Performance Measurement and Government Budgeting

Bibliographical Essay in Independent Study

By Mwabilu L. Ngoyi

Rutgers Newark

TABLE OF CONTENTS Page I. Introduction II. Concern for Citizen Participation and Democratic Governance of Budgeting III. Assessment of Uses of Performance Measurement in Budgeting IV. Past Federal Efforts to Link Performance Measures and Budgeting V. Current State Governments Experience with Performance-Based Budgeting. VI. Theoretical Considerations for the Budgetary Reform and Performance Measurement VII. VIII. Prospects for Using Performance Measures in the Budget Process Conclusion 11 15 18 19 4 7 8 3 3

IX. Selected Annotated Bibliography

I. Introduction The literature on performance measurement is voluminous and may have been growing at very rapid pace in recent years. However, the following bibliography is patently selective in that it includes only those references in the past decade that illuminate a link between performance measurement and government budgeting. It will then seek to capture the issues that will help to shed the light on government budgeting effort to operate like a business and on cautious approaches on how to allay concerns for citizen participation and improving democratic governance of budgeting. In particular, students of public administration would be interested to seek and acquire the knowledge of remedies and strategies that can be applied to overcome obstacles to using rational model of budgeting and to successfully linking performance measures and budgeting. Henceforth, past as well as government current efforts to link performance measures and budgeting, ways including prospects for using performance measures in the budgetary process will be central in this bibliography. Furthermore, the bibliography will contain primarily books, articles, and monographs that are in English and that meet the requirements of scholarly investigation. This quality is important if the bibliography is to be used as a tool for research. II. Concern for Citizen Participation and Democratic Governance of Budgeting Much of the taxpayer revolt and the reinventing government movement have stemmed from the inability of public servants to think out of the box and to communicate how well or poorly their work serves the public. It has been observed, however, as tax revolts and citizen complaints have increased, the issues most citizens, managers and officials are alarmed about have surfaced in the budget. Hence, the budget reflects every issue, with a power to force action in the direction citizens desire. Arguably, the current wave of budgeting reform focuses not only on the performance of government agencies and programs (Melkers and Willoughby, 1998), but also on government accountability (Griefel 1993, 403) and sharing of information as well as decisionmaking in budgeting. Berman (1997) indicated that sharing information about what governments do well, through performance measurement and reporting, can reduce citizens

cynicism. Furthermore, broader participation in the trading-off of various goals and means to achieve them, as well as in developing performance measures will yield better decisions and a sharing of risks between citizens and government officials. It is important to point out that the emphasis on accountability can be explained by the widespread belief in recent years of government being uncontrollable and only responsive to interest groups (Rubin 1996, 112). Hence, the argument that government must be driven by citizen desires and needs rather than by government rules and regulations is now repeatedly being voiced (King, 1995). Government accountability, which was until recently basically an accounting matter and largely established through fiscal and budget reports, has this past decade seen its meaning considerably broadened (Campbell, 1997). Citizens now expect governments not only to be held accountable for its actions and the results of its actions (Campbell, 1997), but also to better communicate to them progress toward goals and objectives (King, 1995). Moreover, citizens strongly feel that they should and will pay only for results and no longer for government efforts (Melkers and Willoughby, 1998). Understandably, this pressure has lead governments across the country to embark on performance-based efforts (Broom and McGuire, 1995) in order to adapt to this new environment by improving the accountability of the budget to the citizenry. This trend is evidenced in the increased interest by governments in not only developing new measurement and reporting systems (Ammons, 1996; Lynch and Day, 1996), but also attempting to link funding to results or outcomes (Gianakis 1996, 127-143). These new systems, which are, generally, called outcome and performance measurement systems (Kravchuck and Schack, 1996), provide data that allow citizens to assess how economically, efficiently, and effectively government uses public funds. Certainly, the expansion to deal with inputs, outputs, and outcomes with citizen participation in understanding and framing issues will characterize budgeting in the 21st century. III. Assessment of Uses of Performance Measurement in Budgeting

