to this assignment were Judith A. Droitcour, Assistant Director, Eric M. Larson, and Penny Pickett. Statistical support was provided by Sid Schwartz, Mark Ramage, and Anna Maria Ortiz. To gain insight into the acceptability of the grouped answers approach, we discussed the approach with numerous experts in immigration studies and immigration issues, including immigrant advocates. Table 5 lists the experts we met with and their organizations. To ensure that we identified immigration experts from varied perspectives, we consulted Demetrios G. Papademetriou, who is among the immigration experts listed in table 5, and Michael S. Teitelbaum, Vice President of the Alfred J. Sloan Foundation. With respect to immigrant advocates, we sought to include advocates who represented (1) immigrants in general, without respect to ethnicity; (2) Hispanic immigrants, as these are the largest group of foreign-born residents; (3) Asian American immigrants, as these are also a large group; and (4) Arab American immigrants, as these have been the target of interior (that is, nonborder) enforcement efforts in recent years. To determine what the 2004 General Social Survey (GSS) test indicated about the acceptability of grouped answers questions to foreign-born respondents and its “generally usability” in large-scale surveys, we obtained the Census Bureau’s report of its analysis of those data, and we assessed the reliability of the GSS data through a comparison of answers to interrelated questions. Then we submitted the Census Bureau’s report of its analysis to Dr. Alan Zaslavsky, an independent expert, for review; developed our own analysis of the GSS data and submitted our paper describing that analysis to the same expert; and summarized the expert’s conclusions and appended his report and the Census Bureau’s report (reproduced in appendixes III and IV), as well summarizing our conclusions. We used these procedures to ensure independence, given that the GSS test was based on our earlier recommendation that the Census Bureau and the Department of Homeland Security (DHS) test the grouped answers approach. To describe additional research that might be needed, we outlined the grouped answers approach and reviewed the main conclusions of the GSS test in meetings with the immigration experts listed in table 5 and with private sector statisticians. Additionally, we discussed the approach with various federal officials and staff at agencies responsible for fielding large- scale surveys. To assess the precision of indirect estimates, we addressed questions to Dr. Zaslavsky, developed illustrative tables showing hypothetical calculations under specified assumptions, and subjected those tables to review. To identify and describe candidate surveys for piggybacking the grouped answers question series, we set minimum criteria for consideration (nationally representative, mainly or only in-person interviews, and data on at least 50,000 persons overall, including native-born and foreign-born). Then we identified surveys that met those criteria, collected documents concerning the surveys, and interviewed officials and staff at federal agencies that sponsored or conducted those surveys. We also talked with experts in immigration about additional key criteria for selecting an appropriate survey. The scope of our work had several limitations. We did not attempt to collect new data from foreign-born respondents in a survey, focus group, or other format. We did not assess census or survey coverage of the foreign-born or undocumented populations. We did not assess nonresponse rates among foreign-born or undocumented persons selected for interview. We did not review alternative methods of obtaining estimates of the undocumented. While we consulted a number of private sector experts and sought to include a range of perspectives, other experts may have other views. Finally, we do not know to what extent the broad range of persons who compose immigrant communities share the views of the immigrant advocates we spoke with. Logically, grouped answers data can be used to estimate subgroups of the undocumented population, using the following procedures: 1. isolate survey data for (a) the subsample 1 respondents who are in the desired subgroup, based on a demographic or other question asked in the survey (for example, if the survey included a question on each respondent’s employment, data could be isolated for foreign-born who are employed), and (b) subsample 2 respondents in that subgroup; 2. calculate (a) the percentage of the subsample 1 subgroup respondents who are in each box of immigration status card 1 and (b) the percentage of subsample 2 subgroup respondents who are in each box of immigration status card 2; and 3. carry out the subtraction procedure (percentage in Box B, Card 1, minus percentage in Box A, Card 2), thus estimating the percentage of the subgroup who are undocumented. The resulting percentage can be multiplied by a census count or an updated estimate of the foreign-born persons who are in the subgroup (for example, multiply the estimate of the percentage of employed foreign-born who are undocumented by the census count or updated estimate of the number of employed foreign-born). These steps can be repeated to indirectly estimate the size of the undocumented population within various subgroups defined by activity, demographics, and other characteristics (such as those with or without health insurance) that are asked about in the survey. Without an extremely large survey, it would be difficult or impossible to derive reliable estimates for subgroups with few foreign-born persons or few undocumented persons. Ongoing surveys conducted annually have sometimes combined 2 or 3 years of data in order to provide more reliable estimates of low- prevalence groups; however, there is a loss of time-specificity. Program cost data are sometimes available on an average per-person basis, and surveys sometimes ask about benefit use. In such cases, the total costs of a program associated with a certain group can be estimated. Program costs associated with the undocumented population might be estimated by either (1) multiplying the estimated numbers of undocumented persons receiving benefits by average program costs or (2) performing the following procedures: 1. Isolate survey data for all foreign-born subsample 1 respondents who said they were in Box B of Card 1 and estimate each individual respondent’s program cost. Then aggregate the individual costs to estimate the total program cost (potentially, millions or billions of dollars) associated with the population of foreign-born persons defined by the group of immigration statuses in Box B, Card 1. 2. Isolate data for all foreign-born subsample 2 respondents who said they were in Box A of Card 2 and, as above, estimate each individual respondent’s program costs, aggregating these to estimate the total program costs associated with the population of foreign-born persons defined by the immigration statuses in Box A, Card 2 (again, potentially millions or billions of dollars). 3. Because the only difference between the immigration statuses in Box B, Card 1, and Box A, Card 2, is the inclusion of the undocumented status in Box B, Card 1, start with the total program cost estimate for all Box B, Card 1, respondents and subtract the corresponding cost estimate for Box A, Card 2, respondents. The result of the subtraction procedure represents an indirect estimate of program costs associated with the undocumented population. A more precise cost estimate can be obtained by calculating an additional “mirror image” cost estimate—this time, starting with costs estimated for respondents in Box B of Card 2 and subtracting costs associated with respondents in Box A of Card 1. The two “mirror image” estimates could then be averaged. The key limitations on such procedures are sample size and the representation of key subgroups—for example, foreign-born respondents residing in small states and local areas. Thus, for example, it is possible that state-level costs associated with undocumented persons might be estimated with reasonable precision for a large state or city with many foreign-born persons and a relatively high percentage of undocumented (potentially, California or New York City) but not for many smaller states or areas, unless very large samples (or samples focused on selected areas of interest) were drawn. Further work could explore the ways that complex analyses could be conducted to help delineate costs. Contributions can be conceptualized as contributions to the economy through work or, potentially, through taxes paid. Such contributions might be estimated by combining grouped answers data with other survey questions to estimate relevant subgroups, such as employed undocumented persons. In complex analyses, these data could potentially be combined with other data to help estimate taxes paid. Logically, other quantitative estimates might be obtained through procedures similar to those outlined above for estimating program costs. For example, the numbers of children in various immigration statuses might be estimated by asking an adult respondent how many foreign-born children (or how many foreign-born school-age children) reside in the household and then—using the 3-box card assigned to the adult respondent—asking how many of these children are in Box A, Box B, and Box C. We note that, thus far, testing has not asked respondents to report children’s immigration status with the grouped answers approach. If subsamples 1 and 2 are sufficiently large, it might also be possible to estimate the portion of the undocumented population represented by “overstays” who were legally admitted to this country for a specific authorized period of time but remained here after that period expired (without a timely application for extension of stay or change of status) and currently undocumented persons who are applicants for legal status and are waiting for DHS to approve (or disapprove) their application. To estimate overstays would require a separate question on whether the respondent had entered the country on a temporary visa. To estimate undocumented persons with pending applications would require a separate question concerning pending applications for any form of legal status (including, for example, applications for U.S. citizenship as well as applications for legal permanent resident status and other legal statuses). The precision of such estimates would depend on factors such as sample size, the percentages of foreign-born who came in on temporary visas or who have pending applications of some kind, and the numbers of undocumented persons within these groups. Appropriately informing each respondent about what information he or she is being asked to provide is a key issue. On one hand, the grouped answers approach logically conveys to each respondent exactly what he or she is being asked to reveal about himself or herself; no one we spoke with suggested otherwise. On the other hand, the grouped answers question series does not indicate that the respondent is being asked to participate in an effort that will result in estimates of all immigration statuses. Therefore, a statement is needed to convey this information. Officials and staff at the National Center for Health Statistics (NCHS) were particularly concerned about this issue and believed that failing to adequately address informed consent issues could be considered unethical. Privacy protection specialists at the Census Bureau said that An introductory statement before the first immigration-related question might be phrased, “The next questions are geared to helping us know more about immigration and the role that it plays in American life.” When each respondent is shown the 3-box training cards, it would be possible to explain to him or her that—while the survey does not ask, and does not want to know, the specifics of which Box B category applies to him or her—there will be other interviews in which other respondents will be asked about some of the Box B categories or statuses. Just before showing each respondent the immigration status card, it should be stated—and, in fact, interviewers stated in the test with Hispanic farmworkers—that “Using the boxes allows us to obtain the information we need, without asking you to give us information that you might not want to.” Further: “Because we’re using the boxes, we WON’T ‘zero in’ on anything somebody might not want to tell us.” It may also be possible to explain that the study’s goal is to allow researchers to broadly estimate all categories or statuses on the card for the population of immigrants—but to indicate that this will be done without ever asking questions that “zero in” on something that some respondents might not want to disclose in an interview. Neither the estimation method (that is, the two cards) nor the specific policy relevance of immigration-status estimates would have to be described to all respondents. However, interviewer statements should be provided for responding to respondents who have doubts or questions. The statistical expression and variance of a grouped answers estimate is as follows, with the starting point being the percentage or proportion of subsample 1 who are in Box B, Card 1, and the procedure being to subtract from this the proportion of subsample 2 who are in Box A, Card 2 (with cards and boxes as defined as in figure 3):Grouped answers estimate = p. where p = the proportion of subsample 1 in Box B, Card 1 p) = [(pq) + (p/n = the proportion of subsample 1 not in Box B, Card 1 q = 1 – p = numbers of respondents in subsamples 1 and 2, respectively. The immigration status cards in figure 3 are designed so that Boxes A and B include all major immigration statuses. This design ensures that, on each card, the Box B categories apply to the largest possible number of legally present respondents. In designing the cards this way, we reasoned that this should reduce the question threat associated with choosing Box B. As a result, few respondents are expected to choose Box C (“some other category not in Box A or Box B”). For example, in the 2004 GSS test, only one foreign-born respondent of more than 200 chose Box C. Therefore, we believe that for purposes of illustrative variance calculations, it is reasonable to assume that no one chooses Box C. Under this assumption, the two mirror-image estimates of the percentage of the foreign-born who are undocumented would necessarily be exactly the same, as explained below. For simplicity, the discussion in this appendix assumes simple random sampling, for both the main sample and the selection of the two subsamples. The alternative, mirror-image estimate can then be defined as follows: As indicated above, q are defined in terms of p and p = (1– q) – (1– q = q In other words, under the assumption that no one chooses Box C, the mirror-image estimates of the percentage undocumented are, by definition, identical. Thus, no precision gain follows from combining them. No additional information is provided by a second, mirror-image estimate. In contrast, quantitative indirect estimates are based on a combination of (1) grouped answers data and (2) additional, separate quantitative data or estimates (for example, per-person estimates of emergency-visit costs based on respondent reports of number of emergency room visits in the past year and other information from hospitals on per-visit costs). If the quantitative data are tallied or totaled for individuals in each box of each card, the result is four different figures, none of which can be derived from the others. (There are different respondents in each box, and each would have separately reported how many emergency room visits, for example, he or she made in the past year.) Thus, for quantitative estimates of this type, calculating two independent mirror-image estimates, and averaging them, may yield a more precise result. Nancy R. Kingsbury, (202) 512-2700 or kingsburyn@gao.gov. Key GAO staff contributing to this report include Judith A. Droitcour, Eric M. Larson, and Penny Pickett. Statistical support was provided by Sid Schwartz, Mark Ramage, and Anna Maria Ortiz. Bird, Ronald. Statement of Ronald Bird, Chief Economist, Office of the Assistant Secretary for Policy, U.S. Department of Labor, before the Committee on the Judiciary, U.S. Senate, July 5, 2006. Boruch, Robert, and Joe S. Cecil. Assuring the Confidentiality of Social Research Data. Philadelphia: University of Pennsylvania Press, 1979. Camarota, Steven A., and Jeffrey Capizzano. “Assessing the Quality of Data Collected on the Foreign Born: An Evaluation of the American Community Survey (ACS): Pilot and Full Study Findings,” Immigration Studies White Papers, Sabre Systems Inc., April 2004. http://www.sabresys.com/whitepapers/CIS_whitepaper.pdf (Sept. 6, 2006). Costanzo, Joseph, and others, “Evaluating Components of International Migration: The Residual Foreign-Born,” Population Division Working Paper 61, U.S. Census Bureau, Washington, D.C., June 2002, p. 22. Droitcour, Judith A., and Eric M. Larson, “An Innovative Technique for Asking Sensitive Questions: The Three-Card Method,” Bulletin de Mèthodologie Sociologique, 75 (July 2002): 5–23. El-Badry, Samia, and David A. Swanson, “Providing Special Census Tabulations to Government Security Agencies in the United States: The Case of Arab-Americans,” paper presented at the 25th International Population Conference of the International Union for the Scientific Study of Population, Tours, France, July 18–23, 2005. Hill, Kenneth. “Estimates of Legal and Unauthorized Foreign-Born Population for the United States and Selected States Based on Census 2000.” Presentation at the U.S. Census Bureau Conference, Immigration Statistics: Methodology and Data Quality, Alexandria, Virginia, February 13–14, 2006. Hoefer, Michael, Nancy Rytina, and Christopher Campbell. Estimates of the Unauthorized Immigrant Population Residing in the United States: January 2005. Washington, D.C.: Department of Homeland Security, Office of Immigration Statistics, August 2006. GAO. Undocumented Aliens: Questions Persist about Their Impact on Hospitals’ Uncompensated Care Costs, GAO-04-472. Washington, D.C.: May 21, 2004. GAO. Illegal Alien Schoolchildren: Issues in Estimating State-by-State Costs, GAO-04-733. Washington, D.C.: June 23, 2004. GAO. Overstay Tracking: A Key Component of Homeland Security and a Layered Defense, GAO-04-82. Washington, D.C.: May 21, 2004. GAO. Record Linkage and Privacy: Issues in Creating New Federal Research and Statistical Information. GAO-01-126SP. Washington, D.C.: April 2001. GAO. Survey Methodology: An Innovative Technique for Estimating Sensitive Survey Items, GAO/GGD-00-30. Washington, D.C.: November 1999. GAO. Immigration Statistics: Information Gaps, Quality Issues Limit Utility of Federal Data to Policymakers, GAO/GGD-98-164. Washington, D.C.: July 31, 1998. Greenberg, Bernard G., and others. “The Unrelated Questions Randomized Response Model: Theoretical Framework.” Journal of the American Statistical Association, 64 (1969): 520–39. Kincannon, Charles Louis, “Procedures for Providing Assistance to Requestors for Special Data Products Known as Special Tabulations and Extracts,” memorandum to Associate Directors, Division Chiefs, Bureau of the Census, Washington, D.C., August 26, 2004. Locander, William, and others. “An Investigation of Interview Method, Threat, and Response Distortion.” Journal of the American Statistical Association, 71 (1976): 269–75. National Research Council, Committee on National Statistics, Local Fiscal Effects of Illegal Immigration: Report of a Workshop. Washington, D.C.: National Academy Press, 1996. Passel, Jeffrey S. “The Size and Characteristics of the Unauthorized Migrant Population in the U.S.: Estimates Based on the March 2005 Current Population Survey.” Research Report. Washington, D.C.: Pew Hispanic Center, March 7, 2006. Passel, Jeffrey S., Rebecca L. Clark, and Michael Fix. “Naturalization and Other Current Issues in U.S. Immigration: Intersections of Data and Policy,” In Proceedings of the Social Statistics Section of the American Statistical Association: 1997. Alexandria, Va.: American Statistical Association, 1997. Robinson, J. Gregory. “Memorandum for Donna Kostanich.” DSSD A.C.E. Revision II Memorandum Series No. PP-36. Washington, D.C.: U.S. Bureau of the Census, December 31, 2002. Rytina, Nancy F. Estimates of the Legal Permanent Resident Population and Population Eligible to Naturalize in 2004. Washington, D.C.: Department of Homeland Security, Office of Immigration Statistics, February 2006. Schryock, Henry S., and Jacob S. Siegel and Associates. The Methods and Materials of Demography. Washington, D.C.: U.S. Government Printing Office, 1980. Siegel, Jacob S., and David A. Swanson. The Methods and Materials of Demography, 2nd ed. San Diego, Calif.: Elsevier Academic Press, 2004. U.S. Census Bureau, “The U.S. Census Bureau’s Intercensal Population Estimates and Projections Program: Basic Underlying Principles,” paper distributed by the Census Bureau at its conference on Population Estimates: Meeting User Needs, Alexandria, Virginia, July 19, 2006. U.S. Commission on Immigration Reform. U.S. Immigration Policy: Restoring Credibility: 1994 Report to Congress. Washington, D.C.: U.S. Government Printing Office, 1994. U.S. Immigration and Naturalization Service, Office of Policy and Planning. Estimates of the Unauthorized Immigrant Population Residing in the United States: 1990 to 2000. Washington, D.C.: January 2003. U.S. Department of Labor, Findings from the National Agricultural Workers Survey (NAWS) 2000–2002: A Demographic and Employment Profile of United States Farm Workers. Research Report 9. Washington, D.C.: March 2005. Warner, Stanley. “Randomized Response: A Survey Technique for Eliminating Evasive Answer Bias.” Journal of the American Statistical Association, 60 (1995): 63–69. Warren, Robert, and Jeffrey S. Passel. “A Count of the Uncountable: Estimates of Undocumented Aliens Counted in the 1980 Census.” Demography, 24:3 (1987): 375–93. | As greater numbers of foreign-born persons enter, live, and work in the United States, policymakers need more information--particularly on the undocumented population, its size, characteristics, costs, and contributions. This report reviews the ongoing development of a potential method for obtaining such information: the "grouped answers" approach. In 1998, GAO devised the approach and recommended further study. In response, the Census Bureau tested respondent acceptance and recently reported results. GAO answers four questions. (1) Is the grouped answers approach acceptable for use in a national survey of the foreign-born? (2) What further research may be needed? (3) How large a survey is needed? (4) Are any ongoing surveys appropriate for inserting a grouped answers question series (to avoid the cost of a new survey)? For this study, GAO consulted an independent statistician and other experts, performed test calculations, obtained documents, and interviewed officials and staff at federal agencies. The Census Bureau and DHS agreed with the main findings of this report. DHS agreed that the National Survey of Drug Use and Health is not an appropriate survey for inserting a grouped answers question series. The grouped answers approach is designed to ask foreign-born respondents about their immigration status in a personal-interview survey. Immigration statuses are grouped in Boxes A, B, and C on two different flash cards--with the undocumented status in Box B. Respondents are asked to pick the box that includes their current status and are told, "If it's in Box B, we don't want to know which specific category applies to you." The grouped answers approach is acceptable to many experts and immigrant advocates--with certain conditions, such as (for some advocates) private sector data collection. Most respondents tested did not object to picking a box. Research is needed to assess issues such as whether respondents pick the correct box. A sizable survey--roughly 6,000 or more respondents--would be needed for 95 percent confidence and a margin of error of (plus or minus) 3 percentage points. The ongoing surveys that GAO identified are not appropriate for collecting data on immigration status. (For example, one survey takes names and Social Security numbers, which might affect acceptance of immigration status questions.) Whether further research or implementation in a new survey would be justified depends on how policymakers weigh the need for such information against potential costs and the uncertainties of future research. |
In August 2004, the President issued HSPD-12, which directed the Department of Commerce to develop a new standard for secure and reliable forms of ID for federal employees and contractors to enable a common standard across the federal government by February 27, 2005. The directive defines secure and reliable ID as meeting four control objectives. Specifically, the identification credentials must be ● based on sound criteria for verifying an individual employee’s or contractor’s identity; ● strongly resistant to identity fraud, tampering, counterfeiting, and ● able to be rapidly authenticated electronically; and issued only by providers whose reliability has been established by an official accreditation process. HSPD-12 stipulates that the standard must include criteria that are graduated from “least secure” to “most secure” to ensure flexibility in selecting the appropriate level of security for each application. In addition, the directive directs agencies to implement, to the maximum extent practicable, the standard for IDs issued to federal employees and contractors in order to gain physical access to controlled facilities and logical access to controlled information systems by October 27, 2005. In response to HSPD-12, NIST published FIPS 201, Personal Identity Verification of Federal Employees and Contractors, on February 25, 2005. The standard specifies the technical requirements for PIV systems to issue secure and reliable ID credentials to federal employees and contractors for gaining physical access to federal facilities and logical access to information systems and software applications. Smart cards are a primary component of the envisioned PIV system. The FIPS 201 standard is composed of two parts, PIV-I and PIV-II. PIV-I sets standards for PIV systems in three areas: (1) identity proofing and registration, (2) card issuance and maintenance, and (3) protection of card applicants’ privacy. There are many steps to the identity proofing and registration process, such as completing a background investigation of the applicant, conducting and adjudicating a fingerprint check prior to credential issuance, and requiring applicants to provide two original forms of identity source documents from an OMB-approved list of documents. The card issuance and maintenance process should include standardized specifications for printing photographs, names, and other information on PIV cards and for other activities, such as capturing and storing biometric and other data, and issuing, distributing, and managing digital certificates. Finally, agencies are directed to perform activities to protect the privacy of the applicants, such as assigning an individual to the role of “senior agency official for privacy” to oversee privacy-related matters in the PIV system; providing full disclosure of the intended uses of the PIV card and related privacy implications to the applicants; and using security controls described in NIST guidance to accomplish privacy goals, where applicable. The second part of the FIPS 201 standard, PIV-II, provides technical specifications for interoperable smart card-based PIV systems. The components and processes in a PIV system, as well as the identity authentication information included on PIV cards, are intended to provide for consistent authentication methods across federal agencies. The PIV-II cards (see example in fig. 1) are intended to be used to access all federal physical and logical environments for which employees are authorized. The PIV cards contain a range of features—including photographs, cardholder unique identifiers (CHUID), fingerprints, and Public Key Infrastructure (PKI) certificates—to enable enhanced identity authentication at different assurance levels. To use these enhanced capabilities, specific infrastructure needs to be in place. This infrastructure may include biometric (fingerprint) readers, personal ID number (PIN) input devices, and connections to information systems that can process PKI digital certificates and CHUIDs. Once acquired, these various devices need to be integrated with existing agency systems, such as a human resources system. Furthermore, card readers that are compliant with FIPS 201 need to exchange information with existing physical and logical access control systems in order to enable doors and systems to unlock once a cardholder has been successfully authenticated and access has been granted. FIPS 201 includes specifications for three types of electronic authentication that provide varying levels of security assurance. ● The CHUID or visual inspection, provides some confidence. ● A biometric check without the presence of a security guard or attendant at the access point, offers a high level of assurance of the cardholders’ identity. ● A PKI check, independently or in conjunction with both biometric and visual authentication, offers a very high level of assurance in the identity of the cardholder. OMB guidance and FIPS 201 direct agencies to use risk-based methods to decide which type of authentication is appropriate in a given circumstance. In addition to the three authentication methods, PIV cards also support the use of PIN authentication, which may be used in conjunction with one of these capabilities. For example, the PIN can be used to control access to biometric data on the card when conducting a fingerprint check. NIST has issued several publications that provide supplemental guidance on various aspects of the FIPS 201 standard. NIST also developed a suite of tests to be used by approved commercial laboratories to validate whether commercial products for the PIV card and the card interface are in conformation with the standard. In August 2005, OMB issued a memorandum to executive branch agencies with instructions for implementing HSPD-12 and the new standard. The memorandum specifies to whom the directive applies; to what facilities and information systems FIPS 201 applies; and, as outlined in the following text, the schedule that agencies must adhere to when implementing the standard. ● October 27, 2005—For all new employees and contractors, adhere to the identity proofing, registration, card issuance, and maintenance requirements of the first part (PIV-I) of the standard. ● October 27, 2006—Begin issuing cards that comply with the second part (PIV-II) of the standard and implementing the privacy requirements. ● October 27, 2007—Verify and/or complete background investigations for all current employees and contractors who have been with the agency for 15 years or less. Issue PIV cards to these employees and contractors, and require that they begin using their cards by this date. ● October 27, 2008—Complete background investigations for all individuals who have been federal agency employees for more than 15 years. Issue cards to these employees and require them to begin using their cards by this date. Figure 2 shows a timeline that illustrates when HSPD-12 and additional guidance was issued as well as the major deadlines for implementing HSPD-12. The General Services Administration (GSA) has also provided implementation guidance and product performance and interoperability testing procedures. In addition, GSA establishe Managed Service Office (MSO) that offers shared service civilian agencies to help reduce the costs of procuring FIPS 201- compliant equipment, software, and services by sharing some of the infrastructure, equipment, and services among participating agencies. According to GSA, the shared service offering—referred to as the USAccess Program—is intended to provide several ser such as producing and issuing the PIV cards. As of October 2007, GSA had 67 agency customers with more than 700,000 government employees and contractors to whom cards would be issued throug shared service providers. In addition, as of December 31, 2007, the MSO had installed over 50 enrollment stations with 15 agencies actively enrolling employees and issuing PIV cards. While there are several services offered by the MSO, it is not intended to provide support for all aspects of HSPD-12 implementation. For example, the MSO does not provide services to help agencies integrate theirphysical and logical access control systems with their PIV systems . In 2006, GSA’s Office of Governmentwide Policy established the interagency HSPD-12 Architecture Working Group, which is intended to develop interface specifications for HSPD-12 system interoperability across the federal government. As of July 200 7, the group had issued 10 interface specification documents, including a specification for exchanging data between an agency and a shared service provider. In February 2006, we reported tha challenges in implementing FIPS 2 time frames and funding uncertainties as well as incomplete implementation guidance. We recommended that OMB monitor agencies’ implementation process and completion of key acti In response to this recommendation, beginning on March 1, 2007, OMB directed agencies to post to their public Web sites quarterly reports on the number of PIV cards they had issued to their employees, contractors, and other individuals. In addition, in Augu 2006, OMB directed each agency to submit an updated implementation plan. t agencies faced several 01, including constrained testing vities. Military servicemembers, federal workers, and industry personnel must obtain security clearances to gain access to classified information. Clearances are categorized into three levels: top secre secret, and confidential. The level of classification denotes t degree of protection required for information and the amount of damage that unauthorized disclosure could reasonably cause tonational security. The degree of expected damage that unauthoriz disclosure could reasonably be expected to cause is “exceptional ly grave damage” for top secret information, “serious damage” for secret information, and “damage” for confidential information. We designated DOD’s personnel security clearance program a high- risk area in January 2005 and continued that designation in the updated list of high-risk areas that we published in 2007. We identified this program as a high-risk area because of long-standing delays in determining clearance eligibility and other challenges. DOD represents about 80 percent of the security clearances adjudicated by the federal government and problems in the clearance program can negatively affect national security. For example, delays in renewing security clearances for personnel who are already doing classified work can lead to a heightened risk of unauthorized disclosure of classified information. In contrast, delays in providing initial security clearances for previously non-cleared personnel can result in other negative consequences, such as additional costs and delays in completing national security-related contracts, lost opportunity costs, and problems retaining the best qualified personnel. DOD’s Office of the Under Secretary of Defense for Intelligence has responsibility for determining eligibility for clearances for servicemembers, DOD civilian employees, and industry personnel performing work for DOD and 23 other federal agencies, and employees in the federal legislative branch. That responsibility includes obtaining background investigations, primarily through the Office of Personnel Management (OPM). Within DOD, government employees use the information in OPM- provided investigative reports to determine clearance eligibility of clearance subjects. Recent significant events affecting the clearance program of DOD and other federal agencies include the passage of the Intelligence Reform and Terrorism Prevention Act of 2004 and the issuance of the June 2005 Executive Order 13381, “Strengthening Processes Relating to Determining Eligibility for Access to Classified National Security Information.” The act included milestones for reducing the time to complete clearances, general specifications for a database on security clearances, and requirements for reciprocity of clearances. Among other things, the executive order established as policy that agency functions relating to determining eligibility for access to classified national security information shall be appropriately uniform, centralized, efficient, effective, timely, and reciprocal and provided that the Director of OMB would ensure the policy's effective implementation. Agencies had made limited progress in implementing and using PIV cards. While the eight agencies we reviewed had generally taken steps to complete background checks on most of their employees and contractors and establish basic infrastructure, such as purchasing card readers, none of the agencies met OMB’s goal of issuing PIV cards by October 27, 2007, to all employees and contractor personnel who had been with the agency for 15 years or less. In addition, for the limited number of cards that had been issued, agencies generally had not been using the electronic authentication capabilities on the cards. A key contributing factor for why agencies had made limited progress in adopting the use of PIV cards is that OMB, which is tasked with ensuring that federal agencies implement HSPD-12, focused agencies’ attention on card issuance, rather than on full use of the cards’ capabilities. Until OMB revises its approach to focus on the full use of card capabilities, HSPD-12’s objective of increasing the quality and security of ID and credentialing practices across the federal government may not be fully achieved. As we have previously described, by October 27, 2007, OMB had directed federal agencies to issue PIV cards and require PIV card use by all employees and contractor personnel who have been with the agency for 15 years or less. HSPD-12 requires that the cards be used for physical access to federally controlled facilities and logical access to federally controlled information systems. In addition, to issue cards that fully meet the FIPS 201 specification, basic infrastructure—such as ID management systems, enrollment stations, PKI, and card readers—will need to be put in place. OMB also directed that agencies verify and/or complete background investigations by this date for all current employees and contractors who have been with the agency for 15 years or less. Agencies had taken steps to complete background checks that were directed by OMB, on their employees and contractors and establish basic infrastructure to help enable the use of PIV capabilities. For example, Commerce, Interior, NRC, and USDA had established agreements with GSA’s MSO to use its shared infrastructure, including its PKI, and enrollment stations. Other agencies, including DHS, HUD, Labor, and NASA—which chose not to use GSA’s shared services offering—had acquired and implemented other basic elements of infrastructure, such as ID management systems, enrollment stations, PKI, and card readers. However, none of the eight agencies had met the October 2007 deadline regarding card issuance. In addition, for the limited number of cards that had been issued, agencies generally had not been using the electronic authentication capabilities on the cards. Instead, for physical access, agencies were using visual inspection of the cards as their primary means to authenticate cardholders. While it may be sufficient in certain circumstances—such as in very small offices with few employees—in most cases, visual inspection will not provide an adequate level of assurance. OMB strongly recommends minimal reliance on visual inspection. Also, seven of the eight agencies we reviewed had not been using the cards for logical access control. Furthermore, most agencies did not have detailed plans in place to use the various authentication capabilities. For example, as of October 30, 2007, Labor had not yet developed plans for implementing the electronic authentication capabilities on the cards. Similarly, Commerce officials stated that they would not have a strategy or time frame in place for using the electronic authentication capabilities of PIV cards until June 2008. Table 1 provides details about the progress each of the eight agencies had made as of December 1, 2007. A key contributing factor to why agencies had made limited progress is that OMB—which is tasked with ensuring that federal agencies implement HSPD-12—had emphasized the issuance of the cards, rather than the full use of the cards’ capabilities. Specifically, OMB’s milestones were not focused on implementation of the electronic authentication capabilities that are available through PIV cards, and had not set acquisition milestones that would coincide with the ability to make use of these capabilities. Furthermore, despite the cost of the cards and associated infrastructure, OMB had not treated the implementation of HSPD-12 as a major new investment and had not ensured that agencies have guidance to ensure consistent and appropriate implementation of electronic authentication capabilities across agencies. Until these issues are addressed, agencies may continue to acquire and issue costly PIV cards without using their advanced capabilities to meet HSPD-12 goals. While OMB had established milestones for near-term card issuance, it had not established milestones to require agencies to develop detailed plans for making the best use of the electronic authentication capabilities of PIV cards. Consequently, agencies had concentrated their efforts on meeting the card issuance deadlines. For example, several of the agencies we reviewed chose to focus their efforts on meeting the next milestone—that cards be issued to all employees and contractor personnel and be in use by October 27, 2008. Understandably, meeting this milestone was perceived to be more important than making optimal use of the cards’ authentication capabilities, because card issuance is the measure that OMB is monitoring and asking agencies to post on their public Web sites. The PIV card and the services involved in issuing and maintaining the data on the card, such as the PKI certificates, are costly. For example, PIV cards and related services offered by GSA through its shared service offering cost $82 per card for the first year and $36 per card for each of the remaining 4 years of the card’s life. In contrast, traditional ID cards with limited or no electronic authentication capabilities cost significantly less. Therefore, agencies that do not implement electronic authentication techniques are spending a considerable amount per card for capabilities that they are not able to use. A more economical approach would be to establish detailed plans for implementing the technical infrastructure necessary to use the electronic authentication capabilities on the cards and time the acquisition of PIV cards to coincide with the implementation of this infrastructure. Without OMB focusing its milestones on the best use of the authentication capabilities available through PIV cards, agencies are likely to continue to implement minimum authentication techniques and not be able to take advantage of advanced authentication capabilities. Before implementing major new systems, agencies are generally directed to conduct thorough planning to ensure that costs and time frames are well understood and that the new systems meet their needs. OMB establishes budget justification and reporting requirements for all major information technology investments. Specifically, for such investments, agencies are directed to prepare a business case—OMB Exhibit 300—which is supported by a number of planning documents that are essential in justifying decisions regarding how, when, and the extent to which an investment would be implemented. However, OMB determined that because agencies had ID management systems in place prior to HSPD-12 and that the directive only directed agencies to “standardize” their systems. the implementation effort did not constitute a new investment. According to an OMB senior policy analyst, agencies should be able to fund their HSPD-12 implementations through existing resources and should not need to develop a business case or request additional funding. While OMB did not direct agencies to develop business cases for HSPD-12 implementation efforts, PIV card systems are likely to represent significant new investments at several agencies. For example, agencies such as Commerce, HUD, and Labor had not implemented PKI technology prior to HSPD-12, but they are now directed to do so. In addition, such agencies’ previous ID cards were used for limited purposes and were not used for logical access. These agencies had no prior need to acquire or maintain card readers for logical access control or to establish connectivity with their ID management systems for logical access control and, consequently, had previously allocated very little money for the operations and maintenance of these systems. For example, according to Labor officials, operations and maintenance costs for its pre-HSPD-12 legacy system totaled approximately $169,000, while its fiscal year 2009 budget request for HSPD-12 implementation is approximately $3 million—17 times more expensive. While these agencies recognized that they are likely to face substantially greater costs in implementing PIV card systems, they had not always thoroughly assessed all of the expenses they are likely to incur. For example, agency estimates may not have included the cost of implementing advanced authentication capabilities where they are needed. The extent to which agencies need to use such capabilities could significantly impact an agency’s cost for implementation. While the technical requirements of complying with HSPD-12 dictated that a major new investment be made, generally, agencies had not been directed by OMB to take the necessary steps to thoroughly plan for these investments. For example, six of the eight agencies we reviewed had not developed detailed plans regarding their use of PIV cards for physical and logical access controls. In addition, seven of the eight agencies had not prepared cost-benefit analyses that weighed the costs and benefits of implementing different authentication capabilities. Without treating the implementation of HSPD-12 as a major new investment by requiring agencies to develop detailed plans based on risk-based assessments of agencies’ physical and logical access control needs that support the extent to which electronic authentication capabilities are to be implemented, OMB will continue to limit its ability to ensure that agencies properly plan and implement HSPD-12. OMB Had Not Provided Guidance for Determining Which PIV Card Authentication Capabilities to Implement for Physical and Logical Access Controls Another factor contributing to agencies’ limited progress is that OMB had not provided guidance to agencies regarding how to determine which electronic authentication capabilities to implement for physical and logical access controls. While the FIPS 201 standard describes three different assurance levels for physical access (some, high, and very high confidence) and associates PIV authentication capabilities with each level, it is difficult for agencies to link these assurance levels with existing building security assurance standards that are used to determine access controls for facilities. The Department of Justice has developed standards for assigning security levels to federal buildings, ranging from level I (typically, a leased space with 10 or fewer employees, such as a military recruiting office) to level V (typically, a building, such as the Pentagon or Central Intelligence Agency headquarters, with a large number of employees and a critical national security mission). While there are also other guidelines that agencies could use to conduct assessments of their buildings, several of the agencies we reviewed use the Justice guidance to conduct risk assessments of their facilities. Officials from several of the agencies we reviewed indicated that they had not been using the FIPS 201 guidance to determine which PIV authentication capabilities to use for physical access because they had not found the guidance to be complete. Specifically, they were unable to determine which authentication capabilities should be used for the different security levels. The incomplete guidance has contributed to several agencies—including Commerce, DHS, and NRC—not reaching decisions on what authentication capabilities they were going to implement. More recently, NIST has begun developing guidelines for applying the FIPS 201 confidence levels to physical access control systems. However, this guidance has not yet been completed and was not available to agency officials when we were conducting our review. Agencies also lacked guidance regarding when to use the enhanced authentication capabilities for logical access control. Similar to physical access control, FIPS 201 describes graduated assurance levels for logical access (some, high, and very high confidence) and associates PIV authentication capabilities with each level. However, as we have previously reported, neither FIPS 201 nor supplemental OMB guidance provides sufficient specificity regarding when and how to apply the standard to information systems. For example, such guidance does not inform agencies how to consider the risk and level of confidence needed when different types of individuals require access to government systems, such as a researcher uploading data through a secure Web site or a contractor accessing government systems from an off-site location. Until complete guidance is available, agencies will likely continue either to delay in making decisions on their implementations or to make decisions that may need to be modified later. As defined by OMB, one of the primary goals of HSPD-12 is to enable interoperability across federal agencies. As we have previously reported, prior to HSPD-12, there were wide variations in the quality and security of ID cards used to gain access to federal facilities. To overcome this limitation, HSPD-12 and OMB guidance direct that ID cards have standard features and means for authentication to enable interoperability among agencies. While steps had been taken to enable future interoperability, progress had been limited in implementing such capabilities in current systems, partly because key procedures and specifications had not yet been developed. As we have previously stated, NIST established conformance testing for the PIV card and interface, and GSA established testing for other PIV products and services to help enable interoperability. In addition, the capability exists for determining the validity and status of a cardholder from another agency via PKI. However, procedures and specifications to enable cross-agency interoperability using the CHUID—which is expected to be more widely used than PKI—had not been established. While PIV cards and FIPS 201-compliant readers may technically be able to read the information encoded on any PIV card—including cards from multiple agencies—this functionality is not adequate to allow one agency to accept another agency’s PIV card, because there is no common interagency framework in place for agencies to electronically exchange status information on PIV credentials. For example, the agency that issued a PIV card could revoke the cardholder’s authorization to access facilities or systems if the card is lost or if there has been a change in the cardholder’s employment status. The agency attempting to process the card would not be able to access this information because a common framework to electronically exchange status information does not exist. The interfaces and protocols that are needed for querying the status of cardholders have not yet been developed. In addition, procedures and policies had not been established for sharing information on contractor personnel who work at multiple federal agencies. Without such procedures and policies, agencies will issue PIV cards to their contractor staff for access only to their own facilities. Contractors who work at multiple agencies may need to obtain separate PIV cards for each agency. GSA recognized the need to address these issues and has actions under way to do so. According to GSA, the Federal Identity Credentialing Committee is developing guidance on the issuance and maintenance of PIV cards to the contractor community. GSA is also developing a standard specification that will enable interoperability in the exchange of identity information among agencies. According to GSA officials, they plan to complete and issue guidance by the end of September 2008. Additionally, NIST is planning to issue an update to a special publication that focuses on interfaces for PIV systems. Such guidance should help enable agencies to establish cross-agency interoperability—a primary goal of HSPD-12. To help ensure that the objectives of HSPD-12 are achieved, we made several recommendations in our report. First, we recommended that OMB establish realistic milestones for full implementation of the infrastructure needed to best use the electronic authentication capabilities of PIV cards in agencies. In commenting on a draft of our report, OMB stated that its guidance requires agencies to provide milestones for when they intend to leverage the capabilities of PIV credentials. However, in order to ensure consistent governmentwide implementation of HSPD-12, it is important for OMB to establish such milestones across agencies, rather than to allow individual agencies to choose their own milestones. Next, we recommended that OMB require each agency to develop a risk-based, detailed plan for implementing electronic capabilities. OMB stated that previous guidance required agencies to provide milestones for when they plan to fully leverage the capabilities of PIV credentials for physical and logical access controls. However, agencies were required to provide only the dates they plan to complete major activities, and not detailed, risk-based plans. Until OMB requires agencies to implement such plans, OMB will be limited in its ability to ensure agencies make the best use of their cards’ electronic authentication capabilities. We also recommended that OMB require agencies to align the acquisition of PIV cards with plans for implementing the cards’ electronic authentication capabilities. In response, OMB stated that HSPD-12 aligns with other information security programs. While OMB’s statement is correct, it would be more economical for agencies to time the acquisition of PIV cards to coincide with the implementation of the technical infrastructure necessary for enabling electronic authentication techniques. This approach has not been encouraged by OMB, which instead measures agencies primarily on how many cards they issue. Lastly, we recommended that OMB ensure guidance is developed that maps existing physical security guidance to FIPS 201 guidance. OMB stated that NIST is in the process of developing additional guidance to clarify the relationship between facility security levels and PIV authentication levels. In March 2008, NIST released a draft of this guidance to obtain public comments. In our previous reports, we have also documented a variety of problems present in DOD’s personnel security clearance program. Some of the problems that we noted in our 2007 high-risk report included delays in processing clearance applications and problems with incomplete investigative and adjudicative reports to determine clearance eligibility. Delays in the clearance process continue to increase costs and risk to national security, such as when new industry employees are not able to begin work promptly and employees with outdated clearances have access to classified documents. Moreover, DOD and the rest of the federal government provide limited information to one another on how they individually ensure the quality of clearance products and procedures. While DOD continues to face challenges in timeliness and quality in the personnel security clearance process, high-level governmentwide attention has been focused on improving the security clearance process. As we noted in February 2008, delays in the security clearance process continue to increase costs and risk to national security. An August 2007 DOD report to Congress noted that delays in processing personnel security clearances for industry have been reduced, yet that time continues to exceed requirements established by the Intelligence Reform and Terrorism Prevention Act of 2004 (IRTPA). The act currently requires that adjudicative agencies make a determination on at least 80 percent of all applications for a security clearance within an average of 120 days after the date of receipt of the application, with 90 days allotted for the investigation and 30 days allotted for the adjudication. However, DOD’s August 2007 report on industry clearances stated that, during the first 6 months of fiscal year 2007, the end-to-end processing of initial top secret clearances took an average of 276 days; renewal of top secret clearances, 335 days; and