The authors examine the enduring difficulties of reforming the budgetary process, explain some of the causes of these difficulties and recommend budget reforms as starting points to improve U. S. federal government budgeting. They also report on initiatives of the Bush Administration and Congress to implement performance measurement in federal management and budgeting.
Over the last fifty years a number of changes have been introduced in the federal government budget process, largely by the executive branch of government. In the 1950s, again in the 1990s, and in 2002 an emphasis on costs, activities and outcomes led to various performance-oriented budget experiments. Sandwiched between these performance budget experiments were such reforms as Program budgeting; the Planning, Programming, and Budgeting System (PPBS); Management by Objectives; Target Based Budgeting; and Zero Based Budgeting. None of these were persuasive enough to gain universal adoption, although remnants of some systems exist in later systems and some, e.g., zero based budgeting, have been re-adopted under a variety of conditions. For example, the ranking system of zero based budgeting with its levels of effort below the current level finds its way in many systems that would not think of doing a comprehensive, zero based, review each year. However, when fiscal conditions are difficult, ZBB has its uses, particularly at state and local levels.
Performance Budgeting, Program Budgeting, and PPBS all were systems originally oriented toward national defense and the Department of Defense. Performance Budgeting focused on activities accomplished, outputs and unit costs. In the 1950s, performance budgeting did not get a lengthy trial in defense, but it did see some success in local government. Its Achilles' heel seemed to be the masses of calculations it took to derive unit cost data; in this it made budgets less rather than more comprehensible. Moreover and more importantly, not all government programs had measurable outputs, or even items that could stand as surrogates.
The Department of Defense went on to focus on programs with program budgeting which presented budgets in terms of programs, rather than activities or objects of expenditure, but it too frustrated analysis at the aggregate level and gave way to PPBS. This is an extremely complex resource allocation system used within the Department of Defense where it combines threat assessment, planning for the appropriate force structure and buying the goods and services and providing appropriate training to the people who will be able to avert or overcome the threat.
Civilian sectors of the federal government had a short flirtation with PPBS, as with management by objectives as a budget system, and Zero Based Budgeting. (1) PPBS was abandoned at about the time when more and more of the federal budget was directed toward mandatory expenditures and the systems which followed it were even less well equipped to answer the important questions about budget policy. Remnants of PPBS survive in attempts to estimate benefits and costs and to do long range planning, but few non-defense functions have a threat based world with which to contend and threat assessment and modifying force structure must be seen as the heart of the PPBS apparatus.
Under the Government Performance and Results Act of 1993 (GPRA), the federal government embarked on a performance measurement experiment that may lead to performance budgeting. In this iteration, strategic planning and customer/clientele involvement have been added to activity groupings, the search for appropriate outcomes and the effort to cost outcomes and evaluate changes in outcomes that might come from adding additional dollars. The strategic planning and customer involvement facets of this iteration are important changes. In the 1950s, performance budgeting seemed to focus on costs of activities, implicitly assuming that cost per unit numbers would eventually lead to more centralization of the budget process and allow a few people to make good judgments about all activities by following the changes in a few key numbers. Under GPRA, this iteration of performance measurement and strategic planning in the 1990s seems more open to decentralization of goals throughout the organization and to forcing customer/client desires to percolate up into consideration at higher levels. The promise of a more useful and effective system seems tantalizingly close. This promise has attracted subordinate levels of government in the U. S. (Florida, Texas, Arizona) and elsewhere (New Zealand). However, the record of budget innovation suggests that one ought not be too optimistic about reform. Ultimately, object of expenditure data still remain linked to the accounting system and accounting methodology--accurate and useful performance assessment against cost generally requires accrual accounting consistent with the Chief Financial Officers Act of 1990 whereas federal accounting is done generally on a cash basis, and strictly on a cash basis with respect to measuring and reporting the budget surplus or deficit. Hence, when worst comes to worst, the data for line item analyses can still be found. Particularly for Congressional appropriations committees and staffs, line-item and functional analysis has an enduring appeal.
The most recent reform introduced in federal budgeting was that implemented by President George W. Bush and his Office of Management and Budget in the FY 2003 President's Budget. This budget proposed to link performance to budgets to relate funding to performance measures and accomplishments for federal departments and agencies. The overall Bush management reform agenda included improvement in the strategic use of human resource capital, competitive sourcing (outsourcing), financial management improvements, initiatives to implement e-government, and linking performance to results and outcomes. In 2001 the President's Office of Management and Budget (OMB) reviewed agency management and performance against multidimensional sets of criteria supporting each of these five initiatives. A "traffic light" performance scorecard (red, yellow, green for poor, needs improvement or satisfactory performance) was used to report on agency performance relative to these five sets of criteria. These performance criteria also were applied in examining agency budget requests when OBM prepared the FY 2003 budget. The federal government Chief Financial Officer, who also serves as Director of OMB's Office of Federal Financial Management responsible for performance evaluation and guiding and monitoring implementation of GPRA, indicated that OMB would continue to evaluate agency performance quarterly and would report its scorecard results to Congress and the public semiannually.
While some critics objected to the perceived absence of methodological rigor or potential utility of the traffic light approach, they failed to understand that the three color grading system was supported by criteria adapted to evaluate the five initiatives for performance assessment. In fact, the General Accounting Office (GAO) evaluated the criteria applied by OMB and reported to Congress that the criteria assessed the appropriate variables of agency performance and their application across most government programs enabled better assessment of government-wide performance on a more comparable basis. On February 15, 2002, Christopher Mihm, GAO Director of Strategic Issues, testified to the House Sub-Committee on Government Reform and to Congress generally (see GAO, 2002) that GAO supported OMB and the Bush administration's performance assessment approach, and that it had the potential to improve implementation of the Government Performance and Results Act passed by Congress and monitored and reviewed by the oversight committees of Congress and GAO. The Chairman of the Sub-Committee, Rep. Stephen Horn expressed satisfaction with the administration's scorecard approach as did other members of Congress. Horn added, however, that he would like the authorization and appropriation committees of Congress that approve agency programs and appropriations to pay as much attention to the performance assessment initiatives of both the executive and Congress when they reviewed budgets. Horn's preference is important. It is one thing for the House and Senate committees responsible for monitoring agency management performance to use the scorecards and another for authorizing and appropriation committees to do so. For meaningful budget reform to take place so that performance criteria are analyzed and applied in congressional passage of the budget, it is imperative that the appropriations committees in particular adopt a performance measurement orientation, something these committees have resisted in the past due, in part, to their perception that to do so would reduce their authority and power to control how and where federal money is spent as a result of surrendering decision making discretion to "budgeting by formula" (McCaffery and Jones, 2001, chapters 1, 3 and 4).
It must be noted that GAO is the audit arm of Congress and seldom sides with the executive branch when evaluating government agency efficiency. Thus, the endorsement from GAO for the President's initiatives seemed to indicate a willingness of some committees of Congress to cooperate with the Bush administration's efforts to implement performance management and budgeting. Thus, in the early part of the FY 2003 budget review cycle it appeared that the President and Congress were generally in pursuit of the same types of information, evaluation and reporting on executive agency performance. The absence of alliance between the executive and Congress has been applied to explain the failure of many past government management and budget reform initiatives. Consequently, in February 2002 it appeared that the chances for meaningful performance...