Efforts to mandate improvements in performance measurement have focused on the budget process. Reforms in the Progressive Era have stressed the ability to base budgetary choices more explicitly on desired results. More recently, through the passage of the Chief Financial Officers (CFO) Act and the passage of legislation on performance measurement, the legislative branch [Congress] is once again addressing the issue of using performance measures in the budget process. The Congressional Budget Office (1993) indicated that there are three possible advantages to linking performance measures to the budget. These include the allocation of resources, improvement of agency management, and financial reporting. Allocation of Resources Governments at all levels could use performance information to make decisions on how to allocate scarce resources among competing priorities. Performance measures will prove most beneficial if they can help in determining how much money should be spent on the various purposes of government. Ideally, these choices would be aided by a better expectation of what a new dollar for one activity would buy compared with spending the same dollar on some other activity. Hence, traditional line-item budgets could be replaced by a system granting program managers greater flexibility in managing their resources, but holding them accountable for achieving program results. Such replacement is the goal of those advocating the use of performance measurement as a tool that can transform the [federal] budget process. It is important to indicate that the use of performance based budgeting, such as outcome budgeting constitutes a substantial deviation from the incremental line-item budgeting. The two methods of budgeting imply substantially different focuses; line-item budgeting centers almost exclusively on how much money is being spent, and performance-based budgeting would concentrate on varying levels of program results that might accompany varying levels of funding. But, using performance measures to allocate resources is difficult. First, while it is hard to develop measures based on results for government programs, it is also perceived as an almost impossible task to establish common denominators of performance among the activities of government. Furthermore, using performance measures for resource allocation implies knowledge of how to compare the measures for those two activities in a way that would inform trade-offs between the two. Unfortunately, trade-offs between government activities are almost exclusively a function of the perceived need and priority for government

action. It must be kept in mind that politics plays an important and legitimate role in budget decisions, even where measures of outcomes exist. Second, the relationship between performance and the budget is not straightforward. Poor results may be caused by the difficulty of the problem being addressed rather than by inadequacies in the design or management of a program. Nevertheless, performance measures may provide useful information for decision-makers. For example, if policy-makers could obtain information about the connection between a given level of resources and a given level of results for a program, they might choose to provide a larger budget only if the incremental improvement in outcomes was judged to be worth the additional expenditure. In this way, performance measures can inform the budget process without dictating budget outcomes. Performancebased budgeting, then, is about shifting the debate over resource allocation from its current focus on inputs to focus on results. This change in focus does not, nor does it intend to, remove politics from the process (CBO, 1993). Agency Management of Internal Resources When government wide decisions on resource allocation cannot be affected, agencies may find measurements valuable for improving their management of a given level of resources, regardless of whether their use results in a significant shift of resources from one program or agency to another. For instance, an agency that is organized geographically could use performance measures to target resources toward those regions where the work-load is greatest or where the problems are most acute. Furthermore, ties may be developed between the measurement of organizational and individual performance following the suggestions formulated by proponents of pay-for-performance schemes (CBO, 1993). Financial Reporting The governments or specific agencies may use performance measures to report their accomplishments to elected officials and decision-makers. At the state and local level, the Governmental Accounting Standards Board has called for service efforts and accomplishments (SEA) reporting. The GASBs Service Efforts and Accomplishments (SEA) initiative, which was implemented in 1987, has devised means of measuring the results of programs at local and state levels and then linked the results with accounting information

(Governmental Accounting Standards Board, 1994; Hatry and Fountain, 1990). At the federal level, the Chief Financial Officers (CFO) Act, which was adopted by the US Congress and the Bush administration in 1990, has focused on the appointment of personnel and the establishment of procedures to improve federal accounting. In addition, it required agency CFOs to include in their financial statements systematic measures of performance for programs(Hendrick and Forrester 1999, 608). IV. Past Federal Efforts to Link Performance Measures and Budgeting According to the Congressional Budget Office (1993), the current effort to link performance measurement to budgeting appears a logical successor to the efforts undertaken by the [federal] government the last four decades to tie performance to budgeting on a government wide scale. Such efforts consist of performance budgeting, program budgeting (including the prototypical program budget format known as the Planning Programming Budgeting System or PPBS), and zero-based budgeting (ZBB). It is important to indicate that these past systems commonly attempted, in part, to increase the level of information for [federal] budgeting and management (CBO, 1993). They represent a top-down approach to budgeting and management, which was instituted by a President and concerned with linking the use of resources and the outcomes resulting from their use. These systems, in particular PPBS (Schick, 1992) and ZBB, attempted to budget on the basis of program effectiveness. These approaches were labeled as rational budgeting systems and characterized by critics as antithetical to political norms and unworkable in a pluralist and individualist society (Gross, 1969; Lindblom, 1979; Wildvasky, 1966, 1969). Critics considered bargaining as an essential element in policy-making (Pilegge 1992, 77). Accordingly, since budgeting, which is about allocation of resource, is inherently political, the assumption of rational allocation of resource was judged unrealistic since it omitted an explicit consideration of political factors (Wildvasky, 1966). Moreover, the political system works best with budgetary schemes aimed at minimizing, but not focusing on, conflict (CBO, 1993). Furthermore, traditional budgeting required fewer explicit decisions in the allocation process as well as little annual review of the large budget compared to these systems. Finally, these approaches not only suffered from the inability to achieve a consensus between top managers and subordinates over goals and