all secret clearances, 208 days. We also noted in February 2008, that delays in clearance processes can result in additional costs when new industry employees are not able to begin work promptly and increased risks to national security because previously cleared industry employees are likely to continue working with classified information while the agency determines whether they should still be eligible to hold a clearance. To improve the timeliness of the clearance process, we recommended in September 2006 that OMB establish an interagency working group to identify and implement solutions for investigative and adjudicative information-technology problems that have resulted in clearance delays. In commenting on our recommendation, OMB’s Deputy Director for Management stated that the National Security Council’s Security Clearance Working Group had begun to explore ways to identify and implement improvements to the process. DOD and the Rest of the Government Provide Limited Information on How to Ensure the Quality of Clearance Products and Procedures As we reported in February 2008, DOD and the rest of the federal government provide limited information to one another on how they individually ensure the quality of clearance products and procedures. For example, DOD’s August 2007 congressionally mandated report on clearances for industry personnel documented improvements in clearance processes but was largely silent regarding quality in clearance processes. While DOD described several changes to the processes and characterized the changes as progress, the department provided little information on (1) any measures of quality used to assess clearance processes or (2) procedures to promote quality during clearance investigation and adjudication processes. Specifically, DOD reported that the Defense Security Service, DOD’s adjudicative community, and OPM are gathering and analyzing measures of quality for the clearance processes that could be used to provide the national security community with a better product. However, the DOD report did not include any of those measures. In September 2006, we reported that while eliminating delays in clearance processes is an important goal, the government cannot afford to achieve that goal by providing investigative and adjudicative reports that are incomplete in key areas. We additionally reported that the lack of full reciprocity—when one government agency fully accepts a security clearance granted by another government agency—is an outgrowth of agencies’ concerns that other agencies may have granted clearances based on inadequate investigations and adjudications. Without fuller reciprocity of clearances, agencies could continue to require duplicative investigations and adjudications, which result in additional costs to the federal government. In the report we issued in February 2008, we recommended that DOD develop measures of quality for the clearance process and include them in future reports to Congress. Statistics from such measures would help to illustrate how DOD is balancing quality and timeliness requirements in its personnel security clearance program. DOD concurred with that recommendation, indicating it had developed a baseline performance measure of the quality of investigations and adjudications and was developing methods to collect information using this quality measure. Recent High-Level Governmentwide Attention Has Been Focused On Improving the Security Clearance Process In February 2008, we reported that while DOD continues to face timeliness and quality challenges in the personnel security clearance program, high-level governmentwide attention has been focused on improving the security clearance process. For example, we reported that OMB’s Deputy Director of Management has been responsible for a leadership role in improving the governmentwide processes since June 2005. During that time, OMB has overseen, among other things, the growth of OPM’s investigative workforce and greater use of OPM’s automated clearance-application system. In addition, an August 9, 2007, memorandum from the Deputy Secretary of Defense indicates that DOD’s clearance program is drawing attention at the highest levels of the department. Streamlining security clearance processes is one of the 25 DOD transformation priorities identified in the memorandum. Another indication of high-level government attention we reported in February 2008 is the formation of an interagency security clearance process reform team in June 2007. Agencies included in the governmentwide effort are OMB, the Office of the Director of National Intelligence, DOD, and OPM. The team’s memorandum of agreement indicates that it seeks to develop, in phases, a reformed DOD and intelligence community security clearance process that allows the granting of high-assurance security clearances in the least time possible and at the lowest reasonable cost. The team’s July 25, 2007, terms of reference indicate that the team plans to deliver “a transformed, modernized, fair, and reciprocal security clearance process that is universally applicable” to DOD, the intelligence community, and other U.S. government agencies. A further indication of high level government attention is a memorandum issued by the President on February 5, 2008 which called for aggressive efforts to achieve meaningful and lasting reform of the processes to conduct security clearances. In the memorandum, the President acknowledged the work being performed by the interagency security clearance process reform team and directed that the team submit to the President an initial reform proposal not later than April 30, 2008. In closing, OMB, GSA, and NIST have made significant progress in laying the foundation for implementation of HSPD-12. However, agencies did not meet OMB’s October 2007 milestone for issuing cards and most have made limited progress in using the advanced security capabilities of the cards that have been issued. These agency actions have been largely driven by OMB’s guidance, which has emphasized issuance of cards rather than the full use of the cards’ capabilities. As a result, agencies are acquiring and issuing costly PIV cards without using the advanced capabilities that are critical to achieving the objectives of HSPD-12. Until OMB provides additional leadership by guiding agencies to perform the planning and assessments that will enable them to fully use the advanced capabilities of these cards, agencies will likely continue to make limited progress in using the cards to improve security over federal facilities and systems. Regarding security clearances, in June 2005, OMB took responsibility for a leadership role for improving the governmentwide personnel security clearance process. The current interagency security clearance process reform team represents a positive step to address past impediments and manage security clearance reform efforts. Although the President has called for a reform proposal to be provided no later than April 30, 2008, much remains to be done before a new system can be implemented. Mr. Chairman and members of the subcommittee, this concludes our statement. We would be happy to respond to any questions that you or members of the subcommittee may have at this time. If you have any questions on matters discussed in this testimony, please contact Linda D. Koontz at (202) 512-6240 or Brenda S. Farrell at (202) 512-3604 or by e-mail at koontzl@gao.gov or farrellb@gao.gov. Other key contributors to this testimony include John de Ferrari (Assistant Director), Neil Doherty, Nancy Glover, James P. Klein, Rebecca Lapaze, Emily Longcore, James MacAulay, David Moser and Shannin O’Neill. This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. | In an effort to increase the quality and security of federal identification (ID) practices, the President issued Homeland Security Presidential Directive 12 (HSPD-12) in August 2004. This directive requires the establishment of a governmentwide standard for secure and reliable forms of ID. GAO was asked to testify on its report, being released today, assessing the progress selected agencies have made in implementing HSPD-12. For this report, GAO selected eight agencies with a range of experience in implementing ID systems and analyzed actions these agencies had taken. GAO was also asked to summarize challenges in the DOD personnel security clearance process. This overview is based on past work including reviews of clearance-related documents. Military servicemembers, federal workers, and industry personnel must obtain security clearances to gain access to classified information. Long-standing delays in processing applications for these clearances led GAO to designate the Department of Defense's (DOD) program as a high-risk area in 2005. In its report on HSPD-12, GAO made recommendations to the Office of Management and Budget (OMB), to, among other things, set realistic milestones for implementing the electronic authentication capabilities. GAO has also made recommendations to OMB and DOD to improve the security clearance process. Much work had been accomplished to lay the foundations for implementation of HSPD-12--a major governmentwide undertaking. However, none of the eight agencies GAO reviewed--the Departments of Agriculture, Commerce, Homeland Security, Housing and Urban Development, the Interior, and Labor; the Nuclear Regulatory Commission; and the National Aeronautics and Space Administration--met OMB's goal of issuing ID cards by October 27, 2007, to all employees and contractor personnel who had been with the agency for 15 years or less. In addition, for the limited number of cards that had been issued, most agencies had not been using the electronic authentication capabilities on the cards and had not developed implementation plans for those capabilities. A key contributing factor for this limited progress is that OMB had emphasized issuance of the cards, rather than full use of the cards' capabilities. Furthermore, agencies anticipated having to make substantial financial investments to implement HSPD-12, since ID cards are considerably more expensive than traditional ID cards. However, OMB had not considered HSPD-12 implementation to be a major new investment and thus had not required agencies to prepare detailed plans regarding how, when, and the extent to which they would implement the electronic authentication mechanisms available through the cards. Until OMB revises its approach to focus on the full use of the capabilities of the new ID cards, HSPD-12's objectives of increasing the quality and security of ID and credentialing practices across the federal government may not be fully achieved. Regarding personnel security clearances, GAO's past reports have documented problems in DOD's program including delays in processing clearance applications and problems with the quality of clearance related reports. Delays in the clearance process continue to increase costs and risk to national security, such as when new DOD industry employees are not able to begin work promptly and employees with outdated clearances have access to classified documents. Moreover, DOD and the rest of the federal government provide limited information to one another on how they individually ensure the quality of clearance products and procedures. While DOD continues to face challenges in timeliness and quality in the personnel security clearance process, high-level government attention has been focused on improving the clearance process. |
A federally recognized tribe is an American Indian or Alaska Native tribal entity that is recognized as having a government-to-government relationship with the United States. Federally recognized tribes and their members are eligible for programs and services the federal government provides to Indians because of their status as Indians. Eligible tribes and certain tribal entities have access to federal resources for child welfare programs under programs administered by HHS. Federal funding provided under titles IV-E and IV-B of the Social Security Act covers a portion of costs for states and participating tribes of operating their foster care, adoption assistance, and kinship guardianship assistance programs (title IV-E) and funding for child welfare services, such as parent support and counseling for children and families (title IV- B). Approximately 20,000 of the 415,000 children in foster care at the end of fiscal year 2014 were identified as American Indian or Alaska Native alone or American Indian or Alaska Native in combination with one or more races. Title IV-E of the Social Security Act is the largest single federal source of child welfare funding, providing nearly $7.5 billion in fiscal year 2015 to child welfare agencies to support foster care and transitional independent living programs for eligible children, adoption assistance for children with special needs, and kinship guardianship assistance. Both states and tribes seeking to operate a title IV-E program must have an approved title IV-E plan. The title IV-E plan provides documentation that state or tribal law, regulations, and policies comply with program requirements. Among other things, the title IV-E program requires that states and tribes: make reasonable efforts, consistent with the health and safety of the child, to preserve and reunify families (1) prior to a child’s placement in foster care, to prevent the need for removing the child; and (2) to make it possible for the child to safely return home; prepare a written case plan for each child receiving foster care maintenance payments and ensure periodic court or administrative review of each such case; and, regularly hold a permanency hearing to determine the case goal for the child (to be held at least every 12 months) and make reasonable efforts to finalize the case goal—reunification, adoption, legal guardianship, placement with a fit and willing relative, or, where appropriate, another planned permanent living arrangement. Case goal options under title IV-E are shown in table 1. Title IV-E funds reimburse participating states and tribes for a portion of their eligible costs of providing foster care, adoption assistance, or kinship guardianship assistance on behalf of eligible children: Title IV-E Foster Care Program: The title IV-E Foster Care Program provides federal funding for state and tribal title IV-E agencies to support out-of-home care for children until the children are safely returned home, placed permanently with adoptive families or placed in other planned arrangements for permanency. Federal funds are available for a portion of monthly maintenance payments for the daily care and supervision of eligible children; administrative costs to manage the program; training of staff and foster care providers; recruitment of foster parents; and costs related to the design, implementation and operation of a state-wide data collection system. Title IV-E Adoption Assistance Program: The Adoption Assistance Program provides funds to state and tribal title IV-E agencies to facilitate the timely placement of children, whose special needs or circumstances would otherwise make placement with adoptive families difficult. Federal funds are available for a one-time payment to families to assist with the costs of adopting an eligible child, and for a portion of monthly subsidies to assist with the care of an eligible adopted child, as well as for certain administrative and other program costs. Title IV-E Guardianship Assistance Program: The Guardianship Assistance Program (GAP) is an optional program under title IV-E that provides funds to support the care of children discharged from foster care to legal guardianship who meet the program’s eligibility requirements. Under the program, state and tribal title IV-E agencies who opt to provide guardianship assistance payments to relatives who have assumed legal guardianship of eligible children receive partial reimbursement from the federal government for the cost of providing these payments. In general, eligible children include those who have been eligible for title IV-E foster care maintenance payments during at least a 6 consecutive month period during which the child resided in the home of the prospective relative guardian who was licensed or approved as meeting the licensure requirements as a foster family home, among other criteria. The reimbursement rate varies by type of expense. The federal share of foster care maintenance payments, adoption assistance payments, and guardianship assistance payments is between 50 percent and 83 percent of costs, with higher federal support going to states and eligible tribes with lower per capita incomes. The federal share of training and administrative costs is 75 percent and 50 percent, respectively. Eligible tribes may receive federal title IV-E funds either by (1) directly operating a foster care title IV-E program pursuant to an approved title IV- E plan; or (2) administering all or part of a title IV-E program on behalf of a state pursuant to a tribal-state agreement. Tribal-state agreements: Tribes may enter into cooperative agreements with their state child welfare agency to administer part of the state’s title IV-E program. Through tribal-state agreements, title IV- E funds are generally passed through the state to the tribes. Whether tribes or states have responsibility for activities such as developing a case plan for the child’s placement and care, or reviewing the case plan on a regular basis, can vary depending on the agreement. HHS considers the state to be the grantee for title IV-E funding and as a result, the state has ultimate responsibility for ensuring compliance with title IV-E program requirements. Direct title IV-E programs: The Fostering Connections to Success and Increasing Adoptions Act of 2008 (Fostering Connections Act) authorized tribes to directly operate a title IV-E program and obtain federal reimbursement for eligible program costs pursuant to an approved title IV-E plan. As of October 2015, there were seven tribes approved to directly operate a title IV-E program. The provisions of title IV-E apply to tribes that operate their own programs in the same manner as they apply to states, with limited exceptions. HHS is responsible for oversight and monitoring of states and tribes that are approved to directly operate a title IV-E program. Tribes also have access to federal funding provided under title IV-B, subparts 1 and 2 of the Social Security Act for various child welfare services—$664 million in fiscal year 2015. These programs fund a range of child welfare services and activities generally aimed at preventing abuse and neglect; preserving and reunifying families; and promoting safety, permanency, and well-being of children in foster or adoptive placements. To receive title IV-B funding, states and tribes must submit a 5-year plan—the Child and Family Services Plan—that sets forth the vision and goals to be accomplished to improve the state’s or tribe’s coordinated child and family service delivery system. It also includes, among other things, the state’s or tribe’s plans for use of funding from title IV-B. Tribes that receive title IV-B funding and operate foster care programs must also comply with certain title IV-E requirements, including those related to case goals. State and tribal title IV-B agencies must submit annual updates—Annual Progress and Service Reports—on the progress made toward accomplishing the goals and objectives in the Child and Family Services plan, which are reviewed by HHS regional office staff. Tribes must also be operating a program under title IV-B, subpart 1 to be eligible to directly administer a title IV-E program. The main mechanism for oversight of title IV-B tribal agencies is HHS’s review of the tribe’s Child and Family Service Plan and annual updates. This review is conducted by HHS regional office staff as part of the joint planning process. In addition, tribes or families involved with tribal child welfare programs may receive other state or federal funding outside of titles IV-E or IV-B to support child welfare services, or they may be supported only through tribal funds. The Adoption and Safe Families Act of 1997 (ASFA) added APPLA as an option for a child’s case goal if none of the other case goals enumerated in the statute would be in the best interests of the child. In addition, ASFA removed language allowing a child to be “continued in foster care on a permanent or long-term basis” because of his or her special needs or circumstances. Under title IV-E, APPLA is permitted as a case goal in cases where the child welfare agency has documented to the court a “compelling reason” for determining that the other case goals would not be in the best interests of the child. The Preventing Sex Trafficking and Strengthening Families Act, enacted in 2014, restricted the use of APPLA as a case goal to children age 16 and older, although it delayed implementation for children in foster care under the responsibility of an Indian tribe. For older children with a case goal of APPLA, the Act places additional requirements on states and tribes. Among other things, at each permanency hearing (to be held at least every 12 months), the child welfare agency is required to (1) document its intensive, ongoing, and unsuccessful efforts to return the child home or secure a placement with a relative, legal guardian, or adoptive parent; and (2) implement procedures to ensure that the court re-determines whether APPLA is the appropriate case goal for the child and asks the child about his or her desired permanency outcome. ACF’s Children’s Bureau is responsible for providing program guidance under titles IV-B and IV-E to grantees. ACF is also responsible for pro- viding technical assistance to tribes on title IV-B and IV-E program requirements through nine of its regional offices. In 2014, HHS consolidated its training and technical assistance provided to states and tribes. The Center for Tribes (the Center), one of the three centers in HHS’s Child Welfare Capacity Building Collaborative, serves as a technical assistance resource for tribes. All tribes that receive title IV-E or title IV-B funds are eligible to receive assistance from the Center. The Center also develops products that all tribes can access from the Children’s Bureau website, including ones that have information on practices such as customary adoption and family group decisionmaking. It also offers services to tribes that are tailored to their particular needs; tribes are responsible for requesting these services from the Center. Available data from HHS show that a lower percentage of Indian children were assigned APPLA as a case goal at the end of fiscal year 2014 compared to non-Indian children. According to AFCARS data, about 6.1 percent of Indian children were assigned APPLA as a case goal (approximately 1,200 children) compared to 8.3 percent of non-Indian children (approximately 33,000 children). The percentage of Indian and non-Indian children assigned other case goals also differed (see table 2). However, in part because these children could be in either state or tribal foster care, these data cannot be used to identify children covered by the 3-year delay for the APPLA age restriction. While data on all children in foster care under the responsibility of a state agency are included in AFCARS, data on some children under the responsibility of a tribal agency are not included in AFCARS. In addition, children identified as Indian in AFCARS are not necessarily enrolled members of a federally-recognized tribe. While a slightly lower percentage of Indian children were assigned APPLA as a case goal than non-Indian children, a higher percentage of Indian children than non-Indian children who were assigned this case goal were younger than 16. Specifically, about 41 percent of Indian children with APPLA were younger than 16 (about 500 children), while about 23 percent of non-Indian children with this case goal were younger than 16 (about 7,500 children) (see fig. 1). Of Indian children assigned APPLA as a case goal, about 9 percent were ages 6 through 10, while 29 percent were ages 11 through 15, compared to approximately 3 percent and 19 percent of non-Indian children assigned APPLA, respectively. A much smaller percentage of both Indian and non-Indian children assigned this case goal were age 5 or younger (see fig. 2). On average, Indian and non-Indian children with APPLA as a case goal moved among foster care homes about the same number of times (see table 3). Both Indian and non-Indian children who were assigned APPLA as a case goal had an average of about six different placements during their most recent stay in foster care. In contrast, both Indian and non-Indian children who were assigned other case goals had an average of around three placements. In AFCARS the number of placements is defined as the number of places the child has lived, including the current setting, during the current removal episode and does not include the number of places a child may have lived during a previous episode in foster care or trial home visits. Our analysis shows that a higher percentage of all children with APPLA as a case goal were placed in either institutions or group homes (about 30 percent) than children with other case goals (between 7 and 11 percent). However, some differences exist between Indian and non-Indian children with APPLA. Although Indian and non-Indian children with APPLA moved among foster homes about the same number of times, a slightly lower percentage of Indian than non-Indian children with APPLA as a case goal were placed in institutional care. While 13.1 percent of Indian children with this case goal were placed in institutional care, 16.8 percent of non-Indian children with APPLA were placed in institutions. Some additional differences exist between the percentage of Indian and non-Indian children placed in group homes, but these differences were small (13.5 percent of Indian compared to 13.9 percent of non-Indians). Data also showed that a higher percentage of Indian children with APPLA were placed in family-style foster homes, either with relatives or non- relatives, than non-Indian children with APPLA (see fig. 3). Most tribal officials we interviewed reported that they did not anticipate challenges associated with limiting the use of the APPLA case goal to children aged 16 and older. Specifically, officials from 27 of 32 tribes said they did not anticipate that the restriction would pose challenges for their foster care programs. Tribal officials we interviewed also did not anticipate challenges with implementing the related provisions in the Preventing Sex Trafficking and Strengthening Families Act that introduced additional case review requirements for foster children to whom the APPLA case goal is assigned. Of the 27 tribes that said they did not anticipate challenges with limiting the use of the APPLA case goal to children aged 16 and older, 17 tribes told us they use APPLA as a case goal for some children under tribal custody. However, these tribes told us that they assign APPLA rarely and only after determining that other case goals, such as reunification, adoption, or guardianship—which are required to be considered before APPLA is used—were not possible or appropriate. Tribal officials reported that most children in their foster care programs are initially assigned reunification as a case goal and are often reunified with their parents. These officials told us that if they determine that reunification is not possible, they will assign adoption or guardianship as a case goal, depending on the specific case. However, some tribal officials also noted certain instances in which these case goals were not possible or appropriate and that they had determined that APPLA was the most appropriate goal. These examples included: Older children—Officials from 8 tribes said they use APPLA as a case goal either primarily or exclusively for older children or teenagers in foster care. Tribes reported that it was often more difficult to achieve permanency for older children due to a number of challenges, such as a lack of services tailored to older children or greater difficulty in recruiting adoptive families. Some tribes reported that they use APPLA as a last resort when they cannot find a permanent home for older children in foster care; officials told us that they use APPLA to allow the child to continue to receive foster care maintenance payments while preparing to age out of the system. The challenges facing older children in foster care in general are well-documented. Children who do not want to be adopted or placed into guardianships—Officials from 5 tribes reported assigning APPLA as a case goal in cases in which older foster children indicated that they did not want to be adopted or placed into guardianships. For example, 1 tribe reported assigning APPLA in the case of two children who were under age 16 for whom reunification was unlikely, but who did not want to be adopted. According to tribal officials, the children were initially placed in a family-like foster care home, briefly transferred to a residential facility, and ultimately placed in an independent living arrangement. Children with severe health issues—Officials from 4 tribes reported using APPLA as a case goal primarily for foster children with significant mental, behavioral, or physical health issues. For example, officials from one tribe told us that they sometimes assign APPLA as a case goal while providing additional services to help foster children through a crisis due to behavioral or mental health issues. These cases could include foster children who might need to stay in residential care for a certain amount of time. These tribal officials told us that APPLA is sometimes used for children who are placed in therapeutic foster homes where they receive specialized care. Some other tribes reported that they believe this case goal could be, in some cases, an appropriate case goal for younger children who need long- term, intensive medical treatment in therapeutic foster homes or residential facilities. Another tribe told us the tribe has recently experienced an influx of tribal children entering the child welfare system who have a number of significant health issues due to prenatal exposure to drugs, for whom APPLA could be the most appropriate case goal. Some tribes we contacted—5 of 32—said they anticipated that the APPLA change would present challenges for their foster care programs. Of these tribes, two reported using the APPLA case goal extensively for children in tribal foster care. One tribe reported that the majority of children in foster care under their custody, most of whom are younger than 16, are assigned long-term foster care as a case goal. A tribal official told us that while some of these children are in family-like foster homes, most move around between residential facilities based on their behavioral needs or that have specific age or gender restrictions. This official noted that the tribe was concerned about the impact of the APPLA age restriction given the number of their children in long-term foster care arrangements. However, this official said that tribal officials had not discussed the restriction in detail with their tribal attorney, given the 3- year delayed implementation period. The second tribe told us that of the approximately 50 children under tribal custody, the majority are assigned long-term foster care and are placed in stable, family-like foster care homes in the care of extended family members. The tribal official we spoke with said that when the APPLA restriction takes effect, they will have to pursue guardianships for children younger than 16 who had previously been assigned an APPLA case goal. However, the official said that the tribe has faced challenges establishing guardianships. She told us that guardianships limit families’ ability to access other resources, such as educational and training vouchers in some cases, and that the tribe has a number of foster families who would participate in guardianship arrangements but do not want to be licensed for a variety of reasons. While most tribes we contacted did not anticipate challenges in limiting the use of APPLA as a case goal to older children in foster care, many tribal officials reported other challenges that made finding permanent homes for children under tribal custody difficult. Specifically, tribes reported facing challenges regarding cultural opposition to termination of parental rights and differences between state and tribal definition of a relative; licensing requirements; and resource constraints (see fig. 4). Each of these challenges potentially limits the use of adoption, living with a relative, and guardianship as alternative case goals for younger children who have been assigned APPLA when the age limitation on APPLA takes effect. Several tribes we contacted reported that the federal foster care requirements related to termination of parental rights were in conflict with their cultural values and presented challenges to their efforts to establish permanency for children in foster care. Specifically, 7 tribes we spoke with reported that termination of parental rights is not acceptable in their tribal culture. We previously reported that tribes developing title IV-E plans encountered challenges associated with incorporating required language on termination of parental rights into their tribal codes or policies, which made it difficult to obtain internal approval and successfully complete the title IV-E plan. HHS officials acknowledged that termination of parental rights may not align with Indian tribes’ cultural values and in April 2015, HHS updated its Child Welfare Policy Manual to modify its interpretation of the statute to clarify that tribal title IV-E agencies may develop an alternative to termination of parental rights, such as a modification of parental rights, so long as the alternative satisfies other applicable requirements. Several tribes we interviewed practice tribal customary adoptions, which, unlike traditional adoptions, modify but do not terminate parental rights and also maintain family connections. Of the 29 tribes that discussed their use of customary adoption, 9 tribes reported using such adoptions, 12 tribes reported that they did not, and 8 tribes reported that they did not practice customary adoptions currently but might pursue them in the future. HHS officials told us that they support customary adoption as a culturally-appropriate method of achieving permanency for children. According to HHS officials, families that adopt a child through customary adoption are eligible for title IV-E adoption subsidies. Several tribal officials we contacted told us that another challenge facing their foster care programs were state definitions of who is considered a “relative” for purposes of foster care; these definitions can contradict tribes’ cultural traditions. Specifically, officials from 9 tribes we spoke with in individual and group interviews said their tribes broadly defined the term relative and that the tribe’s definition was not always compatible with state definitions. For tribes administering title IV-E on behalf of the state pursuant to a tribal-state agreement, this difference could affect tribes’ use of the case goal of “placement with a fit and willing relative” or whether tribal guardianships with relatives would qualify for GAP payments under a state’s definition of “relative.” For example, officials from one tribe told us that tribal children are born into an extensive clan system and that a tribal member’s “extended family” is very large, and their clan affiliation is a permanent and necessary component to their identity. Another tribe reported that nearly all of their children in foster care are placed with individuals who are considered to be relatives under the tribe’s broad definition, which encompasses not only blood relatives but also fictive kin—tribal members who may not be blood related but who have a close relationship with the child. Tribal officials told us that they believed federal title IV-E requirements did not provide for such a broader definition of “relative” and that some tribal members who would be considered relatives under the tribe’s definition may not qualify for certain services, such as GAP. According to HHS policy, there is no federal definition of “relative” for the purpose of the title IV-E GAP. Tribes that administer title IV-E on behalf of the state pursuant to a tribal-state agreement generally follow the state title IV-E plan. HHS officials acknowledged that tribes frequently define “relative” differently than states. However, in some cases, a state’s definition of a relative could be more restrictive than a tribe’s, and this narrower definition could be included in the tribe’s title IV-E agreement with the state. According to HHS, tribes approved to directly operate a title IV-E program have discretion to define the term relative for the purposes of the title IV-E GAP. HHS officials told us that HHS will accept a title IV-E plan or amendment that contains a reasonable interpretation of a relative, including a plan that limits the term to include biological and legal familial ties or a plan that more broadly includes tribal kin, extended family and friends, or other “fictive kin”. More than one-third of tribes we contacted—15 of 36 tribes—told us that satisfying federal and state licensing requirements often made it more difficult to license potential foster care homes and in some cases also prevented tribes from establishing potential permanent placements for their children. For example, in order to be eligible for GAP payments, a child must have resided with a licensed relative foster caretaker for at least 6 continuous months. Several tribal officials we interviewed said that licensing requirements regarding the number of family members living in a home, the size of the home, or the availability of separate bedrooms presented challenges when attempting to license potential foster homes, given the prevalence of multi-generational homes in tribal communities. An official from one tribe also told us that background checks conducted as part of the licensing process will sometimes disqualify otherwise suitable relatives from becoming licensed and therefore a possible permanent caregiver. For example, some tribal officials noted that an otherwise qualified relative may not pass a background check because of a past drug offense. Participants in a 2011 survey by the National Child Welfare Resource Center for Tribes reported that substance abuse not only contributed to difficulties some families had when they tried to reunify with their children but also created challenges in finding suitable foster homes. However, other tribes we interviewed reported that they have been able to incorporate some flexibility into their tribal licensing requirements to address licensing challenges or account for other circumstances unique to tribal communities. For example, some tribes we interviewed said that their tribe’s licensing requirements provide more flexibility with regard to the size of a home or the number of bedrooms available. The Fostering Connections Act authorized states to waive, on a case-by-case basis, non-safety licensing standards for relative foster family homes and gave states discretion to determine what constitutes a non-safety standard. According to HHS officials, tribes approved to directly operate title IV-E programs are allowed to define and waive non-safety standards for relative foster family homes, as are states. However, some officials we spoke with from tribes approved to directly operate title IV-E programs were not aware that they were able to waive non-safety licensing standards. In addition, two of the six states we reviewed for our study— Michigan and New Mexico—were not among those that reported in 2014 allowing waivers of non-safety licensing standards for relative foster family homes and defining such standards in state law, regulations, or agency policy, while four states—Arizona, Montana, Oklahoma, and Washington—reported in 2014 that they did allow such waivers. According to HHS, tribes operating programs under title IV-E tribal-state agreements are bound by the terms in those agreements, which could include licensing requirements. HHS officials we spoke with said that tribes have not expressed to them any questions or concerns about challenges they have experienced with non-safety licensing standards. More than half of tribes we spoke with—21 of 36 tribes—told us that resource constraints presented challenges for their foster care programs and for establishing permanency for children under their custody. These constraints include a lack of foster care homes, limited availability of needed services, and limited financial support for permanent guardianships. Officials from 7 tribes reported that a significant challenge for their programs is a shortage of licensed foster care homes, particularly those that are willing and capable of caring for children with special needs. Similarly, officials from 4 tribes told us that they had limited access to therapeutic foster homes or inpatient care facilities because their tribes were located in rural areas far from these homes and facilities or because these resources were already being used by state foster care programs. Tribal officials also reported that the financial assistance available to foster care providers or permanent caregivers for children in foster care is often insufficient to fully cover all of the associated costs. Officials from one tribe reported that the majority of tribal members struggle to make ends meet and feed members of their current households. While the tribe offers financial assistance to foster families, tribal officials said that amount is insufficient to fully cover the costs of fostering a child. In addition, several tribes told us that tribal foster care programs generally do not have access to the range of services available to states, such as independent living services that can help children in foster care transition to permanency or resources and services for children with special needs. HHS officials we spoke with acknowledged that there is great demand among tribes for resources for these high-needs children. Some tribal officials also told us that families entering into guardianships did not receive any financial assistance, such as title IV-E GAP, which, according to tribal officials, made it more difficult to establish permanency through guardianships. HHS has provided information and some assistance to tribes on implementing the change to APPLA, including policy guidance and technical assistance through HHS regional offices, and some support for establishing permanency for children in tribal foster care. Despite reporting resource constraints, however, most tribes we interviewed reported that they did not participate in title IV-E GAP to receive federal funding to support their use of guardianship as a permanent placement. Without greater tribal participation in GAP, tribes may not be receiving available funding that could help them encourage the use of and support guardianship, particularly for older children or those with special needs for whom finding permanent placements may be difficult. The Children’s Bureau has sent guidance to regional offices, state child welfare agencies, and tribal organizations concerning the provisions in the Preventing Sex Trafficking and Strengthening Families Act, including the change to the APPLA case goal. For example, the Children’s Bureau used a listserv to disseminate program instructions and information memoranda on changes related to the Act. Staff in the five regions that serve the tribes we interviewed told us that they discussed the guidance with their respective tribal child welfare agencies. Staff in three of these regions reported that because of frequent changes in tribal staff, they maintain an up-to-date listserv of tribal social services directors, who will be responsible for implementing the APPLA change. They used the listserv to disseminate guidance, including program instructions and information memoranda about the change in APPLA. Some tribal officials told us that they were aware of the APPLA change and had received information from HHS, while others were uncertain whether they had received information, and some in our group interviews did not specify whether they had received it. The regional offices and the Center for Tribes also provide technical assistance to tribes that can include help in implementing the APPLA change and developing other permanency options. Staff in the five regional offices reported that as part of the joint planning process for assisting tribes with developing their 5-year title IV-B plans, regional officials worked with the tribes on how the tribes could address the change in APPLA. An official with the Center for Tribes told us that as of March 2016, one tribe had requested assistance with incorporating all of the elements of the Preventing Sex Trafficking and Strengthening Families Act into their policies and procedures. In addition, officials from one regional office and the Center told us that they help tribes with developing other permanency options such as kinship guardianship (which may be supported by title IV-E GAP) and customary adoption. One regional official reported helping tribes develop provisions on customary adoption that could be added to their tribal code. A Center official also said that they have received inquiries about GAP and have had requests for assistance with customary adoption and licensing. Despite the resource constraints reported by many of the tribes we interviewed, most tribal officials we spoke with that use guardianship as a case goal reported that their tribes do not participate in title IV-E GAP, which provides federal funding for a portion of assistance payments paid to relative guardians on behalf of eligible children. Of the 27 tribes that reported using guardianship as a case goal, 20 tribes reported that they do not participate in GAP and do not receive federal funding to support those guardianships. Of these 20 tribes, 6 do not participate because they are located in states that do not offer the program. According to HHS officials, tribes in states that do not offer GAP would need to directly operate a title IV-E program in order to receive federal guardianship assistance. As of March 2016, 33 states and 6 tribes had been approved to operate GAP, according to HHS (see fig. 5). However, 14 tribes who use guardianship as a case goal are in states that offer GAP, but tribal officials told us they did not participate in GAP for several reasons. Officials from 5 tribes reported that because GAP was not included as an option in their title IV-E tribal-state agreements, families with guardianship arrangements do not receive GAP payments. Officials from 3 other tribes said they did not receive title IV-E funding so their tribes were not eligible for GAP payments. Officials from an additional 5 tribes told us that they did not participate due to access challenges at the state level, such as age restrictions or difficulty meeting state requirements. For example, in one state, according to tribal officials, a child must be 12 years old before a guardianship would be approved. Finally, officials from 1 tribe said that they were unaware of title IV-E GAP (see fig. 6). Given that tribes cited different reasons for not participating in GAP, HHS, states, and tribes may need to take different actions to address those reasons. In addition, there may be other reasons why tribes that we did not interview may not be participating in GAP. HHS officials said that because the program provides additional support for foster children’s transition to permanency, participating in title IV-E GAP and providing tribes GAP payments is in a state’s interest. However, we previously reported that some states opted not to participate in GAP because of limited funding for the state’s portion of the GAP payment to relative guardians. HHS officials told us that they were not aware of any instances in which a state participated in GAP but did not provide access to tribes through title IV-E agreements. However, officials also noted that HHS generally does not have information on the contents of title IV-E tribal-state agreements, as these agreements are not required to be submitted as part of a state’s title IV-E plan. Officials said that it is the responsibility of tribes and states to discuss GAP and whether to include the program in their IV-E agreements. They noted that Title IV-E calls on states to negotiate in good faith with tribes to develop a tribal-state agreement to administer the IV-E program, including GAP assistance payments. They also told us that the lack of guardianship funding to tribes means that tribes may be forced to rely on long-term foster care to fund permanent placements. ACF’s strategic plan provides a basis for the Children’s Bureau to support tribes’ participation in GAP. As part of its overarching goal to support underserved and underrepresented populations, ACF’s 2015-2016 strategic plan includes a goal for working with tribes to increase their capacity to promote child safety, permanency, and well-being. It also includes a goal to promote and facilitate improved tribal-state relations and policy at the regional and state levels to improve outcomes for Indian children. In addition, federal standards for internal control state that an agency’s management system should assure that monitoring is performed continually as part of agency operations. Without more awareness by HHS on why tribes may not be participating in GAP, including why some tribal-state agreements do not include GAP as an option for tribes, some tribes that could benefit from GAP may not participate in the program. The limited participation in GAP among the tribes selected for our study suggests that tribes may not be making use of available financial support for guardianships that could assist them in establishing permanency for their children and preventing them from remaining in long-term foster care. In 2012, a collaborative project of six child welfare organizations identified title IV-E GAP as one method of helping child welfare agencies establish permanency for children in foster care, particularly for older or other children for whom finding permanent placements may be difficult. Several states surveyed by the project mentioned that they were targeting, although not exclusively, special groups of children through GAP, including children with the APPLA case goal. The project report noted that financial assistance for guardians raising children who were in foster care is an essential component of the supports relatives need to offer these children permanent families. By taking steps that might increase tribal participation in GAP, HHS could better support tribes’ efforts to find secure and stable homes for their children in foster care. Changes made by the Preventing Sex Trafficking and Strengthening Families Act to the use of the APPLA case goal were expected by child welfare advocates to help ensure permanent adult connections for older children. While most tribal officials we spoke with did not consider the restriction on the use of APPLA to older children as a challenge, more than half noted that they faced resource constraints—including limited financial resources—which affect their ability to find permanent placements for children under their care. In 2008, title IV-E GAP was established to provide federal dollars to assist states and tribes with payments to relatives who are willing to provide permanent homes for children. However, most tribes we interviewed were not participating in the program. Some reported that their tribal-state title IV-E agreements did not include access to GAP while other tribes faced access challenges at the state level. Others did not receive any title IV-E funding or were unaware of the program, and tribes may have additional reasons for not participating. Guardianship arrangements can be a useful case goal to pursue in situations where reunification is not possible and adoption conflicts with tribal cultural values. GAP could help tribes by providing the financial assistance that tribal relatives need in order to provide a home for a child in their care. When the APPLA restriction takes effect, it can also provide an alternative to APPLA for those tribes trying to place a younger child who has special needs and requires more resources for care and for whom other permanency goals are not viable. As our analysis of the fiscal year 2014 AFCARS data indicated, some of these younger children may now be in institutional care and may need placement in a family setting if they are assigned a new case goal instead of APPLA. In addition, the program can provide needed resources for relatives willing to provide a stable home for children over the age of 16 who might otherwise be assigned APPLA. Without greater tribal participation in title IV-E GAP, some tribes may not have critical financial support to help children exit foster care to homes with relatives. By taking steps to increase participation in GAP, HHS could better support tribes’ efforts to care for their children. To help improve tribes’ ability to maintain safe, stable, and permanent care for children, the Secretary of Health and Human Services should direct the Children’s Bureau to explore the reasons for low tribal participation and identify actions to increase this participation in the title IV-E Guardianship Assistance Program. We provided a draft of this report to HHS for review and comment. HHS provided written comments that are reproduced in appendix II. HHS also provided technical comments that we incorporated, as appropriate. HHS concurred with our recommendation that the Children’s Bureau explore the reasons for low tribal participation and identify actions to increase participation in the title IV-E GAP. HHS noted that some tribal- state agreements might predate 2008, when GAP was established. HHS also said that regional office staff participate annually in joint planning for the title IV-B and IV-E programs with their respective states and tribes and that participation in GAP is a topic covered in joint planning activities. In addition, HHS said that regional office staff are available to assist states and tribes with discussions about GAP participation when tribal- state agreements are renegotiated and that technical assistance is available to tribes, if needed. According to HHS, the agency plans to add information to the Children’s Bureau website about direct federal funding for tribal title IV-E agencies and about tribal-state partnership agreements and plans to distribute issue briefs on GAP and best practices for tribal- state agreements. We agree that HHS has the planning process, technical assistance resources, and regional staff in place to discuss GAP participation with title IV-E state and tribal agency officials. However, our review found that tribal participation in GAP remains low, which suggests that HHS needs to identify actions to increase participation in this program. We believe that the additional actions HHS plans to take—providing information on the Children’s Bureau website about direct funding and distributing issue briefs on GAP and best practices for tribal-state partnerships and agreements—could support tribes’ participation in GAP either by helping tribes to directly operate a title IV-E program or to negotiate a tribal-state agreement that includes a provision for GAP participation. Because some tribes reported challenges at the state level to participating in GAP and several tribes reported that the state where they are located does not participate in the program, we encourage HHS to engage title IV-E state agency officials in discussions about tribal participation in GAP during the annual review of the their title IV-E state plan. As HHS noted in its response, regional office staff are a resource to states and tribes as they renegotiate agreements to help them explore participation in GAP. Through such discussions, HHS could identify ways that regional office staff might help state agencies resolve any challenges to GAP participation that tribes experience at the state level. HHS has taken several steps over the past few years to help tribes with their title IV-E programs, including hiring a tribal coordinator to facilitate communication between regional offices and tribal title IV-E agencies. Taking additional steps to ensure that tribes have the opportunity to participate in GAP could go a long way toward helping tribes gain more resources for children under their care and better support tribes’ efforts to care for children exiting foster care to permanent homes. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies of this report to the appropriate congressional committees, the Secretary of the Department of Health and Human Services, and other interested parties. We will also make copies available to others on request. In addition, the report will be available at no charge on GAO’s website at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-7215 or brownke@gao.gov. Contact points for our Office of Congressional Relations and Public Affairs can be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix III. This report examines: (1) what available data show about Indian children in foster care compared to all other children in foster care; (2) the extent to which selected tribal child welfare agencies face challenges in addressing recent changes to APPLA and in establishing permanent homes for children in tribal foster care; and (3) the extent to which HHS has assisted tribal child welfare agencies in implementing the APPLA change and addressing any challenges to establishing permanent homes. To determine what available data show about Indian children in foster care compared to all other children in foster care, we analyzed relevant national data on foster care case plans from HHS’s Adoption and Foster Care Analysis and Reporting System (AFCARS) for fiscal year 2014, the most recent year for which data are available. Through AFCARS, state title IV-E agencies are required to report to HHS semi-annually data on (1) all children in foster care for whom the agency has responsibility for placement, care, or supervision, and (2) all adopted children who were placed by the agency or for whom the agency is providing adoption assistance, care, or services. HHS in 2012 issued regulations that generally apply the same requirements for data collection and reporting to tribal title IV-E agencies. AFCARS contains information on placement settings, case plans, and length of time in foster care, among other information. Currently, AFCARS provides the following options for reporting the child’s most recent “case plan goal” (referred to in this report as a “case goal”): reunify with parent(s) or principal caretaker(s); live with other relative(s); long-term foster care; Options for reporting the child’s current placement setting in foster care include: foster family home (relative); foster family home (non-relative); trial home visit. Specifically, we reviewed AFCARS data for the 415,129 children who were in foster care on the last day of the fiscal year. Our analysis included a number of key data elements, including sex, whether the child was an American Indian/Alaska Native, number of placement settings during current foster care episode, current placement setting, case goal, and length of stay of latest removal episode. For the purposes of our data analysis discussion, we used the term Indian for children identified as American Indian/Alaska Native in AFCARS. Because the fiscal year 2014 AFCARS dataset does not have a measure for the APPLA case goal, we used long-term foster care and emancipation as proxy measures for APPLA. The AFCARS regulations were first issued in 1993, and HHS published a proposed rule in February 2015 to revise AFCARS to, among other things, incorporate statutory changes made since then. AFCARS contains data on children under the responsibility of tribal title IV-E agencies operating under tribal-state title IV-E agreements but does not yet contain data for the 7 tribes approved to directly operate title IV-E programs. Therefore, some children under the responsibility of tribal title IV-E agencies—and covered by the 3-year delayed implementation period for the APPLA restriction—are not represented in AFCARS. Data for certain other children under the care and custody of tribal child welfare agencies are not submitted into AFCARS. In addition, while AFCARS provides data on most children in foster care in the United States, there are several limitations to using the data available to study Indian children in foster care, particularly those under the care and placement responsibility of Indian tribes. In general, in AFCARS a person’s race is determined by how they define themselves or by how others define them; in the case of young children, parents determine the race of the child. In AFCARS, “American Indian or Alaska Native” (AI/AN) is defined as “a person having origins in any of the original peoples of North or South America (including Central America), and who maintains tribal affiliation or community attachment.” Children identified as Indian in AFCARS are not necessarily enrolled members of a federally-recognized tribe or eligible for enrollment. Before deciding to use the data provided by HHS, we conducted a data reliability assessment. We assessed the reliability of the data file that HHS provided by (1) performing electronic data testing for obvious errors in accuracy and completeness, (2) reviewing existing information about the data and the system that produced the data, and (3) interviewing agency officials and child welfare experts knowledgeable about these data. We determined that the data were sufficiently reliable for the purposes of this report. To obtain the perspectives of tribal officials, we conducted semi- structured interviews (2 in person and 13 by telephone) with tribal child welfare officials from 15 federally recognized tribes and held six group interviews with representatives from 21 other federally recognized tribes. We selected the 15 tribes for individual interviews from six states that have high numbers of tribes receiving title IV-E or title IV-B funding: Arizona, Michigan, Montana, New Mexico, Oklahoma, and Washington. Tribes in these states were selected to achieve variation in tribal population under age 21 and type of title IV funding—via a direct title IV-E program, a title IV-E tribal-state agreement, or a title IV-B program only. We conducted one telephone group interview and convened five in- person group interviews with officials from 21 additional tribes who were assembled for regional or national meetings at the following venues: Session convened by Casey Family Programs prior to a tribal consultation session with the Administration for Children and Families, Washington, D.C., September 2015. Oklahoma Indian Child Welfare Association Conference, Norman, OK, November 2015 - two discussion groups. Casey Family Programs National Title IV-E Summit, Denver, CO, December 2015. Michigan Quarterly Tribal-State Partnership meeting, Lansing, MI, January 2016. All tribes in attendance at these events were invited to participate in our group interviews. In total, our tribal interviewees included representatives of 6 tribes that were approved to operate a title IV-E program directly through HHS; representatives of 15 tribes that received title IV-E funding through a tribal-state agreement as well as title IV-B funding; and representatives of 13 tribes that received title IV-B funding only. The information obtained in our interviews is not generalizable to the population of federally recognized tribes, but provides examples of tribes’ experience using the APPLA case goal and any challenges they may face establishing permanency for children in their care when the changes go into effect. In addition, in the report we use qualifiers, such as “some” and “several” in some cases to quantify responses from tribal officials across our individual and group interviews with officials from 36 tribes in total. These qualifiers are defined as follows: “Most” more than 18 tribes “Many” represents 11-17 “Several” represents 6-10 “Some” represents 2-5 We also reviewed relevant federal laws, regulations and HHS guidance to states and tribes. In addition, we interviewed state child welfare officials from the six selected states, as well as officials from the Administration on Children and Families’ (ACF) Children’s Bureau at HHS; officials from five of HHS’s regional offices that work with the states where the selected tribes were located; and the Center for Tribes, an HHS-funded technical assistance provider. Kay E. Brown, Director, (202) 512-7215 or brownke@gao.gov. In addition to the contact named above, Elizabeth Morrison (Assistant Director), Sara Edmondson (Analyst-in-Charge), Kelly Snow, and Kate Blumenreich made key contributions to this report. Also contributing to this report were Ed Bodine, James Bennett, David Chrisinger, Sarah Cornetto, Jean McSween, and Daren Sweeney. | The Preventing Sex Trafficking and Strengthening Families Act, enacted in 2014, limited the use of APPLA as a case goal to children aged 16 and older. The Act made this provision effective 3 years after enactment for children under tribal responsibility. Some experts raised concerns that tribes may use the APPLA case goal to retain tribal connections for hard-to-place children, such as younger children with special needs. GAO was asked to explore tribes' views on these matters. This report examines: (1) data comparing Indian and non-Indian children in foster care; (2) challenges selected tribal child welfare agencies may face in addressing changes to APPLA and establishing permanent homes for children in tribal foster care; and (3) HHS assistance to tribes in implementing the APPLA change and addressing any challenges to establishing permanent homes. GAO reviewed relevant federal laws, regulations, and HHS guidance; analyzed HHS's fiscal year 2014 data on child welfare agencies' foster care case plans; and interviewed officials from 36 tribes that receive federal child welfare funding from six states with high numbers of tribes receiving this funding. GAO also interviewed HHS officials and officials at six state child welfare agencies. To receive federal child welfare funding, state and tribal child welfare agencies must comply with certain requirements, including developing a permanency plan for the child that identifies how the child will exit the foster care system to a permanent home (“case goal”). If other case goals, such as reunifying with parents, adoption, or guardianship are not possible or appropriate, a child may be assigned “another planned permanent living arrangement” (APPLA) as a case goal. Unlike other case goals, children assigned an APPLA case goal are normally expected to remain in foster care until they reach adulthood, which could result in young children remaining in foster care for many years. Because available foster care data do not include a measure for the APPLA case goal, GAO used long-term foster care and emancipation as proxy measures for this case goal. These data show that in 2014 about 6.1 percent of Indian children had APPLA as a case goal, compared to 8.3 percent of non-Indian children. Of the approximately 1,200 Indian children who were assigned an APPLA case goal, 41 percent were younger than 16, while of the approximately 33,000 non-Indian children with this case goal, 23 percent were younger than 16. These data also show, on average, that Indian and non-Indian children with APPLA as a case goal moved among foster homes about the same number of times. Most tribal officials GAO interviewed reported that they did not anticipate challenges in implementing the limitation on the use of APPLA to children age 16 and older, but many reported other challenges to establishing permanent homes for children in tribal foster care. Some organizations expressed the view that the APPLA age restriction would compel tribes to pursue other arrangements with non-Indian homes if they could not allow a child to remain in foster care with an Indian family. However, tribal officials GAO interviewed said that they rarely use APPLA and instead pursue reunification with family members or other case goals, such as guardianship. At the same time, tribal officials reported challenges with licensing foster family homes and resource constraints that may make establishing permanent homes—including guardianships—difficult. The Department of Health and Human Services (HHS) has provided information on APPLA through a listserv and information memoranda and some assistance to tribes in establishing permanent homes for children in foster care. However, many tribes GAO interviewed indicated that they were not receiving Guardianship Assistance Program funds under title IV-E of the Social Security Act which provide support for children exiting foster care to relative guardianships. Guardianship can be a useful alternative to APPLA when reunification and adoption are not viable options. Of the 36 tribes that GAO contacted, 14 reported that they did not participate in the program because it was not included in their title IV-E tribal-state agreements or the tribe faced challenges at the state level, among other reasons. One of HHS's strategic goals is to work with tribes to increase their capacity to promote child safety, permanent homes, and well-being. By taking actions to support tribes' participation in the Guardianship Assistance Program, HHS could help them receive funds to help establish permanent homes for children in tribal foster care, including those who might be affected by the APPLA change. GAO recommends that HHS explore the reasons for low tribal participation in the federal guardianship program and identify actions to increase tribal participation. HHS agreed with our recommendation. |
Since 2011, we have identified 11 areas across DHS where fragmentation, overlap, or potential duplication exists, and suggested 24 actions to the department and Congress to help strengthen the efficiency and effectiveness of DHS operations. In some cases, there is sufficient information available to show that if actions are taken to address individual issues, significant financial benefits may be realized. In other cases, precise estimates of the extent of potential unnecessary duplication, and the cost savings that can be achieved by eliminating any such duplication, are difficult to specify in advance of congressional and executive branch decision making. However, given the range of areas we identified at DHS and the magnitude of many of the programs, the cost savings associated with addressing these issues could be significant. In April 2013, we identified 2 new areas where DHS could take actions to address fragmentation, overlap, or potential duplication. First, we found that DHS does not have a department-wide policy defining research and development (R&D) or guidance directing how components are to report R&D activities. As a result, the department does not know its total annual investment in R&D, a fact that limits DHS’s ability to oversee components’ R&D efforts and align them with agency-wide R&D goals and priorities. DHS’s Science and Technology Directorate, Domestic Nuclear Detection Office, and the U.S. Coast Guard—the only DHS components that report R&D-related budget authority to the Office of Management and Budget (OMB) as part of the budget process—reported $568 million in fiscal year 2011 R&D budget authority. However, we identified at least 6 components with R&D activities and an additional $255 million in R&D obligations in fiscal year 2011 by other DHS components that were not reported to OMB in the budget process. To address this issue, we suggested that DHS develop and implement policies and guidance for defining and overseeing R&D at the department. Second, we reported that the fragmentation of field-based information sharing can be disadvantageous if activities are uncoordinated, as well as if opportunities to leverage resources across entities are not fully exploited. We suggested that DHS and other relevant agencies develop a mechanism that will allow them to hold field-based information-sharing entities accountable for coordinating with each other and monitor and evaluate the coordination results achieved, as well as identify characteristics of entities and assess specific geographic areas in which practices that could enhance coordination and reduce unnecessary overlap could be adopted. DHS generally agreed with our suggestions and is reported taking steps to address them. Moving forward, we will monitor DHS’s progress to address these actions. Concurrent with the release of our 2013 annual report, we updated our assessments of the progress that DHS has made in addressing the actions we suggested in our 2011 and 2012 annual reports.outlines the 2011-2012 DHS-related areas in which we identified Table 1 fragmentation, overlap, or potential duplication, and highlights DHS’s and Congress’s progress in addressing them. In our March 2011 and February 2012 reports, in particular, we suggested that DHS or Congress take 21 actions to address the areas of overlap or potential duplication that we found. Of these 21 actions, 2 (approximately 10 percent) have been addressed, 13 (approximately 62 percent) have been partially addressed, and the remaining 6 (approximately 29 percent) have not been addressed. For example, to address the potential for overlap among three information-sharing mechanisms that DHS funds and uses to communicate security-related information with public transit agencies, in March 2011, we suggested that DHS could identify and implement ways to more efficiently share security-related information by assessing the various mechanisms available to public transit agencies.We assessed this action as partially addressed because TSA has taken steps to streamline information sharing with public transit agencies, but the agency continues to maintain various mechanisms to share such information. In March 2011, we also found that TSA’s security assessments for hazardous material trucking companies overlapped with efforts conducted by the Department of Transportation’s (DOT) Federal Motor Carrier Safety Administration (FMCSA), and as a result, government resources were not being used effectively. After we discussed this overlap with TSA in January 2011, agency officials stated that, moving forward, they intend to only conduct reviews on trucking companies that are not covered by FMCSA’s program, an action that, if implemented as intended, we projected could save more than $1 million over the next 5 years. We also suggested that TSA and FMCSA could share each other’s schedules for conducting future security reviews, and avoid scheduling reviews on hazardous material trucking companies that have recently received, or are scheduled to receive, a review from the other agency. We assessed this action as addressed because in August 2011, TSA reported that it had discontinued conducting security reviews on trucking companies that are covered by the FMCSA program. Discontinuing such reviews should eliminate the short-term overlap between TSA’s and FMCSA’s reviews of hazardous material trucking companies. Although the executive branch and Congress have made some progress in addressing the issues that we have previously identified, additional steps are needed to address the remaining areas and achieve associated benefits. For example, to eliminate potential duplicating efforts of interagency forums in securing the northern border, in March 2011, we reported that DHS should provide guidance to and oversight of interagency forums to prevent duplication of efforts and help effectively utilize personnel resources to strengthen coordination efforts along the northern border. Further, the four DHS grant programs that we reported on in February 2012—the State Homeland Security Program, the Urban Areas Security Initiative, the Port Security Grant Program, and the Transit Security Grant Program—have multiple areas of overlap and can be sources of potential unnecessary duplication. These grant programs, which FEMA used to allocate about $20.3 billion to grant recipients from fiscal years 2002 through 2011, have similar goals and fund similar activities, such as equipment and training, in overlapping jurisdictions. To address these areas of overlap, we reported that Congress may want to consider requiring DHS to report on the results of its efforts to identify and prevent unnecessary duplication within and across these grant programs, and consider these results when making future funding decisions for these programs. Such reporting could help ensure that both Congress and FEMA steer scarce resources to homeland security needs in the See appendix I, table 4, for a most efficient, cost-effective way possible.summary of the fragmentation, overlap, and duplication areas and actions we identified in our 2011-2013 annual reports that are relevant to DHS. Our 2011-2013 annual reports also identified 13 areas where DHS or Congress should consider taking 29 actions to reduce the cost of operations or enhance revenue collection for the Department of the Treasury. Most recently, in April 2013, we identified 4 cost-savings and revenue enhancement areas related to DHS. Table 2 provides a summary of the 2011-2012 DHS-related areas in which we identified opportunities for cost savings or revenue enhancement, as well the status of efforts to address these areas. In addition, in April 2013 we also reported on the steps that DHS and Congress have taken to address the cost savings and revenue enhancement areas identified in our 2011 and 2012 annual reports. Table 3 provides a summary of the 2011-2012 DHS-related areas in which we identified opportunities for cost savings or revenue enhancement, as well the status of efforts to address these areas. Of the 21 related actions we suggested that DHS or Congress take in our March 2011 and February 2012 reports to either reduce the cost of government operations or enhance revenue collection, as of March 2013, 3 (about 14 percent) have been addressed, 11 (about 52 percent) have been partially addressed, and 7 (about 33 percent) have not been addressed. For example, in February 2012, we reported that to increase the likelihood of successful implementation of the Arizona Border Surveillance Technology Plan, minimize performance risks and help justify program funding, the Commissioner of CBP should update the agency’s cost estimate for the plan using best practices. This year, we assessed this action as partially addressed because CBP initiated action to update its cost estimate, using best practices, for the plan by providing revised cost estimates in February and March 2012 for the plan’s two largest projects. However, CBP has not independently verified its life cycle cost estimates for these projects with independent cost estimates and reconciled any differences with each system’s respective life cycle cost estimate, consistent with best practices. Such action would help CBP better ensure the reliability of each system’s cost estimate. Further, in March 2011, we stated that Congress may wish to consider limiting program funding pending receipt of an independent assessment of TSA’s Screening of Passengers by Observation Techniques (SPOT) program. This year, we assessed this action as addressed because Congress froze the program funds at the fiscal year 2010 level and funded less than half of TSA’s fiscal year 2012 request for full-time behavior detection officers. Although DHS and Congress have made some progress in addressing the issues that we have previously identified that may produce cost savings or revenue enhancements, additional steps are needed. For example, in February 2012, we reported that FEMA should develop and implement a methodology that provides a more comprehensive assessment of a jurisdiction’s capability to respond to and recover from a disaster without federal assistance. As of March 2013, FEMA had not addressed this action. In addition, in the 2012 report, we suggested that Congress, working with the Administrator of TSA, may wish to consider increasing the passenger aviation security fee according to one of many options, including but not limited to the President’s Deficit Reduction Plan option ($7.50 per one-way trip by 2017) or the Congressional Budget Office, President’s Debt Commission, and House Budget Committee options ($5 per one-way trip). These options could increase fee collections over existing levels from about $2 billion to $10 billion over 5 years. However, as of March 2013, Congress had not passed legislation to increase the passenger security fee. For additional information on our assessment of DHS’s and Congress’s efforts to address our previously reported actions, see GAO’s Action Tracker . Following its establishment in 2003, DHS focused its efforts primarily on implementing its various missions to meet pressing homeland security needs and threats, and less on creating and integrating a fully and effectively functioning department. As the department matured, it has put into place management policies and processes and made a range of other enhancements to its management functions, which include acquisition, information technology, financial, and human capital management. However, DHS has not always effectively executed or integrated these functions. The department has made considerable progress in transforming its original component agencies into a single cabinet-level department and positioning itself to achieve its full potential; however, challenges remain for DHS to address across its range of missions. DHS has also made important strides in strengthening the department’s management functions and in integrating those functions across the department. As a result, in February 2013, we narrowed the scope of the high-risk area and changed the focus and name from Implementing and Transforming the Department of Homeland Security to Strengthening the Department of Homeland Security Management Functions. Of the 31 actions and outcomes GAO identified as important to addressing this area, DHS has fully or mostly addressed 8, partially addressed 16, and initiated 7. Moving forward, continued progress is needed in order to mitigate the risks that management weaknesses pose to mission accomplishment and the efficient and effective use of the department’s resources. For example: Acquisition management: Although DHS has made progress in strengthening its acquisition function, most of DHS’s major acquisition programs continue to cost more than expected, take longer to deploy than planned, or deliver less capability than promised. We identified 42 programs that experienced cost growth, schedule slips, or both, with 16 of the programs’ costs increasing from a total of $19.7 billion in 2008 to $52.2 billion in 2011—an aggregate increase of 166 percent. We reported in September 2012 that DHS leadership has authorized and continued to invest in major acquisition programs even though the vast majority of those programs lack foundational documents demonstrating the knowledge needed to help manage risks and measure performance. We recommended that DHS modify acquisition policy to better reflect key program and portfolio management practices and ensure acquisition programs fully comply with DHS acquisition policy. DHS concurred with our recommendations and reported taking actions to address some of them. Moving forward, DHS needs to, for example, validate required acquisition documents in a timely manner, and demonstrate measurable progress in meeting cost, schedule, and performance metrics for its major acquisition programs. Information technology management: DHS has defined and begun to implement a vision for a tiered governance structure intended to improve information technology (IT) program and portfolio management, which is generally consistent with best practices. However, the governance structure covers less than 20 percent (about 16 of 80) of DHS’s major IT investments and 3 of its 13 portfolios, and the department has not yet finalized the policies and procedures associated with this structure. In July 2012, we recommended that DHS finalize the policies and procedures and continue to implement the structure. DHS agreed with these recommendations and estimated it would address them by September 2013. Financial management: DHS has, among other things, received a qualified audit opinion on its fiscal year 2012 financial statements for the first time since the department’s creation. DHS is working to resolve the audit qualification to obtain an unqualified opinion for fiscal year 2013. However, DHS components are currently in the early planning stages of their financial systems modernization efforts, and until these efforts are complete, their current systems will continue to inadequately support effective financial management, in part because of their lack of substantial compliance with key federal financial management requirements. Without sound controls and systems, DHS faces challenges in obtaining and sustaining audit opinions on its financial statement and internal controls over financial reporting, as well as ensuring its financial management systems generate reliable, useful, and timely information for day-to-day decision making. Human capital management: In December 2012, we identified several factors that have hampered DHS’s strategic workforce planning efforts and recommended, among other things, that DHS identify and document additional performance measures to assess workforce planning efforts.recommendations and stated that it plans to take actions to address them. In addition, DHS has made efforts to improve employee morale, such as taking actions to determine the root causes of morale DHS agreed with these problems. Despite these efforts, however, federal surveys have consistently found that DHS employees are less satisfied with their jobs than the government-wide average. In September 2012, we recommended, among other things, that DHS improve its root cause analysis efforts of morale issues. DHS agreed with these recommendations and noted actions it plans to take to address them. In conclusion, given DHS’s significant leadership responsibilities in securing the homeland, it is critical that the department’s programs and activities are operating as efficiently and effectively as possible; that they are sustainable; and that they continue to mature, evolve, and adapt to address pressing security needs. Since it began operations in 2003, DHS has implemented key homeland security operations and achieved important goals and milestones in many areas. These accomplishments are especially noteworthy given that the department has had to work to transform itself into a fully functioning cabinet department while implementing its missions. However, our work has shown that DHS can take actions to reduce fragmentation, overlap, and unnecessary duplication to improve the efficiency of its operations and achieve cost savings in several areas. Further, DHS has taken steps to strengthen its management functions and integrate them across the department; however, continued progress is needed to mitigate the risks that management weaknesses pose to mission accomplishment and the efficient and effective use of the department’s resources. DHS has indeed made significant strides in protecting the homeland, but has yet to reach its full potential. Chairman Duncan, Ranking Member Barber, and members of the subcommittee, this concludes my prepared statement. I would be pleased to respond to any questions that members of the subcommittee may have. For further information regarding this testimony, please contact Cathleen A. Berrick at (202) 512-3404 or berrickc@gao.gov. In addition, contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this statement. Individuals who made key contributions to this testimony are Kathryn Bernet, Assistant Director; Elizabeth Luke; and Meg Ullengren. This enclosure presents a summary of the areas and actions we identified in our 2011-2013 annual reports that are relevant to the Department of Homeland Security (DHS). It also includes our assessment of the overall progress made in each of the areas and the progress made on each action that we identified in our 2011 and 2012 annual reports in which Congress and DHS could take actions to reduce or eliminate fragmentation, overlap, and potential duplication or achieve other potential financial benefits. As of April 26, 2013, we have not assessed DHS’s progress in addressing the relevant 2013 areas. Table 4 presents our assessment of the overall progress made in implementing the actions needed in the areas related to fragmentation, overlap, or duplication. Table 5 presents our assessment of the overall progress made in implementing the actions needed in the areas related to cost savings or revenue enhancement. | Since beginning operations in 2003, DHS has become the third-largest federal department, with more than 224,000 employees and an annual budget of about $60 billion. Over the past 10 years, DHS has implemented key homeland security operations and achieved important goals to create and strengthen a foundation to reach its potential. Since 2003, GAO has issued more than 1,300 reports and congressional testimonies designed to strengthen DHSs program management, performance measurement efforts, and management processes, among other things. GAO has reported that overlap and fragmentation among government programs, including those of DHS, can cause potential duplication, and reducing it could save billions of tax dollars annually and help agencies provide more efficient and effective services. Moreover, in 2003, GAO designated implementing and transforming DHS as high risk because it had to transform 22 agencies into one department, and failure to address associated risks could have serious consequences. This statement addresses (1) opportunities for DHS to reduce fragmentation, overlap, and duplication in its programs; save tax dollars; and enhance revenue, and (2) opportunities for DHS to strengthen its management functions. Since 2011, GAO has identified 11 areas across the Department of Homeland Security (DHS) where fragmentation, overlap, or potential duplication exists and 13 areas of opportunity for cost savings or enhanced revenue collections. In these reports, GAO has suggested 53 total actions to the department and Congress to help strengthen the efficiency and effectiveness of DHS operations. In GAOs 2013 annual report on federal programs, agencies, offices, and initiatives that have duplicative goals or activities, GAO identified 6 new areas where DHS could take actions to address fragmentation, overlap, or potential duplication or achieve significant cost savings. For example, GAO found that DHS does not have a department-wide policy defining research and development (R&D) or guidance directing components how to report R&D activities. Thus, DHS does not know its total annual investment in R&D, which limits its ability to oversee components R&D efforts. In particular, GAO identified at least 6 components with R&D activities and an additional $255 million in R&D obligations in fiscal year 2011 by DHS components that was not centrally tracked. GAO suggested that DHS develop and implement policies and guidance for defining and overseeing R&D at the department. In addition, GAO reported that by reviewing the appropriateness of the federal cost share the Transportation Security Administration (TSA) applies to agreements financing airport facility modification projects related to the installation of checked baggage screening systems, TSA could, if a reduced cost share was deemed appropriate, achieve cost efficiencies of up to $300 million by 2030 and be positioned to install a greater number of optimal baggage screening systems. GAO has also updated its assessments of the progress that DHS and Congress have made in addressing the suggested actions from the 2011 and 2012 annual reports. As of March 2013, of the 42 actions from these reports, 5 have been addressed (12 percent), 24 have been partially addressed (57 percent), and the remaining 13 have not been addressed (31 percent). Although DHS and Congress have made some progress in addressing the issues that GAO has previously identified, additional steps are needed to address the remaining areas to achieve associated benefits. While challenges remain across its missions, DHS has made considerable progress since 2003 in transforming its original component agencies into a single department. As a result, in its 2013 biennial high-risk update, GAO narrowed the scope of the area and changed its focus and name from Implementing and Transforming the Department of Homeland Security to Strengthening the Department of Homeland Security Management Functions. To more fully address this area, DHS needs to further strengthen its acquisition, information technology, and financial and human capital management functions. Of the 31 actions and outcomes GAO identified as important to addressing this area, DHS has fully or mostly addressed 8, partially addressed 16, and initiated 7. Moving forward, DHS needs to, for example, validate required acquisition documents in a timely manner, and demonstrate measurable progress in meeting cost, schedule, and performance metrics for its major acquisition programs. In addition, DHS has begun to implement a governance structure to improve information technology management consistent with best practices, but the structure covers less than 20 percent of DHSs major information technology investments. While this testimony contains no new recommendations, GAO previously made about 1,800 recommendations to DHS designed to strengthen its programs and operations. The department has implemented more than 60 percent of them and has actions under way to address others. |
The United States has approximately 360 commercial sea and river ports that handle more than $1.3 trillion in cargo annually. A wide variety of goods, including automobiles, grain, and millions of cargo containers, travel through these ports each day. While no two ports are exactly alike, many share certain characteristics, like their size, general proximity to a metropolitan area, the volume of cargo being processed, and connections to complex transportation networks designed to move cargo and commerce as quickly as possible, that make them vulnerable to physical security threats. Entities within the maritime port environment are also vulnerable to cyber- based threats because maritime stakeholders rely on numerous types of information and communications technologies to manage the movement of cargo throughout ports. Examples of these technologies include the following: Terminal operating systems: These are information systems used by terminal operators to, among other things, control container movements and storage. For example, the terminal operating system is to support the logistical management of containers while in the terminal operator’s possession, including container movement and storage. To enhance the terminal operator’s operations, the system can also be integrated with other systems and technologies, such as financial systems, mobile computing, optical character recognition, and radio frequency identification systems. Industrial control systems: In maritime terminals, industrial control systems facilitate the movement of goods throughout the terminal using conveyor belts or pipelines to various structures (e.g., refineries, processing plants, and storage tanks). Business operations systems: These are information and communications technologies used to help support the business operations of the terminal, such as communicating with customers and preparing invoices and billing documentation. These systems can include e-mail and file servers, enterprise resource planning systems,networking equipment, phones, and fax machines. Access control and monitoring systems: Information and communication technology can also be used to support physical security operations at a port. For example, camera surveillance systems can be connected to information system networks to facilitate remote monitoring of port facilities, and electronically enabled physical access control devices can be used to protect sensitive areas of a port. See figure 1, an interactive graphic, for an overview of the technologies used in the maritime port environment. See appendix III for a printable version. Move mouse over blue system names to get descriptions of the systems. See appendix III for noninteractive version of this graphic. The location of the entity that manages these systems can also vary. Port facility officials we interviewed stated that some information technology systems used by their facilities are managed locally at the ports, while others are managed remotely from locations within and outside the United States. In addition, other types of automated infrastructure are used in the global maritime trade industry. For example, some ports in Europe use automated ground vehicles and stacking cranes to facilitate the movement of cargo throughout the ports. Like threats affecting other critical infrastructures, threats to the maritime information technology (IT) infrastructure can come from a wide array of sources. For example, advanced persistent threats—where adversaries possess sophisticated levels of expertise and significant resources to pursue their objectives—pose increasing risk. Threat sources include corrupt employees, criminal groups, hackers, and terrorists. These threat sources vary in terms of the capabilities of the actors, their willingness to act, and their motives, which can include monetary or political gain or mischief, among other things. Table 1 describes the sources of cyber- based threats in more detail. These sources of cyber threats may make use of various cyber techniques, or exploits, to adversely affect information and communications networks. Types of exploits include denial-of-service attacks, phishing, Trojan horses, viruses, worms, and attacks on the IT supply chains that support the communications networks. Table 2 describes the types of exploits in more detail. Similar to those in the United States, ports elsewhere in the world also rely on information and communications technology to facilitate their operations, and concerns about the potential impact of cybersecurity threats and vulnerabilities on these operations have been raised. For example, according to a 2011 report issued by the European Network and Information Security Agency, the maritime environment, like other sectors, increasingly relies on information and communications systems to optimize its operations, and the increased dependency on these systems, combined with the operational complexity and multiple stakeholders involved, make the environment vulnerable to cyber attacks. In addition, Australia’s Office of the Inspector of Transport Security reported in June 2012 that a cyber attack is probably the most serious threat to the integrity of offshore oil and gas facilities and land-based production. In addition, a recently reported incident highlights the risk that cybersecurity threats pose to the maritime port environment. Specifically, according to Europol’s European Cybercrime Center, a cyber incident was reported in 2013 (and corroborated by the Federal Bureau of Investigation) in which malware was installed on a computer at a foreign port. The reported goal of the attack was to track the movement of shipping containers for smuggling purposes. A criminal group used hackers to break into the terminal operating system to gain access to security and location information that was leveraged to remove the containers from the port. Port owners and operators are responsible for the cybersecurity of their operations, and federal plans and policies specify roles and responsibilities for federal agencies to support those efforts. In particular, the National Infrastructure Protection Plan (NIPP), a planning document originally developed pursuant to the Homeland Security Act of 2002 and Homeland Security Presidential Directive 7 (HSPD-7), sets forth a risk management framework to address the risks posed by cyber, human, and physical elements of critical infrastructure. It details the roles and responsibilities of DHS in protecting the nation’s critical infrastructures; identifies agencies that have lead responsibility for coordinating with the sectors (referred to as sector-specific agencies); and specifies how other federal, state, regional, local, tribal, territorial, and private-sector stakeholders should use risk management principles to prioritize protection activities within and across sectors. In addition, NIPP sets up a framework for operating and sharing information across and between federal and nonfederal stakeholders within each sector that includes the establishment of two types of councils: sector coordinating councils and government coordinating councils. The 2006 and 2009 NIPPs identified the U.S. Coast Guard as the sector-specific agency for the maritime mode of the transportation sector.and resilience strategies for the maritime environment. In this role, the Coast Guard is to coordinate protective programs Under NIPP, each critical infrastructure sector is also to develop a sector- specific plan to detail the application of its risk management framework for the sector. The 2010 Transportation Systems Sector-Specific Plan includes an annex for the maritime mode of transportation. The maritime annex is considered an implementation plan that details the individual characteristics of the maritime mode and how it will apply risk management, including a formal assessment of risk, to protect its systems, assets, people, and goods. In February 2013, the White House issued Presidential Policy Directive 21, which shifted the nation’s focus from protecting critical infrastructure against terrorism toward protecting and securing critical infrastructure and increasing its resilience against all hazards, including natural disasters, terrorism, and cyber incidents. The directive identified sector-specific agency roles and responsibilities to include, among other things, serving as a day-to-day federal interface for the prioritization and coordination of sector-specific activities. In December 2013, DHS released an updated version of NIPP. The 2013 NIPP reaffirms the role of various coordinating structures (such as sector coordinating councils and government coordinating councils) and integrates cyber and physical security and resilience efforts into an enterprise approach for risk management, among other things. The 2013 NIPP also reiterates the sector-specific agency roles and responsibilities as defined in Presidential Policy Directive 21. In addition, in February 2013 the President signed Executive Order 13636 for improving critical infrastructure cybersecurity. The executive order states that, among other things the National Institute of Standards and Technology shall lead the development of a cybersecurity framework that will provide technology-neutral guidance; the policy of the federal government is to increase the volume, timeliness, and quality of cyber threat information sharing with the U.S. private sector; agencies with responsibility to regulate the security of critical infrastructure shall consider prioritized actions to promote cyber security; and DHS shall identify critical infrastructure where a cybersecurity incident could have a catastrophic effect on public health or safety, economic security, or national security. The primary laws and regulations that establish DHS’s maritime security requirements include the Maritime Transportation Security Act of 2002 (MTSA), the Security and Accountability for Every Port Act of 2006 (SAFE Port Act),laws. and Coast Guard’s implementing regulations for these Enacted in November 2002, MTSA requires a wide range of security improvements for protecting our nation’s ports, waterways, and coastal areas. DHS is the lead agency for implementing the act’s provisions and relies on its component agencies, including the Coast Guard and FEMA, to help implement the act. The Coast Guard is responsible for security of U.S. maritime interests, including completion of security plans related to geographic areas around ports with input from port stakeholders. These plans are to assist the Coast Guard in the protection against transportation security incidents across the maritime port environment. The Coast Guard has designated a captain of the port within each of 43 geographically defined port areas across the nation who is responsible for overseeing the development of the security plans within his or her respective geographic region. The MTSA implementing regulations, developed by the Coast Guard, require the establishment of area maritime security committees across all port areas. The committees for each of the 43 identified port areas, which are organized by the Coast Guard, consist of key stakeholders who (1) may be affected by security policies and (2) share information and develop port security plans. Members of the committees can include a diverse array of port stakeholders, including federal, state, local, tribal, and territorial law enforcement agencies, as well as private sector entities such as terminal operators, yacht clubs, shipyards, marine exchanges, commercial fishermen, trucking and railroad companies, organized labor, and trade associations. These committees are to identify critical port infrastructure and risks to the port, develop mitigation strategies for these risks, and communicate appropriate security information to port stakeholders. The area maritime security committees, in consultation with applicable stakeholders within their geographic region, are to assist the Coast Guard in developing the port area maritime security plans. Each area maritime security plan is to describe the area and infrastructure covered by the plan, establish area response and recovery protocols for a transportation security incident, and include any other information DHS requires. In addition, during the development of each plan, the Coast Guard is to develop a risk-based security assessment that includes the identification of the critical infrastructure and operations in the port, a threat assessment, and a vulnerability and consequence assessment, among other things. The assessment is also to consider, among other things, physical security of infrastructure and operations of the port, existing security systems available to protect maritime personnel, and radio and telecommunication systems, including computer systems and networks as well as other areas that may, if damaged, pose a risk to people, infrastructure, or operations within the port. Upon completion of the assessment, a written report must be prepared that documents the assessment methodology that was employed, describes each vulnerability identified and the resulting consequences, and provides risk reduction strategies that could be used for continued operations in the port. MTSA and its associated regulations also require port facility owners and operators to develop facility security plans for the purpose of preparing certain maritime facilities, such as container terminals and chemical processing plants, to deter a transportation security incident. The plans are to be updated at least every 5 years and are expected to be consistent with the port’s area maritime security plan. The MTSA implementing regulations require that the facility security plans document information on security systems and communications, as well as facility vulnerability and security measures, among other things. The implementing regulations also require port facility owners and operators, as well as their designated facility security officers, to ensure that a facility security assessment is conducted and that, upon completion, a written report is included with the corresponding facility security plan submission for review and approval by the captain of the port. The facility security assessment report must include an analysis that considers measures to protect radio and telecommunications equipment, including computer systems and networks, among other things. Enacted in October 2006, the SAFE Port Act created and codified new programs and initiatives related to the security of the U.S. ports, and amended some of the original provisions of MTSA. For example, the SAFE Port Act required the Coast Guard to establish a port security exercise program. MTSA also codified the Port Security Grant Program, which is to help defray the costs of implementing security measures at domestic ports. According to MTSA, funding is to be directed towards the implementation of area maritime security plans and facility security plans among port authorities, facility operators, and state and local government agencies that are required to provide port security services. Port areas use funding from the grant program to improve port-wide risk management, enhance maritime domain awareness, and improve port recovery and resiliency efforts through developing security plans, purchasing security equipment, and providing security training to employees. FEMA is responsible for designing and operating the administrative mechanisms needed to implement and manage the grant program. Coast Guard officials provide subject matter expertise regarding the maritime industry to FEMA to inform grant award decisions. DHS and the other stakeholders have taken limited steps with respect to maritime cybersecurity. In particular, the Coast Guard did not address cybersecurity threats in a 2012 national-level risk assessment. In addition, area maritime security plans and facility security plans provide limited coverage of cybersecurity considerations. While the Coast Guard helped to establish mechanisms for sharing security-related information, the degree to which these mechanisms were active and facilitated the sharing of cybersecurity-related information varied. Also, FEMA had taken steps to address cybersecurity through the Port Security Grant Program, but it has not taken additional steps to help ensure cyber-related risks are effectively addressed. Other federal stakeholders have also taken some actions to address cybersecurity in the maritime environment. According to DHS officials, a primary reason for limited efforts in addressing cyber- related threats in the maritime environment is that the severity of cyber- related threats has only recently been recognized. Until the Coast Guard and FEMA take additional steps to more fully implement their efforts, the maritime port environment remains at risk of not adequately considering cyber-based threats in its mitigation efforts. While the Coast Guard has assessed risks associated with physical threats to port environments, these assessments have not considered risks related to cyber threats. NIPP recommends sector-specific agencies and critical infrastructure partners manage risks from significant threats and hazards to physical and cyber critical infrastructure for their respective sectors through, among other things, the identification and detection of threats and hazards to the nation’s critical infrastructure; reduction of vulnerabilities of critical assets, systems, and networks; and mitigation of potential consequences to critical infrastructure if incidents occur. The Coast Guard completes, on a biennial basis, the National Maritime Strategic Risk Assessment, which is to be an assessment of risk within the maritime environment and risk reduction based on the agency’s efforts. Its results are to provide a picture of the risk environment, including a description of the types of threats the Coast Guard is expected to encounter within its areas of responsibility, such as ensuring the security of port facilities, over the next 5 to 8 years. The risk assessment is also to be informed by numerous inputs, such as historical incident and performance data, the views of subject matter experts, and risk models, including the Maritime Security Risk Analysis Model. However, the Coast Guard did not address cybersecurity in the fourth and latest iteration of the National Maritime Strategic Risk Assessment, which was issued in 2012. While the assessment contained information regarding threats, vulnerabilities, and the mitigation of potential risks in the maritime environment, none of the information addressed cyber- related risks. The Coast Guard attributed this gap to its limited efforts to develop inputs related to cyber threats, vulnerabilities, and consequences to inform the assessment. Additionally, Coast Guard officials stated that the Maritime Security Risk Analysis Model, a key input to the risk assessment, did not contain information regarding cyber-related threats, vulnerabilities, and potential impacts of cyber incidents. The Coast Guard plans to address this deficiency in the next iteration of the assessment, which is expected to be completed by September 2014, but officials could provide no details on how cybersecurity would be specifically addressed. Without a thorough assessment of cyber-related threats, vulnerabilities, and potential consequences to the maritime subsector, the Coast Guard has limited assurance that the maritime mode is adequately protected against cyber-based threats. Assessments of cyber risk would help the Coast Guard and other maritime stakeholders understand the most likely and severe types of cyber-related incidents that could affect their operations and use this information to support planning and resource allocation to mitigate the risk in a coordinated manner. Until the Coast Guard completes a thorough assessment of cyber risks in the maritime environment, maritime stakeholders will be less able to appropriately plan and allocate resources to protect the maritime transportation mode. MTSA and the SAFE Port Act provide the statutory framework for preventing, protecting against, responding to, and recovering from a transportation security incident in the maritime environment. MTSA requires maritime stakeholders to develop security documentation, including area maritime security plans and facility security plans. These plans, however, do not fully address the cybersecurity of their respective ports and facilities. Area maritime security plans do not fully address cyber-related threats, vulnerabilities, and other considerations. The three area maritime security plans we reviewed from the three high-risk port areas we visited generally contained very limited, if any, information about cyber-related threats and mitigation activities. For example, the three plans reviewed