Seven Years of GPRA: Has the Results Act Provided Results?
by Guest Blogger, 7/8/2002
Ellen Taylor testified before the House Committee on Government Reform Subcommittee on Government Management, Information and Technology
of the House Committee on Government Reform on "Seven Years of GPRA: Has the Results Act Provided Results?"
Statement of
Ellen Taylor
Policy Analyst
OMB Watch
Before the House Committee on Government Reform
Subcommittee on Government Management, Information and Technology
of the
House Committee on Government Reform
On
"Seven Years of GPRA: Has the Results Act Provided Results?"
July 20, 2000
Mr. Chairman and members of the Subcommittee, thank you for the opportunity to submit testimony on the implementation of the Government Performance and Results Act.
My name is Ellen Taylor, and I am a policy analyst at OMB Watch, a nonprofit research and advocacy organization that seeks to promote government accountability and citizen participation. For the past three years, OMB Watch has actively sought to increase the participation of nonprofit groups in the implementation of GPRA. In that effort, we conducted surveys of nonprofits, executive agencies, and congressional staff, and hosted meetings, especially in the environmental area, to monitor implementation and opportunities for nonprofit involvement. We also maintain an email list, involving hundreds of people, to discuss issues surrounding GPRA. Based on our work-and especially our lack of success in involving nonprofits in the implementation of GPRA-the following are some specific thoughts and suggestions about the future of GPRA.
First off, we applaud you for holding this hearing. Considering GPRA's promise for strengthening government, there have been too few such hearings. Without the concerted and constructive involvement of Congress (particularly through oversight), GPRA cannot live up to its potential, and is in danger of becoming just another bureaucratic paper exercise in the long history of government reform efforts.
To make sure that doesn't happen, our recommendations are threefold: 1) Congress should mandate stakeholder involvement in the performance planning process and encourage active solicitation of comments on performance measures; 2) federal agencies must make greater use of the Internet to make data and performance measures publicly accessible to strengthen transparency and accountability in government; and, 3) Congress and Executive agencies must work together in partnership to create an environment within which agencies can effectively implement GPRA.
There is no doubt that GPRA can be a powerful tool. We believe these three changes may help GPRA to strengthen government accountability and to enhance public trust in its functions by highlighting government's achievements.
I. CONTEXT OF THE GOVERNMENT PERFORMANCE AND RESULTS ACT
- GPRA's success depends on government's commitment to it. It is important to keep in mind that, historically, GPRA is but one in a series of government reform efforts, extending back to before the Hoover Commission and continuing today. Each Administration and many Congresses like to put their particular stamp of reform on government. As a result, there is often a lack of consistency and commitment to a particular reform effort. Over the years, the civil servants of government have been skeptical of reform because of this. As Steven Kellman describes in this month's Government Executive magazine, reform efforts simply become another "drill." Paul Light, in his book, The Tides of Reform, Making Government Work 1945-19951, lists twelve pages of laws enacted by Congress with the purpose of improving government operations. Paul Posner also discussed past initiatives to make government more effective at length in the GAO report, Performance Budgeting: Past Initiatives Offer Insights for GPRA Implementation2. GPRA has some important differences from the efforts that went before it, but it cannot be divorced from them. If GPRA is to succeed and not become yet another partially successful, or failed, or abandoned effort - Congress and the Executive Branch, in partnership, must build on past efforts as we move to a new Administration and devote time and resources to the full implementation of GPRA.