objectives, they also had an unintended consequence of increasing the demand for analysis, the capacity to do it, including its use in devising solutions to societal policy issues. The Congressional Budget Office (1993) further asserted that, by a general consensus, these past efforts did not alter the resource allocation process. Important lessons, which emerged from attempt to design system purported to link measurement and budgeting, are that: 1). All personnel agency must be committed to the measurement system; 2) Collected data must be used; 3) Challenges in linking outcome measurement to budgeting must be accounted in the design of budget system and that time must be devoted to sort clearly out these linkages. V. Current State Governments Experience with Performance-Based Budgeting Florida, Louisiana, Minnesota, Oregon, and Texas are among the most well-known states concerning their implementation of performance-based budgeting and application of management activities (Broom and McGuire, 1995; Melkers and Willoughby, 1998). Melkers and Willoughby contend that, in the ideal, performance-based budgeting requires measurement of results, outcomes, and impacts. The Congressional Budget Office (1993) indicated that these states did not show that performance measures had a significant impact on their budget process. However, Oregon stands out to be the first state in the nation to have implemented outcome budgeting (Blanche-Kappler and Lissman, 1996). Outcome budgeting not only looks similar to these earlier fiscal practices when viewed from the measurement perspective (Schick, 1990), but it also built upon them in that: 1) its use of strategic planning and program structure draws upon the Planning-ProgrammingBudgeting-System; 2) the setting of objectives and targets is derived from the Management by Objectives; and 3) its designation of expected levels of performance for each level of expenditure was introduced with the Zero-Based Budgeting (Southern Growth Policies Board, 1996). However, Schick (1990) asserts that unlike past innovations, which aimed to optimize programs, this current development seeks to improve organizations. He further indicated that current management-oriented reforms are concerned with what organizations do and produce and place emphasis on means of holding them accountable for performance. These reforms aim to provide managers financial and other incentives that make them more aware of and accountable for costs and performance and more willing and able to shift resources to more

productive activities (Schick, 1990). Outcome budgeting, thus, appears to be a product of a new historical paradigm shift. Such system of budgeting, which affords less control over line items in exchange for agency accountability for results, constitutes a fundamental change from the existing budgeting practice. Oregon experience with outcome budgeting

Martin (1997, 119) indicated that the State of Oregons approach to outcome budgeting can be described as a combination of both the linking and purchase approaches operating at the community/state level. Oregon started experimenting with performance measurement in 1988 and its legislature formally adopted the Oregon benchmarks in 1993. The Oregon Progress Board (undated, 23-24) reports that the Oregon law states in chapter 724, it is the policy of this state to create and administer programs and services designed to attain societal outcomes such as the Oregon Benchmarks and to promote the efficient and measured use of resources. In 1994, the State of Oregon presented to the federal government the Oregon Option, a proposal which sought to pilot test new forms of intergovernmental cooperation based on managing for results (Martin 1997, 121). The terms were such that the state would receive federal funding related to mutually agreeable measurable results in the areas of workforce development, childhood health, and family stability. As part of this pilot test, the federal Department of Health and Human Services (DHHS) agreed to grant greater decisionmaking authority to the Oregon Adult and Family Services Division (AFS) in exchange for agreeing to be held accountable for results using the Oregon Benchmarks. This agreement made Oregon the first state in the nation to receive outcome-based funding from the federal government; henceforth, AFS could receive as much as $15 million in additional federal matching funds (Blanche-Kappler and Lissman 1996, 8).