included information about the types of information and communications technology systems that would be used to communicate security information to prevent, manage, and respond to a transportation security incident; the types of information that are considered to be Sensitive Security Information; and how to securely handle and transmit this information to those with a need to know. However, the MTSA-required plans did not identify or address any other potential cyber-related threats directed at or vulnerabilities in the information and communications systems or include cybersecurity measures that port area stakeholders should take to prevent, manage, and respond to cyber-related threats and vulnerabilities. Coast Guard officials we met with agreed that the current set of area maritime security plans, developed in 2009, do not include cybersecurity information. This occurred in part because, as Coast Guard officials stated, the guidance for developing area maritime security plans did not require the inclusion of a cyber component. As a result, port area stakeholders may not be adequately prepared to successfully manage the risk of cyber-related transportation security incidents. Coast Guard officials responsible for developing area maritime security plan guidance stated that the implementing policy and guidance for developing the next set of area maritime security plans includes basic considerations that maritime stakeholders should take into account to address cybersecurity. Currently, the area maritime security plans are formally reviewed and approved on a 5-year cycle, so the next updates will occur in 2014 and will be based on recently issued policy and guidance. Coast Guard officials stated that the policy and guidance for developing the area security plans was updated and promulgated in July 2013 and addressed inclusion of basic cyber components. Examples include guidance to identify how the Coast Guard will communicate with port stakeholders in a cyber-degraded environment, the process for reporting a cyber-related breach of security, and direction to take cyber into account when developing a port’s “all hazard”-compatible Marine Transportation System Recovery Plan. Our review of the guidance confirmed that it instructs preparers to generally consider cybersecurity issues related to information and communication technology systems when developing the plans. However, the guidance does not include any information related to the mitigation of cyber threats. Officials representing both the Coast Guard and nonfederal entities that we met with stated that the current facility security plans also do not contain cybersecurity information. Our review of nine facility security plans from the organizations we met with during site visits confirmed that those plans generally have very limited cybersecurity information. For example, two of the plans had generic references to potential cyber threats, but did not have any specific information on assets that were potentially vulnerable or associated mitigation strategies. According to federal and nonfederal entities, this is because, similar to the guidance for the area security plans, the current guidelines for facility security plans do not explicitly require entities to include cybersecurity information in the plans. Coast Guard officials stated that the next round of facility security plans, to be developed in 2014, will include cybersecurity provisions. Since the plans are currently in development, we were unable to determine the degree to which cybersecurity information will be included. Without the benefit of a national-level cyber-related risk assessment of the maritime infrastructure to inform the development of the plans, the Coast Guard has limited assurance that maritime-related security plans will appropriately address cyber-related threats and vulnerabilities associated with transportation security incidents. Although the Coast Guard helped to establish mechanisms for sharing security-related information, the degree to which these mechanisms were active and shared cybersecurity-related information varied. As the DHS agency responsible for maritime critical infrastructure protection-related efforts, the Coast Guard is responsible for establishing public-private partnerships and sharing information with federal and nonfederal entities in the maritime community. This information sharing is to occur through formalized mechanisms called for in federal plans and policy. Specifically, federal policy establishes a framework that includes government coordinating councils—composed of federal, state, local, or tribal agencies—and encourages the voluntary formation of sector coordinating councils, typically organized, governed by, and made up of nonfederal stakeholders. Further, federal policy also encourages sector-specific agencies to promote the formulation of information sharing and analysis centers (ISAC), which are to serve as voluntary mechanisms formed by owners and operators for gathering, analyzing, and disseminating information on infrastructure threats and vulnerabilities among owners and operators of the sectors and the federal government. The Maritime Modal Government Coordinating Council was established in 2006 to enable interagency coordination on maritime security issues. Coast Guard officials stated that the primary membership consisted of representatives from the Departments of Homeland Security, Transportation, Commerce, Defense, and Justice. Coast Guard officials stated that the council has met since 2006, but had only recently begun to discuss cybersecurity issues. For example, at its January 2013 annual meeting, the council discussed the implications of Executive Order 13636 for improving critical infrastructure cybersecurity for the maritime mode. In addition, during the January 2014 meeting, Coast Guard officials discussed efforts related to the development of a risk management framework that integrates cyber and physical security resilience efforts. In 2007, the Maritime Modal Sector Coordinating Council, consisting of owners, operators, and associations from within the sector, was established to enable coordination and information sharing within the sector and with government stakeholders. However, the council disbanded in March 2011 and is no longer active. Coast Guard officials attributed the demise of the council to a 2010 presidential memorandum that precluded the participation of registered lobbyists in advisory committees and other boards and commissions, which includes all Critical Infrastructure Partnership Advisory Council bodies, including the Critical Infrastructure Cross-Sector Council, and all sector coordinating councils, according to DHS. The former chair of the council stated that a majority of the members were registered lobbyists, and, as small trade associations, did not have non-lobbyist staff who could serve in this role. The Coast Guard has attempted to reestablish the sector coordinating council, but has faced challenges in doing so. According to Coast Guard officials, maritime stakeholders that would likely participate in such a council had viewed it as duplicative of statutorily authorized mechanisms, such as the National Maritime Security Advisory Committee and area maritime security committees. As a result, Coast Guard officials stated that there has been little stakeholder interest in reconstituting the council. While Coast Guard officials stated that these committees, in essence, meet the information-sharing requirements of NIPP and, to some extent, may expand the NIPP construct into real world “all hazards” response and recovery activities, these officials also stated that the committees do not fulfill all the functions of a sector coordinating council. For example, a key function of the council is to provide national-level information sharing and coordination of security-related activities within the sector. In contrast, the activities of the area maritime security committees are generally focused on individual port areas. In addition, while the National Maritime Security Advisory Committee is made up of maritime-related private-sector stakeholders, its primary purpose is to advise and make recommendations to the Secretary of Homeland Security so that the government can take actions related to securing the maritime port environment. Similarly, another primary function of the sector coordinating council may include identifying, developing, and sharing information concerning effective cybersecurity practices, such as cybersecurity working groups, risk assessments, strategies, and plans. Although Coast Guard officials stated that several of the area maritime security committees had addressed cybersecurity in some manner, the committees do not provide a national-level perspective on cybersecurity in the maritime mode. Coast Guard officials could not demonstrate that these committees had a national-level focus to improve the maritime port environment’s cybersecurity posture. In addition, the Maritime Information Sharing and Analysis Center was to serve as the focal point for gathering and disseminating information regarding maritime threats to interested stakeholders; however, Coast Guard officials could not provide evidence that the body was active or identify the types of cybersecurity information that was shared through it. They stated that they fulfill the role of the ISAC through the use of Homeport—a publicly accessible and secure Internet portal that supports port security functionality for operational use. According to the officials, Homeport serves as the Coast Guard’s primary communications tool to support the sharing, collection, and dissemination of information of various classification levels to maritime stakeholders. However, the Coast Guard could not show the extent to which cyber-related information was shared through the portal. Though the Coast Guard has established various mechanisms to coordinate and share information among government entities at a national level and between government and private stakeholders at the local level, it has not facilitated the establishment of a national-level council, as recommended by NIPP. The absence of a national-level sector coordinating council increases the risk that critical infrastructure owners and operators would not have a mechanism through which they can identify, develop, and share information concerning effective cybersecurity practices, such as cybersecurity working groups, risk assessments, strategies, and plans. As a result, the Coast Guard would not be aware of and thus not be able to mitigate cyber-based threats. Under the Port Security Grant Program, FEMA has taken steps to address cybersecurity in port areas by identifying enhancing cybersecurity capabilities as a funding priority in fiscal years 2013 and 2014 and by providing general guidance regarding the types of cybersecurity-related proposals eligible for funding. DHS annually produces guidance that provides the funding amounts available under the program for port areas and information about eligible applicants, the application process, and funding priorities for that fiscal year, among other things. Fiscal year 2013 and 2014 guidance stated that DHS identified enhancing cybersecurity capabilities as one of the six priorities for selection criteria for all grant proposals in these funding cycles. FEMA program managers stated that FEMA added projects that aim to enhance cybersecurity capabilities as a funding priority in response to the issuance of Presidential Policy Directive 21 in February 2013. Specifically, the 2013 guidance stated that grant funds may be used to invest in functions that support and enhance port-critical infrastructure and key resources in both physical space and cyberspace under Presidential Policy Directive 21. The 2014 guidance expanded on this guidance to encourage applicants to propose projects to aid in the implementation of the National Institute of Standards and Technology’s cybersecurity framework, established pursuant to Executive Order 13636, and provides a hyperlink to additional information about the framework. In addition, the guidance refers applicants to the just-established DHS Critical Infrastructure Cyber Community Voluntary Program for resources to assist critical infrastructure owners and operators in the adoption of the framework and managing cyber risks. While these actions are positive steps towards addressing cybersecurity in the port environment, FEMA has not consulted individuals with cybersecurity-related subject matter expertise to assist with the review of cybersecurity-related proposals. Program guidance states that grant applications are to undergo a multi-level review for final selection, including a review by a National Review Panel, comprised of subject matter experts drawn from the Departments of Homeland Security and Transportation. However, according to FEMA program managers, the fiscal year 2013 National Review Panel did not include subject matter experts from DHS cybersecurity and critical infrastructure agencies—such as the DHS Office of Cybersecurity and Communications, the DHS Office of Infrastructure Protection, or the Coast Guard’s Cyber Command. As a result, the National Review Panel had limited subject matter expertise to evaluate and prioritize cybersecurity-related grant proposals for funding. Specifically, according to FEMA guidance, the proposal review and selection process consists of three levels: an initial review, a field review, and a national-level review. During the initial review, FEMA officials review grant proposals for completion. During the field review, Coast Guard captains of the port, in coordination with officials of the Department of Transportation’s Maritime Administration, review and score proposals according to (1) the degree to which a proposal addresses program goals, including enhancing cybersecurity capabilities, and (2) the degree to which a proposal addresses one of the area maritime security plan priorities (e.g., transportation security incident scenarios), among other factors. The captains of the port provide a prioritized list of eligible projects for funding within each port area to FEMA, which coordinates the national review process. In March 2014, FEMA program managers stated that cybersecurity experts were not involved in the National Review Panel in part because the panel has been downsized in recent years. For the future, the officials stated that FEMA is considering revising the review process to identify cybersecurity proposals early on in the review process in order to obtain relevant experience and expertise from the Coast Guard and other subject matter experts to inform proposal reviews. However, FEMA has not documented this new process or its procedures for the Coast Guard and FEMA officials at the field and national review levels to follow for the fiscal year 2014 and future cycles. In addition, because the Coast Guard has not conducted a comprehensive risk assessment for the maritime environment that includes cyber-related threats, grant applicants and DHS officials have not been able to use the results of such an assessment to inform their grant proposals, project scoring, and risk-based funding decisions. MTSA states that, in administering the program, national economic and strategic defense concerns based on the most current risk assessments available shall be taken into account. Further, according to MTSA, Port Security Grant Program funding is to be used to address Coast Guard-identified vulnerabilities, among other purposes. FEMA officials stated that the agency considers port risk during the allocation and proposal review stages of the program funding cycle. However, FEMA program managers stated that the risk formula and risk-based analysis that FEMA uses in the allocation and proposal review stages do not assess cyber threats and vulnerabilities. Additionally, during the field-level review, captains of the port score grant proposals according to (1) the degree to which a proposal addresses program goals, including enhancing cybersecurity capabilities, and (2) the degree to which a proposal addresses one of the area maritime security plan priorities (e.g., transportation security incident scenarios), among other factors. However, as Coast Guard officials stated, and our review of area maritime security plans indicated, current area maritime security plans generally contain very limited, if any, information about cyber- related threats. Further, a FEMA Port Security Grant Program section chief stated that he was not aware of a risk assessment for the maritime mode that discusses cyber-related threats, vulnerabilities, and potential impact. Using the results of such a maritime risk assessment that fully addresses cyber-related threats, vulnerabilities, and consequences, which—as discussed previously—has not been conducted, to inform program guidance could help grant applicants and reviewers more effectively identify and select projects for funding that could enhance the cybersecurity of the nation’s maritime cyber infrastructure. Furthermore, FEMA has not developed or implemented outcome measures to evaluate the effectiveness of the Port Security Grant Program in achieving program goals, including enhancing cybersecurity capabilities. As we reported in November 2011, FEMA had not evaluated the effectiveness of the Port Security Grant Program in strengthening critical maritime infrastructure because it had not implemented measures to track progress toward achieving program goals. Therefore, we recommended that FEMA—in collaboration with the Coast Guard— develop time frames and related milestones for implementing performance measures to monitor the effectiveness of the program. In response, in February 2014 FEMA program managers stated that the agency developed and implemented four management and administrative measures in 2012 and two performance measures to track the amount of funds invested in building and sustaining capabilities in 2013. According to a FEMA program manager, FEMA did not design the two performance measures to evaluate the effectiveness of the program in addressing individual program goals, such as enhancing cybersecurity capabilities, but to gauge the program’s effectiveness in reducing overall maritime risk in a port area based on program funding. While these measures can help improve FEMA’s management of the program by tracking how funds are invested, they do not measure program outcomes. In addition, in February 2012, we found that FEMA had efforts under way to develop outcome measures for the four national preparedness grant programs, including the Port Security Grant Program, but that it had not completed these efforts. Therefore, we recommended that FEMA revise its plan in order to guide the timely completion of ongoing efforts to develop and implement outcome-based performance measures for all four grant programs. In January 2014, FEMA officials stated that they believe that the implementation of project-based grant application tracking and reporting functions within the Non-Disaster Grant Management System will address our February 2012 recommendation that the agency develop outcome measures to determine the effectiveness of the Port Security Grant Program. However, the officials did not provide details about how these functions will address the recommendation. While the development of the Non-Disaster Grant Management System is a positive step toward improving the management and administration of preparedness grants, FEMA officials stated that the deployment of these system functions has been delayed due to budget reductions, and the time frame for building the project-based applications and reporting functions is fiscal year 2016. Therefore, it is too early to determine how FEMA will use the system to evaluate the effectiveness of the Port Security Grant Program. Until FEMA develops outcome measures to evaluate the effectiveness of the program in meeting program goals, it cannot provide reasonable assurance that funds invested in port security grants, including those intended to enhance cybersecurity capabilities, are strengthening critical maritime infrastructure—including cyber-based infrastructure—against risks associated with potential terrorist attacks and other incidents. In addition to DHS, the 2010 Transportation Systems Sector-Specific Plan identified the Departments of Commerce, Defense, Justice, and Transportation as members of the Maritime Modal Government Coordinating Council. Many agencies, including others within DHS, had taken some actions with respect to the cybersecurity of the maritime subsector. For more details on these actions, see appendix II. Disruptions in the operations of our nation’s ports, which facilitate the import and export of over $1.3 trillion worth of goods annually, could be devastating to the national economy. While the impact of a physical event (natural or manmade) appears to have been better understood and addressed by maritime stakeholders than cyber-based events, the growing reliance on information and communications technology suggests the need for greater attention to potential cyber-based threats. Within the roles prescribed for them by federal law, plans, and policy, the Coast Guard and FEMA have begun to take action. In particular, the Coast Guard has taken action to address cyber-based threats in its guidance for required area and facility plans and has started to leverage existing information-sharing mechanisms. However, until a comprehensive risk assessment that includes cyber-based threats, vulnerabilities, and consequences of an incident is completed and used to inform the development of guidance and plans, the maritime port sector remains at risk of not adequately considering cyber-based risks in its mitigation efforts. In addition, the maritime sector coordinating council is currently defunct, which may limit efforts to share important information on threats affecting ports and facilities on a national level. Further, FEMA has taken actions to enhance cybersecurity through the Port Security Grant Program by making projects aimed at enhancing cybersecurity one of its funding priorities. However, until it develops procedures to instruct grant reviewers to consult cybersecurity-related subject matter experts and uses the results of a risk assessment that identifies any cyber-related threats and vulnerabilities to inform its funding guidance, FEMA will be limited in its ability to ensure that the program is effectively addressing cyber-related risks in the maritime environment. To enhance the cybersecurity of critical infrastructure in the maritime sector, we recommend that the Secretary of Homeland Security direct the Commandant of the Coast Guard to take the following actions: work with federal and nonfederal partners to ensure that the maritime risk assessment includes cyber-related threats, vulnerabilities, and potential consequences; use the results of the risk assessment to inform how guidance for area maritime security plans, facility security plans, and other security- related planning should address cyber-related risk for the maritime sector; and work with federal and nonfederal stakeholders to determine if the Maritime Modal Sector Coordinating Council should be reestablished to better facilitate stakeholder coordination and information sharing across the maritime environment at the national level. To help ensure the effective use of Port Security Grant Program funds to support the program’s stated mission of addressing vulnerabilities in the maritime port environment, we recommend that the Secretary of Homeland Security direct the FEMA Administrator to take the following actions: in coordination with the Coast Guard, develop procedures for officials at the field review level (i.e., captains of the port) and national review level (i.e., the National Review Panel and FEMA) to consult cybersecurity subject matter experts from the Coast Guard and other relevant DHS components, if applicable, during the review of cybersecurity grant proposals for funding and in coordination with the Coast Guard, use any information on cyber- related threats, vulnerabilities, and consequences identified in the maritime risk assessment to inform future versions of funding guidance for grant applicants and reviews at the field and national levels. We provided a draft of this report to the Departments of Homeland Security, Commerce, Defense, Justice, and Transportation for their review and comment. DHS provided written comments on our report (reprinted in app. IV). In its comments, DHS concurred with our recommendations. In addition, the department stated that the Coast Guard is working with a variety of partners to determine how cyber- related threats, vulnerabilities, and potential consequences are to be addressed in the maritime risk assessment, which the Coast Guard will use to inform security planning efforts (including area maritime security plans and facility security plans). DHS also stated that the Coast Guard will continue to promote the re-establishment of a sector coordinating council, and will also continue to use existing information-sharing mechanisms. However, DHS did not provide an estimated completion date for these efforts. In addition, DHS stated that FEMA will work with the Coast Guard to develop the recommended cyber consultation procedures for the Port Security Grant Program by the end of October 2014, and will use any information on cyber-related threats, vulnerabilities, and consequences from the maritime risk assessment in future program guidance, which is scheduled for publication in the first half of fiscal year 2015. Officials from DHS and the Department of Commerce also provided technical comments via e-mail. We incorporated these comments where appropriate. Officials from the Departments of Defense, Justice, and Transportation stated that they had no comments. We are sending copies of this report to interested congressional committees; the Secretaries of Commerce, Defense, Homeland Security, and Transportation; the Attorney General of the United States; the Director of Office of Management and Budget; and other interested parties. In addition, the report is available at no charge on the GAO website at http://www.gao.gov. If you or your staff have any questions about this report, please contact Gregory C. Wilshusen at (202) 512-6244 or wilshuseng@gao.gov or Stephen L. Caldwell at (202) 512-9610 or caldwells@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix V. Our objective was to identify the extent to which the Department of Homeland Security (DHS) and other stakeholders have taken steps to address cybersecurity in the maritime port environment. The scope of our audit focused on federal agencies that have a role or responsibilities in the security of the maritime port environment, to include port facilities. We focused on the information and communications technology used to operate port facilities. We did not include other aspects of the maritime environment such as vessels, off-shore platforms, inland waterways, intermodal connections, systems used to manage water-based portions of the port, and federally managed information and communication technology. To identify federal agency roles and select the organizations responsible for addressing cybersecurity in the maritime port environment, we reviewed relevant federal law, regulations, policy, and critical infrastructure protection-related strategies, including the following: Homeland Security Act of 2002; Maritime Transportation Security Act of 2002; Homeland Security Presidential Directive 7—Critical Infrastructure Identification, Prioritization, and Protection, December 2003; Security and Accountability for Every Port Act of 2006; 2006 National Infrastructure Protection Plan; 2009 National Infrastructure Protection Plan; 2013 National Infrastructure Protection Plan; 2010 Transportation Systems Sector-Specific Plan; Presidential Policy Directive 21—Critical Infrastructure Security and Resilience, February 12, 2013; Executive Order 13636—Improving Critical Infrastructure Title 33, Code of Federal Regulations, Chapter 1, Subchapter H. We analyzed these documents to identify federal agencies responsible for taking steps to address cybersecurity in the maritime environment, such as developing a risk assessment and information-sharing mechanisms, guiding the development of security plans in response to legal requirements, and providing financial assistance to support maritime port security activities. Based on our analysis, we determined that the U.S. Coast Guard (Coast Guard) and Federal Emergency Management Agency (FEMA), within DHS, were relevant to our objective. We also included the Departments of Transportation, Defense, Commerce, and Justice as they were identified as members of the Maritime Modal Government Coordinating Council in the 2010 Transportation Systems Sector-Specific Plan. We also included other DHS components, such as U.S. Customs and Border Protection, National Protection and Programs Directorate, Transportation Security Administration, and United States Secret Service, based on our prior cybersecurity and port security work and information learned from interviews during our engagement. To determine the extent to which the Coast Guard and FEMA have taken steps to address cybersecurity in the maritime port environment, we collected and analyzed relevant guidance and reports. For example, we analyzed the Coast Guard’s 2012 National Maritime Strategic Risk Assessment, Coast Guard guidance for developing area maritime security plans, the 2012 Annual Progress Report—National Strategy for Transportation Security, the Transportation Sector Security Risk Assessment, and FEMA guidance for applying for and reviewing proposals under the Port Security Grant Program. We also examined our November 2011 and February 2012 reports related to the Port Security Grant Program and our past work related to FEMA grants management for previously identified issues and context. In addition, we gathered and analyzed documents and interviewed officials from DHS’s Coast Guard, FEMA, U.S. Customs and Border Protection, Office of Cybersecurity and Communications, Office of Infrastructure Protection, Transportation Security Administration, and United States Secret Service; the Department of Commerce’s National Oceanic and Atmospheric Administration; the Department of Defense’s Transportation Command; the Department of Justice’s Federal Bureau of Investigation; and the Department of Transportation’s Maritime Administration, Office of Intelligence, Security and Emergency Response, and the Volpe Center. To gain an understanding of how information and communication technology is used in the maritime port environment and to better understand federal interactions with nonfederal entities on cybersecurity issues, we conducted site visits to three port areas—Houston, Texas; Los Angeles/Long Beach, California; and New Orleans, Louisiana. These ports were selected in a non-generalizable manner based on their identification as both high risk (Group I) ports by the Port Security Grant Program, and as national leaders in calls by specific types of vessels— oil and natural gas, containers, and dry bulk—in the Department of Transportation Maritime Administration’s March 2013 report, Vessel Calls Snapshot, 2011. For those port areas, we analyzed the appropriate area maritime security plans for any cybersecurity-related information. We also randomly selected facility owners from Coast Guard data on those facilities required to prepare facility security plans under the Maritime Transportation Security Act’s implementing regulations. For those facilities whose officials agreed to participate in our review, we interviewed staff familiar with Coast Guard facility security requirements or information technology security, and analyzed their facility security plans for any cybersecurity-related items. We also included additional nonfederal entities such as port authorities and facilities as part of our review. The results of our analysis of area maritime security plans and facility security plans at the selected ports cannot be projected to other facilities at the port areas we visited or other port areas in the country. We also met with other port stakeholders, such as port authorities and an oil storage and transportation facility. We met with the following organizations: APM Terminals Axiall Cargill Domino Sugar Company Harris County, Texas, Information Technology Center Louisiana Offshore Oil Port Magellan Terminals Holdings, L.P. Metropolitan Stevedoring Port of Houston Authority Port of Long Beach Port of Los Angeles Port of New Orleans SSA Marine St. Bernard Port Trans Pacific Container Service We determined that information provided by the federal and nonfederal entities, such as the type of information contained within the area maritime security plans and facility security plans, was sufficiently reliable for the purposes of our review. To arrive at this assessment, we corroborated the information by comparing the plans with statements from relevant agency officials. We conducted this performance audit from April 2013 to June 2014 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objective. We believe the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objective. This appendix summarizes cybersecurity-related actions, if any, taken by other agencies of the departments identified as members of the Government Coordinating Council of the Maritime Mode related to the nonfederally owned and operated maritime port environment. Under Executive Order 13636, the Secretary of Homeland Security is to use a risk-based approach to identify critical infrastructure where a cybersecurity incident could reasonably result in catastrophic regional or national effects on public health or safety, economic security, or national security. The Secretary is also to apply consistent, objective criteria in identifying such critical infrastructure. Sector-specific agencies were to provide the Secretary with information necessary to identify such critical infrastructure. To implement Executive Order 13636, DHS established an Integrated Task Force to, among other things, lead DHS implementation and coordinate interagency and public- and private-sector efforts. One of the eight working groups that made up the task force was assigned the responsibility for identifying cyber-dependent infrastructure. Officials from DHS’s Office of Infrastructure Protection who were responsible for the working group stated that, using the defined methodology, the task force examined the maritime mode as part of its efforts. Office of Cybersecurity and Communications The Office of Cybersecurity and Communications, among other things, is responsible for collaborating with public, private, and international partners to ensure the security and continuity of the nation’s cyber and communications infrastructures in the event of terrorist attacks, natural disasters, and catastrophic incidents. One division of the Office of Cybersecurity and Communications (Stakeholder Engagement and Cyber Infrastructure Resilience) offers to partner with critical infrastructure partners—including those in the maritime port environment—to conduct cyber resilience reviews. These reviews are voluntary and are based on the CERT® Resilience Management Model, a process improvement model for managing operational resilience. They are facilitated by field-based Cyber Security Advisors. The primary goal of this program is to evaluate how critical infrastructure and key resource providers manage the cybersecurity of significant information. In addition, the Industrial Control Systems Cyber Emergency Response Team——a branch of the National Cybersecurity and Communications Integration Center division within the Office of Cybersecurity and Communications—directed the development of the Cyber Security Evaluation Tool, which is a self-assessment tool that evaluates the cybersecurity of an automated industrial control or business system using a hybrid risk- and standards-based approach, and provides relevant recommendations for improvement. We observed one maritime port entity engage with Office of Cybersecurity and Communications staff members to conduct a cyber resilience review. According to data provided by Office of Cybersecurity and Communications officials, additional reviews have been conducted with maritime port entities. In addition, three maritime port entities informed us they conducted a self-assessment using the Cyber Security Evaluation Tool. The Office of Infrastructure Protection is responsible for working with public- and private-sector critical infrastructure partners and leads the coordinated national effort to mitigate risk to the nation’s critical infrastructure. Among other things, the Office of Infrastructure Protection has the overall responsibility for coordinating implementation of NIPP across 16 critical infrastructure sectors and overseeing the development of 16 sector-specific plans. Through its Protective Security Coordination Division, the Office of Infrastructure Protection also has a network of field-based protective security advisors, who are security experts that serve as a direct link between the department and critical infrastructure partners in the field. Two nonfederal port stakeholders identified protective security advisors as a resource for assistance in cybersecurity issues. Officials from Infrastructure Protection’s Strategy and Policy Office supported the Coast Guard in developing the sector-specific plan and annual report for the maritime mode. U.S. Customs and Border Protection (CBP) is responsible for securing America’s borders. This includes ensuring that all cargo enters the United States legally, safely, and efficiently through official sea ports of entry; preventing the illegal entry of contraband into the country at and between ports of entry; and enforcing trade, tariff, and intellectual property laws and regulations. In addition, CBP developed and administered the Customs-Trade Partnership Against Terrorism program, a voluntary program where officials work in partnership with private companies to review the security of their international supply chains and improve the security of their shipments to the United States. Under this program, CBP issued minimum security criteria for U.S.-based marine port authority and terminal operators that include information technology security practices (specifically, password protection, establishment of information technology security policies, employee training on information technology security, and developing a system to identify information technology abuse that includes improper access). Among other things, the Secret Service protects the President, Vice President, visiting heads of state and government, and National Special Security Events; safeguards U.S. payment and financial systems; and investigates cyber/electronic crimes. In support of these missions, the Secret Service has several programs that have touched on maritime port cybersecurity. The Electronic Crimes Task Force initiative is a network of task forces established in the USA PATRIOT Act for the purpose of preventing, detecting, and investigating various forms of electronic crimes, including potential terrorist attacks against critical infrastructure and financial payments systems. The Secret Service also conducts Critical Systems Protection advances for protective visits. This program identifies, assesses, and mitigates any risks posed by information systems to persons and facilities protected by the Secret Service. It also conducts protective advances to identify, assess, and mitigate any issues identified with networks or systems that could adversely affect the physical security plan or cause physical harm to a protectee. The advances support all of the Secret Service’s protective detail offices by implementing network monitoring, and applying cyber intelligence analysis. Additionally, the program supports full spectrum protective visits, events, or venues domestically, in foreign countries, special events, and national special security events. In addition, Secret Service personnel in Los Angeles have engaged with maritime port stakeholders in Los Angeles and Long Beach in several ways. For example, Secret Service staff gave a general cybersecurity threat presentation to port stakeholders, though no specific cyber threats to the maritime port environment were discussed. In addition, Secret Service was requested by a local governmental entity to assist in assessing the cyber aspects of critical infrastructure. Secret Service officials stated that they are still very early on in this process and are currently working with the entity to identify the critical assets/components of the cyber infrastructure. The process is still in the information-gathering phase, and officials do not expect to release any sort of summary product until mid-2014 at the earliest. Officials stated that the end product would detail any potential vulnerabilities identified during the assessment and make recommendations for mitigation that the stakeholder could implement if it chooses. Secret Service officials also stated that an evaluation was conducted under the Critical Systems Protection Program with a maritime port stakeholder in the Houston area, but did not provide details regarding this evaluation. The Transportation Security Administration (TSA) is the former lead sector-specific agency for the transportation systems sector. TSA currently co-leads the sector with the Department of Transportation and Coast Guard, and it supports, as needed, the Coast Guard’s lead for maritime security. TSA also uses the Transportation Sector Security Risk Assessment to determine relative risks for the transportation modes. However, according to TSA officials, Coast Guard and TSA agreed in 2009 that the maritime modal risk assessment would be addressed in a separate report. TSA also established the Transportation Systems Sector Cybersecurity Working Group, whose meetings (under the Critical Infrastructure Partnership Advisory Council framework) have discussed maritime cybersecurity issues. Although components of the Department of Commerce do have maritime- related efforts under way, none are directly related to the cybersecurity of the port environment. Further, the National Institute of Standards and Technology (NIST) has not developed any specific standards related to the cybersecurity of maritime facilities within our scope. NIST has started to work with private sector stakeholders from different critical infrastructure sectors to develop a voluntary framework for reducing cyber risks to critical infrastructure, as directed by Executive Order 13636. It is developing this voluntary framework in accordance with its mission to promote U.S. innovation and industrial competitiveness. The framework has been shaped through ongoing public engagement. According to officials, more than 3,000 people representing diverse stakeholders in industry, academia, and government have participated in the framework’s development through attendance at a series of public workshops and by providing comments on drafts. On February 12, 2014, NIST released the cybersecurity framework. Though representatives from numerous critical infrastructure sectors provided comments on the draft framework, only one maritime entity provided feedback, in October 2013. The entity stated that the framework provided a minimum level of cybersecurity information, but may not provide sufficient guidance to all relevant parties who choose to implement its provisions and suggestions. Additionally, the entity stated that it found the framework to be technical in nature and that it does not communicate at a level helpful for business executives. Department of Commerce officials stated that NIST worked to address these comments in the final version of the framework. The mission of the Department of Transportation is to serve the United States by ensuring a fast, safe, efficient, accessible, and convenient transportation system that meets our vital national interest and enhances the quality of life of the American people. The department is organized into several administrations, including the Research and Innovative Technology Administration, which coordinates the department’s research programs and is charged with advancing the deployment of cross-cutting technologies to improve the nation’s transportation networks. The administration includes the Volpe Center, which partners with public and private organizations to assess the needs of the transportation community, evaluate research and development endeavors, assist in the deployment of state-of-the-art transportation technologies, and inform decision- and policy-making through analyses. Volpe is funded by sponsoring organizations. In 2011, Volpe entered into a 2-year agreement with DHS’s Control Systems Security Program to evaluate the use of control systems in the transportation sector, including the maritime mode. Under this agreement, Volpe and DHS developed a road map to secure control systems in the transportation sector in August 2012. The document discussed the use of industrial control systems in the maritime mode, and described high-level threats. It also established several goals for the entire transportation sector with near- (0-2 years), mid- (2-5 years), and long-term (5-10 years) objectives, metrics, and milestones. Volpe and DHS also developed a cybersecurity standards strategy for transportation industrial control systems, which identified tasks for developing standards for port industrial control systems starting in 2015. Volpe also conducted outreach to various maritime entities. According to Volpe officials, this study was conducted mostly at international port facilities and vessels (though U.S. ports were visited under a different program). The officials stated that the agreement was canceled due to funding reductions resulting from the recent budget sequestration. DHS officials gave two reasons why funding for Volpe outreach was terminated after sequestration. First, as part of a reorganization of the Office of Cybersecurity and Communications, there is a heightened focus on “operational” activities, and DHS characterized Volpe’s assistance under the agreement as outreach and awareness. Second, the officials stated that because the demand for incident management and response continues to grow, a decision was made to stop funding Volpe to meet spending cuts resulting from sequestration and increase funding for cyber incident response for critical infrastructure asset owners and operators who use industrial control systems. Although components of the Department of Justice have some efforts under way, most of those efforts occur at the port level. Specifically, the department’s Federal Bureau of Investigation is involved in several initiatives at the local level, focused on interfacing with key port stakeholders as well as relevant entities with state and local governments. These initiatives are largely focused on passing threat information to partners. Additionally, the Bureau’s Infragard program provides a forum to share threat information with representatives from all critical infrastructure sectors, including maritime. While the Department of Defense has recognized the significance of cyber-related threats to maritime facilities, the department has no explicit role in the protection of critical infrastructure within the maritime sub- sector. Officials also said that the department had not supported maritime mode stakeholders regarding cybersecurity. In addition, though the Department of Defense was identified as a member of the Maritime Modal Government Coordinating Council in the 2010 Transportation Systems Sector-Specific Plan, the department was not listed as a participant in the 2013 or 2014 council meetings. Further, DHS, including the U.S. Coast Guard, had not requested support from Defense on cybersecurity of commercial maritime port operations and facilities. Figure 2 provides an overview of the technologies used in the maritime port environment (see interactive fig. 1) and includes the figure’s rollover information. In addition to the contacts named above, key contributions to this report were made by Michael W. Gilmore (Assistant Director), Christopher Conrad (Assistant Director), Bradley W. Becker, Jennifer L. Bryant, Franklin D. Jackson, Tracey L. King, Kush K. Malhotra, Lee McCracken, Umesh Thakkar, and Adam Vodraska. National Preparedness: FEMA Has Made Progress, but Additional Steps Are Needed to Improve Grant Management and Assess Capabilities. GAO-13-637T. Washington, D.C.: June 25, 2013. Communications Networks: Outcome-Based Measures Would Assist DHS in Assessing Effectiveness of Cybersecurity Efforts. GAO-13-275. Washington, D.C.: April 3, 2013. High Risk Series: An Update. GAO-13-283. Washington, D.C.: February 14, 2013. Cybersecurity: National Strategy, Roles, and Responsibilities Need to Be Better Defined and More Effectively Implemented. GAO-13-187. Washington, D.C.: February 14, 2013. Information Security: Better Implementation of Controls for Mobile Devices Should Be Encouraged. GAO-12-757. Washington, D.C.: September 18, 2012. Maritime Security: Progress and Challenges 10 Years after the Maritime Transportation Security Act. GAO-12-1009T. Washington, D.C.: September 11, 2012. Information Security: Cyber Threats Facilitate Ability to Commit Economic Espionage. GAO-12-876T. Washington, D.C.: June 28, 2012. IT Supply Chain: National Security-Related Agencies Need to Better Address Risks. GAO-12-361. Washington, D.C.: March 23, 2012. Homeland Security: DHS Needs Better Project Information and Coordination among Four Overlapping Grant Programs. GAO-12-303. Washington, D.C.: February 28, 2012. Critical Infrastructure Protection: Cybersecurity Guidance Is Available, but More Can Be Done to Promote Its Use. GAO-12-92. Washington, D.C.: December 9, 2011. Port Security Grant Program: Risk Model, Grant Management, and Effectiveness Measures Could Be Strengthened. GAO-12-47. Washington, D.C.: November 17, 2011. Coast Guard: Security Risk Model Meets DHS Criteria, but More Training Could Enhance Its Use for Managing Programs and Operations. GAO-12-14. Washington, D.C.: November 17, 2011. Information Security: Additional Guidance Needed to Address Cloud Computing Concerns. GAO-12-130T. Washington, D.C.: October 6, 2011. Cybersecurity: Continued Attention Needed to Protect Our Nation’s Critical Infrastructure. GAO-11-865T. Washington, D.C.: July 26, 2011. Critical Infrastructure Protection: Key Private and Public Cyber Expectations Need to Be Consistently Addressed. GAO-10-628. Washington, D.C.: July 15, 2010. Cyberspace: United States Faces Challenges in Addressing Global Cybersecurity and Governance. GAO-10-606. Washington, D.C.: July 2, 2010. Critical Infrastructure Protection: Current Cyber Sector-Specific Planning Approach Needs Reassessment. GAO-09-969. Washington, D.C.: September 24, 2009. Cyber Analysis and Warning: DHS Faces Challenges in Establishing a Comprehensive National Capability. GAO-08-588. Washington, D.C.: July 31, 2008. Homeland Security: DHS Improved its Risk-Based Grant Programs’ Allocation and Management Methods, But Measuring Programs’ Impact on National Capabilities Remains a Challenge. GAO-08-488T. Washington, D.C.: March 11, 2008. Maritime Security: Coast Guard Inspections Identify and Correct Facility Deficiencies, but More Analysis Needed of Program’s Staffing, Practices, and Data. GAO-08-12. Washington, D.C.: February 14, 2008. Cybercrime: Public and Private Entities Face Challenges in Addressing Cyber Threats. GAO-07-705. Washington, D.C.: June 22, 2007. Risk Management: Further Refinements Needed to Assess Risks and Prioritize Protective Measures at Ports and Other Critical Infrastructure. GAO-06-91. Washington, D.C.: December 15, 2005. | U.S. maritime ports handle more than $1.3 trillion in cargo annually. The operations of these ports are supported by information and communication systems, which are susceptible to cyber-related threats. Failures in these systems could degrade or interrupt operations at ports, including the flow of commerce. Federal agencies—in particular DHS—and industry stakeholders have specific roles in protecting maritime facilities and ports from physical and cyber threats. GAO's objective was to identify the extent to which DHS and other stakeholders have taken steps to address cybersecurity in the maritime port environment. GAO examined relevant laws and regulations; analyzed federal cybersecurity-related policies and plans; observed operations at three U.S. ports selected based on being a high-risk port and a leader in calls by vessel type, e.g. container; and interviewed federal and nonfederal officials. Actions taken by the Department of Homeland Security (DHS) and two of its component agencies, the U.S. Coast Guard and Federal Emergency Management Agency (FEMA), as well as other federal agencies, to address cybersecurity in the maritime port environment have been limited. While the Coast Guard initiated a number of activities and coordinating strategies to improve physical security in specific ports, it has not conducted a risk assessment that fully addresses cyber-related threats, vulnerabilities, and consequences. Coast Guard officials stated that they intend to conduct such an assessment in the future, but did not provide details to show how it would address cybersecurity. Until the Coast Guard completes a thorough assessment of cyber risks in the maritime environment, the ability of stakeholders to appropriately plan and allocate resources to protect ports and other maritime facilities will be limited. Maritime security plans required by law and regulation generally did not identify or address potential cyber-related threats or vulnerabilities. This was because the guidance issued by Coast Guard for developing these plans did not require cyber elements to be addressed. Officials stated that guidance for the next set of updated plans, due for update in 2014, will include cybersecurity requirements. However, in the absence of a comprehensive risk assessment, the revised guidance may not adequately address cyber-related risks to the maritime environment. The degree to which information-sharing mechanisms (e.g., councils) were active and shared cybersecurity-related information varied. Specifically, the Coast Guard established a government coordinating council to share information among government entities, but it is unclear to what extent this body has shared information related to cybersecurity. In addition, a sector coordinating council for sharing information among nonfederal stakeholders is no longer active, and the Coast Guard has not convinced stakeholders to reestablish it. Until the Coast Guard improves these mechanisms, maritime stakeholders in different locations are at greater risk of not being aware of, and thus not mitigating, cyber-based threats. Under a program to provide security-related grants to ports, FEMA identified enhancing cybersecurity capabilities as a funding priority for the first time in fiscal year 2013 and has provided guidance for cybersecurity-related proposals. However, the agency has not consulted cybersecurity-related subject matter experts to inform the multi-level review of cyber-related proposals—partly because FEMA has downsized the expert panel that reviews grants. Also, because the Coast Guard has not assessed cyber-related risks in the maritime risk assessment, grant applicants and FEMA have not been able to use this information to inform funding proposals and decisions. As a result, FEMA is limited in its ability to ensure that the program is effectively addressing cyber-related risks in the maritime environment. GAO recommends that DHS direct the Coast Guard to (1) assess cyber-related risks, (2) use this assessment to inform maritime security guidance, and (3) determine whether the sector coordinating council should be reestablished. DHS should also direct FEMA to (1) develop procedures to consult DHS cybersecurity experts for assistance in reviewing grant proposals and (2) use the results of the cyber-risk assessment to inform its grant guidance. DHS concurred with GAO's recommendations. |