- Satisfaction with government services is nearly on par with private sector services, but trust in government is low, albeit rising. Curiously, satisfaction with the performance of specific government services is measurably improving, while overall trust in government is only slightly increasing (but overall remains low). This makes for a confusing context for an effort aimed at improving public confidence in the capacity of the federal government in terms of performance and results. It seems that performance, measured as efficiency and customer service, are not the whole story. A Pew Research Center Survey, "Performance and Purpose; Constituents Rate Government Agencies," released in April 2000, studied five federal agencies and found that citizens gave high marks to these agencies' performance but that judgments about government "rely as much on attitude about an agency's purpose as on evaluations of its performance." They also noted only a slow rise in overall trust in government in February 2000 from 1992, still not rising to the level of trust exhibited in 1988. The American Customer Satisfaction Index, released in December 1999, rated customer satisfaction with government as compared to customer satisfaction in the private section, finding that the public sector is on par with the private sector.3 (Public services had a range of 51 to 87 on a 0-100 scale, with a weighted aggregate score of 68.6; private services had a range of 53 to 86 with a weighted aggregate score of 73.) ACSI also noted that trust in government is not the result of customer satisfaction with government; rather since we don't trust government, we are so pleased when it does something well that we accord it higher ratings. Our own focus groups, conducted by Lake, Snell, Perry and Associates, found great interest in investing in our future during this period of economic prosperity. Voters identified issues and programs run by the federal government as examples of investments that should be made. At the same time they negatively described politicians and bureaucracy, which rubs off on government in general. This is not surprising since we frequently see news stories concentrate on "waste, fraud and abuse" in government programs, while rarely does the media report success stories. For the purposes of evaluating GPRA, improving the performance of agencies and showing results may not, in fact, lead to increased citizen confidence in government as a whole. But, then again, it may help. The fact is that improved performance and measurable results are not the whole picture, and the context of citizen satisfaction with government services within an overall mistrust or ambivalence about government must be recognized. We recommend that Congress focus on the successes of government programs and agencies, which are many, and on the ways that services are improving, rather than maintain a focus on failures.
- Although GPRA was conceived as a tool to strengthen government performance, it has increasingly been used as a partisan tool to push ideological perspectives. GPRA was passed on a bipartisan basis with the hope that improving government performance would strengthen government and make it more accountable to the public. Unfortunately, as GPRA implementation proceeded some have used it to advance ideological goals-for example, to privatize, to downsize, or to cut agency budgets. If the goal of good government performance is put in the same hopper with political rhetoric and partisan polarization, there is no possibility that GPRA will be anything more than another failed attempt at government reform. GPRA should be about government performance. Nor was GPRA intended to be perfect in the beginning. The full implementation of GPRA was conceived as an iterative process. It will take time to get it right. While constructive criticism is necessary and important during the learning stage, it should not be framed as agency bashing, or as evidence that GPRA cannot ever achieve the goals for which it was designed. There must be an improved dialogue between Congress and Executive Branch agencies as partners in this process. A partnership between Congress and agencies, with the goal of improving performance, should ensure more meaningful and ambitious performance measures that improve as time passes. Another problem that we see with the implementation of GPRA is that while agencies are faced with GPRA requirements, such as developing performance measures and identifying performance baselines, they are simultaneously required to conform to the provisions of other laws that may conflict with GPRA. For example, the Paperwork Reduction Act, which mandates a reduction in paperwork burden, is at odds with GPRA's mandate to collect data to verify and validate performance. To satisfactorily address performance, it may be necessary to collect new data. Another example: While agencies are required to consult with stakeholders, at least some feel that it would be too expensive since they think they are required to comply with the Federal Advisory Committee Act, which costs money. Agencies must be given the resources to meet GPRA requirements without undue burdens. Congress has not raised or considered the difficulties that agencies have in fully implementing GPRA without being given any additional resources to do so. This has affected not only the ability of agencies to meaningfully involve stakeholders, but more broadly, the ability of agencies to accomplish the fundamental tasks of GPRA. Finally, GPRA will not provide a magical solution to difficult budget decisions. The use of GRPA and its performance reports in making budget decisions puts agencies in a precarious position. It could easily lead to performance goals that are set so low that meeting them will be a given, rather than setting goals that are high and striving to achieve them, and sometimes failing. If failure to achieve goals is used as a reason to cut budgets, the only performance report will be a good performance report, yet it will be meaningless in terms of the original purpose. Does bad performance always mean that a program's budget should be cut? Or might it mean that the program was operating with too little money to be effective? Does good performance automatically deserve more money? Or, might it indicate that an agency could, in fact, perform as well with a little less money? We recognize that one of the distinguishing and powerful features of GPRA is its linkage to the budget, but this is also a complicated matter and does not guarantee better budgetary decisions. While some have argued for a greater linkage between performance and resources, our point is that performance measurement may be a helpful tool, but quantified measures of performance will never reveal the whole story. Program performance should be just one of several factors influencing budgets. Congress will still be faced with difficult political decisions.