Use of performance measures in an outcome-based budget

Outcome budgeting, described by Osborne and Gaebler (1992, 61) as a system that focuses on the outcomes of the funded activity, has received little attention in the literature of either public administration or human services. Hendrick and Forrester (1999, 577) point out that outcome budgeting represents a contract between principal (elected officials and citizens)

10

and agents (agencies), which purports to deliver services and meet performance in exchange for funding. In contrast to other performance budgeting, which places an emphasis on outputs, economy and efficiency, outcome budgeting is rather concerned with outcomes and effectiveness (Hendrick and Forrester 1999, 569). This format involves the analysis of results, accomplishments, or impact, which sets it apart from all other budget types (Martin 1997). Moreover, while other budget systems targeted internal stakeholders (government managers and administrators), outcome budgeting attempts to communicate with and educate external stakeholders (elected officials, citizens, clients, advocacy groups and others) about government accomplishments and the costs of achieving these results (Martin, 1997). Such scope sets apart outcome budgeting as a truly new species of budgeting. Its implementation depends on the ability of policy-makers to ensure that the contractual agreement is upheld (Hendrick and Forrester 1999, 577). Martin (1997) indicated that the implementation of this format in government human service agencies involved the selection of basic approach (i.e. linking and purchase approaches) and a unit of analysis (i.e. at program or service level; agency or organizational level; and state or community level or any combination thereof). Osborne and Gaebler (1992) indicated that the linking approach involves the inclusion of outcomes as part of agencys budget documents and budget processes as outputs are included in performance budgets. Such approach allows external stakeholders to see what resources are allocated to achieve a planned outcome. contractual way (Martin, 1997). In the purchase approach, specific resources are allocated to achieve specific outcomes in a quasiThis approach is more sophisticated and difficult to implement when compared to linking approach. Martin further indicated that while outcome budgeting may be easy to implement at the program/service level, it is difficult to implement at the community or state level despite its appeal. This is because relating outcomes and resources to changes in state and community indicators is difficult at best and raises issues not encountered at other levels. Martin also reported that outcome budgeting at the state level is compatible with state and communitys benchmarking efforts. Budgets are generally linked to geographically determined priorities or to purchase specific desired increases or decreases in social indicators.

11

Gianakis (1996, 141) asserted that outcome-based performance measurement systems help to realize the promise of the executive budget and that of professional public management. He pointed out that the exercise of policy and financial controls [which are unavoidable elements of public management] through the oversight of program inputs can compromise the effectiveness of program managers. Using program outcome measures in the resource allocation process enable these oversight functions to focus on program results. He sees such focus as less intrusive to public managers and allows them greater discretion in deciding how inputs will be used to accomplish targeted results. Thus, managers can focus their energies on achieving those results instead of concentrating them on the political games necessary to secure inputs in the absence results-oriented measures of performance. Such promise, he added, can act as catalyst for program managers to participate in a meaningful way in the development effort (1996, 142). VI. Theoretical Considerations for the Budgetary Reforms and Performance Measurement Despite all the efforts in the past decades to reform budgeting in order to improve the efficiency, effectiveness, responsiveness as well as accountability in governmental actions, criticism about budgeting has not been silenced and the pursuit of better budgetary theories continues. In an essence, incrementalism has remained the major theory in budgeting as explained by Wildavsky in his The Politics of the Budgetary Process (1979). Lindbloom (1979), in his Still Muddling, Not Yet Through, claims that incrementalism, albeit its limitation and the fact that it represents non-comprehensive change, was the most common method of policy making. Research conducted by Wildavsky (1964 [1979]) and Fenno (1966) has shown that very little new policy occurred during the annual budget process due to many factors, which included among other things: 1) the lack of formal analysis; 2) the reliance on the political judgment and procedural expertise of key participants; 3) the acceptance of past decisions; and 4) the decentralization and fragmentation of the budget process that impaired the formulation of any integrated policy. Gianakis (1996) indicated that the normative aspect of incrementalism encourages national public policy to be made through marginal adjustment to existing line-items considering the peoples limited capacity of understanding and solving societal problems.

12

Furthermore, he also reported that incrementalists favor planning and analysis to occur at the margin of the policy process because of time constraint in reaching consensus on annual federal budget that precluded thorough analysis of policy options. However, Rubin (1990) indicated that incrementalism is a political theory of the budget process and the different political environments and institutions of state as well as local governments have made them less accessible to theory building in this area. Rubin explained that the executive organization at the federal level was represented by incrementalist independent agencies whereas state and local governments were characterized by integrated service delivery organizations. Such situation had made state and local governments more prone than the federal government not only in experimenting with budget reforms, but also in adopting management techniques as well as efficiency tools developed by budget reformers. Thus, for several decades, reform in budgeting has been associated with the use of analytical techniques, which provided information to improve public delivery systems. Reformers also considered that formal analysis could inform the political nature of the budget process and that planning should be made more manifest in budgeting (Gianakis 1996, 129). While planning was increasingly merging with budgeting at the state and local levels (Rubin 1990, 182), indications were that performance measurement systems were less developed and inadequately integrated with organizational decision making processes (Grizzle, 1987; Poister and McGowan, 1984). effectiveness and performance generally lacked. Schick and Hatry (1982) indicated that the lack of information on programmatic implications of resource allocation alternatives affected the efficiency and effectiveness of budget choices. They further argued that the development of outcome or results-oriented program measures could effectively enhance the integration of planning and budgeting and thus provide a common ground for program comparison. They also indicated that input could not improve the situation whereas output measures, which facilitated comparisons of alternatives within programs, provided for neither the comparisons of programs required by the integration of planning and budgeting nor the use of formal analysis to improve the efficiency of the resource allocation process. But, Gianakis (1996) pointed out that outcome measures were difficult to develop for programs in the public sector. He further questioned whether outcome measures could clarify the means-end relationship between public programs Thus, the information on program