The standard of living of the elderly depends on total retirement income, which includes Social Security, pensions, income from assets, and earnings from employment. While Social Security provides a foundation for retirement income, savings through pension and retirement savings plans, as well as by individuals on their own behalf, can contribute substantially to ensuring a secure retirement. For example, the Social Security Administration reports that while 39 percent of income for persons 65 and over came from Social Security income in 2001, 18 percent of their income came from pensions (see fig. 1). To encourage employers to establish and maintain pension plans for their employees, the federal government provides preferential tax treatment under the Internal Revenue Code (IRC) for plans that meet certain requirements. A purpose of tax preferences for employer-sponsored pensions is to encourage savings for workers’ retirement. Pension tax preferences are structured to strike a balance between providing incentives for employers to start and maintain voluntary, tax-qualified pension plans and ensuring participants receive an equitable share of the tax-favored benefits. A qualified pension plan is a retirement plan that satisfies certain requirements set forth in the Internal Revenue Code. In order to be tax- qualified, private pension plans must satisfy a number of requirements, including minimum requirements on coverage and benefits. Private pension plans must also generally meet the requirements of the Employee Retirement Income Security Act of 1974 (ERISA). Title I of ERISA, among other requirements, contains requirements regarding information that plan sponsors must provide to participants and defines the obligations of the individuals who administer employer-sponsored plans. There has been a continuing trend from DB to DC plans over the last 2 decades. DOL reports that private sector employers sponsored over 56,000 tax-qualified DB plans in 1998 down from over 139,000 in 1979, while the number of tax-qualified DC plans sponsored by private employers more than doubled from over 331,000 to approximately 674,000 during this same period. Along with this continuing trend to sponsoring DC plans, there has also been a shift in the type of plans that private sector workers participate in. DOL reports that the percentage of private sector workers who participated in a primary DB plan has decreased from 37 percent in 1979 to 21 percent by 1998, while the percentage of such workers who participated in a primary DC plan has increased from 7 to 27 percent during this same period. Moreover, these same data show that, by 1998, the majority of active participants (workers participating in their employer’s plan) were in DC plans, whereas nearly 20 years earlier most of them were in DB plans. Employers who sponsor DB plans are responsible for making contributions that are sufficient for funding the promised benefit, investing and managing the plan assets, and bearing the investment risk because the employer, as plan sponsor, agrees to make future payments during retirement. However, under DC plans, workers bear the investment risk because there is no promise made by the employer that money will be available during retirement. Thus, as a result of this shift from DB to DC plans, an increasing share of the responsibility for providing for one’s retirement income has shifted from the employer to the employee. DB plans sponsored by private employers are required to offer joint and survivor annuities to married participants beginning at the plan’s normal retirement age. These annuity payouts, called qualified joint and survivor annuities (QJSA), guarantee a benefit for the life of the participant and the participant’s surviving spouse. DB plans may also offer a single-life annuity to unmarried participants. With respect to DC plans, there is no uniform requirement regarding benefit payouts they must offer. Rather, certain DC plans must adhere to the QJSA requirements because, similar to DB plans, they are subject to minimum funding standards. Other DC plans are not subject to the QJSA requirements if the plan provides for payment in full of the participant’s accrued benefit under the plan to the spouse on the death of the participant and the participant has not elected to receive a life annuity. Plans subject to the QJSA requirements must provide participants with a written explanation of the terms and conditions of the QJSA. As part of this notice, participants must be furnished with a description of other benefit payouts that the plan offers as options to the QJSA—including information on their features and value of a participant’s benefits under such options. In addition, the plan must provide participants with an explanation of the participant’s right to make, and the effect of, an election to waive the QJSA form of benefit, the rights of the participant’s spouse, and the right to revoke an election to waive the QJSA form of benefit. Because of concerns that participants who are offered QJSA benefits do not have adequate information to compare these benefits with other optional payouts, IRS has proposed regulations to strengthen the requirements regarding the written explanation of a QJSA that plans must provide. Specifically, the proposed regulations provide additional guidance regarding information that plans furnish to describe the value of a participant’s benefits under optional payouts compared with the value of a participant’s QJSA benefits. The comparison must show what the participant would receive under each optional payout relative to the QJSA (including for those benefit payouts that are subsidized) in a way that is meaningful. Additionally, this comparison must include information that is more readily understandable to participants. There is also required information that plans must provide retiring participants about lump sum payouts. Plans that offer a lump sum payout must provide a rollover notice to retiring participants. The rollover notice must discuss the participant’s ability to have such payouts directly transferred by the plan to another eligible retirement plan. Rollover notices must also include information about the tax consequences of choosing various payout options, such as rolling the assets to another account or taking a lump sum directly as a cash settlement. Retiring participants who have the option to receive benefits as a lump sum amount (i.e., cash settlement) may also choose to directly “roll over” their assets to another qualified retirement plan, such as an IRA. DB plans that permit lump sums must adhere to certain rules regarding the calculation of lump sum amounts. Lump sums must be at least as large as the actuarial equivalent (i.e., present value) of a participant’s accrued benefit (i.e., the value of the deferred annuity that the participant is entitled to receive or at the plan’s normal retirement age). DC plan participants also have the option to defer receipt of benefits by leaving assets in their individual accounts. Both directly rolling over assets into another qualified retirement plan and leaving benefits in the plan preserve pension assets at the point of retirement. Also, DC plan participants may have the option to receive their benefits as a series of installment payments at retirement that he or she may spend or save as desired. However, unlike a DB plan, a DC plan cannot itself provide a life annuity, but can offer to purchase an annuity from an insurance company. Retirees that do not choose or purchase annuities at the point of retirement assume personal responsibility for managing their pension and retirement savings plan assets to provide retirement income. In particular, these retirees must decide how pension assets are saved or invested and determine the timing and amount of withdrawals. Increasingly, retirees will—on average—need to balance income and expenditures over a longer period of time than in the past. This is in part due to the long-term trend towards earlier retirement throughout most of the twentieth century. Nearly half of all men now leave the labor force by age 62 and almost half of all women are out of the workforce by age 60. Moreover, the decline in the average retirement age has occurred in an environment of rising longevity for the elderly. Falling mortality rates have added almost 4 years to the expected life span of a 65-year-old man and more than 5 years to the life expectancy of a 65-year-old woman since 1940. Individuals face a variety of risks in managing their assets, income, and expenditures at and during retirement. For example, retirees may outlive their pension or retirement savings plan assets. In addition, inflation may erode the purchasing power of their income, investments may yield returns that are less than expected or decline in value, and large unplanned expenses, such as those to cover long-term care, may occur at some point during retirement. Annuities offer a means to mitigate much of the financial uncertainty that accompanies living to very old ages, but may not necessarily be the best approach for all retirees. For example, an individual with a life shortening illness might not be concerned about the financial needs that accompany living to a very old age. Also, some individuals may want to continue to accumulate assets during retirement and could invest their pension assets in IRAs or financial products, in which such assets could be more heavily invested in equities. Strategies to manage risk during retirement, when most are decumulating rather than accumulating assets will necessarily be highly individualized. The Internal Revenue Service, DOL’s Employee Benefits Security Administration (EBSA), and the Pension Benefit Guaranty Corporation are primarily responsible for enforcing laws that govern private pension plans. The Internal Revenue Service enforces provisions of the IRC that apply to tax-qualified pension plans. EBSA enforces ERISA’s reporting and disclosure provisions and fiduciary responsibility standards, which among other things concern the type and extent of information provided to plan participants. The Pension Benefit Guaranty Corporation insures the benefits of participants in certain tax-qualified private sector defined benefit plans. Recognizing the importance of retirement savings, the Congress enacted the Savings Are Vital to Everyone’s Retirement (SAVER) Act of 1997 to advance the public’s knowledge and understanding of the importance of retirement savings. The act requires DOL to, among other things, maintain an ongoing outreach program to the public to effectively promote retirement savings and to disseminate specific educational materials related to retirement savings and the principles of saving and investment. DOL’s Retirement Savings Education Campaign (RSEC), which began 2 years prior to passage of the SAVER Act, is an outreach program that targets owners of small businesses, women, minorities, and youth to change the way they think about, and act on, their retirement saving needs. As part of its campaign, DOL is partnering with outside organizations to develop informational materials and tools to help individuals understand their retirement benefit options and make informed decisions about retirement, including managing assets during retirement. DB plans are more likely to make annuities available to retiring participants because they are required to do so, while DC plans are more likely to make lump sums available. Additionally, nearly half of private sector workers who participated in a DB plan have a lump sum available at retirement, while over a third of DC plan participants have annuities available. Pension plan sponsors we spoke with provide participants with applicable notices about their benefit payout options available under the plan. Some plan sponsors we spoke with provide information beyond these notices, but this information primarily focuses on saving for retirement and not on issues related to managing pension assets at and during retirement. DB plans are more likely than DC plans to make annuities available at retirement, while DC plans are more likely than DB plans to make lump sums available. The most recent BLS data (2000) show that, not unexpectedly, all private sector workers who participated in DB plans had an annuity option available at retirement, while only 38 percent of their DC counterparts had this option. Almost all private sector workers who participated in DC plans (94 percent) had a lump sum option available and just under half (46 percent) of their DB counterparts had this option available. Additionally, over half of these workers in DC plans had an installment payment option (55 percent) available. Some private sector workers in each type of plan also had more than one benefit payment option available at retirement (see fig. 2). BLS data show that 46 percent of private sector workers in DB plans had both an annuity and lump sum option available. Also, 32 percent of all private sector workers who participated in DC plans had a lump sum, an annuity and an installment payout option available at retirement. Two surveys conducted on the incidence of payout options plans make available found similar results. These surveys indicated that almost all DC plans offer a lump sum to retiring participants, and all DB plans offer an annuity. In 2001, a study by Hewitt Associates on certain DB plans found that 40 percent of these plans offered all participants a lump sum option. A survey of certain DC plans by the Profit Sharing Council of America shows that, in 2001, about 99 percent of the DC plans surveyed offered a lump sum at retirement, 56 percent offered an installment option, and 28 percent offered an annuity. Plan sponsors we spoke with also confirmed our findings from the BLS data. Several factors may affect the benefit payout options plans made available to retiring participants. Our expert panel suggested that DC plans do not offer an annuity because of certain challenges associated with providing this payout option. For example, members of the expert panel suggested that current QJSA regulations—which require plans that offer an annuity to offer a QJSA annuity and adhere to spousal consent rules—may be administratively burdensome to plan sponsors. Also, some plans do not offer an annuity in part because of the concern about being held liable for any losses to participants in the event the annuity provider cannot meet its financial obligations. Our expert panel also identified worker preferences as an important factor affecting the pension benefit payout options plans offer to retiring participants. In part, plan sponsors offer lump sums in response to employee demand for this option and choose not to offer annuities absent employee demand for them. Also, pension experts and plan sponsors we spoke with agreed that plans offer lump sums because workers generally prefer them to annuities. A funding provider for defined contribution plans, which used to only offer annuity payouts at retirement, expanded the payout options it makes available to retiring participants in response to participants’ desire for more options and control in managing their pension and retirement savings plan assets. Plan sponsors we spoke with indicated that they provide retiring participants with applicable notices about benefit payout options available under the plan. For example, those sponsors that offer an annuity payout told us that they provide the required QJSA and spousal consent notices to participants. Also, those plan sponsors that offer a lump sum told us that they provide participants with the required rollover notice that reviews the tax consequences of choosing various payout options. Additionally, these sponsors provide retiring participants with various accompanying materials to the notices that further describe all the benefit payout options available under their plans. While plan sponsors we spoke with provided some additional information on saving for retirement, they generally did not provide information on considerations relevant to managing pension and retirement savings plan assets at and during retirement. For example, some of these sponsors provide information on investment alternatives and the potential impacts of various investment strategies on accumulating assets for retirement. Some provide calculators or annual reports that, based on a participant’s current account balance, estimate retirement income using various saving strategies and age scenarios. However, the information provided by the plan sponsors we spoke with generally does not discuss considerations relevant to managing pension and retirement savings plan assets at and during retirement. These plan sponsors generally do not discuss the potential pros and cons of available payout options as related to managing pension assets during retirement. For example, these sponsors do not provide information that shows income payments a retiring participant could receive from an annuity compared with income payments the participant might receive from personally investing and drawing down a cash settlement. Additionally, they typically do not discuss risks retirees may face in managing their assets during retirement, or provide information on how to assess needs at or during retirement. Plan sponsors are hesitant to provide information and education on managing assets during retirement because of liability concerns. Although plan sponsors are permitted to provide information and education to participants, there is no specific guidance that plan sponsors may follow to provide retiring participants with information on issues related to managing their pension assets during retirement. If a plan sponsor provides what is considered to be advice, the sponsor may be held liable for any monetary losses a participant experiences for making an unfavorable decision--with respect to choosing a benefit payout at retirement or managing pension assets—based on information they provide. Plan sponsors are, therefore, careful not to provide information that may be perceived as advice and could result in litigation if participants choose benefit payout options or assets management strategies that ultimately reduce their retirement income. A funding provider for defined contribution plans we spoke with does, however, provide information and education on potential risks retirees face and other considerations for managing assets during retirement. This organization provides information on three risks to retirement income, including eroding purchasing power due to inflation, outliving one’s pension and retirement assets, and the possibility of getting lower than anticipated investment returns due to market conditions. Participants near and at retirement also receive information to better understand how each of the various payout options they may choose from could be most useful in meeting their individual retirement income needs and preferences. Information and education is provided at key stages of an employee’s participation in the plan using multiple communication approaches, including seminars, Web-based planning tools, written materials, worksheets, and one-on-one counseling. Representatives from this organization cited several reasons for providing this type of information, including an increase in payout options available and participant demand for more information and education. When we analyzed the types of payout employees received at retirement, we found retirees increasingly selecting benefit payouts other than annuities. Although we found that about 60 percent of retirees received annuities, over time an increasing percentage of more recent retirees chose to directly roll over their lump sum benefits into an IRA or to defer their receipt by leaving them in the plan. On the basis of our statistical analysis, we found that retirees who received benefits from DB plans were most likely to receive annuities, while those who received benefits from DC plans were most likely to directly roll over these assets into an IRA or defer the receipt of benefits. We found that participation in a DB plan was the primary factor in choosing to receive an annuity. Our analysis of recently retired workers with pensions indicates that while most received annuities, many received other types of payouts. As shown in figure 3, from 1992 to 2000 about 60 percent of new retirees with pensions received an annuity. However, about 40 percent reported directly rolling over benefits into an IRA or deferring receipt by leaving benefits in their plan, and approximately 14 percent of retirees took pension assets as a cash settlement. While three-fifths of all retirees took annuities, over time an increasing percentage of more recent retirees received other types of payouts. Specifically, the percentage of all retirees who either directly rolled over benefits into an IRA or deferred their receipt increased from about 32 to 47 percent, while the percentage who received cash settlements directly from their plan changed little. In contrast, the percentage of retirees receiving annuities ranged from a peak of about 65 percent to about 57 percent. Most retirees participated in DB plans between 1992 and 2000, and payouts received by retirees with DB benefits tended to be markedly different from payouts received by retirees with DC benefits, which helps to explain why most retirees received annuities. About 77 percent of retirees with DB plan benefits received an annuity from those plans (see fig. 4), while only 8 percent of retirees with DC plan benefits received or purchased annuities with their benefits (see fig. 5). Conversely, over three-quarters of retirees with DC plan benefits directly rolled over assets into an IRA or deferred receipt of benefits by leaving assets in their plan. The growing trend towards payouts other than annuities may reflect, at least in part, the continuing trend in coverage towards DC plans. Among all retiring participants who received pension benefits, the percentage who had participated in DC plans increased considerably over time, while the percentage who had participated in DB plans decreased somewhat after peaking in 1994. Also, little change occurred in the types of payouts received by those with benefits from either DB or DC plans. For example, about 90 percent of retirees with DC plan benefits received payouts other than annuities over the entire period we examined. Similarly, during this same period, retirees with DB plan benefits received their payouts largely through annuities, with little change. One likely reason why many retirees with DB plans receive annuities is that many DB retirees do not have other payout options available. About 38 percent of the DB retirees we analyzed reported having a choice of receiving a payout other than an annuity. We narrowed our analysis to examine the payout choices made by retirees, eliminating from our analysis DB retirees with no payout choice other than an annuity. Thus, in addition to examining how all retirees receive their pensions, we also analyzed only retirees who report having a choice of receiving benefits as an annuity or as a lump sum amount. Over the period, the percentage of retirees who chose to directly roll over their lump sum benefits to an IRA or to defer receipt of benefits rose substantially, from around 44 percent of retirees to about 66 percent (see table 2 in app. II). The percentage choosing annuities and cash settlements, however, remained flat, indicating that more retirees chose multiple payouts. Additionally, we found that 64 percent of DB retirees with a choice still choose annuities over other options. As with the full sample of retirees, the changes over time among those with a choice of payout appears to be attributable largely to the trend toward participation in DC plans, as we do not observe many changes in payout choices within either plan type. Currently, pension experts and plan sponsors we spoke with told us that most retirees do not choose annuities when they have a choice of payout options. For example, one plan administrator we spoke with indicated that about two-thirds of retirees in the DB plans they administer choose payouts other than an annuity when a choice of payout options is offered. Additionally, a funding provider for DC plans reports that the percentage of their participants who chose an annuity (single or joint life) as their initial income selection fell from 78 percent to 45 percent from 1995 to 2001. Nevertheless, our analysis is not necessarily inconsistent with this information. Although we found that most retirees with pension benefits received an annuity, we also found that a growing percentage of all retirees with a choice of payout options received benefit payouts other than an annuity. Moreover, the majority of those retirees who received DC plan benefits and who had a choice of payout options choose to receive payouts other than an annuity (see table 2 in app. II). It is also possible that the receipt of lump sums from DB plans, whether as cash settlements or through directly rolling over lump sum benefits to an IRA, has increased since 2000. Recently, 30-year Treasury bond rates, which DB plan sponsors must use to determine lump-sum amounts, have fallen from their overall 1990’s levels. As a result, lump sums from DB plans have increased in value relative to participants’ annuity benefits, and retiring participants may find lump sums more attractive when they are available under their plan. We further analyzed retirees with a choice of payouts to determine statistically which factors may influence retirees to choose annuities. Not unexpectedly, participation in a DB plan was the strongest predictor of annuity choice. Also, retirees with lower total household assets (excluding pensions and other retirement assets) and retirees with more years in the workforce were more likely to choose an annuity, all else being equal. In addition, different factors seemed to influence payout choices for retirees with DC benefits as compared with those with DB benefits. For example, retirees with DC plan benefits were more likely to be influenced by the price of an annuity and by their perceived health status (with those reporting they were in better health more likely to choose an annuity), while these factors did not appear to affect retirees with DB plan benefits. Pension experts we spoke with suggested that annuities may appeal to individuals with certain characteristics or preferences, while others prefer to have control of their assets. Annuities may appeal to individuals who expect to live long, are concerned about outliving their resources in retirement, value a predictable, guaranteed income stream, or do not wish or expect to leave a bequest. However, some retirees may decline to consider annuities because such payouts are generally irrevocable, instead preferring the flexibility that other payouts offer. Such retirees may believe they can receive more income and better protect themselves against inflation by investing assets on their own. Also, some retirees may want to manage their pension assets because they have difficulty comparing the value of a lump sum amount to its equivalent annuity income stream. As such, retirees may believe lump sum amounts are worth more than income payments from annuities. Although not all retirees have the option to receive their pension as an annuity, they may purchase individual annuities to pay income during retirement directly from insurance companies. However, few retirees use their pension benefits or other assets to purchase individual life annuities. Some retirees may choose not to purchase an individual life annuity because the availability of these annuities is limited, individual life annuities have associated administrative and other expenses, and such annuities generally do not provide protection against inflation. Additionally, one individual annuity provider told us that the average premium, or one-time payment an individual makes, to purchase an individual life annuity is approximately $130,000. The demand for individual annuities may grow as the market continues to develop innovative annuity products that appeal to consumer preferences. For example, demand for individual variable annuities, which offer the potential for higher income payouts based on investment returns is growing. Additionally, as more individuals will be approaching retirement with responsibility for managing a larger share of their pension assets, the demand for individual life annuities may increase. Pension experts identified a range of actions that could help retiring participants preserve their pension and retirement savings plan assets. Some policy options would require plans to payout or offer annuities to retiring participants, while others would make it easier for plans to offer annuities at retirement, or encourage retiring participants to choose annuities. Additionally, pension experts generally agreed that information and education could be provided to participants to help them make better decisions regarding how they manage their pension assets during retirement. For example, they noted that participants need to be aware of various risks that may affect how participants manage and draw down their pension assets to provide income during retirement. At present, public education focuses primarily on saving for retirement. Some actions that could help retiring participants preserve their pension and retirement savings plan assets include options intended to increase or encourage the receipt of annuities. While annuities are not the only way plan participants can preserve their pension assets at retirement, they can provide guaranteed income throughout retirement. Thus, we asked an expert panel to identify options that could encourage more annuitization of pension and retirement savings plan assets at retirement. Some options are intended to increase the number of retiring participants who receive annuities by imposing new requirements on plan sponsors. One option is to require that all pension and retirement savings plans pay life annuities to retiring participants. Such a mandate would ensure all retiring participants have pension income for their remaining lifetimes. A variation on mandatory annuitization would make life annuities the default payout option in all DC plans. This could be achieved by requiring all tax- qualified DC plans to offer a life annuity at retirement and retiring participants to actively choose to receive some other payout (e.g., a cash settlement) instead of an annuity. While annuities would not be compulsory, it is likely that more retiring participants would choose annuities because choosing to receive some other payout would require an affirmative choice. This is because some retiring participants may not take the necessary steps to choose another type of payout available under their plan. Another somewhat less stringent option than mandatory annuitization is to require tax-qualified DC plans to offer annuities to retiring participants like DB plans are required to do. This option would provide more retirees with the opportunity to preserve retirement savings by choosing an annuity from their plan without requiring that they do so. Also, concerns that some participants might have about the expense of purchasing an individual annuity and potential difficulty in searching for an annuity product could be mitigated if annuities were available under their plan. Although these actions would increase the number of retirees who receive annuities thus ensuring retirement income throughout their lives, they also have drawbacks. For example, requiring that pension plans provide life annuities to retiring participants would reduce people’s ability to tailor the receipt of benefits to their particular circumstances. Depending on one’s individual circumstances and preferences, an annuity may not be the best payout option for managing pension assets during retirement. A retiring participant who is in poor health or needs cash to cover certain expenses may not want to receive an annuity. Also, some retirees might increase their income by rolling over benefits directly to an IRA, thus enabling them to invest and draw down their pension assets during retirement. Requiring all tax-qualified DC plans to offer annuities to retiring participants—as the default payout option or not—might not substantially increase the number of retirees who choose annuities. Our analysis shows that recent experience with retiring participants who have a choice of payout options indicates that these retirees increasingly choose to directly roll over their lump sum benefits into an IRA or defer the receipt of benefits. Also, some plan sponsors and pension experts we spoke with indicated that retiring participants generally do not choose annuities when they have other payout options available. A common drawback of such requirements on plans to offer annuities at retirement is that they could increase the administrative and regulatory burdens plan sponsors face. DC plans would have to comply with applicable laws and regulations that must be satisfied when annuities are provided, such as offering annuities that provide income for the life of the participant and spouse or beneficiary. And those plans that do not currently offer an annuity payout option at retirement would have to contract with an annuity provider. Imposing new requirements on plans to pay or offer annuities at retirement represent prescriptive approaches that do not necessarily help retiring participants understand their pension payout options or make decisions suited to their individual needs or preferences. While requiring plan sponsors to pay or offer annuities represents one set of options for increasing annuitization, other options involve modifying certain requirements to make it easier for qualified plans to offer annuities. One such modification could be providing regulatory relief to plan sponsors from potential fiduciary liability they assume in selecting an annuity provider. Because plan sponsors must generally select the safest available annuity for participants, those that do not offer annuities may be concerned about being held liable for any losses to participants in the event the annuity provider cannot meet its financial obligations. Another modification could be exempting DC plans that are not required to offer an annuity but choose to do so from having to make a joint and survivor annuity the default payout or from satisfying associated spousal consent requirements. These modifications could encourage more DC plans to offer annuities to the extent they reduce administrative or regulatory burdens that plans would incur otherwise. However, these modifications could lessen certain protections available to plan participants that receive or choose annuities. In addition to options that focus on plan sponsors, our expert panel identified other policy options that focus on encouraging more retiring participants to choose annuities or purchase them with their pension assets. For example, lowering taxes on annuity income from qualified plans could encourage some retiring participants, who would not otherwise do so, to choose or purchase an annuity. Such an incentive would not involve any new requirements on plan sponsors to payout or offer annuities at retirement. Nor does it constrain an individual’s choice because retiring participants who receive lump sum benefits could partially annuitize their pension assets and maintain some assets they could more easily access to cover immediate and/or large expenses. Also, a tax incentive for qualified plan annuities could potentially help to reduce the long-term burden on government assistance programs, such as the Supplemental Security Income and Medicaid programs, to the extent that fewer retirees deplete their assets during retirement. However, a tax incentive for income provided by qualified plan annuities could also have drawbacks. For example, such an incentive might be limited to only those assets held in qualified retirement plans. Also, annuities could be made more attractive to some retiring participants for whom another payout option might be more advantageous, such as those in ill health who need large sums of cash to cover medical expenses. Moreover, some participants, who would elect an annuity from their plan even in the absence of such an incentive, would benefit. To the extent this option encourages more retirees to choose or purchase annuities, it would result in the federal government forgoing some amount of revenue. Another option that could encourage more retiring participants to choose annuities would involve modifying the mandatory interest rate that DB plan sponsors must use to calculate lump sums. By law, DB plans must use the interest rate on 30-year Treasury bonds, which some pension experts and plan sponsors consider to be low and thus inflate the value of lump sums relative to annuities. A higher mandatory interest rate would generally decrease and potentially equalize the value of lump sums from DB plans relative to participants’ annuity benefits. As a result, smaller lump sums may not be as economically attractive to some retirees. However, the extent to which more retiring participants with DB benefits would choose annuities instead of lump sums when they are both offered is uncertain. Alternatively, the taxes that apply to lump sums (i.e., cash settlements) received directly by individuals prior to retirement could also be applied to cash settlements received by retiring participants. Currently, lump sums that are received directly by participants as cash settlements (prior to attaining age 59-½) are subject to certain taxes. Less favorable tax treatment of cash settlements, while not requiring retiring participants to take an annuity or any other type of payout, could encourage retirees to preserve their pension benefits. However, this option could be disadvantageous for some retiring participants. For example, those participants who are in poor health and need cash to pay for medical expenses may want the access to large sums of cash and flexibility that a cash settlement provides. Moreover, to the extent retiring participants have difficulty comparing the value of annuity income payments with lump sum amounts, options that seek to influence the payouts chosen by retiring participants might have limited impact. Beyond options focusing exclusively on annuities, pension experts we spoke with generally agreed that retiring participants need information and education to help them make decisions about how to manage their pension assets during retirement. While annuities reduce the risk of outliving one’s assets, they may not always be the best choice for addressing individual retirement needs and preferences. Moreover, retiring participants may have a choice of benefit payout options, and the payouts they choose may or may not address their individual retirement income needs and preferences. According to our expert panel, retiring participants need information and education on various risks that affect the level of income needed during retirement. These risks include outliving one’s assets during retirement (i.e., longevity risk) and financial risks, such as declining purchasing power of retirement income (i.e., inflation risk) that affect how retirees balance income and expenses. Almost all of the respondents from our expert panel rated information on the financial risks individuals face in retirement (96 percent) and the risk of outliving one’s assets in retirement (91 percent) as very or extremely effective in helping retiring participants make decisions about how to manage pension assets. Furthermore, a recent study by the Society of Actuaries on retirement risks indicates that both retirees and individuals approaching retirement age tend to underestimate the average life expectancy of individuals at age 65. This study also reports that 63 percent of pre-retirees and 55 percent of retirees surveyed are somewhat or very concerned about not being able to keep the value of their savings and investments growing faster than inflation. Our expert panel also noted the importance of information and education on other considerations relevant to managing pension and retirement savings plan assets during retirement. Such considerations include how to assess needs in retirement, how to compare annuity and lump sum amounts, the value of expected benefit from DB and DC plans, how annuities provide retirement income, and strategies for drawing down pension assets during retirement. At least 60 percent of our expert panel participants rated such considerations as very or extremely effective in helping retiring participants make decisions about managing their pensions during retirement. Overall, federal efforts to provide information and education on retirement planning have focused on accumulating pension assets and not on how to manage these assets to provide income throughout retirement. Under its authority to implement the SAVER Act, current DOL outreach efforts are primarily aimed at advancing public awareness and understanding about the importance of saving for retirement. For example, DOL convened two national summits focusing on challenges to saving for retirement. Also, as we previously reported, DOL conducts a range of outreach activities, including developing and distributing publications and using public service announcements. DOL has begun to broaden the focus of its education initiatives to include managing assets during retirement. For example, DOL is developing a tool kit for those near retirement that will include some information on considerations relevant to managing retirement assets during retirement. However, some pension experts told us that there is a need for more focus on managing pension and retirement savings plan assets during retirement. These experts generally agreed that the federal government could improve public awareness and understanding about issues related to managing pension assets during retirement. Also, pension experts we spoke with generally agreed that participants need information and education in several areas to help them make decisions about how to manage their pension assets during retirement. Some of these experts told us that many participants do not accurately assess the risk that they could live to very old age and have little income to meet their needs. Others indicated that retiring participants do not understand how annuities provide retirement income or how to assess retirement income needs. At present, federal pension law does not generally address managing pension and retirement savings plan assets during retirement. Disclosures plan sponsors must provide to participants about their pension benefits are intended to give them information about rights and obligations under the plan. There are no additional requirements on plan sponsors to provide information and education to participants regarding managing pensions during retirement. Also, while DOL issued regulatory guidance for plan sponsors who want to provide investment information and education to their participants, it has not issued similar guidance regarding the provision of education on retirement planning. Recognizing the need for more information on retirement planning, DOL’s Employee Retirement Income Security Act Advisory Council Working Group on Planning for Retirement issued a report that recommended DOL explore regulatory measures to encourage employers to provide retirement planning advice to their employees. The decreasing number of employer-sponsored pension plans that offer only life annuities at retirement and the increasing percentage of retiring participants who choose benefit payouts other than annuities suggest that, in the future, fewer retirees may receive pension income guaranteed to last throughout retirement. The growth in the number of DC plans, along with the increasing availability of lump sums from DB plans, means that retirees will face greater responsibility and choices for managing their pension and other assets at and throughout retirement. Depending on their choices, retirees could be at greater risk of outliving their pension and retirement savings plan assets or ultimately having insufficient income to maintain their standard of living through their retirement years. Such risks underscore the need for providing enhanced information and education to participants about their available payout options, the issues they may face in managing retirement assets, and how different options may mitigate, or increase, these risks. As part of their responsibility, retirees will have to weigh certain pros and cons of different ways to manage and preserve pension assets. Currently, the notices that plan sponsors must furnish to retiring participants are not sufficient to help them choose payout options that suit their individual circumstances, while assuring adequate levels of such income to the extent possible. Our expert panel suggested that providing several types of information, such as on risks that could affect retirement income security, could help retiring participants make more informed decisions regarding how they balance income and expenditures during retirement. To improve public awareness and understanding of important considerations related to managing pension and retirement savings plan assets at and during retirement, the Congress should consider amending ERISA so that it specifically requires plan sponsors to provide participants with a notice on risks that individuals face when managing their income and expenditures at and during retirement. Also, the Congress could consider stipulating that this notice must be provided to participants at certain key milestones, such as at enrollment in the plan, when participants receive or when changes are made to certain plan documents, when participants reach various years of service, when a participant separates from service, and/or at retirement among other instances. We provided a draft of this report to the Department of Labor and the Department of the Treasury. We received technical comments from both agencies that we incorporated as appropriate. In the draft of this report we sent to agency for review, we recommended that the Secretary of Labor direct the Assistant Secretary, Employee Benefits Security Administration, to require plan sponsors to provide participants with information on risks that individuals face when managing their income and expenditures during retirement. DOL officials said that the Secretary does not currently have the legal authority under ERISA to require plan sponsors to provide such information. Consequently, we changed our recommendation to a matter for consideration for the Congress to amend ERISA so that it requires plan sponsors to provide a notice to participants on risks that may affect an individual’s ability to manage income and expenditures at and during retirement. In addition, we received a letter from the Department of the Treasury that neither agrees nor disagrees with our findings and conclusions. Instead, the letter highlights the Administration’s proposal to replace the 30-year Treasury rate as the mandated discount rate used in many pension calculations. Of relevance to this report, the letter notes that the Administration’s proposal would affect the calculation of lump sum payments (see app. V). We are sending copies of this report to the Secretary of Labor, the Secretary of the Treasury, and interested congressional committees. We will also make copies available to others on request. In addition, the report will be available at no charge on GAO’s Web site at http://www.gao.gov. If you have any questions concerning this report, please contact me at (202) 512-7215 or George A. Scott at (202) 512-5932. Other major contributors to this report include Jeremy Citro, Mark M. Glickman, Gene Kuehneman, Luann Moy, Nyree M. Ryder, Patrick DiBattista, Joseph Applebaum, and Roger Thomas. We used a variety of data sources to examine the pension payouts plans make available to retiring participants and the benefit payouts they receive, as well as to identify what available actions could help retiring participants preserve their pension and retirement savings plan assets. We used National Compensation Survey (NCS) data from the Bureau of Labor Statistics (BLS) to determine the availability of various pension payout options. Further, we analyzed Health and Retirement Study (HRS) data covering individual respondents that retired between 1992 and 2000 to determine the pension payouts retirees receive and what factors influenced their choice of payout. Generally, the estimates in this report of the availability and the receipt of pension payouts are derived from a sample of usable responses (i.e., NCS and HRS) and therefore are subject to sampling and nonsampling errors. Sampling errors are the differences that can arise between results derived from a sample and those computed from observations of all units in the population being studied. When probability techniques are used to select a sample, statistical measures called “standard errors” can be calculated to measure possible sampling errors. Nonsampling errors also affect survey results. They can be attributed to many sources: inability to obtain information about all establishments in the sample; definitional difficulties; differences in the interpretation of questions; inability or unwillingness of respondents to provide correct information; mistakes in recording or coding the data; and other errors of collection, response, processing, coverage, and estimation for missing data. Computer edits of the data and professional review of both individual and summarized data reduce the nonsampling errors in recording, coding, and processing the data. These nonsampling errors can influence the accuracy of information presented in the report, although the magnitude of their effect is not known. Finally, we convened a virtual expert panel—using a Delphi method—to identify and evaluate the actions available to help retiring participants preserve their pension and retirement savings plan assets at and during retirement. We performed our work between August 2002 and July 2003 in accordance with generally accepted government auditing standards. BLS collects information covering incidence and detailed provisions of selected employee benefit plans as part of the NCS. The portion of the NCS from which reported estimates on employee benefits were made covers all private-sector establishments in the United States, with the exception of farms and private households. The most recent (2000) NCS obtained data from 1,436 private industry establishments, representing over 107 million workers; of this number, nearly 86 million were full-time workers and the remainder—nearly 22 million—were part-time workers. NCS collects incidence and provisions data for both defined benefit and defined contribution retirement plans. Excluded from the survey are self- employed persons, proprietors, major stockholders, members of a corporate board who are not otherwise officers of the corporation, volunteers, unpaid workers, family members who are paid token wages, the permanently disabled, partners in unincorporated firms, and U.S. citizens working overseas. BLS statistics based on NCS data are estimates derived from a sample of usable occupation quotes selected from the responding establishments. They are not tabulations based on data from all employees in private establishments within the scope of the survey. BLS did not calculate estimates of sample error for these statistics. Summary, data collection, and survey methodology information for the NCS is publicly available through the Bureau of Labor Statistics’ World Wide Web site at http://www.bls.gov/ncs/home.htm. HRS is a national panel study intended to provide data related to retirement, health insurance, saving and economic well-being. The HRS began with an initial (1992) sample of over 12,600 persons in 7,600 households. The HRS baseline is drawn from in-home, face-to-face survey interviews conducted in 1992 for the 1931-1941 birth cohort (and their spouses, if married, regardless of age); and in 1998 for newly added 1924-1930 and 1942-1947 birth cohorts. Follow-ups are administered by telephone every second year, with proxy interviews after death. Future data collections will largely replicate the 1998 HRS in design, format, coverage, structure, and measurement. Data is collected by the Institute for Social Research, University of Michigan, and is supported by funding from the National Institute on Aging (NIA), the Social Security Administration (SSA), the Department of Labor, the state of Florida Department of Elder Affairs, and the Assistant Secretary for Planning and Evaluation at the Department of Health and Human Services. HRS is an ongoing survey that plans to be continually representative of the complete U.S. population over the age of 50 by adding additional cohorts every 6 years while continuing to follow up with existing cohorts. Further information on the design, history, content, and use of HRS study components is available at http://hrsonline.isr.umich.edu/intro/sho_intro.php?hfyle=uinfo. The RAND HRS data file is a cleaned and streamlined version of the Health and Retirement Study with derived variables covering a broad, though not complete range of measures and which are named consistently across waves. NIA and SSA support the development and continued maintenance of the RAND HRS data. As of late 2001, RAND HRS data included the HRS cohort (1931-1941 birth cohort, plus spouses) and is based on 1992, 1994, and 1996 public releases and the 1998 and 2000 preliminary releases. We reviewed applicable laws and regulations to identify benefit payout options plan sponsors must and may provide at retirement and the types of accompanying information they must furnish to participants. We obtained data from NCS on the types of payout options available to participants. Specifically, we tabulated supplementary NCS data published by BLS in the Monthly Labor Review (April 2003). We recalculated the percent of participants with each payout option to include only those for which benefit options were determinable. Also, we interviewed plan sponsors and practitioners to supplement BLS data and to determine what benefit payout options DB and DC plan sponsors typically make available to participants at retirement. Further, we asked our Delphi panel to identify factors that affect the benefit payout options offered to retiring participants, as well as conducted interviews with plan sponsors and practitioners to determine some of the factors that affect the options offered. We determined what information DB and DC plan sponsors must provide to retiring participants about their benefit payout options by reviewing relevant provisions of pension laws and regulations. We also interviewed plan sponsors to determine the information plan sponsors do and do not provide to retiring participants and some factors that influence the types of information they provide. While we did not have a specific selection criteria for interviewing pension plan sponsors, we sought a range in terms of the type of company (i.e., we interviewed an insurance, a manufacturing company, a pharmaceutical, a tobacco company, a funding provider for educational institutions, and a law firm) and in the types of plans (i.e., we interviewed both DB and/or DC plan sponsors). We analyzed HRS data to determine the benefit payouts pension plan participants receive at retirement. We examined benefit payouts for 1,523 HRS respondents that reported being covered by a pension on a job they held in the preceding survey wave and left to enter retirement. This information was collected for HRS in 1994, 1996, 1998, and 2000 (survey waves 2-5) and included respondents that retired between 1992 and 2000. We used information reported by individual respondents on the type of plan they participated in and on the corresponding pension payouts received. We only examined pension payouts received at the earliest time at which a respondent reported leaving a job to retire. For example, a respondent who reported being retired in 1994, reported resuming work in 1995, and again reported retiring in 2000 would be categorized as a 1994 retiree and not as a 2000 retiree. Our analyses depend upon the accuracy of reported plan type among the recently retired who received or deferred a pension payout in connection with their recent retirement. Some experts have expressed concerns regarding the accuracy of HRS respondents with respect to pension availability and type of pension. Workers who are years away from retirement may not have good information about their plan type. To mitigate this concern, we limit our analyses to respondents who leave a job they held in the previous wave to retire. We believe that these respondents are likely to have more accurate information about their pension plans because they likely will have received recent information on their plans and payout choices. Accordingly, we confirm a respondent’s reported plan type and choice of payouts by examining the respondent’s actual payout from a DB or DC plan, and where discrepancies exist between a respondent’s plan type description and actual receipt of benefits from a type of plan we use the information from the actual receipt. There is a range of sampling errors for the estimated percentages of retirees that receive each type of pension payout as reported in tables 2-4. Except as noted in these tables, all estimated percentages had sampling errors less than plus or minus 6 percentage points at the 90 percent confidence level. To determine the payouts plan participants receive at retirement, we tabulated the number and percent of participants for four benefit payout option categories. These four categories include receiving or purchasing an immediate annuity, rolling over assets directly into an Individual Retirement Account (IRA), deferring receipt of benefits by leaving them in the plan, or receiving benefits directly from their plan as a cash settlement directly from their plan (i.e., lump sum amount). We tabulated figures for the receipt of pension payouts, as well as payouts elected by those retirees with a choice of payout options, for HRS waves two through five. The numbers and percent receiving any given pension payout may exceed the totals because individuals may have more than one pension and because respondents may receive more than one payout from a pension (e.g., a respondent with a DB pension may take a partial cash settlement and receive an annuity for the remainder). There