- There is considerable uncertainty about whether GPRA is working or not. There is no doubt that agencies are treating GPRA seriously. Lots of hard work has been put into agency strategic and performance plans. But what are they measuring? Are they using the correct measures? And is it appropriate to place such a heavy emphasis on quantitative measures? There are government activities that do not lend themselves to quantification. To assess the performance of Head Start, for instance, in any substantive way is a complex undertaking and requires a deep understanding of the program. How do you measure the improved well being of children who participate in Head Start? While qualitative measures are allowed, the emphasis is on quantifying performance. Other tools for measuring results, besides numbers, need to be tested and used. There is also a concern the emphasis on specific and quantifiable outcomes in GPRA could lead to a shift away from broader values and concerns that could have detrimental effects on the environment, health and safety, and consumers. For instance, Public Citizen issued a report in September 1999 finding a decline in OSHA workplace inspections. That decline was not linked directly to regulations under GPRA-rather to the "Reinventing Government" program-but it is an example of the potential negative effects that efforts to make a program more efficient by emphasizing measurable outcomes might have. Quantifiable results have a place, but should by no means be the only or even primary way of measuring results under GPRA. So far, GPRA generally has not provided meaningful information on the quality of government programs. Instead, the emphasis has been on efficiency and measurable results. Unfortunately, those may run roughshod over important values of equity and effectiveness. Moreover, program performance, in many cases, cannot be reduced to annual reports. Program evaluation is a complex process that takes many years. Head Start, if measured on an annual basis during its early years, may not have been perceived to be as successful as the long-term studies have now shown it to be. The work that the federal government performs may not always be cost-effective and capable of achieving a business model of efficiency, but then again, government is not a business. Government gets the problems that the private sector can't or won't handle.
- There is not a lot of public awareness about GPRA. The release of the first performance reports in March 2000 ought to have made headlines. Perhaps what is most striking right now is the remarkable lack of interest in GPRA, except by government agencies and selected congressional committees such as this one. In our study, Measuring the Measurers: A Nonprofit Assessment of the Government Performance and Results Act, released in December 1998, only 93 of 241 nonprofit respondents said they had even heard of GPRA before receiving our survey, in spite of a very high level of awareness of and experience with performance measurement generally. This lack of knowledge about GPRA is supported by the fact that a majority of those on our "government performance" email list are from government agencies and departments. We have consistently noted that government, not the public, shows the greatest interest in GRPA. In June 1999, we did a survey of congressional staff, who should be at the heart of the GPRA process, and sent out over 200 questionnaires, followed up with at least one phone call, and had a return rate of only eleven responses. Neither the public, nor Congress (except for the General Accounting Office, which has shown sustained effort in following and commenting on GPRA), seems involved in GPRA. On the other hand, agencies have been spending a lot of time and effort to fulfill the requirements of GPRA. One respondent from the Food and Nutrition Service said that the agency had gone through a sea change-from thinking of GPRA as another reporting requirement, to seeing it as something useful and trying to integrate GPRA as a management tool, a plan for priorities, and a way of measuring performance. But if agencies are the only entities taking GPRA seriously, it cannot succeed.