are payout categories where the effect on receipt of pension benefits differs between retirees with benefits from DB plans and those with benefits from DC plans. The payout category “deferring benefit receipt” for retirees with benefits from DB plans generally means delaying the receipt of an annuity, while for retirees with benefits from DC plans this payout category generally means maintaining the DC account balance with the plan sponsor. Also, while retirees with annuities from DB plans receive an immediate annuity from their plan sponsor, retirees with annuity payouts from DC plans may have converted their account balance into an annuity through their pension plan sponsor or used their account balance to purchase an annuity privately. The HRS data on annuity payouts received by retirees with DC plan benefits do not permit us to determine whether these annuities are from plans or were purchased privately. A further possibility is that retirees with DC plan benefits that receive a cash settlement and privately purchase an annuity might categorize this payout as a cash withdrawal or as an annuity. We categorize DC participants pension payouts based on the retiree’s survey responses. We further tabulated pension payouts separately for DB and DC pensions. We excluded 49 respondents from this tabulation because their survey responses did not distinguish whether the pension payouts they reported corresponded to their DB or DC pension. We tabulated DB and DC pension payouts using the same four categories as for the overall tabulations. Additionally, we tabulated benefit payouts for retirees with a choice of benefit payouts. This group includes all retirees participating in a DC pension as well as DB participants that reported having a choice or demonstrated having such a choice by receiving all or part of their DB pension in a form other than an annuity. We also conducted logistic regressions on retiree payout choices to evaluate factors that might influence retirees to choose an annuity versus other payouts. We augmented the main HRS information with accompanying information from the RAND HRS data set. Survey sample weights were used throughout our analysis because HRS data is collected from a stratified sample. We used the individual sampling weights from the first survey wave in our regressions and to calculate associated standard errors. We used STATA software to estimate logistic regression parameters and associated standard errors. Results of our regression analyses are presented in appendix II. To supplement results from our analysis of HRS data, we interviewed plan sponsors and practitioners to obtain testimony and data on the payouts participants receive at retirement, as well as the benefit payouts retiring participants choose when offered a choice of payout options. In addition, we obtained plan sponsors and practitioners’ views on why retiring participants choose (or do not choose) certain payout options, such as annuities or lump sums. We also asked our Delphi panel to identify the most significant factors that affect the payout options retiring participants elect. We convened a virtual panel on the Internet of 27 experts in the area of pensions and retirement to address the third study objective. The panelists were asked to identify factors affecting benefit payout options offered to and/or elected by retirees, policy options that could encourage more annuitization of pension and retirement plan savings, and the role that education and information could play in helping retirees make optimal decisions about retirement income management. We employed a modified version of the Delphi method to organize and gather opinions from experts in the area of pensions and retirement using a Web-based forum. The panel was selected from a list of experts, including from participants in the Comptroller General’s Retirement Advisory Panel, referrals from interviews, experts cited in the literature, and representatives of other important players in the pension and retirement field. To ensure we had a range of views, we asked participants from several different backgrounds including: academic, practitioners, legal experts, plan sponsors, consumer and public interest groups, and insurance providers, to participate in our survey. Of the 30 experts we contacted, 27 agreed to participate. The identity of respondents, as well as their comments and answers, remained anonymous to other participants. Our Delphi process entailed 3 questionnaire phases. Phase I asked the panel to identify the most significant factors that affect pension and retirement savings plan benefit payout options offered to and elected by retiring participants; identify options that could be considered to encourage more annuitization of pension and retirement plan savings and the likely effects and tradeoffs of these options; and discuss the role of information and education. Phase II presented 7 follow-up questions where respondents were asked to either rank or rate the responses from phase I (all responses were included in follow-up questions). The Phase III survey provided panelists with some of the key findings from phase II and solicited their feedback about these findings. Phase III also asked the panel to identify options to encourage retiring participants to preserve their pension assets at retirement by deferring the receipt of benefits (i.e., leaving assets in an account balance), or rolling over assets directly to an IRA at retirement. A full discussion of this expert panel, including the process we employed and methodology, and highlights of results from phases I and II are presented in appendix III. A copy of the phase II questionnaire can be viewed at http://www.gao.gov/cgi-bin/getrpt?gao-03- 990sp. This appendix presents more detailed descriptive statistics for our analysis of the relationship between the choices to receive an annuity versus other pension payouts. It includes further discussion of pension payout categories and retiree choice, characteristics of retirees by payout choice, and regression statistics. For each respondent in the Health and Retirement Study (HRS) who reports leaving a job to retire, HRS collects information on how the respondent received his DB and/or DC pension payouts. Retirees with DB pensions are asked whether they (1) expect future benefits, (2) are receiving benefits now, (3) received a cash settlement, (4) rolled benefits into an IRA, or (5) lost benefits. Retirees with DC pensions are asked whether they (1) withdrew the money, (2) rolled assets into an IRA, (3) left plan assets to accumulate, or (4) converted assets to an annuity. We characterize these pension payouts in four categories: Annuities include DB respondents receiving benefits now and DC respondents who converted assets to an annuity. Cash settlement includes DB respondents who received a cash settlement and DC respondents who withdrew the money from their plan. Direct rollover into IRA and/or deferred benefits includes DB and DC respondents who rolled plan assets into an IRA directly from their plan, includes DB respondents who expect future benefits and includes DC respondents who left plan assets to accumulate. We used these categories for all HRS retirees receiving one or more pension payout. These payouts are reported in table 1. In addition to tabulating the form in which retirees receive their pensions, we also analyzed the pension payouts received by retirees who had a choice among different pension payout options. We identify this subset of retirees from HRS answers about the available options for payout of a pension associated with the job from which a respondent retired. For DB participants, we used information from HRS waves prior to retirement, since questions about options for pension payouts are asked only when the respondent has a current job, not retrospectively about a job from which a respondent has retired. We define “choice” as having the option to take a pension as either a lump sum amount (i.e., as a cash settlement or as a direct rollover to an IRA) or as an annuity. We include in this definition all DC participants. This is because all DC participants have the option to take a rollover IRA, almost all have the option to take a cash settlement, and all have the option of purchasing an annuity on the private market. We also include DB participants who annuitize and who report in the prior wave that they had the option of taking a pension as a lump sum or in installments. Additionally, we consider DB participants whom we observe a cash settlement or IRA rollover to have had a choice, because almost all DB plans must offer an annuitized payout of benefits. Thus, the only retirees we categorized as not having a disposition choice are DB participants who elect to receive an annuity and who, in the previous wave, report that they did not have a lump sum option. Using our definitions of choice, we analyzed pension payouts for those retirees who choose the form of their pension over other available forms. These payouts are reported in table 2. For retirees who made different pension payout choices, we calculated means for several descriptive variables for each category. We included only those retirees with a choice of payouts and present the means for three categories: (1) those who chose an annuity; (2) those who did not choose an annuity; and (3) all retirees with a choice of payout. We performed logistical regressions to ascertain the contributions of different factors to the probability of choosing an annuity. Specifically, we calculated logistic regressions of the form: log(p/1-p) = β*X + ε. Where the β are coefficients that represent the effect that our explanatory variables have on the log odds of having an annuity versus not having an annuity, and X represents a series of retiree characteristics; and ε an error term. Only retirees with information available on all explanatory variables were included in these regressions. We calculated this regression for all retirees, and separately for those who took a pension from a DB plan and those who took a pension from a DC plan (see tables 4, 5, and 6). This appendix presents the results from the expert panel on options to encourage the preservation of pension and retirement savings. Included here are the questions and some of the results from the three questionnaires that were completed by members of the panel selected for this study (referred to as “phase I,” “phase II,” and “phase III”). We obtained a pledge of confidentiality from our requesters that they would not request any of the responses obtained during this Delphi survey process. A complete set of descriptive statistics from the survey can be found at http://www.gao.gov/cgi-bin/getrpt?gao-03-990sp. We administered the questionnaires for phases I and II over the Internet; we administered phase III via E-mail. In the first phase of the expert panel, which ran from February 11 to February 28, 2003, we asked the panelists to respond to three open-ended questions about the preservation of pension and retirement plan savings. We developed these questions based on our study objectives. We pre- tested the questions on the on-line version with two individuals to ensure that the questionnaire (1) was clear and unambiguous, (2) did not place undue burden on individuals completing it, and (3) was independent and unbiased. We made relevant changes before we deployed the first questionnaire to all participants on the Internet. Phase I consisted of open-ended questions on themes related to the preservation of pension and retirement plan savings. The questions addressed the following themes. 1. Factors that affect pension and retirement savings plan benefit payout options offered to and elected by retiring participants. 2. Options that could be considered to encourage more annuitization of pension and retirement plan savings and the likely effects and tradeoffs of these options. 3. The role of information and education in managing pension and retirement plan savings during retirement. After panelists completed the first questionnaire, we performed a content analysis on the responses to the open-ended questions in order to compile a list of the most important factors affecting preserving pension and retirement savings, as well as identify options that may encourage more annuitization of pension and retirement assets, and the type of education and information that could assist retirees in making optimal decisions regarding their retirement income. We coded panelists’ responses, and similar responses were given the same code. To maintain standards of methodological integrity, two team members coded each of the participant’s responses together and, when necessary, codes were updated to reflect participants’ responses. Any disagreements in coding decisions were discussed until consensus was reached. We had a third person review some of the coded responses to ensure that our coding decisions were valid. We contacted respondents, if necessary, when a response was unclear. We reviewed and coded answers to each of the three questions to develop close-ended questions for phase II of the survey. Twenty-four of the 27 panelists selected completed phase I of the survey (about 89 percent response rate). Those that did not complete this phase were dropped from subsequent phases. Below are lists of the categories for the responses from the phase I open-ended questions. Categories are presented in order of frequency from most frequently to least frequently provided responses for each of the questions. Worker preferences for the type of plan they want and /or how they receive benefits [employers include certain benefits in the plan because workers want them or workers prefer to receive benefits in a certain way]. Lack of consumer knowledge/understanding about annuitization and/or key risks they will face in retirement. Challenges to offering an annuity, such as administrative cost/burden, or compliance with applicable rules (including QJSA, PBGC premiums, etc.). Adequacy of available annuity product types (i.e., variety, pricing, value of payments, lack of inflation protection, etc.). The value of lump sums from DB plans has increased [low 30-year Treasury rate makes lump sums more valuable]. Trends in types of employer-sponsored plans. Participants’ expectations about needs in retirement (e.g., income, expenses, longevity). Preferences of owners/executives who start plans. Individuals believe they can do better managing the money than with an annuity. The role of financial advisors [financial planners prefer lump sums because they receive better commission]. Changes in plan design/features within plans. Impact of laws and regulations on employer decisions (i.e., impact of ERISA or the tax code). Trends in workforce demographics and retirement [greater worker mobility, people are living longer]. Changes in the availability and/or election of various payout options. Amount of retirement/saving plan assets of future retirees. Concerns and trust issues about annuity providers and/or employers ability to provide annuity payments (i.e., solvency issues). Widespread media, investment community, and employee focus on account balances [the focus for pensions have been on saving and accumulating]. Competitive pressures: attract workers, minimize/stabilize costs. Workers/retirees already have annuity income (i.e., from social security, from a DB plan, from a spouse’s plan). Household decisions about retirement income. Bequest motives. Retirees/employees don’t have adequate information to make benefit elections. Participants’ lack of understanding about the value of certain DB plan benefit features (e.g., early retirement subsidies). PBGC guaranties qualified DB annuity payouts. Participants do not understand investments and/or how to invest their retirement savings. The taxation of distributions from various types of retirement plan vehicles. Change in employee attitude about employers’ role in providing retirement security. Concerns of higher-income DB participants about potential loss of benefits as a result of PBGC guarantee limits. Inertia-the real and/or perceived cost of changing the status quo in terms of options offered. Increase information and education to participants/ retirees. Provide tax incentives for employees who receive qualified annuity income (i.e., favorable tax treatment of annuity income). Mandating pension/retirement saving plan benefits be paid as annuities (partial or full). Change related regulations (e.g., interest rate for DB lump sum calculations, PBGC premium requirements, etc.) that affect pension obligations or payout options. Require qualified DC plans to offer an annuity option. Modify rules/regulations that currently apply when plans offer an annuity (e.g., limit QJSA provisions). Mandating qualified DC plans offer an annuity as a default option of pension benefits (i.e., apply QJSA provisions). Have PBGC or another government agency provide annuities to employers and/or employees (i.e., as a competitor to provide or sell annuities). Develop more adequate annuity products (not a policy option per se). Provide tax incentives for employers and/or insurance providers to provide annuities to retirees. Apply the same tax penalties for taking a lump sum at retirement as are applied for pre-retirement lump sum distributions. Simplify various DB plan rules to level the playing field with DC plans. Amend ERISA Investment Advisor rules to clarify that plan sponsors may provide information/education on managing income during retirement. Change benefit portability rules/regulations. Allow employer plans to distribute a certain amount of pension benefits as annuity income and the remainder with participant discretion. Allow plan sponsors or employers to form or join purchasing pools to offer annuities. Set minimum standards for state insurance guaranty funds. Enable government to act as an insurer for commercial annuity providers (i.e., federal guaranty program). Require pension/retirement plans that allow retirees to elect lump sums to also offer the option to annuitize some benefits at a later date. Require pension/retirement plans offering distributions in the form of an annuity to offer an inflation-indexed annuity option. Require all DC plans that do not normally pay out in the form of an annuity to roll out all lump sum distributions to a new type of IRA that pays benefits in the form of a J&S annuity. Helping participants to understand longevity risk (i.e., risk of outliving assets). Strategies/advice for managing retirement income during retirement (i.e., decumulation). Helping participants/retirees understand financial risks that they will face in retirement (e.g., inflation, lower standard of living, investment). Helping participants assess needs in retirement (i.e., health, income, etc.). Annuities-what are they? How do they work?, etc. Improving financial literacy. Payout options plans make available to retiring participants (e.g., description and/or value of retirement benefits under available options). Seeking financial “advice,” and other resources for retirement income planning. How to project potential retirement income from pensions/retirement plan savings. The value of expected DB and/or DC plan benefits (i.e., what a participant’s accumulation is likely to provide). How to compare annuity and lump sum amounts (i.e., how to compare equivalent amounts). De-emphasize information and education on investing/investments vis-à- vis retirement income needs. Available annuity products employers could offer. The tradeoffs of extending one’s working life. The pricing of annuity products (i.e., administrative fees). How guaranteed lifetime income from a participant’s retirement plan could enhance government provided retirement income. How various types of retirement savings plans are taxed. How to take inventory of retirement income sources. We analyzed the responses to the questions above to develop the phase II questionnaire. The purpose of the second phase was to provide the panelists with the opportunity to consider the other panelists’ responses to the first phase and to respond in a structured, quantifiable way. Phase II, which ran from April 3 to April 18, 2003, consisted of several closed-ended questions on the categorized responses to phase I (all response codes/categories were included in follow-up questions). In phase II, panelists rated these items on various dimensions (e.g., major/minor factor, effectiveness of options, help/hinder coverage, ease of compliance) depending on the theme. We also asked the experts to rank responses to phase I questions one and three. We pretested the questions for the second phase; using the same methods as in phase I. Twenty-two of the 24 panelists that completed the phase I survey also completed phase II (about 92 percent response rate for those included in phase II). Those that did not complete this phase were dropped from subsequent phases. As part of the analysis, we calculated the frequency of responses to identify the highest rated items for phase II. The results in this section are displayed based on responses that were rated in the top two rating categories for questions 1 and 3-6, as well as the top five responses identified most frequently in the top five for questions 2 and 7. To be included in the top five for the rating questions, at least 85 percent of panelists had to respond to the question. For the questions with a five- point scale, we collapsed the scale to a three-point scale by combining the top two available responses and combing the bottom two available responses. For example, if the five-point scale included extremely effective, very effective, moderately effective, somewhat effective, slightly or not effective, the three-point scale will be: extremely or very effective; moderately effective; and somewhat, slightly or not effective. For the ranking questions (2 and 7), we identified the most frequent responses ranked in the top five by calculating the frequency in which they were in the top five. We report the top five responses for all phase II questions in this appendix. In the phase I questionnaire, we asked each member of the panel “What do you consider to be the top 5 factors in pensions and retirement affecting the payout options offered to retiring participants and/or elected by retirees? (In your response, you might consider trends in employer pensions, worker preferences, workforce coverage and participation, retirement, the economy, or any other trends you believe are important. Please identify the most significant first).” We compiled a list of the factors that experts identified and categorized them. We then presented the list of factors to the experts in phase II and asked them to rate how great a factor, if at all, are each of the trends were in affecting payout options offered to retiring participants and/or elected by retirees. The ratings were made on a four-point scale ranging from “major factor” to “not a factor” (panelists were also given the option of responding “no answer”). We also asked panelists to rank the factors identified as at least moderate in question 1. Responses in the top five for the question, “among the factors that you checked as ‘at least moderate,’ what would you rank as the top 5 factors affecting plan payout options offered and/or elected by retiring participants?” are shown in table 8. In phase I, we asked panelists: “What options, if any, could policymakers consider that could encourage more annuitization of pension and retirement plan savings at retirement? What are the likely effects and tradeoffs associated with each of these options with respect to plan sponsors, participants, the pensions and investment community, and the federal government? (Please consider such options as mandates, incentives, other government actions, information and education, etc. in your response.)” After categorizing responses to this question, we asked the following series of questions in phase II. The ratings were made on a five-point scale for each of these questions (panelists were also given the option of responding “no answer”). 1. How effective, if at all, would each of the following options be in encouraging more annuitization of pension and retirement plan savings? 2. In your opinion, would the following options help or hinder pension and retirement plan coverage? 3. How easy or difficult would it be for plan sponsors to comply with and/or act on the following options? In phase I, we asked panelists, “What types of information and education could help retiring participants make more optimal decisions regarding the use (i.e., saving and spending) of pension and retirement plan savings during retirement? How and in what form could each type of information or education be delivered?” After categorizing the responses to that question, we followed up with a rating and ranking question about the effectiveness of each type of information and education. We then presented the list of types to the experts in phase II and asked them to rate how effective, if at all, each type of information and education would be in helping retiring participants make more optimal decisions. The ratings were made on a five-point scale ranging from “extremely effective” to “slightly or not effective” (panelists were also given the option of responding “no answer”). We calculated the frequency of responses for the types rated in the phase II questionnaire. We also asked panelists to rank the types of information and education identified as at least moderately effective in phase I. The top five most commonly ranked responses to the question, “Among the types of information and education that you rated ‘at least moderately effective,’ what would you rank as the five most effective types to help retirees make more optimal decisions?” are shown in table 13. The third phase, which was conducted via e-mail, ran from May 6 to May 13, 2003. The purpose of this phase was to provide panelists with some of the key findings from phase II and obtain feedback about the results, as well as to identify other ways that a retiree could preserve their retirement savings. We conducted a pretest of the questionnaire and made changes as necessary. Ten experts (45 percent of the 22 panelists that completed phase II) responded with comments or responses to our questions. In the third phase, we asked panelists the following questions. “For each of the options below please discuss what actions (policy or otherwise), if any, could encourage more retirees to preserve their pension and retirement savings plan assets. Please discuss some of the potential tradeoffs, such as the effect on plan coverage, plan compliance, and effectiveness for preserving pension and retirement savings plan assets, of the options identified.” 1. Options to encourage retiring participants to preserve their pension assets at retirement by deferring the receipt of benefits (i.e., leaving assets in an account balance), or rolling over assets directly to an IRA at retirement. 2. Options to assist retirees in managing their assets personally with the objective of providing an income stream to help them balance income and expenditures. 3. What other options, if any, should be considered to help retiring participants preserve their pension and retirement savings plan assets at retirement? Originally, we asked the panelists to respond to these three questions about actions that could encourage the preservation of pension and retirement savings plan assets. Based on feedback about the length of and time commitment needed to respond to the phase III questionnaire, we narrowed the focus and gave panelists the option of only responding to question one. Some respondents provided answers for all three of the questions and others only responded to question one. Responses to this questionnaire are presented at http://www.gao.gov/cgi-bin/getrpt?gao-03- 990sp. | The decisions that retiring workers make about how and when to draw down their pension plan assets determine in part whether they will have pension income that lasts throughout retirement. Individuals will need pension and other retirement income to sustain them over a longer period of time than in the past. Moreover, the continuing trend towards pension plans with individual accounts has increased participants' responsibility for managing their pension assets during retirement. As such, our objectives were to determine: (1) what benefit payout options and accompanying information pension plans make available to participants at retirement, (2) what benefit payouts plan participants receive at retirement, and (3) the actions available to help retiring participants preserve their pension and retirement savings plan assets. Defined benefit (DB) plans make annuities available to all participants at retirement, while defined contribution (DC) plans make lump sums available to almost all participants. Recent data also show that about half of private sector workers who participated in DB plans had a lump sum option at retirement, and over a third of their counterparts in DC plans had an annuity option. Plan sponsors GAO spoke with provide retiring participants with applicable notices about their benefit payout options available under the plan. Additional information provided by plan sponsors GAO spoke with primarily focused on saving for retirement. Risks that can affect retirement income, or other considerations relevant to managing pension assets at and during retirement were generally not discussed. According to GAO's analysis, while 60 percent of recent retirees received annuities, an increasing percentage from 1992 to 2000 directly rolled over lump sum benefits into an individual retirement account or deferred their receipt by leaving these assets in the plan, a trend in part explained by the shift toward retirees with DC plan benefits. Additionally, GAO found that a growing percentage of those retirees who reported having a choice of benefit payouts chose to directly roll over their lump sum benefits or leave benefits in the plan rather than receive annuities. Actions available to help retiring participants preserve their pension and retirement savings plan assets range from options that would encourage the receipt of annuities to providing information to help participants make better decisions about managing their pension assets at and during retirement. According to an expert panel GAO used as part of this study, retirees need to be aware of the risk of outliving one's assets in retirement and the financial risks individuals face in retirement. Over 90 percent of GAO's panelists rated providing information on such risks as very or extremely effective in helping retiring participants make decisions about managing their pension assets. |
According to IRS data, about 60,000 FCCs and about 2.3 million USCCs filed income tax returns in 1995. FCCs and USCCs may not pay U.S. income tax for a variety of reasons. For instance, some corporations may have zero tax liabilities because of current-year operating losses; losses carried forward from preceding tax years; or sufficient tax credits available to offset tax liabilities. Other corporations may report no taxable income because of the improper pricing of intercompany transactions. Any company that has a related company with which it transacts business needs to establish transfer prices for those intercompany transactions. The pricing of intercompany transactions affects the distribution of profits and ultimately the taxable income of the companies. Research efforts have attempted to explain why, on average, USCCs appear to be more profitable than FCCs. Some researchers have concluded that part of the difference is attributable to differences in the characteristics of FCCs and USCCs, while acknowledging that transfer price abuses may also explain some of the difference. The 1997 study, for example, focused on why FCCs tended to report a lower ratio of net income to gross receipts—a measure of profitability— than USCCs did. According to this study, differences in investment income, industrial classification, age, and amount of interest expense explain much of the difference in the profitability between FCCs and USCCs. That study also found that corporations whose largest foreign shareholders owned only 25 to 50 percent of the corporations’ stock had low profitability, which was similar to corporations that were 100-percent owned by a single foreign shareholder. The author suggested that, if income-shifting through transfer price abuses were an important factor in explaining the differences in profitability across corporations, then one would expect single shareholder corporations to be less profitable because they would seem to have less difficulty in shifting income between affiliates. We were unable to identify any studies that were able to control for all potential factors, other than transfer price abuse, that may explain the difference in profitability. To meet our objectives, we obtained data from IRS’ Statistics of Income (SOI) Division on corporate tax returns for 1989 through 1995. We used these data to determine the percentages of corporations that did not pay taxes in each year and to obtain information about their characteristics. We did not audit the SOI data; however, we conducted reliability tests to ensure the consistency of the data with selected FCC and USCC corporate statistics published by the SOI Division. In this report, we defined a corporation as being large if its reported total assets in tax year 1995 were at least $250 million or its reported gross receipts totaled at least $50 million. For years preceding tax year 1995, we deflated the $250 million asset and $50 million receipt definition of large corporate size by the gross domestic product price deflator for those years. We made this adjustment in the dollar magnitude of the definition because changing price levels, over time, alter the purchasing power of gross receipts and assets. Our report also compares “new” and “older” corporations. New corporations are those whose income tax returns showed incorporation dates within 3 years of the tax year date. For example, for tax year 1995, new corporations are those with incorporation dates no earlier than 1993. Other corporations are “older” corporations. A limitation on the precision of our comparison between “new” versus “older” FCCs and USCCs is that the definition of “new” relies on the dates of incorporation indicated on corporate tax returns. In some cases, corporations that merge may also reincorporate under a new corporate name. While the “new” entity may represent the combination of two mature corporations, our definition would count them as one “new” corporation. The SOI data in this report are based on SOI’s probability sample of taxpayer returns and thus are subject to some imprecision owing to sampling variability. Using SOI’s sampling weights, we estimated confidence intervals for the percentage of nontaxpaying FCCs and USCCs for tax years 1989 through 1995. We requested comments on a draft of this report from the Commissioner of Internal Revenue and the Director of the Department of the Treasury’s Office of Tax Analysis. IRS and Treasury’s comments are discussed near the end of this letter. We did our review from July through December 1998 in accordance with generally accepted government auditing standards. In each year from 1989 through 1995, a majority of corporations, both foreign- and U.S.-controlled, paid no U.S. income tax. However, in each of these years, a higher percentage of FCCs than USCCs paid no taxes. The percentage of FCCs not paying taxes ranged from 67 percent to 73 percent during those years, while the percentage of USCCs not paying taxes ranged from 59 to 62 percent, as shown in figure 1 and appendix I, table I.1. Large corporations, both FCCs and USCCs, were more likely to pay taxes than smaller corporations. Among large corporations, the percentage of FCCs that paid no tax exceeded that for USCCs from 1989 to 1993. However, in 1994, the difference between the two groups was not statistically significant, and in 1995, the percentage of large FCCs that paid no U.S. income tax was slightly less than that of large USCCs. In 1989, 33 percent of large FCCs and 27 percent of large USCCs paid no tax, while in 1995, 29 percent of large FCCs and 32 percent of large USCCs paid no tax. (See fig. 1 and app. I, table I.2.) Although nontaxpaying corporations, both foreign- and U.S.-controlled, were the majority of all corporations that filed tax returns in 1995, they accounted for well under half of all corporate assets and receipts. The 67 percent of FCCs that paid no federal income tax in 1995 accounted for 24 percent of the assets and 25 percent of the gross receipts of all FCCs in that year, as shown in figure 2. Similarly, the 61 percent of USCCs that paid no U.S. income tax in 1995 accounted for 21 percent of the assets owned by all USCCs and 17 percent of their receipts. (See also, table I.3. in app. I.) Large nontaxpaying FCCs and USCCs filed a small percentage of all returns filed by nontaxpaying corporations, yet they accounted for most of the assets of those corporations in 1995. Specifically, large nontaxpaying FCCs made up about 2 percent of all nontaxpaying FCCs but accounted for 84 percent of the assets of all nontaxpaying FCCs. Similarly, large nontaxpaying USCCs made up only about four-tenths of 1 percent of all nontaxpaying USCCs, yet they accounted for 80 percent of all nontaxpaying USCC assets. Also in 1995, large nontaxpaying FCCs accounted for 86 percent of the receipts generated by all nontaxpaying FCCs, while large nontaxpaying USCCs accounted for 48 percent of the receipts generated by all nontaxpaying USCCs. (See fig. I.1 in app. I.) The concentration of assets and receipts in the large nontaxpaying FCCs and USCCs was similar during the earlier years of our study period. This concentration was not unique to nontaxpaying corporations; taxpaying corporations were similarly concentrated. Other ways to compare large FCCs and USCCs include examining (1) the percentage of large FCCs and USCCs that paid relatively little tax and (2) the taxes paid relative to gross receipts by large corporations, as shown in table 1. In 1995, the percentage of large FCCs and USCCs that paid less than $100,000 in tax was 42 percent for FCCs and 40 percent for USCCs. In 1995, large FCCs, as a whole, paid significantly less tax per $1,000 in gross receipts than did large USCCs, despite the fact that a greater percentage of large USCCs paid no tax. The reason for this is that the large FCCs that paid relatively little or no tax had significantly greater average gross receipts than did the large USCCs that paid little or no tax. An earlier study of the relative profitability of FCCs and USCCs suggested that the lower relative age of FCCs partially explained their lower reported profitability. That study also showed that the reported profitability of both FCCs and USCCs varied across industrial sectors. From 1989 to 1993, a greater percentage of large FCCs than large USCCs were new (i.e., incorporated for 3 years or less). However, as shown in figure 3, this relationship was reversed for 1994 and 1995, and although not statistically significant, the percentage of new large USCCs exceeded the percentage of new large FCCs. This change could explain, in part, why the percentage of nontaxpaying large FCCs declined relative to the percentage of nontaxpaying large USCCs over the same time period. The IRS data that we examined for tax years 1989-95 also showed that, in each of those years, the percentage of new large corporations paying no tax exceeded the percentage of older large corporations paying no tax. This relationship held for large FCCs, large USCCs, and all large corporations together. (See table I.4 in app. I.) Large FCCs were more heavily concentrated in the manufacturing and wholesale trade sectors and less heavily concentrated in the financial services sector than were large USCCs. These differences in industrial concentration were found whether one compared large nontaxpaying FCCs to large nontaxpaying USCCs or all large FCCs to all large USCCs. Table 2 shows these comparisons for 1995. The ratios of the costs of goods sold and other costs to receipts varied significantly across industries, which could account for some of the difference between the amount of taxes that large FCCs paid per dollar of receipts and the amount that large USCCs paid. The ratio of taxable income per dollar of receipts should be inversely related to the ratio of costs per dollar of receipts. Corporations in the manufacturing, wholesale trade, and retail trade industries, on average, had significantly higher ratios of costs of goods sold to receipts than corporations in the financial and nonfinancial service industries. The largest component of costs of goods sold is purchases from other businesses, which, as table 3 on p. 10 indicates, are relatively unimportant for the two service industries. In contrast, corporations in the financial services industry, on average, had significantly higher ratios of interest expenses to receipts. This pattern of differences in cost ratios across industries was similar for both all large FCCs and all large USCCs. The pattern also was similar for large nontaxpaying FCCs and USCCs, with one exception: the ratio of interest expenses to receipts for nontaxpaying USCCs in the financial services sector was not significantly higher than that for nontaxpaying USCCs overall. (See table I.5 in app. I.) Wholesale trade, the industry with the highest ratio of costs of goods sold to receipts, had the lowest ratio of taxes paid to receipts, while financial services, the industry with the lowest ratio of costs of goods sold to receipts, had the highest ratio of taxes paid to receipts among the major industries. These relationships were similar for both large FCCs and large USCCs. The fact that a significantly larger percentage of all large USCCs were in the financial services industry and a significantly smaller percentage of them were in the wholesale trade industry (compared to large FCCs) may, in part, explain why the aggregate ratio of taxes paid to receipts, shown in table 3, was significantly higher for USCCs. Nevertheless, within each industry, the ratio of taxes paid to receipts was higher for large USCCs than for large FCCs in 1995. Moreover, in every major industry except the financial services industry, a greater percentage of large FCCs than large USCCs paid no tax at all for all the years that we examined. (See tables I.6 through I.8 in app. I.) The data in table 3 do not reveal any logical relationships across industries between (1) the ratios of either costs or taxes paid to receipts and (2) the percentage of corporations paying tax in each industry. For example, even though corporations in the financial services industry, on average, had the lowest ratio of costs of goods sold to receipts and the highest ratio of taxes paid to receipts, a higher percentage of corporations in that industry paid no tax compared with all the other industries. We requested comments on a draft of this report from the Commissioner of Internal Revenue and the Director of the Department of the Treasury’s Office of Tax Analysis. On March 8, 1999, we received comments prepared by IRS’ Chief Operations Officer through the Office of the National Director for Legislative Affairs. The Director of the Office of Tax Analysis and his staff provided comments in a February 25, 1999, meeting. Both IRS and Treasury were in overall agreement with the draft report. Both elaborated on issues we had raised, and both provided some technical comments, which we incorporated where appropriate. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copies to Senator William V. Roth, Jr., Chairman, and Senator Daniel Patrick Moynihan, Ranking Minority Member, Senate Committee on Finance; Representative Bill Archer, Chairman, and Representative Charles B. Rangel, Ranking Minority Member, House Committee on Ways and Means; Representative Amo Houghton, Chairman, and Representative William J. Coyne, Ranking Minority Member, Subcommittee on Oversight, House Committee on Ways and Means; and other interested congressional committees. We will also send copies to The Honorable Robert E. Rubin, Secretary of the Treasury; The Honorable Charles O. Rossotti, Commissioner of Internal Revenue; and other interested parties. Copies will also be made available to others upon request. This report was prepared under the direction of Charlie W. Daniel, Assistant Director. Other major contributors are listed in appendix II. If you have any questions, please call Mr. Daniel or me on (202) 512-9110. The tables and figure in this statistical compendium supplement those in the letter. All the values were obtained from IRS’ SOI corporate data files for tax years 1989-95. Other includes transportation and public utilities; mining; construction; agriculture, forestry, and fishing; and other trades. Shirley A. Jones, Senior Attorney The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touch-tone phone. A recorded menu will provide information on how to obtain these lists. | Pursuant to a congressional request, GAO provided an update to its report on the nonpayment of U.S. income taxes by foreign-controlled corporations (FCC) and U.S.-controlled corporations (USCC), focusing on comparisons of: (1) the percentages of FCCs and USCCs that filed income tax returns showing no tax liabilities for 1989 through 1995, the latest years for which data were available; and (2) selected characteristics, including age, industrial sector, and certain cost ratios, of large corporations--those with assets of $250 million or more or gross receipts of $50 million or more. GAO noted that: (1) in each year between 1989 and 1995, a majority of corporations, both foreign- and U.S.-controlled, paid no U.S. income tax; (2) among large corporations, the percentage of FCCs that paid no tax exceeded that for USCCs from 1989 through 1993; (3) in 1994, the difference between the two groups was not statistically significant, and in 1995, the percentage of large FCCs that paid no U.S. income tax was slightly less than that of large USCCs; (4) differences in the characteristics of large FCCs and USCCs may account for part of the differences in the amount of taxes paid by the two groups; (5) one difference was the percentage of new corporations--3 years old or less--in each group; (6) the Internal Revenue Service data GAO reviewed indicate that newer corporations were less likely than older corporations to pay taxes; (7) from 1989 to 1993, a greater percentage of large FCCs than large USCCs were new, but from 1994 to 1995, a greater percentage of large USCCs than large FCCs were new; (8) another significant difference between large FCCs and large USCCs was in their distribution across industrial sectors; (9) in 1995, in comparison to large USCCs, large FCCs were more heavily concentrated in the manufacturing and wholesale trade sectors and less concentrated in the financial services sector; (10) aggregate ratios of costs to receipts for all large corporations differed significantly across industrial sectors; (11) the difference in cost ratios across industries, combined with the fact that large FCCs and USCCs were concentrated in different industries, could account for some of the difference in the amount of taxes that large FCCs paid per dollar of receipts and that large USCCs paid; and (12) the ratio of taxable income per dollar of receipts should be inversely related to the ratio of costs per dollar of receipts. |
Mortgage servicers are the entities that manage payment collections and other activities associated with mortgage loans. Servicing duties can involve sending borrowers monthly account statements, answering borrowers’ inquiries, collecting monthly mortgage payments, and maintaining escrow accounts for property taxes and insurance. In the event that a borrower becomes delinquent on loan payments, servicers also initiate and conduct foreclosures. Errors, misrepresentations, and deficiencies in foreclosure processing can result in a number of harms to borrowers ranging from inappropriate fees to untimely or wrongful foreclosure. A number of federal regulators share responsibility for regulating the banking industry in relation to the origination and servicing of mortgage loans. OCC has authority to oversee nationally chartered banks and federal savings associations. The Federal Reserve oversees insured state-chartered banks that are members of the Federal Reserve System, bank and thrift holding companies, and entities that may be owned by federally regulated depository institution holding companies but are not federally insured depository institutions. In September 2010, allegations surfaced that several servicers’ documents accompanying judicial foreclosures may have been inappropriately signed or notarized. In response to this and other servicing issues, federal banking regulators directed servicers to complete self-assessments of their foreclosure processes. In addition, banking regulators conducted a coordinated on-site review of 14 of the largest mortgage servicers to evaluate the adequacy of controls over servicers’ foreclosure processes and to assess servicers’ policies and procedures for compliance with applicable federal and state laws. On the basis of their findings from the coordinated review, OCC, OTS, and the Federal Reserve issued in April 2011 formal consent orders against the 14 servicers under their supervision. To comply with the consent orders, each of the 14 servicers is required to, among other things: enhance its vendor management, training programs and processes, and compliance with all applicable federal and state laws, rules, regulations, court orders, and servicing guidelines. In addition, the consent orders required each servicer to retain an independent firm to review certain foreclosure actions on primary residences from January 1, 2009, to December 31, 2010. Third-party consultants were permitted to retain outside counsel to provide necessary legal expertise in completing the foreclosure review. Through the foreclosure review, consultants were to identify borrowers who suffered financial injury as a result of errors, misrepresentations, or other deficiencies in foreclosure actions, and recommend remediation for harms suffered by borrowers, as appropriate. In general, the consent orders identified seven areas for consultants to review: (1) whether the servicer had proper documentation of ownership of the loan; (2) whether the foreclosure was in accordance with applicable state and federal laws; (3) whether a foreclosure sale occurred while a loan modification was under consideration; (4) whether nonjudicial foreclosures followed the terms of the loan and state law requirements; (5) whether fees charged to the borrower were permissible, reasonable, and customary; (6) whether loss-mitigation activities were handled in accordance with program requirements and policies; and (7) whether any errors, misrepresentations, or other deficiencies resulted in financial injury to the borrower. Servicers proposed third-party consultants to conduct the foreclosure reviews. After regulators reviewed the independence of proposed third-party consultants, servicers and third-party consultants submitted engagement letters outlining their foreclosure review processes to the regulators for their review and approval. OCC and the Federal Reserve posted approved engagement letters between the servicers and third-party consultants on their respective websites. Regulators also posted on their websites a financial remediation framework that provided examples of errors covered by the consent orders and the corresponding compensation or other assistance that consultants could recommend based on their findings. The foreclosure review had two components: a process for eligible borrowers to request a review of their particular circumstances (referred to as the borrower outreach process) and a review of categories of files (referred to as the look-back review). The borrower outreach process was intended to complement the look-back review and help identify borrowers who may have suffered financial injury. The regulators required the servicers to inform borrowers who believed they might have been financially harmed due to inappropriate foreclosure that they could submit a request for review of their particular circumstances. The servicers conducted this outreach through advertising and direct mail. After several extensions, the final deadline for submission of these requests was December 31, 2012. For the look-back review, the consent orders allowed third-party consultants to use statistical sampling techniques to select samples of files for review from various categories of loans, pursuant to regulators’ guidance and approval as well as 100 percent review of certain loan categories. Regulators established minimum sampling requirements for third-party consultants to conduct statistically valid sampling of the target population and expectations for additional sampling methods, if warranted, to identify as many harmed borrowers as possible for remediation. Bank examiners often use sampling methods to review the files of a financial institution’s operations. The use of sampling by examiners has been a regularly accepted practice and is commonly used as a means to review a bank’s files when a full review of all files is not practicable. OCC states in its handbook on sampling methodologies, which is geared toward sampling loan portfolios, that it is impractical or impossible for bank examiners to review all items or files when examining an area of a bank’s operations, especially if the volume of information is large. Examiners use sampling to identify a random subset of files to learn about the multitude of items from which those files are drawn. Upon drawing appropriate statistical inferences from this subset, they can state with a certain level of confidence that the inferences apply to the population as a whole. The benefits of using statistical sampling include the ability to generalize results to the sampled populations and to quantify uncertainty in estimates attributable to sampling. In our prior report on the borrower outreach component of the foreclosure review, we found that regulators and servicers had gradually improved the communication materials for borrowers but that regulators could make further enhancements to the outreach efforts. First, we found that regulators did not consider best practices, such as using tests or focus groups, to assess the readability of the outreach materials and did not solicit input from consumer groups when reviewing initial communication materials. As a result, we reported that the materials at that time might have been too complex to be widely understood. Second, we found that, although the communication materials included information about the purpose, scope, and process for the foreclosure review and noted that borrowers may be eligible for compensation, the materials did not provide enough specific information about remediation, which best practices suggest could have encouraged more borrower responses. Third, we found that the outreach planning and evaluation targeted all eligible borrowers with limited analysis conducted to tailor the outreach to specific subgroups within the population. Therefore, we noted that some underrepresented borrowers may not have been apparent to regulators without further analysis of the characteristics of respondents compared to nonrespondents. To help ensure that all borrowers had a fair opportunity for review, we recommended that OCC and the Federal Reserve enhance the language on the foreclosure review website, include specific remediation information in the outreach, and require servicers to analyze trends in borrowers who have not responded and, if warranted, take additional steps to reach underrepresented groups. In response to the recommendations that we made in our previous report, OCC and the Federal Reserve took steps to enhance the independent foreclosure review website and communication materials and conducted more targeted outreach. We determined that regulators implemented our first recommendation by including on the foreclosure review website a help sheet for the request-for-review form that provides tips in plain language, an explanation of key terms, and additional instructions to help borrowers fill out the form. In response to our second recommendation, OCC and the Federal Reserve publicly released a financial remediation framework and included ranges of potential payment amounts or categories in their communication materials and other outreach. For example, the regulators released 60-second radio and television advertisements stating that if consultants find errors, homeowners may be eligible for compensation or other remedies, such as refunding fees, stopping a foreclosure action, or making payments that could range from $500 to $125,000. In addition, at the regulators’ instruction, servicers included similar language on ranges of potential payment amounts or categories in other outreach materials. In response to our third recommendation, OCC and the Federal Reserve tailored their recent outreach actions to target communities based on audience characteristics, response data, and consumer research. Regulators included a wide range of minority media and a broad mix of media formats, including print advertisements, radio, television, and Internet. To tailor this outreach, a market analysis was completed to identify areas and ethnic groups with the greatest opportunity for increased awareness. Outreach materials also were made available in eight different languages. Regulators solicited numerous community groups for their input and assistance on the outreach and identified effective messengers by using leaders of community groups that represent minorities to deliver radio and television public service announcements. In January 2013, OCC and the Federal Reserve issued joint press releases stating that they had reached agreements with 11 of the 14 mortgage servicing companies subject to the April 2011 consent orders to discontinue the foreclosure review conducted by third-party consultants and to provide almost $8.8 billion in cash payments and foreclosure- prevention assistance to borrowers. With this change in direction from the foreclosure review to an agreed-upon payment process, regulators and servicers moved away from identifying the types and extent of harm borrowers may have experienced and focused instead on assigning borrowers into categories based on objective criteria and issuing payments in what they expect will be a shorter amount of time than would have occurred under the foreclosure review. To explain their rationale for pursuing the agreements, OCC and Federal Reserve staff said they considered a variety of factors, including delays in payments to harmed borrowers, the total remediation payments expected to be made to borrowers, and costs of the reviews. Under the agreements, servicers will provide approximately $3.4 billion in direct payments to eligible borrowers. Using a framework provided by regulators and characteristics of borrowers’ loans, servicers will categorize borrowers, and regulators will develop a distribution plan and direct a payment administrator to distribute cash payments. As a result, all borrowers who were eligible for foreclosure reviews under the consent orders are expected to receive payments ranging from hundreds of dollars up to $125,000, depending on the borrower’s category. Under the agreements, servicers will also provide approximately $5.4 billion in foreclosure-prevention assistance to borrowers, such as loan modifications. To the extent practicable, servicers are to prioritize such assistance for borrowers eligible for the foreclosure review. Eligible borrowers are expected to receive notice about the payments by the end of March 2013, whether or not they filed a request-for-review form, and borrowers will not need to take further action to be eligible for compensation. Nearly 4 million borrowers, or about 90 percent of the eligible borrower population, are covered by the servicers that signed the agreements. In late February 2013, regulators released amendments to the April 2011 consent orders that incorporated the provisions of the agreements. The amended orders are publicly available on the regulators’ respective websites. Consultants for the servicers that did not reach agreements with the regulators—Ally Financial (GMAC Mortgage), Everbank, and OneWest—continue their foreclosure review activities. More than 450,000 borrowers fall under the three servicers that do not have amended consent orders. In April 2011, regulators entered into consent orders with 14 mortgage servicers. The consent orders require the servicers to conduct foreclosure reviews by engaging third-party consultants to review the servicers’ loan files to identify and remediate the financial injuries suffered by borrowers through errors, misrepresentations, or other deficiencies in their foreclosure processes in 2009 and 2010. In January 2013, the regulators announced agreements in principle with 11 of the 14 servicers subject to the April 2011 consent orders. The parties agreed to replace the foreclosure review with a compensation framework that does not rely on determinations of whether borrowers suffered financial harm, instead requiring participating servicers to provide cash payments and foreclosure-prevention assistance to borrowers. In February 2013, the regulators publicly released amended consent orders that formally replaced the requirements related to the foreclosure review for the servicers participating in the agreements. According to regulators, the goals of the foreclosure review were for consultants to identify as many harmed borrowers as possible, to treat similarly situated borrowers across all 14 servicers similarly, and to help restore public confidence in the mortgage market. However, regulators faced various challenges in accomplishing these goals. According to regulator staff and third-party consultants, coordinating the foreclosure review process was challenging because of the large number of actors directly involved in the review process and the large number of borrowers eligible for the review. Specifically, the foreclosure review process included 14 servicers; 14 third-party consultant teams from 7 different consulting and accounting firms, some with subcontractors; and more than 10 third-party law firms. Further, the reviews were overseen by local examination teams for OCC or the Federal Reserve as well as headquarters staff for both regulators. In addition, there were as many as 4.3 million borrowers whose files could have been reviewed. One examination team informed us that the size of the review population presented challenges beyond those that accompany a typical consent order. Another challenge examiners and consultants noted was that, due to the deadline extensions for borrowers to submit their requests-for- review, the volume of the file review work kept changing and the total number of files requiring review was unknown. The original deadline for requests-for-review was April 2012. Regulators reported that more than 160,000 borrowers had submitted requests-for-review as of April 30, 2012. The deadline was extended three times to December 31, 2012. Regulators reported that more than 510,000 requests were submitted and received as postmarked on or before the December 31, 2012 deadline. According to OCC, roughly 300,000 additional files were selected for review through the consultants’ look-back review process. OCC staff told us that more files likely would be selected if errors were found and consultants needed to conduct additional sampling to identify as many harmed borrowers as possible. Third-party consultants and law firms told us that the size of the loan files and the scope of the file review made the process complicated and time- consuming. Consultants from whom we obtained information told us that a typical loan file is large with many types of documents. For example, documents may include servicer notes on communication with the borrower, documents collected from the state foreclosure attorneys responsible for the foreclosure activities, records of fees charged and payments made, and, in many cases, documents assessing a borrower’s eligibility for loan modification and loss mitigation activities. Consultants told us that some files may contain as many as 50 documents, potentially comprising more than 2,000 pages. Consultants also stated that the reviews were challenging because they covered such a wide variety of complex issues, including different state foreclosure laws, federal laws and regulations, and guidelines for federal and servicers’ proprietary loan modification programs. To assess each of these areas, consultants developed a series of test questions—generally yes or no questions—to identify potential errors. The number of test questions used by third-party consultants to conduct the file reviews varied. For example, one consultant told us that they had approximately 2,600 test questions with more than 4,000 discrete steps, while another consultant told us they had 16,000 test questions. Further, third-party consultants from whom we obtained information stated that their reviewers spent as many as 50 hours to complete a full file review, although review times varied depending on the issue and type of review. OCC and the Federal Reserve stated that the uniqueness of each servicer’s borrower population and process for recording and storing information on borrowers’ loan files posed challenges for defining the review parameters and developing a uniform review structure for all the consultants. For example, Federal Reserve staff told us that servicers’ systems have different ways to identify borrowers who are current on a loan modification and that the systems have varying capabilities to provide information about the last payment received from a borrower. As a result, consultants had to take different approaches to identify borrowers with the same characteristics. In addition, Federal Reserve staff told us that servicers had different concentrations of loans in geographic areas and that, therefore, consultants might select samples differently based on these concentrations. For example, if a servicer did not have high numbers of foreclosures in a particular state, choosing a large sample of loans in that state would not have been appropriate. As a result, OCC and Federal Reserve staff said that it was not feasible to design one file review process that would apply to all servicers and that it was necessary for third-party consultants to tailor their file review processes, particularly the review questions used to conduct the file review, to the unique characteristics of a given servicer. Faced with various complexities and challenges that the foreclosure review posed, regulators told us that they issued guidance and took a number of oversight steps. First, the sections of the consent orders issued to servicers supervised by both OCC and the Federal Reserve outlining the purpose of the foreclosure review were nearly identical. According to OCC and Federal Reserve staff, the similarity in the consent orders was intended to ensure that the reviews covered the same issues and resulted in similar results for similarly situated borrowers. Consultants we interviewed said that they designed their reviews to address the issues as they were identified in the consent orders. Second, regulators issued guidance to third-party consultants to help frame the file review process and promote consistency in its implementation. Between May 2011 and October 2012, regulators issued 29 joint pieces of guidance to third-party consultants on various topics. For example, OCC and the Federal Reserve jointly issued a financial remediation framework that was designed to be a unifying factor among all the reviews by helping to ensure that similarly harmed borrowers received similar remediation. Regulator staff said that they issued guidance in response to similar questions they received from multiple consultants or examination teams, which oversaw the reviews at the local level. For example, one third-party consultant we interviewed said that the reviews of issues related to the Servicemembers Civil Relief Act (SCRA) would likely provide fairly consistent results for borrowers due, in part, to the clear guidance provided by regulators. In addition to consistent consent orders and guidance, OCC and Federal Reserve staff implemented regular communication mechanisms to help foster consistency in the reviews. Regulators had a robust system of regular meetings involving third-party consultants, servicers, examination team staff overseeing the consultants’ work, and OCC headquarters and Federal Reserve Board staff to discuss challenges with the file review process and help promote consistency among the reviews. OCC and the Federal Reserve met with the parties both separately and as a group and received weekly status reports from the consultants. For example, regulator staff said that weekly calls with third-party consultants provided consultants with an opportunity to raise areas of inconsistency that they had identified. These calls were also used to disseminate new guidance and discuss comments on any pending guidance. According to third-party consultants and examination team staff we interviewed, these meetings were helpful for sharing information among the reviews, developing a similar understanding of the file review process, and discussing challenges consultants encountered in reviewing files that may have affected the consistency of the results for borrowers. In addition to meeting with consultants and servicers, OCC and Federal Reserve staff also held a separate weekly meeting, which included only headquarters staff of each regulator, to discuss new and emerging issues or requests for clarification on guidance that each regulator received from its respective local examination teams. In addition to the weekly calls, OCC headquarters staff and Federal Reserve Board staff visited each of the consultants to observe the file review processes. The April 2011 consent orders expressly allowed third-party consultants to use sampling techniques to identify harmed borrowers. According to regulator staff, the large number of loans in the review population as well as the complexity of the file review process made it difficult for consultants to review all the eligible loan files for errors. As a result, the review relied on sampling, a process of selecting units—in this case, foreclosure files—in a manner that regulators envisioned would allow consultants to identify patterns in errors. In May 2011 regulator-issued guidance on sampling, regulators expressly allowed third-party consultants to use sampling and noted that the consultants’ sampling methodologies should take into consideration public perception as well as the need to provide a high degree of certainty that borrowers who were financially harmed would be identified and obtain remediation. The May 2011 guidance outlined broad parameters for how third-party consultants should approach the sampling methodology to identify harmed borrowers and support a consistent review, such that similarly situated borrowers would have similar results. As part of sampling, regulators anticipated that third-party consultants would segment the review population into loan categories as some loan categories had a potentially higher likelihood of errors. For example, the guidance identified some potential high-risk loan categories, including certain states, foreclosure law firms, and servicing or foreclosure processing activities (e.g., rescinded foreclosures or foreclosure that occurred after loan modification) that could be associated with a higher likelihood of servicing or foreclosure-related errors. Regulator staff stated that differences among the servicer’s loan portfolios and servicing practices made requiring specific loan categories for review inappropriate, with the exception of three categories in which regulators anticipated that consultants would identify a relatively large number of errors in the files. For these three categories—bankruptcies, SCRA loans, and agency- referred foreclosure cases—regulators required 100 percent review of files. The May 2011 guidance also indicated that consultants should be prepared to conduct a second stage of additional analysis of files with a certain number of servicing and foreclosure-related errors that were identified during the initial sampling. Based on our analysis of the sampling plans developed by third-party consultants, we found that the sampling methodologies used by consultants varied among the reviews. Expected population error rate. The sampling plans varied in their expected population error rate from 0 percent—that is, consultants expected to find few or no servicing or foreclosure processing errors in the sampled loan category—to 10 percent, with 8 of the 14 reviews assuming an error rate of zero, three assuming an expected population error rate of 3 percent, and three a rate of 10 percent. Loan category sample size. As a result of different expected population error rates, the plans varied in the size of the samples for analyzing various loan categories identified by the consultant. For example, some consultants selected approximately 100 loans in a sampled loan category for a review, and some consultants selected approximately 370 loans in the sampled loan categories. Loan categories. Based on our analysis, the loan categories used by consultants for their analysis varied from review to review. For instance, although all the consultants analyzed the files for errors related to loan modifications, some categorized loans by the loan modification program (e.g., Home Affordable Modification Program (HAMP) or proprietary) and others categorized loans for reasons modifications were denied. Review parameters: For similar loan categories, some consultants anticipated conducting 100 percent review of all files in that category whereas other consultants planned to sample files. For example, some third-party consultants planned to review all rescinded foreclosures, whereas others proposed reviewing a sample of those loans. According to regulator staff, differences in the sampling plans reflected differences in the size and characteristics of the servicers’ loan portfolios and data systems. Regulator staff explained that they reviewed each proposed sampling plan to help ensure it met the parameters outlined in the guidance and would result in statistically valid results. However, according to OCC staff, they recognized that some consultants had not fully implemented the sampling approach as expected, and OCC is taking steps to address these differences for one of the servicers that is not subject to an amended consent order and must continue its review. Our analysis of the May 2011 sampling guidance provided by regulators found that the guidance was ambiguous about a key parameter that affected consultants’ sampling methodologies and contributed to differences in those methodologies. Specifically, in their May 2011 guidance, regulators did not indicate whether consultants should explicitly set an expected population error rate or provide consultants with direction on the factors they should consider when setting an appropriate expected population error rate to determine the size of the sample used for their analysis. GAO’s Financial Audit Manual and the standards of the American Institute of Certified Public Accountants (AICPA) have found that a key element of effective sampling is the determination of an appropriate sample size based on specified precision and reliability levels and an expected population error rate or frequency of errors. Regulators’ May 2011 guidance on sampling specified that consultants should use 3 percent precision and 95 percent reliability levels to determine their sample size. According to regulator staff, these precision and reliability levels were selected to provide a high-level of confidence in the sampling results. However, the guidance did not specify an expected population error rate for consultants to use in determining sample size. Generally, the expected population error rate, like precision and reliability levels, is determined based on professional judgment and includes consideration of factors such as results of prior reviews and knowledge about any potential risks in servicing and foreclosure processing errors. According to regulators, they expected consultants to find errors in their sampled files, because the consent orders that required the foreclosure review process arose out of assessments conducted by regulators that identified the potential for servicing and foreclosure-related errors. Similar to the precision and reliability levels, there is a relationship between the expected population error rate and sample size needed to attain a specific precision level (margin of error), where the size of the sample generally expands as the expected population error rate approaches 50 percent. When an expected population error rate is not specified, it can be interpreted as implying that the error rate is low or even zero—that is, that few or no errors are expected to be found in the sample regardless of whether the expected population error rate of zero is appropriate for the sample goals. OCC staff told us that their handbook on sampling methodologies—a reference their staff use for sampling and one that both regulators suggested consultants consider in designing their sampling approaches—generally assumes few or no errors will be found in a sample. Variations among the sample sizes used to analyze the various loan categories identified by consultants result from differences in the consultants’ expected population error rates and led consultants to use different triggers to determine when to conduct additional analysis of an error or errors found in a loan category. As shown in figure 1, Consultant A would conduct additional analysis of their sampled loan categories if one or more errors were found. In contrast, Consultant B would conduct additional analysis only when five or more errors were found in their sampled categories. Both of these approaches use the precision and reliability levels specified in the May 2011 guidance, with an expected maximum threshold of a 3-percent error rate, but relied on different sampling approaches and different expected population error rates to calculate sample size. Therefore, with the same number of errors identified through file reviews, some consultants would conduct additional analysis and some would not. According to regulator staff, when errors were found, they expected consultants to conduct additional sampling of the loan category or to review all of the loans in that category. OCC staff explained that after approving the original sampling methodologies, they recognized that their review of the plans had missed some aspects and that consultants were using various expected population error rates to calculate sample size. OCC staff told us that they were considering steps to try and address these differences for one of the servicers that had not joined the agreements to end the foreclosure review. As a result of this variation, the reviews could have produced inconsistent results for similarly situated borrowers, thereby potentially limiting achievement of one of the goals of the foreclosure review process. Our analysis also found that the May 2011 guidance on sampling did not include a discussion of regulators’ expectations for reporting on sampling, and variations among the sampling plans would have limited the types of information that regulators could report. For example, the guidance did not specify if regulators expected consultants, based on their sampling methodology, to be able to provide information on the error rate for the servicer’s whole population of eligible borrowers or for certain characteristics, such as high-risk loan categories (e.g., states). Our analysis found that 1 of the 14 reviews explicitly designed samples that would potentially allow consultants to generate reliable estimates of the error rate for certain states. Other consultants may have been able to report counts of errors by state, but some consultants sampled as few as one or five cases per state, and the counts would not allow regulators to determine whether reported errors were relatively frequent or infrequent in specific states. As a result, regulators could not have reported results by state because some reviews did not have enough cases, selected in a generalizable manner, from individual states to report statistically reliable estimates. According to regulators, reporting error rates was not a consideration in developing the sampling approach outlined in the guidance; rather, the sampling methodology was intended to find as many harmed borrowers as possible. However, it would be feasible to develop sampling approaches that could both produce error rates and identify characteristics of harmed borrowers. Similarly, differences in the loan categories analyzed by the third-party consultants would have limited regulators’ ability to easily aggregate results among the reviews or present comparable servicer-specific information. For example, differences in how consultants organized the loan modification category could have impeded regulators’ ability to aggregate the results of the loan modification analyses conducted by third-party consultants or to present equivalent information among the servicers to the public. In addition, these differences in the loan category definitions as well as other differences in the consultants’ sampling methodologies would have made it difficult for regulators to aggregate results to represent the full population of eligible loans. One consultant told us that they anticipated calculating an error rate based on the number of files reviewed compared to the number of files with errors. This method might have provided some information on errors, but could be subject to bias if the sample of loans reviewed were not representative of the population, regardless of whether the estimates from the samples were appropriately weighted for probability of selection or included appropriate confidence intervals to reflect sampling error. However, this information would not necessarily provide a statistically valid description of the extent of errors in the full population that regulators and other stakeholders would need to understand and assess the results. According to regulator staff, they expected consultants to continue reviewing files until as many harmed borrowers as possible were identified, a goal of the foreclosure review process. Regulator staff told us that if third-party consultants’ initial analyses of sampled loans identified errors within a loan category, consultants were expected to analyze the characteristics of loans with errors to identify any patterns and use this analysis as the basis for a second sampling phase or to review all the files in a loan category. Although not explicitly stated in the May 2011 guidance, according to OCC staff, additional analysis was warranted when the errors in the sampled loan category exceeded 3 percent with a 95 percent confidence level. Although the May 2011 guidance on sampling did not specify the characteristics to consider for this analysis, Federal Reserve staff told us that identifying the characteristics to include would require consultants to use their judgment and they anticipated it would include things like the loan category itself—such as an error related to a certain foreclosure attorney that might warrant additional review of other loans that had used the same attorney—or some other characteristic about the loan. According to a few consultants, they were considering characteristics such as the loan product or date of the loan modification solicitation or foreclosure sale, as potential characteristics. Regulators told us that based on the results of the consultants’ analyses of the initially sampled loans with errors, consultants were to develop a methodology—including conducting additional sampling or reviewing all files with those shared characteristics—to identify other loans that had similar characteristics that could have had similar errors to use as a basis for a second review phase. According to Federal Reserve staff, this process would have been iterative—where consultants would have found errors, analyzed those errors, resampled and then repeated the process until as many files with errors as possible had been found. Third-party consultants were expected to discuss the results of their initial file reviews, their analysis of error patterns, and their proposed approach for conducting additional file reviews with regulators prior to conducting additional reviews. Our analysis found that the regulators’ sampling approach did not include mechanisms to facilitate their oversight of the extent to which consultants would have reached as many harmed borrowers as possible. For example, the regulators’ sampling approach did not provide an objective method for regulators to use in determining if consultants had conducted sufficient reviews and could stop their review activities, except in those cases where there were few or no errors. As we described earlier, consultants were using different triggers of the number of errors to determine if additional analysis of loan categories was required. Without a mechanism for regulators to use in assessing the extent to which each consultant had found as many harmed borrowers as possible or to compare the review results among the consultants, assessing which consultants had done enough work to identify a sufficient portion of harmed borrowers and which consultants needed to conduct additional analysis would have been difficult for regulators. According to Federal Reserve staff, they anticipated that consultants would continue reviewing files until the sampling found no additional errors. However, unless a large majority of the population is examined for errors, reviewing files until sampling finds no more errors does not necessarily imply that all or most errors would be identified. OCC staff told us that at the time of the agreements that led to the amended consent orders, regulators were considering developing such mechanisms. Additional sampling activities—specifically, the use of baseline samples— could have provided regulators with a mechanism to use in monitoring the extent to which consultants had identified as many harmed borrowers as possible. The Office of Management and Budget (OMB) standards for statistical surveys state that where sampling is used, it should include protocols to monitor activities and provide information on the quality of the analyzed data. A baseline sample could have been used to establish an estimate of the number of harmed borrowers among the servicer’s full population as part of their sampling methodology and would have provided an objective and consistent measure for regulators to use to gauge if a third-party consultant had identified a sufficient portion of harmed borrowers through their reviews. Specifically, regulators could have compared a statistically valid estimate of the number of harmed borrowers in the servicer’s total population of eligible borrowers with the number of harmed borrowers the consultant identified through file reviews. Discrepancies between the estimated number of harmed borrowers and the number found in the review would have helped to indicate the extent to which there may have been additional harmed borrowers who had not been identified. Without this type of comparison, regulators did not have an objective measure to help determine when a consultant had completed sufficient review of the servicers’ files. In addition, the regulators’ sampling approach did not provide a clear mechanism for regulators to assess the extent to which consultants had identified the appropriate high- and low-risk loan categories to confirm that those categories were accurate and to signal if there were additional potential high-risk loan categories that had not been identified, but warranted additional sampling and review. As we noted earlier, the regulators’ sampling guidance suggested a number of high-risk loan categories consultants should consider in designing their sampling approach—such as rescinded foreclosure or foreclosures following loan modification—that were potentially associated with a higher likelihood of error and, depending on the servicer, warranted targeted sampling to verify the extent to which errors had occurred among the loans in those categories. However, there is evidence to suggest that other loan categories not included in the guidance, such as certain loan categories based on demographic characteristics may also be associated with higher likelihood of servicer errors. Without a mechanism to assess the accuracy and sufficiency of each consultant’s high- and low-risk loan categories, regulators would not have been able to assess the extent to which consultants were targeting the appropriate high- and low-risk categories to find as many harmed borrowers as possible. An estimate of the overall error rate for the population—a rate that could have been generated using information from the baseline sample—could have been compared with the error rates consultants found for high- and low-risk loan categories to confirm that those categories were accurate and to signal if there were additional potential high-risk loan categories that had not yet been identified. OMB has suggested that where different characteristics are being compared, agencies should conduct additional testing to help ensure appropriate statistical conclusions are derived from the data. In our prior work we have used this type of statistical testing to compare the performance (such as error rate) in a total population with the performance for subgroups—such as loan categories—to help distinguish among those groups. This can be part of conducting an overall risk assessment of the population to enable users to better focus on high- and low-risk areas. Without a mechanism to gauge the extent to which high-risk loan categories had been identified, regulators could have had difficulty assessing whether the consultants’ proposed activities for additional analysis were targeted at the appropriate high-risk loan categories to identify as many harmed borrowers as possible. The May 2011 sampling guidance did not include a requirement that consultants develop a baseline sample, but most consultants included one in their sampling methodology. According to regulator staff, they had not considered requiring consultants to include a baseline sample in their methodology because the purpose of sampling was to help find harmed borrowers by identifying concentrations of servicing and foreclosure- related errors as described in the consent orders, not to establish an error rate for the servicer. Our analysis found that most consultants included a baseline sample for the full review population in their sampling methodology. However, due to differences among the consultants’ baseline samples, the ability of regulators to use consultants’ baseline sampling results to assess consultants’ activities varies. For example, the samples were not designed to estimate actual population error rates and regulators’ guidance did not direct consultants to statistically test for differences in the servicer’s total eligible population and among high- and low-risk loan categories. Our analysis also found that the regulators’ approach to conducting repeated additional analyses to find as many harmed borrowers as possible potentially involved timeliness trade-offs. OMB’s standards for statistical analysis state that the sampling design should be appropriate to achieve the sampling goals; in this case regulator staff told us that the phase-one sampling was designed to identify patterns from among loans with errors to facilitate a second phase of analysis to find as many harmed borrowers as possible. Where the goal of sampling is to identify patterns from among loans with errors in a sample, using a larger sample size has the potential to provide more information to use in analyzing patterns and determining the appropriate next steps. For those consultants that used smaller sample sizes in their phase one analysis, for example, a sample size of 100 loans where one error would trigger additional analysis, finding patterns among the characteristics of those loans with errors may have been difficult in cases where there were few errors. As a result, these consultants may have had to conduct potentially time-consuming repeated sampling of certain loan categories where errors were found to find enough loans with errors to be able to analyze the findings and identify patterns that would have allowed them to conduct additional sampling or review. OCC staff told us that some consultants were considering using statistical modeling as part of their analysis of loans with errors to identify patterns from among those loans. However, in cases where the sample size is small and the number of loans with errors is small, the value of statistical modeling to identify patterns could be limited. According to Federal Reserve staff, due to concerns about the timeliness of the sampling process, regulators anticipated that they may have had to cut short the sampling process and provide remediation based on certain borrower profiles—that is, borrowers who shared characteristics associated with loans with a higher likelihood of error—without analyzing each of the loan files that shared that profile. In contrast, a larger phase one sample size designed with the goal of identifying patterns among errors may have provided consultants with more information on loans with errors to use as the basis for their additional analysis. In particular, this may have been valuable for loan categories with a relatively large number of loans—for example, our analysis found that in some cases a loan category had more than 10,000 loans and, for one review, as many as 50,000 loans—and having more information available to use when analyzing the results for these categories and determining next steps may have been helpful. Although a larger sample size does not guarantee a more timely review, a larger statistical sample may have provided additional opportunities for analysis even with a similar proportion of errors. In the case of the foreclosure review, where the goal was to use sampling to find as many harmed borrowers as possible by analyzing patterns from among loans with errors, larger sample sizes may have resulted in a more timely process and potentially may have identified harmed borrowers more quickly. According to regulators, they designed their sampling to have a low tolerance for errors so as to find as many harmed borrowers as possible. According to third-party consultants, regulators’ guidance did not address certain aspects of the foreclosure review and consultants had to use additional judgment and interpretation when applying certain guidance, increasing the risks of inconsistency among review results. Consultants also noted that regulators issued critical guidance throughout the review process and frequently updated guidance, which expanded the scope of the reviews and contributed to delays. For example, the Federal Reserve issued three clarifications of loan modification guidance and OCC provided seven responses to frequently asked questions on reviews of loan modifications. According to regulator staff, developing the foreclosure review process was intentionally iterative where they responded to the most immediate need and used their evolving knowledge to help refine the guidance. Consultants said that changes to guidance required them to develop new test questions, re-train reviewers, and redo file reviews. Although regulators issued numerous pieces of formal guidance and informally responded to specific questions about review procedures that examination teams or consultants raised, consultants said that some of the guidance issued was not specific, leaving room for their interpretation and potentially contributing to inconsistent interpretations. Guidance on fees: Guidance from regulators generally directed consultants to consider whether the fees servicers charged for actions such as property inspections or lawn care services were permissible under the terms of the loan, followed applicable state and federal law, and were reasonable and customary. However, regulators provided additional explanation in response to requests from consultants for clarification and guidance on defining the terms “reasonable” and “customary.” Consultants for 13 of the 14 reviews indicated that they used investor guidelines to determine if the fees charged were reasonable; however, one consultant told us consultants were using different versions of these guidelines. In some cases, consultants used additional methods to evaluate whether fees charged to borrowers were customary. For example, one consultant evaluated fees against a set of benchmarks created from multiple servicers’ actual fee charges, while others said they benchmarked only against the investor guidelines. Guidance on remediation: Although regulators issued guidance to consultants on how to determine the appropriate remediation for different financial harms, 4 out of the 13 remediation categories would have required consultants to make remediation determinations on a case-by-case basis, risking inconsistent treatment of borrowers. For example, consultants would have had to determine remediation on a case-by-case basis if they found that a servicer had initiated foreclosure or foreclosed on a borrower who was protected by federal bankruptcy law. Consultants reported taking a range of approaches to these case-by-case determinations, including awaiting further guidance, developing their own guidelines, or relying on recommendations from their third-party law firms. According to Federal Reserve staff, at the time of the agreements that led to the amended consent orders, regulators were considering options to provide additional guidance to consultants for work in these areas, including determining remediation amounts. Guidance on missing documentation: Guidance issued to consultants in March 2012 on how to determine whether a borrower suffered financial harm when key documents—such as documents that evidenced that foreclosure actions were taken or loan modification application documents—were missing indicated that consultants should treat those instances as errors on the part of servicers. However, this guidance noted that consultants should defer decisions on how to determine borrower remediation until regulators provided further guidance. In the absence of additional guidance, consultants indicated that they took a variety of approaches, including waiting to make remediation decisions, making preliminary considerations of whether the error might have caused financial harm, or working with their third-party law firm to review any applicable legal precedents. According to OCC staff, at the time of the agreements that led to the amended consent orders, they were planning to issue further guidance to consultants on remediation for borrowers with missing documents and they had informally provided additional direction about treatment of files with missing documents during their regular meetings with consultants. According to third-party consultants, regulators missed key opportunities to increase the likelihood of consistent outcomes for borrowers by not requiring development of common criteria or reference materials that served as the basis for the foreclosure review process. Third-party consultants and their respective law firms we interviewed told us that they each developed their own test questions used by their file reviewers to determine whether any errors or financial harm occurred. Consultants developed the test questions based on analyses of state foreclosure laws, loan modification guidelines, and bank policies, among other references. According to OCC staff, the state law references were fairly straightforward and they had confidence that the third-party consultants and law firms would provide fairly consistent interpretations. However, according to third-party consultants and law firms we interviewed, compiling these references and using them to develop review questions was challenging and time consuming and, in some cases, required judgment or interpretation of the laws or guidelines. For example, they noted that certain areas of relevant state law were unsettled and continued to evolve as courts issued decisions. In addition, representatives of law firms involved in the reviews told us that each firm developed its own list of state foreclosure requirements and often came to interpretations different from those of other law firms involved with the foreclosure review. Consultants also indicated that some law firms had different interpretations of which laws were applicable. According to OCC staff, the scope of the reviews was limited to federal and state laws, but several consultants reported that they reviewed servicers’ compliance with certain county- or court-level requirements, whereas other consultants said they generally did not include these requirements. Although OCC and Federal Reserve staff told us that law firms were selected for their independence and capacity to interpret these documents and make these types of decisions, with multiple law firms developing their own interpretations of the applicable laws and requirements, consultants may have based their test questions on different interpretations of laws, which could have hindered regulators’ ability to achieve one of their goals for the foreclosure review, similar treatment for similarly situated borrowers. Our analysis indicates that regulators missed another opportunity to standardize reference materials in the area of loss mitigation and loan modification requirements. Consultants noted challenges in compiling relevant loan modification program guidelines. For example, according to consultants, they had to compile requirements for multiple loan modification programs and representatives of one consultant we interviewed said that they had to compile requirements of 40 different loan modification programs that the servicer used during the 2009 to 2010 period. In addition to numerous different programs, the guidelines for a single program may have changed multiple times. One consultant noted that they had to track and apply 28 program changes the Department of the Treasury (Treasury) made to HAMP between 2009 and 2010. Our prior work on HAMP identified instances of servicers interpreting HAMP guidelines inconsistently, and a consumer advocacy group report noted similar challenges with servicers implementing HAMP according to the guidelines. In addition, several consultants from whom we obtained information told us that they had to make some judgment calls when interpreting regulatory guidance and program guidelines. For example, one consultant noted that HAMP guidelines were unclear in critical respects, particularly regarding the nature and extent of servicers’ obligations to notify borrowers about steps in the HAMP process. OCC staff said they were aware of Treasury’s HAMP guidance and other servicing guidelines and made an effort to make their guidance to consultants consistent with these materials. OCC and Federal Reserve staff also stated that certain staff members had general discussions with Treasury staff to understand the HAMP guidelines. Consultants we interviewed told us that they largely relied on their internal subject-matter experts to interpret the relevant guidelines and did not consult with program experts, such as Treasury staff who designed and developed HAMP. As a result, third-party consultants and law firms may have applied different interpretations of the legal or program requirements for the same programs to the reviews, and the use of different reference materials could have reduced the likelihood of achieving the goal of treating similarly situated borrowers consistently. Regulators took steps to monitor potential inconsistencies among the reviews, but these steps were limited and likely would have resulted in delays in providing remediation to borrowers. First, according to regulators, they closely monitored weekly reports provided by consultants to identify any differences in their progress that may have indicated some inconsistency in the foreclosure review processes. These initial reports included information on the number of mailings; requests for review; and high-level counts of file reviews started, in process, and completed and the number of files with borrower harm, but they did not include information on the specific types of errors identified in the reviews or the test criteria used to review files. Therefore, the usefulness of the reports for identifying inconsistencies was limited. According to OCC staff, they planned to begin requiring consultants to report on additional information, such as the types of errors associated with financial harm found in the reviews and proposed remediation amounts, which would have helped identify inconsistencies among reviews. Second, OCC and Federal Reserve staff said that they would identify potential inconsistencies among the reviews by having staff from one examination team assist with another team’s oversight of the foreclosure review, but these rotations among the staff were not systematically organized and did not include rotations across regulators. Further, some examination team members we interviewed said that they had participated in one or two reviews for other servicers, but examiners overseeing reviews at a larger servicer noted that they were unable to participate in multiple reviews because of time constraints. In addition, OCC and the Federal Reserve did not provide the examination team members conducting these rotations with guidance on the types of issues to consider in assessing inconsistencies. This unsystematic approach limited the extent to which regulators would have been able to identify trends or inconsistencies. Third, according to regulators, they reviewed some test questions but did not compare them across reviews to identify inconsistencies. OCC staff acknowledged that inconsistencies were inherent in the foreclosure review because of the large number of actors and decision points and the subjective nature of some of the decisions consultants had to make. Similarly, Federal Reserve staff told us that inconsistencies among reviews were inevitable due to differences among the servicers’ policies, procedures, and systems, including their loan modification and loss mitigation programs. Regulator staff said they had planned to conduct assessments of the extent of inconsistencies affecting the outcomes for borrowers across the reviews after the reviews and recommendations for remediation were completed. However, conducting such an assessment after the reviews were completed could have resulted in delays in remediation because third-party consultants may have needed to change their file review questions and redo file reviews if regulators identified inconsistencies. In addition, regulators would have had to wait until all consultants had completed their reviews to conduct such an assessment. Consultants reported that they had estimated completing their reviews in different time periods and an OCC official estimated that the reviews would not have been completed until 2014. Our analysis of the foreclosure review process identified challenges in regulators’ planning for the foreclosure review. These challenges underscored the importance of the following three aspects of planning and monitoring: (1) identifying the type and amount of information to report as final results during the design of data analysis; (2) consulting with stakeholders; and (3) assessing how well the reviews were tracking their goals of identifying as many harmed borrowers as possible, achieving consistent reviews for borrowers, and helping to restore public confidence in the mortgage market. As described earlier, variations among the 14 sampling methodologies used by third-party consultants for analysis of loan files limited the types of information that regulators would have been able to report. According to regulator staff, they did not want to make final decisions on the types of information they may have needed for public reporting before they had seen the preliminary results from the file reviews. OMB has found that in designing data analysis activities, including sampling, agencies should consider the types of data they need to collect to be able to report useful information on the results of their activities to the intended audience. The resulting sample design should have these data output requirements built into the structure. GAO’s internal control standards state that producing reliable and relevant data is important for oversight and management, including oversight provided by Congress. In addition, our prior work has found that public reporting of results can be important for strengthening public confidence in a process, and the data analysis strategy should be designed to include information that can be reported publicly. Providing useful and relevant public reporting of the review results also was a key element in renewing public confidence in the mortgage servicing market, a goal of the foreclosure review process. However, regulators were limited in what they would have been able to report because they did not plan for reporting in the design of the reviews. As described earlier, the scope of the foreclosure review was broad and regulators experienced challenges in issuing guidance on the wide variety of complex issues covered by the reviews, resulting in delays in completing file reviews and potentially contributing to inconsistencies in the file review process. Regulators may have been able to better define the scope of activities and issue more complete guidance prior to commencing the foreclosure review process, thereby potentially reducing the number of revisions to the scope or guidance, by consulting with organizations directly responsible for or familiar with particular aspects of the review before initiating the foreclosure review process. For example, although regulators consulted with Treasury staff for their technical expertise on the HAMP requirements during their development of the loan modification and loss mitigation guidance, regulators did not have the benefit of additional consultations with Treasury compliance officials and discussions with other agencies responsible for overseeing federal loan modification and loss mitigation programs to help them more clearly define the programs covered by the consent order requirements and the elements to consider in assessing servicers’ evaluation of loan modification and loss mitigation activities—areas where regulators issued clarifying guidance to third-party consultants. Regulators issued additional guidance to clarify that the review of HAMP and proprietary loan modification programs should include programs overseen by the U.S. Department of Housing and Urban Development, U.S. Department of Agriculture, and U.S. Department of Veterans Affairs. In addition, regulators did not consult with community groups, such as national organizations representing housing counselors that have worked with individual borrowers on their loan modification and loss mitigation applications. These consultations might have provided input on challenges specific servicers and borrowers experienced with the range of loss mitigation and loan modification activities which could have assisted in identifying high-risk loan categories or program elements to consider. For example, our prior work surveying housing counselors found that while assisting borrowers with HAMP applications, counselors experienced challenges with servicers, including missing documentation, lengthy decision-making processes, and miscalculations of borrowers’ incomes. Our prior work that establishes generally accepted project planning practices identified consulting with stakeholders as one of seven generally accepted practices. In addition, our internal control standards have found that consulting with external stakeholders can have a significant impact on the achievement of goals. In contrast to the process used to develop the loan modification and loss mitigation guidance, OCC and the Federal Reserve consulted with consumer groups while developing the remediation framework. In addition, they consulted with the U.S. Department of Justice in developing the foreclosure review guidelines related to SCRA. Although this consultation occurred later in the process after third-party consultants had begun SCRA reviews, one consultant cited the guidance that resulted from the consultations as evidence of a strong practice that helped promote consistent results among the reviews. As we described earlier, regulators’ sampling approach did not include mechanisms that would allow them to objectively measure the extent to which consultants were on track to identify as many harmed borrowers as possible. Similarly, as we previously described, regulators had a limited process in place to identify inconsistencies among consultants’ file review processes that could have affected results for borrowers. OCC and Federal Reserve staff told us that they were aware of some areas where there may have been inconsistencies, but according to Federal Reserve staff these areas would not have led to significant differences. However, in the absence of mechanisms to systematically monitor consistency, regulators did not have the information to verify their understanding or identify areas of the review with an increased likelihood of inconsistency. Our prior work has identified using intermediate activities or measures to assess progress toward intended results as an effective management practice to understand the extent to which activities are on track to reach stated goals. We have also identified practices that can help agencies successfully implement the Government Performance and Results Act and related results-oriented management initiatives, such as establishing activities during program planning and design to monitor performance toward these goals and using intermediate activities to analyze the gap between where the performance is and where it needs to be to achieve desired outcomes. We found that such activities can help management target areas in need of improvement and select appropriate methodologies to realize that improvement. In the absence of systematic processes to monitor