- There must be meaningful participation of stakeholders in the development of performance measures. There is no requirement in GPRA that stakeholders be consulted in what is, perhaps, the real meat and potatoes of GPRA-the development of performance goals and measures. While the law requires agencies to consult with stakeholders on strategic plans, no such requirement exists for performance plans. The process of actually implementing GPRA-deciding what must be measured to show results and developing valid and verifiable methods of measurement-is not open to public participation. While we understand that the public cannot be privy to federal budget deliberations, the performance plans are simply too important to disallow public engagement on the selection of benchmarks and performance measures. We think that the only way measures can be verified and validated and yield useful results is with public engagement and participation. Developing measures and indicators at the performance plan level must be accomplished with stakeholder involvement. Not only should GPRA be amended to require stakeholder participation in preparation of the performance plan, but also agencies must affirmatively reach out to a variety of stakeholders to get their input. This is especially true with federal grantees and state and local governments who are partners in the process and who will be directly affected by federal action.
- GPRA reports need to be more easily accessible to the public and written in ways that are understandable by the public. While most (but not all) agencies do put their GPRA plans and reports on their websites, they are frequently difficult to locate. Even if they are found, many of the reports are not particularly helpful to laypersons. For example, they are sometimes laden with acronyms and jargon that is impossible for anyone outside government to understand. Because of the value of stakeholder participation in the GRPA implementation, it is vital that the reports be accessible and comprehensible to the public, and organized in ways that make sense. Information should be reported in a way that makes sense to people-whether by watershed, or congressional district, or state, local or regional divisions. Public accountability can have a positive effect on performance, and can shift some of the burden of improving standards to outside the agency. OMB Watch has long worked towards the goal that government information should be transparent-intelligible and useful to citizens. GPRA reports should not be an exception.
- The public must have access not only to the plans and reports, but also to the information used to determine performance measures. A key element of GPRA is that performance must be verified and validated. Generally, this has been interpreted as an effort to get the agencies to report on meeting their performance objectives and measures. However, making the underlying data used to benchmark performance publicly accessible would be a great boost in verifying and validating agency actions. The data need not be perfect-it is recognized that there are areas where agencies lack the baselines and historical data to measure improvement. There are some who would argue that the problems with data quality suggest that the information should not be disclosed. Our experience runs counter to that. As agencies disclose information, the quality of the data improves. Public exposure and input can actually positively affect the quality of the data. In fact, the performance report ought to be an annual report containing or linking to all of the underlying data used by an agency that is relevant to its goals and objectives, so that the public can easily locate, understand, and analyze that information. The availability of information itself can be a powerful force for results. Citizen participation through nonprofit input can be a valuable means of verifying and validating. Nonprofits should be able to go to an agency and say, "this is the way you should be doing measures," but they can only do this if the underlying information is available. GPRA can, in fact, achieve a public right-to-know purpose promoting a whole new level of accountability.
- If the government is successful in meeting the public's right-to-know, then it is conceivable that agency performance measures will be bypassed by public demand for accountability. Although EPA never identified specific amounts of reduction in emission of toxic chemicals, the public accessibility of the Toxics Release Inventory helped to create a 45 percent decline in the release of those chemicals.4 The point is that if the government makes information available in formats that are easy to find and use, such activity alone may force greater accountability and performance. We are not suggesting that a comprehensive right-to-know agenda is a replacement for GPRA, regulatory enforcement, or other mechanisms of government. But it can be a shot in the arm in building a public connection to government and confidence in the work of our federal agencies. OMB Watch has been developing a platform for strengthening public accessibility to government information. The platform provides specific recommendations for EPA in their efforts to promote the public's right-to-know. If this Committee would like a copy of the platform when it is completed, we would be happy to provide it.
- We support a greater partnership between Congress and the Executive Branch. We encourage Congress to continue requesting and using research and analyses by the General Accounting Office, since GOA reports have been some of the most informative and useful sources of information about GPRA. In addition, we encourage Congressional Committees to exercise their oversight authority. We say this with a certain amount of trepidation, because if oversight is done in a context of partisanship and ideological goals, it will be detrimental to the implementation of GPRA. However, if oversight were done in a spirit of partnership and cooperation, it would be a powerful force toward achieving the goal of improved government. Congress also has the ability through oversight not only to examine problems in government, but also to highlight those government programs and services that work, are effective, and accomplish important goals. This could have a huge impact on public trust, which, as previously discussed, does not always go hand in hand with improved performance.