the extent to which the foreclosure review was progressing toward its goals, regulators did not have an early warning mechanism to help identify problem areas where interventions, such as the issuance of clarifying guidance or other support, might have helped reorient activities and address concerns. Regulators publicly released information on the foreclosure review process beyond what is typically disclosed in connection with a consent order, including engagement letters between servicers and consultants and some guidance provided to the consultants. By law, federal banking regulators must disclose any formal enforcement actions entered into under the Federal Deposit Insurance Act. On a case-by-case basis, banking regulators may consider the release of information beyond the mandatory disclosures. In an effort to promote transparency in the foreclosure review process, OCC and the Federal Reserve publicly disclosed some information related to the April 2011 consent orders. For example, in November 2011, OCC released redacted engagement letters between the servicers under its jurisdiction and the consultants contracted to conduct the foreclosure review. With the exception of one servicer, the Federal Reserve released by February 2012 redacted engagement letters for servicers under its jurisdiction. OCC and the Federal Reserve also released the remediation framework for consultants to use that provided examples of situations in which compensation or other remediation is required for financial injury due to servicer errors, misrepresentations, or other deficiencies. Despite these disclosures, some stakeholders perceived gaps in key information about how the file reviews were conducted. Regulators released documents, such as the redacted engagement letters and remediation framework, which generally described the design and intended outcomes of the foreclosure review, but they did not disclose the more detailed guidance and tests consultants relied on to perform their reviews. As previously discussed, third-party consultants developed thousands of test questions to determine error and harm, and regulators issued a number of guidance documents to promote consistency among the reviews and clarify issues raised by consultants, such as how consultants were to construct their sample populations and how to address issues related to borrowers covered by SCRA. Although regulators released their remediation framework and provided answers to frequently asked questions on remediation categories and calculations, they did not release any additional guidance documents nor did they publicly disclose consultants’ test questions, which according to OCC, contained proprietary and supervisory information. To increase the transparency and credibility of the foreclosure review for borrowers, policy makers, and the public, among other stakeholders, consumer groups recommended that regulators release such information. According to consumer groups, without such information, the public would have questions and doubts about how the reviews were executed. OCC and the Federal Reserve staff said that they considered releasing additional guidance to the public, but both regulators refrained from doing so because of concerns that releasing detailed information risked disclosure of confidential or proprietary information. Moreover, test questions developed by consultants were numerous and complex, and Federal Reserve staff stated that review processes were too dissimilar to provide a comprehensive summary. Borrowers who requested reviews under the foreclosure review process initially received limited information about the status of their individual file review. Borrowers received a letter acknowledging their request was received, but some did not receive updates until almost a year after the outreach program was launched, when they received a letter informing them of the continuing nature of the review. In letters to OCC and the Federal Reserve, consumer groups indicated that these borrowers were frustrated by the lack of information on their particular file review. Eligible borrowers could submit requests as early as November 2011. According to OCC staff, borrowers who submitted an accepted request-for-review through June 30, 2012 received a status update letter in September 2012. OCC staff said that the letters communicated to borrowers that their requests were being reviewed but that the results of the review might not be available for several more months. They said that the letters also provided a brief summary of the foreclosure review process, an Internet link to the interagency remediation framework, and notice of other help available through nonprofit organizations approved by the Department of Housing and Urban Development. Regulators indicated that additional status letters would be sent to borrowers with outstanding requests-for- review. Before the regulators halted the foreclosure review, draft letters to be sent to borrowers on the results of their file reviews were in development. However, regulators were still uncertain about specific information they would require be shared with both borrowers who would receive remediation and those who would not. Regulators have acknowledged the importance of transparency, but when announcing the agreements that led to the amended consent orders, they had not yet determined what information to convey beyond that which was included in their press releases and public websites, nor had they determined whether additional information would be provided to borrowers who submitted a request-for-review. During the foreclosure review process, OCC released two interim reports that provide the public with information on the organization and conduct of the file review process and preliminary results, such as the number of requests-for-review received, for institutions it supervises. The OCC reports—issued in November 2011 and June 2012—summarized the status of actions taken to correct deficiencies in mortgage servicing and foreclosure processing identified in the April 2011 consent orders, including activities related to the foreclosure review. For example, the June 2012 report included the number of files selected for review, the number of requested reviews, and the number of reviews completed, among other items. These reports, according to OCC, were intended to build transparency in the process. The Federal Reserve did not issue interim reports on the foreclosure review process for institutions it supervised. According to Federal Reserve staff, they did not do so because they determined that their public release of servicers’ action plans provided sufficient information about how servicers were addressing the requirements of the consent orders and their public release of servicers’ engagement letters provided sufficient information about how the foreclosure review would be conducted. Prior to the announcement of the agreements that led to the amended consent orders and ended the foreclosure review for most servicers, OCC staff told us they had planned to release a final report on the results of the foreclosure review. After the agreements were announced, Federal Reserve staff indicated they expected to publish additional relevant information related to the foreclosure review and the agreements. However, as of February 2013, regulators had not decided what specific information will be made available on the work conducted under the foreclosure review prior to the agreements. While OCC and the Federal Reserve acknowledged the importance of transparency in the foreclosure review process, the absence of timely and useful communications at certain stages of the process—for individual borrowers as well as the general public—hindered transparency and undermined public confidence in the processes and results. In a December 2011 letter to consultants, OCC noted the importance of public confidence in the foreclosure review process. According to the Federal Reserve, the agency endorsed OCC’s letter. However, as previously discussed, regulators did not publicly release detailed information that described how consultants were to determine errors and remediation, and borrowers and the general public received limited information about the status of the reviews. As a result, consumer groups raised concerns about the level of transparency of the foreclosure review process and indicated that the absence of public information undermined credibility and public confidence in the process. Our internal control standards state the importance of relevant, reliable, and timely communications within an organization as well as with external stakeholders. As illustrated in Treasury’s implementation of the Troubled Asset Relief Program (TARP), external communications can include posting information on its website and regular, public reporting. Our previous work on TARP described the importance of public reporting as a means to improve transparency and address potential questions of whether similarly situated borrowers are being treated fairly. For example, Treasury periodically issued public reports on the progress and performance of its TARP housing programs. These reports have provided information to a range of stakeholders, including Congress and the general public. Similarly, consumer groups recommended that regulators provide regular, public reports on the progress and findings of the foreclosure review to increase transparency. Consumer groups also suggested that regulators provide additional information to address transparency-related issues for borrowers eligible for remediation through the foreclosure review. For example, they indicated that borrowers should have had access to information about the status of the review of their file and receive a thorough explanation of how decisions were reached. In addition, they recommended that the review guidelines issued by the regulators be made public, similar to Treasury’s release of guidance for HAMP. According to consumer groups, these actions, among others, could have increased the public’s understanding of the reviews and allowed borrowers to ensure their files were properly reviewed. The foreclosure review, unlike TARP and TARP-funded programs such as HAMP, was not a government program. However, as OCC described, it was part of a larger set of government-directed actions in response to the housing and mortgage crises. More publicly disclosed information about processes and regular reporting about the status of the reviews would have increased transparency and thereby public confidence in the reviews, given that one of the goals regulators articulated for the foreclosure review was to restore public confidence in mortgage markets. The foreclosure review revealed three key lessons that could help inform regulators’ implementation of the amended consent orders: (1) designing project features during the initial stages of the process to influence the efficiency of file reviews, (2) monitoring progress to better ensure the goal of achieving intended results, and (3) promoting transparency to enhance public confidence. These key lessons on planning and implementation could help contribute to an effective process for distributing direct payments and other assistance as prescribed by the amended consent orders between servicers and regulators. In addition, these lessons could inform the foreclosure review process that is continuing for the servicers that did not reach agreements with regulators. The foreclosure review experience suggests that a planning process to determine key project features, such as guidance and necessary data elements, for activities conducted under the amended consent orders could lessen the risk of changes to the planned activities, future delays, or rework. Our work on designing evaluations, including financial audits, has found that systematic and comprehensive planning enhances the quality, credibility, and usefulness of the results and contributes to a more effective use of time and resources. We found that one of the first steps in designing a review should be to define the purpose and scope of the review, including defining what data will be collected and what comparisons will be made. In addition, evaluation questions should be clear and specific, and should use terms that can be defined and measured so that the purpose and scope are readily understood and feasible. Our prior work establishing a lessons-learned process also has found that assessing and using lessons learned from previous experience can provide a powerful method of ensuring that beneficial information is factored into the planning and work processes of future activities. Key practices of assessing lessons learned include collecting and analyzing information on prior activities and applying that information to future activities. As the foreclosure review experience suggests, some implementation challenges were inevitable due to the unprecedented nature of the review. Nevertheless, the broad and expanding scope of the reviews and delays in defining key concepts could have been mitigated by more advanced planning from regulators, resulting in more efficient and effective reviews. The April 2011 consent orders provide a general description of the scope of the file reviews consultants were to conduct. According to third-party consultants, these parameters resulted in a broad review scope that generally covered all instances of noncompliance with applicable law and servicing guidelines as they related to the post-default residential mortgage loan borrower experience, including instances of noncompliance that resulted in financial harm to borrowers requiring remediation. Regulators may have missed opportunities to potentially narrow and refine the project scope—for example, through earlier definition of a harmed borrower or agreement on errors not resulting in remediation that may not have warranted additional review. Changes to guidance also expanded the scope of the reviews. For example, as we noted previously, changes to loan modification guidance contributed to an expanded scope for those reviews and delays in completing the work, including, in some cases, redoing file reviews. According to regulator staff, developing the file review process was iterative; they learned as the reviews progressed, and regulators used that knowledge to help refine the review process. In our work on designing evaluations and audits, we recognize that a review process can be iterative and that the scope and activities of the review may change as work progresses and data limitations or new information arise. Nevertheless, conducting a planning process that involves all stakeholders provides an opportunity to examine preliminary information and pilot-test processes and procedures to help further define the scope of potential activities and hedge against the risk of future changes. In addition, assessing lessons learned by using project critiques and discussions with key participants and stakeholders—such as local examination team staff, third-party consultants and law firms, and external groups—could identify the root causes of strengths and weaknesses of the foreclosure review that could apply to the amended consent order activities. According to regulator staff, they are meeting with examination staff and third-party consultants to discuss challenges with the servicers’ data that may make it difficult for servicers to determine borrowers’ direct payment amounts under the amended consent orders. OCC staff also said that they had discussions with other federal agencies and consumer advocacy groups before announcing the agreements that led to the amended consent orders. As regulators prepare to implement the amended consent orders and compensate borrowers, they may encounter delays and inconsistencies similar to those associated with the foreclosure review if they miss opportunities to make key project planning decisions. Regulators told us that they anticipated borrowers would be contacted by the end of March 2013 and that they also expected the issuance of checks to begin in April. However, regulators still need to make some key decisions about these activities in order to meet their goal of beginning borrower payments in April 2013, and clear guidance in several areas will facilitate the process. According to regulator staff, prior to contacting borrowers each servicer will assign borrowers to direct payment categories. OCC staff said that the servicers were instructed to use objective data and definitions and guidance from regulators in placing borrowers in the categories and that the examination teams would use the guidance provided to servicers to conduct their validation process. Regulator staff said that they agreed that clear guidance on the categorization and file review process and validation by examination teams are tools that will help the servicers and examination teams navigate this complex process. In most cases, servicers, with regulators’ approval, have engaged the third-party consultants to review borrowers’ files in two categories (SCRA and foreclosed borrowers who were not in default) to determine whether borrowers experienced those specific types of harm. According to one third-party consultant, at the time of the agreements that led to the amended consent orders, consultants were waiting on additional guidance from regulators to complete aspects of these reviews. In addition, regulators have not yet determined payment amounts for the different categories of borrowers. Regulator staff said that they could not make these decisions until servicers had completed their categorization process and regulators knew the number of borrowers in each category to allow them to divide up the total payment amount among the borrowers. Regulators also have not determined what to do with any funds that might be left because borrowers refused payments, did not cash their checks, or could not be located. According to regulators, any remaining funds will not be returned to the servicers. Further, regulators told us that the agreements specify that over the next 2 years servicers are required to take loss mitigation and foreclosure prevention actions for which each servicer will earn credit toward fulfilling a specified obligation. These activities will provide assistance to borrowers covered under the consent orders and other borrowers. Regulator staff said they provided a list of eligible activities to servicers and advised them to prioritize borrowers covered under the consent orders for assistance. However, according to OCC staff, they did not provide servicers with criteria for identifying borrowers to whom they will offer mortgage assistance, which would help ensure that eligible borrowers have a consistent opportunity to be considered for these funds. Without assessing past lessons learned and making decisions on key features of the amended consent order activities in advance, regulators risk having to implement changes in the planned activities or publicly announced timelines, which could decrease the efficiency of the process to distribute direct payments and other assistance. The foreclosure review experience suggests that regulators’ process for monitoring third-party consultant, servicer, and examination team activities, including holding regular meetings and reviewing weekly progress reports, could provide a useful model for monitoring activities under the amended consent orders. According to regulators, examination teams will play a critical role in overseeing servicer activities under the amended consent orders, including testing and validating the results of the servicers’ categorization of borrowers. Regularly collecting and reviewing information on the activities and approaches of the examination teams could provide an opportunity to identify challenges and take steps to address them as the activities progress. Similarly, instituting a process to monitor the progress of the servicers’ loan categorization and track the payment administrator’s distribution of payments could help regulators assess the extent to which they are on target to reach their goal of providing notification to borrowers about their payment by the end of March 2013. OCC staff said that they continue to hold regular meetings with the servicers and examination teams to answer questions and share ideas. In addition, OCC staff told us that they intend to conduct site visits with each servicer, review servicers’ draft categorizations, and implement a reporting mechanism to oversee servicers’ activities under the amended consent orders. Regulators’ experience with the foreclosure review suggests that identifying comparative oversight mechanisms to promote consistency could help achieve consistent results among key actors. As discussed earlier, regulators had limited and unsystematic centralized control mechanisms to monitor consistency among the foreclosure review processes and did not have the information to assess the implications of any differences. According to regulators, achieving consistent results for borrowers, so that similarly situated borrowers receive similar payment amounts, is a goal of the amended consent orders, as it was of the foreclosure review process. OCC staff stated that the direct payments provided under the amended consent orders will likely be more consistent than what would have occurred under the foreclosure review because servicers are using a standard framework and objective criteria to categorize borrowers and all borrowers in a particular category will receive the same payment amount. According to regulator staff, the categorization instructions provided to servicers, regular meetings with servicers and examination teams, site visits, and examination team verification processes will provide opportunities to review and discuss the results of each servicer’s categorization of borrowers. However, whether there is a similar process that will compare results across OCC and Federal Reserve supervised servicers is unclear. Furthermore, to what extent regulators will assess the implications of any inconsistencies among the reviews that consultants are continuing to conduct for the servicers that did not sign agreements with regulators is unknown. Applying lessons from the foreclosure review process and our internal control standards to the amended consent order activities would suggest that mechanisms to centrally promote consistency and monitor agreement activities could help achieve consistent results for borrowers. GAO’s internal control standards state that agencies should take steps to comprehensively identify and analyze program operations to determine if risks exist to achieving goals—such as risks to the regulators’ goal of providing similar results for similarly situated borrowers. In our prior work, we found that using a horizontal review mechanism is an option to help mitigate risks of inconsistent results for activities conducted by multiple entities. For example, comparing servicer decision-making processes, including any criteria used by the servicers to categorize borrowers, could identify any potential differences and the extent to which these differences may result in different direct payment decisions for similarly situated borrowers. Similarly, mechanisms to provide clear and specific guidance—such as guidance on testing and validation of servicer activities—for local examination teams to use in their oversight of individual servicer activities could help regulators to more easily monitor and compare servicer activities and the results for borrowers among the reviews. Without using mechanisms to centrally monitor the consistency of servicers’ activities to categorize borrowers, regulators may risk delays in providing direct payments to borrowers and inconsistent results. In addition, without monitoring potential inconsistencies in the foreclosure reviews that are continuing for servicers that are not party to the amended consent orders, regulators will not have the information to assess whether those servicers’ borrowers are being treated consistently. Lessons from foreclosure review activities conducted to date suggest that developing and implementing an effective communication strategy that includes public reporting goals could enhance the transparency of the activities under the amended consent orders. GAO’s internal control standards state the importance of relevant, reliable, and timely communications within an organization as well as with external stakeholders. As a means to strengthen communication with external stakeholders and improve transparency and accountability, our work on TARP has underscored the importance of a communication strategy. Moreover, our prior work on organizational transformation demonstrates that public and private-sector leaders view establishing a communication strategy as especially crucial in the public sector, where policymaking and program management demand transparency and stakeholders are concerned not only with what results are to be achieved, but also with which processes are to be used to achieve those results. In addition, as previously discussed, public reporting is also a mechanism for external communication that can enhance transparency. Experiences with current government initiatives that are aimed at assisting struggling homeowners and involve institutions and mortgage-related issues similar to those of the foreclosure review also highlight the benefits of regular performance reporting. Specifically, periodic reports on the performance of and participation in TARP programs and scheduled reports on servicers’ compliance with requirements of the National Mortgage Settlement are intended to promote transparency and build public confidence. Because the foreclosure review and the subsequent activities under the amended consent orders—like TARP and the National Mortgage Settlement—are part of the larger governmental response to the housing and mortgage crises, a communication strategy which incorporates plans for periodic public reporting may enhance transparency in the distribution of direct payments and other assistance and help restore confidence in the mortgage market. Regulators announced the agreements that led to the amended consent orders without a clear communication strategy. As such, what information will be provided to individual borrowers and the general public about processes, progress, and results of activities under the amended consent orders is unclear. Although OCC and the Federal Reserve have provided some information on the amended consent orders and have plans to release additional information, regulators have not made key decisions on communicating directly with individual borrowers and the extent to which they will report on activities related to the amended consent orders and continuing foreclosure reviews. Regulators provided limited information on the amended consent orders through press releases and updates on their websites, among other ways. For example, OCC and the Federal Reserve issued joint press releases announcing the agreements and related amended consent orders, and OCC and the Federal Reserve posted answers to frequently asked questions on their websites on the agreements that led to the amended consent orders. In addition, regulators plan to release information about the distribution plan after payment amounts are determined, but staff did not describe plans to release additional information about the procedures servicers are using to categorize borrowers. As of January 2013, OCC staff said that they planned to release two public reports, one in April 2013 to discuss the direct payment process, and one in the summer of 2013 to discuss the foreclosure review results of servicers not covered under the amended consent orders. OCC staff told us, however, that they have not decided on the specific content of these reports. As of February 2013, the Federal Reserve also plans to issue public reports about servicers’ activities under the amended consent orders and the results of the foreclosure reviews at servicers not subject to the amended consent orders. Regulators had not decided what, if any, information will be made available on the results of the work conducted under the foreclosure review prior to the agreements. Further, OCC and Federal Reserve staff told us they had not made decisions about the form and specific content of communications, if any, directly to individual borrowers. While the amended consent orders terminate the foreclosure review for most of the servicers, transparency of past and current efforts continues to be important to stakeholders, including Congress and consumer groups. In particular, members of Congress have expressed concerns about the lack of public reporting on the foreclosure review and a lack of information about how the amounts of payments and other assistance in the amended consent orders were determined. In addition, consumer groups expressed concerns about transparency and urged regulators to take additional steps to increase transparency and public confidence in the implementation of the amended consent orders, including robust data collection and reporting. In the absence of a clear communication strategy to direct external communications, including public reporting and direct communication with individual borrowers, regulators face risks to transparency and public confidence similar to those experienced in the foreclosure review process. The foreclosure review was intended to identify as many harmed borrowers as possible, ensure that similarly situated borrowers received similar results, and help restore public confidence in the mortgage market. Ultimately, the complexity of the foreclosure reviews and limitations in regulators’ guidance and monitoring of the foreclosure review challenged their ability to achieve the stated goals. OCC and the Federal Reserve took a number of steps to foster consistency of this unprecedented review, including issuance of nearly identical consent orders and joint issuance of guidance documents for third-party consultants. However, other efforts related to guidance and monitoring may have exacerbated the challenges and complexities inherent in the process. In particular, existing guidance on sampling was ambiguous, leading to inconsistent sampling methodologies used by consultants, and did not include key oversight mechanisms to facilitate assessment of the extent to which consultants had identified as many harmed borrowers as possible. In addition, regulators’ limited monitoring of consistency of the consultants’ sampling methodologies and review processes increased risks that similarly situated borrowers would receive different results. As a result, the regulators risked not achieving the intended goals of identifying as many harmed borrowers as possible and treating similarly situated borrowers similarly, and this remains a challenge for the servicers continuing the foreclosure review. Our prior work has identified practices, such as assessing progress toward goals and designing such monitoring during the planning stage of a project, as effective management practices. In addition, OMB has found that in planning data analysis activities, such as sampling, agencies should take necessary steps to ensure that they have collected the appropriate data from which to draw conclusions. Assessing the review processes of the continuing reviews for consistency and implementing a mechanism to assess the sufficiency of additional sampling activities could help mitigate risk of similarly situated borrowers receiving different results and facilitate oversight of the extent to which consultants have reached as many harmed borrowers as possible. Although the regulators have terminated activity related to the foreclosure review for the servicers with amended consent orders, the foreclosure review process offers an opportunity for the regulators to leverage this experience to help ensure that similar difficulties are better addressed in future efforts. In general, identifying, assessing, and using lessons learned can help ensure that beneficial information is factored into the work processes of future activities, among other things. Therefore, consideration of lessons from the foreclosure review activities, such as advance design of data analysis activities and assessment of inconsistencies in the file review processes, is one way to help regulators improve the clarity of their guidance and ensure effective monitoring of progress toward results for the activities under the amended consent orders and the three remaining servicers still subject to the foreclosure review requirement that will cover 450,000 eligible borrowers. Specific activities include the following. Sound planning and additional oversight mechanisms can enhance project design and help ensure achievement of program goals. Our internal control standards on monitoring, risk assessments, and consultations with stakeholders, as well as generally accepted project management practices for agencies to use in implementing the Government and Performance Results Act and related management initiatives, emphasize the importance of planning, monitoring, and assessing lessons learned. Assessing the strengths and weaknesses of the foreclosure review process by using project critiques and discussions with key participants and stakeholders—such as local examination team staff, third-party consultants, law firms, and external groups—could help ensure better design and more efficient implementation of activities under the amended consent orders. As such, consideration of lessons from the foreclosure review processes would enhance the design, implementation, and oversight of the activities under the amended consent orders. The foreclosure review activities to date also highlight the importance of effective communication. We found that limited communication with individual borrowers and the general public hindered transparency and public confidence in the reviews. Further, regulators announced agreements that led to the amended consent orders and ended the foreclosure review with 11 servicers without a clear communication strategy, and the information that will be provided to borrowers and the general public remains unclear. Our internal control standards state the importance of external communication, and our key practices in implementing transformations that were identified by public and private sector leaders underscore the importance of a communication strategy. As such, the development and implementation of an effective communication strategy could enhance regulators’ efforts to ensure transparency and public confidence in the results of the foreclosure review as well as processes to implement the activities under the amended consent orders. We are making three recommendations to the regulators: (1) To better ensure that the goals of the foreclosure review are realized for servicers that are not subject to amended consent orders, we recommend that the Comptroller of the Currency and the Chairman of the Board of Governors of the Federal Reserve System, as appropriate, improve oversight of sampling methodologies and mechanisms to centrally monitor consistency, such as assessment of the implications of inconsistencies on remediation results for borrowers in the remaining foreclosure reviews. (2) To better ensure that the goals of the amended consent orders related to the distribution of direct payments and other assistance are realized, we recommend that the Comptroller of the Currency and the Chairman of the Board of Governors of the Federal Reserve System identify and apply lessons from the foreclosure review process, such as enhancing planning and monitoring activities to achieve goals, as they develop and implement the activities under the amended consent orders. (3) To better ensure transparency and public confidence in the activities under the amended consent orders and results of the continuing foreclosure reviews, we recommend that the Comptroller of the Currency and the Chairman of the Board of Governors of the Federal Reserve System develop and implement a communication strategy to regularly inform borrowers and the public about the processes, status, and results of the activities under the amended consent orders and continuing foreclosure reviews. We requested comments on a draft of this report from OCC and the Federal Reserve. OCC and the Federal Reserve provided written comments that are presented in appendixes II and III, respectively. The regulators also provided technical comments, which we have incorporated into the report, as appropriate. In commenting on the report, OCC and the Federal Reserve both identified actions that they have taken or planned to implement the recommendations. Specifically, OCC stated that it plans to continue to ensure that its sampling guidance is used in the continuing reviews and that the agency plans to monitor for consistency. As we discussed in the report, developing and using objective measures to monitor and assess consistency among the continuing reviews is also important. With respect to the second recommendation, OCC stated that it plans to apply lessons learned from the foreclosure review to the activities under the amended consent orders. In response to the third recommendation, OCC noted that it recognized the importance of providing additional information to the public about the processes, status, and results of the continuing reviews and activities under the amended consent orders. As such, OCC said it plans to issue at least two public reports, as we noted in the draft report. The Federal Reserve stated that it plans to continue to coordinate with OCC to provide consistent guidance during the continuing reviews. As we discussed in the report, developing and using objective measures to monitor and assess consistency among the continuing reviews is important. With respect to the second recommendation, the Federal Reserve stated that it has expanded its planning and monitoring efforts during the course of the foreclosure review and plans to continue to devote resources to planning and monitoring as it implements the amended consent orders. In responding to the third recommendation, the Federal Reserve outlined a number of steps that it and OCC are taking to communicate about the continuing foreclosure reviews and activities under the amended consent orders. These steps, such as developing a letter to explain why borrowers are receiving a payment and updating borrowers covered under the continuing reviews who submitted a request-for-review, will help provide information to affected borrowers. Moreover, the Federal Reserve and OCC said that they have committed to providing public reports that detail the implementation of the agreements. They also said that they anticipate these reports will include information about the findings of completed reviews, the number of requests for review, and the status of other activities under the consent orders. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies to interested congressional committees, the Board of Governors of the Federal Reserve System, Office of the Comptroller of the Currency, and other interested parties. In addition, the report will be available at no charge on the GAO website at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-8678 or evansl@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix IV. The objectives of this report were to assess: (1) challenges to the achievement of the goals of the foreclosure review, (2) the extent of transparency in the foreclosure review process, and (3) lessons that could be useful for activities under the amended consent orders and continuing reviews. The scope of our work was limited to the foreclosure review at the 14 servicers that are subject to the April 2011 consent orders. We were in the process of reviewing various aspects of the foreclosure review when the Office of the Comptroller of the Currency (OCC) and the Board of Governors of the Federal Reserve System (Federal Reserve) announced agreements with 11 of the 14 servicers to discontinue the foreclosure review and replace it with a broad payment process. To assess challenges to achieving the goals of the foreclosure review process, we identified three areas of the review process to use as a basis for our analysis. Specifically, we analyzed consultants’ application of guidance to conduct the reviews and recommend remediation; consultants’ development of test questions related to loan modifications, state foreclosure laws, and fee reasonableness; and consultants’ sampling methodologies for selecting files to review. Based on our prior work on the foreclosure review, these were areas we identified as particularly challenging for ensuring consistent reviews. To obtain information on consultants’ development of test questions and application of regulatory guidance, we reviewed the guidance and conducted site visits and in-person interviews with five consultant engagement teams. To identify third-party consultants to interview, we selected consultants that were engaged to conduct reviews for servicers that are overseen by each regulator. In addition, we selected consultants engaged by servicers with a range in sizes of eligible population for the review, including some of the largest servicers. During the site visits we also observed demonstrations of the file review process and systems. In addition, we obtained responses to a standardized questionnaire from all of the consultants. We compared consultants’ processes for compiling reference materials, developing test questions, and applying guidance. We reviewed documents regulators used to monitor the file reviews, such as status reports and meeting agendas. We also interviewed regulator staff and examination teams on steps taken to promote and assess consistency in the reviews and the effects of any differences. To obtain information on third-party consultants’ sampling plans, we used a data collection instrument created by our statisticians. We analyzed information contained in engagement letters between the servicers and consultants for the sampling parameters consultants used to select files. We confirmed key observations of our analysis in interviews and site visits with officials responsible for developing the sampling plans at five third- party consultant engagement teams, interviews with regulator staff, and through examination teams that reviewed the plans. In addition, we obtained information on consultants’ plans for additional analytical methods and confirmed other observations of our analysis through a standardized written questionnaire to consultants. We compared the information and these parties’ actions to criteria such as the regulators’ standard practices, stated goals for the foreclosure review, and our internal control standards. In addition, we reviewed our prior work on the Troubled Asset Relief Program (TARP), Home Affordable Modification Program, and reports where we established criteria on lessons-learned procedures, generally accepted project management practices, and effective management practices. We also referred to our sampling standards and several other references on sampling standards from OCC, the American Institute of Certified Public Accountants (AICPA), and the Office of Management and Budget (OMB). We considered these practices applicable to the foreclosure review because they concern the use of data from samples to draw inferences about the populations from which the samples are drawn. To further evaluate and asses the sampling plans, we also compiled a list of common sampling terms based on the work cited above, as well as other analyses, and vetted the definitions for these terms internally with our methodological staff, including staff responsible for sampling procedures related to financial audits and program audits. Any data we obtained on the number of borrowers in the scope of the reviews or the status of the reviews were used for background purposes only and were not used to support our findings and conclusions. As such, we obtained information from regulator staff and the data administrator about how the data were obtained and compiled, but we did not assess the reliability of the data. Statistical sampling can be a powerful tool for drawing inferences about populations when a full census of cases is infeasible. Well designed and executed samples allow researchers to make estimates of population characteristics and to specify uncertainty due to sampling error associated with those estimates. Parameter estimates and associated measures of precision (such as confidence intervals or margins of error) are developed from the sample data using estimation formulas consistent with the sampling plan actually used. For example, if the sample design were complex (such as using differing probabilities of selection for different portions of the population) and if estimation formulas for simple random sampling were used to produce estimates, those estimates and the associated confidence intervals would likely be incorrect. Additionally, estimates projected to a population of interest may be subject to bias when a sample fails to cover the population of interest, such as when the list used for sampling (the sampling frame) excludes relevant units or when data from only a portion of the sample are collected. The design of an appropriate sample is intricately linked with the goals of the sample. These goals can include the importance of establishing the estimate within a narrow confidence interval, the need to conduct statistical testing against a threshold or for differences, the ability to make estimates to subpopulations, and the desire to report specific results. A sample that is designed to test against a specified tolerance level may use different sample sizes and a different approach to statistical testing than a sample that is designed to estimate the level of an attribute in the population. Some uncertainty is implicit in sampling as a tradeoff for such factors as the cost and time required to examine all of the data. However, according to AICPA guidelines, sampling is inappropriate if there is no tolerance for risk of possible erroneous decisions as a result of examining only a sample of the data. For example, figure 2 demonstrates that these two consultants interpreted the “precision” requirement differently. Consultant A interpreted precision as being the threshold for the tolerable error rate, whereas Consultant B interpreted this as requiring a margin of error within plus or minus 3 percentage points of the estimate. Because Consultant A is testing against a threshold using the methodology specified in the OCC handbook on sampling methodologies (based on a Poisson distribution), it uses the upper bound of its one-sided 95 percent confidence interval to draw inferences about whether the population error rate is less than an established threshold value. Consultant B, in contrast, uses a two-sided 95 percent confidence interval because it is developing point estimates. Although both consultants achieve the 95 percent reliability and 3 percent precision guidance from regulators, the consultants’ different approaches to sampling resulted in different sample sizes and different decision rules for when additional sampling or full review of a population would be required, even when the same number, or even proportion, of errors are found in the sample. For example, if 2 errors were found in Consultant A’s sample, one could conclude with 95 percent confidence that the error rate in the population was below 6.3 percent and that the 3 percent precision threshold had not been met. Additional sampling would be required in this case. If 2 errors were found in Consultant B’s sample, one could conclude with 95 percent confidence that the error rate in the population is between 0.1 and 1.9 percent and no additional sampling would be required. If the proportion of errors in Consultant B’s sample was similar to Consultant A’s at approximately 2 percent, or 8 errors, one could conclude with 95 percent confidence that the error rate in the population was between 1 and 4.3 percent, a much narrower estimate than that from Consultant A. Neither of the samples illustrated in figure 2 was designed with the goal of estimating the number of harmed borrowers in the population. However, had samples of this size been drawn with the purpose of estimating the number of harmed borrowers, the width of the two-sided 95 percent confidence intervals around each estimate would vary. Assuming a population size of 100,000 cases and a binomial distribution, and that each loan with an error corresponds to one harmed borrower, samples similar to those in figure 2 would result in different ability to concisely estimate the number of harmed borrowers. If two errors were found in a sample of 100 files, a 95 percent confidence interval for the estimated number of harmed borrowers in the population would run between approximately 240 and 7,040 borrowers. For a sample of 370 cases, a similar proportion of errors (8 errors in a sample of 370) would result in a 95 percent confidence interval around the estimated number of harmed borrowers between approximately 940 and 4,220, a much narrower confidence interval. To effectively use a baseline sample from the full population of loans to monitor whether the foreclosure reviews have identified the majority of harmed borrowers would require setting a realistic expected error rate to calculate sample size and establishing guidelines for an appropriate margin of error around the estimate to ensure that sample estimates were sufficiently precise to meet regulators’ goals. To assess the extent of transparency in the foreclosure review process, we reviewed press releases and documents from regulators related to the foreclosure review. In particular, we reviewed what documents related to the consent orders were available on the regulators’ websites, such as speeches by agency officials, engagement letters, outreach materials, and press releases, and analyzed the content of these documents. We compared this documentation against agency policies on public disclosure of enforcement action information and our previous work on transparency in government programs, such as our work on TARP to identify any similarities and differences. Further, we reviewed reports and summary documentation from the National Mortgage Settlement and interviewed the monitor of the settlement to provide the context of a current example of a large-scale settlement involving similar stakeholders and issues similar to those of the foreclosure review. We also conducted interviews with regulator staff, selected third-party consultants, and consumer groups. To identify lessons learned that might be useful for the activities under the amended consent orders, we compared our findings from the first two objectives with the processes and goals regulators outlined for the activities under the amended consent orders and identified areas with similar challenges. We reviewed press releases on the agreements and amended consent orders from the regulators and interviewed regulator staff about the goals and rationale for the agreements and their plans for designing and implementing the activities under the amended consent orders. We also reviewed our internal control standards, our prior work identifying key practices during agency transformations, our prior work covering TARP and other government programs and agencies, and documents related to the National Mortgage Settlement. In particular, we focused on information in these reports on project design, oversight of progress, and communication with stakeholders. Although the foreclosure review and the activities under the amended consent orders are not government programs, we considered the steps identified in this work as applicable because planning, monitoring, and communication are key principles of any effective process. We conducted this performance audit from July 2012 through March 2013 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. In addition to the contact named above, Karen Tremba (Assistant Director), Bethany M. Benitez, John Karikari, Charlene J. Lindsay, Patricia MacWilliams, Marc Molino, Jill Naamane, Anna Maria Ortiz, Robert Rieke, Jennifer Schwartz, Andrew Stavisky, Sonya Vartivarian, James Vitarello, and Monique Williams made key contributions to this report. | Since April 2011, OCC and the Federal Reserve had been overseeing the foreclosure review, a requirement of consent orders entered into by 14 mortgage servicers. This undertaking involved a review of loan files by thirdparty consultants to identify errors in servicing and foreclosure practices. More than 4 million borrowers were eligible for reviews. In January 2013, the regulators announced agreements with 11 of the servicers that replaced the reviews with a broad payment process to compensate borrowers in a more timely manner. Reviews continue for three remaining servicers. GAO has been reviewing various aspects of the foreclosure review process. This report addresses: (1) challenges to the achievement of the goals of the foreclosure review, (2) transparency of the process, and (3) lessons that could be useful for carrying out activities under the amended consent orders and continuing reviews. GAO analyzed third-party consultants' sampling plans, reviewed regulatory guidance and other documents, and interviewed representatives of third-party consultants and law firms, consumer groups, and regulators. Complexity of the reviews, overly broad guidance, and limited monitoring for consistency impeded the ability of the Office of the Comptroller of the Currency (OCC) and the Board of Governors of the Federal Reserve System (Federal Reserve) to achieve the goals of the foreclosure review--to identify as many harmed borrowers as possible and ensure similar results for similarly situated borrowers. Regulators said that coordinating among foreclosure review participants was challenging, and consultants said that the reviews were complex. In spite of regulators' steps to foster consistency, broad guidance and limited monitoring reduced the potential usefulness of data from consultants and increased risks of inconsistency. For example, GAO found that guidance was revised throughout the process, resulting in delays. Other guidance did not specify key sampling parameters for the file reviews and regulators lacked objective monitoring measures, resulting in difficulty assessing the extent of borrower harm. Good planning and collecting objective data during monitoring provide a basis for making sound conclusions. Without using objective measures to assess sampling or comparing review methods across consultants, regulators' ability to monitor progress toward achievement of foreclosure review goals was hindered. Although regulators released more information than is typically associated with consent orders, limited communication with borrowers and the public adversely impacted transparency and public confidence. To promote transparency, regulators released redacted engagement letters and guidance on remediation. In addition, OCC released two interim progress reports. However, some stakeholders perceived gaps in key information and wanted more detailed information about how the reviews were carried out. Regulators stated they considered publicly releasing additional information, but expressed concerns that releasing detailed information risked disclosure of confidential or proprietary information. Further, borrowers who requested reviews experienced gaps in communication. For example, borrowers who submitted requests when the submission period opened waited nearly a year before receiving an update. The foreclosure review activities to date highlight key lessons related to planning, monitoring, and communication. GAO's prior work shows that assessing and using lessons learned from previous experience can benefit the planning of future activities. The foreclosure review produced lessons in advanced planning and establishing mechanisms to monitor progress toward goals. Without assessing and applying relevant lessons learned, regulators might not address challenges in the continuing reviews or similar challenges in activities under the amended consent orders. In particular, regulators announced the agreements that led to the amended consent orders without a clear communication strategy. Although the regulators plan to release reports on the results of the amended consent orders and the continuing foreclosure reviews, neither regulator had made decisions about what information to provide to borrowers. GAO's internal control standards and best practices indicate that an effective communication strategy and timely reporting can enhance transparency and public confidence. Absent a clear strategy to guide regular communications with individual borrowers and the general public, regulators face risks to transparency and public confidence similar to those experienced in the foreclosure review. OCC and the Federal Reserve should improve oversight of sampling and consistency in the continuing reviews, apply lessons in planning and monitoring from the foreclosure review, as appropriate, to the activities of the continuing reviews and amended consent orders, and implement a communication strategy to keep stakeholders informed. In their comment letters, the regulators agreed to take steps to implement the recommendations. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.