13

and societal problems, which was poorly understood. Likewise, Downs and Larkey (1985) contend that this lack of substantive knowledge undermined the use of performance measurement systems in budgeting and planning and that such systems could not point the way to intelligent actions in the budget process. The budget reformers also sought to improve the management of the budget process by centralizing budget development in a single executive (Gianakis, 1996), which was deemed more prone to apply modern management tools and analytical techniques to the budget process. It was believed that such centralization would yield greater accountability, efficiencies, and possibly more rational outcomes. However, Gianakis (1996) indicated that goal consensus was difficult to achieve, considering the lack of a common bottom line as in the private sector, due to conflict over political values, which disregarded the substantive knowledge on the most effective means. It is well known that the executive and components of the legislative have struggled over decades for dominance of the budget process. The basic budgeting problem of allocation decisions was conceptualized by Key (1940, 1137) in his famous question On what basis shall it be decided to allocate x dollars to activity A instead of activity B? This question could potentially be addressed and outcome measures could be expected to play a role in the resource allocation decision-making process. Key suggested however that the answer could only be derived from political philosophy, since he considered that the best use of public funds constituted a matter of value preferences between ends without a common denominator (1940, 1140). But, Lewis (1952) described institutional and procedural systems, upon which marginal economic analysis could be applied to successive comparison of alternative budget requests, which required measures of program outcome to establish the marginal utility of alternative resource allocation between different programs. He asserted that the impacts of outcome measures could constitute the only bottom line that, such as in private sector, the programs have in common. Such single criterion could thus serve as the basis for decision-makers to allocate resources. It is also important to point out that attempt by legislative bodies to micro-manage service agencies and to drive public policy through control of agencies line-item allocations have compromised the efficiency and effectiveness in the delivery of services (Rubin, 1990, Osborne, 1993). In a theoretical sense, Cothran (1993, 448) argued that line-item constituted a method of centralized control and that budgeting for results involved some centralization.

14

On the other hand, he contends that recent reforms of budgeting appear grounded in the research on organization theory, and especially on an analysis of the virtues of decentralization. Thompson (1991) stressed the importance of devolution of decision making about means and accountability through responsibility in budgeting and accounting. Indeed, he pointed that: It has been demonstrated to the satisfaction of most students of management that the effectiveness of large and complex organizations improves when authority is delegated down into organization along with responsibility (p.53). But, study of developments in state budgeting conducted by Lee (1991) over several decades, while showing that states are now likely to request that agencies provide measures of productivity and effectiveness in their budget request, indicated that policy makers and central budget officers have increased their control of goals and limits. The exercise of greater control by central budget offices over spending details has been facilitated by the computerization of state accounting system. While it appears paradoxical that this new reform calls for centralization and decentralization at the same time, Perrow (1977) argued that organizations often centralize in order to decentralize. Usually, the top management will grant greater flexibility and discretion only when it is confident that those below will effectively accomplish its goals. Cothran (1993) contends that this new reform aims to stimulate motivation and seek to achieve organizational goals through decentralized incentives that allow program managers greater authority and exercise of their creativity in the use of detailed use of resources while holding them accountable for the results. Policy makers or top managers hope that, while they have more time to think about goals and monitor performance, this approach will lead to greater efficiency and effectiveness in the use of available funds. Cothran (1993) further asserts that decentralization is in sink with the near consensus in todays management theory, in that a decentralized organization can be efficient, accountable, and satisfying to workers. Thus, the likelihood for this new reform to have a lasting impact on budgeting is real even if it fails to fundamentally alter the budgetary process.

VII.

Prospects for Using Performance Measures in the Budget Process

15

According to the Congressional Budget Office (1993), experience and current practice can provide useful insights in understanding the prospects for performance measurement and performance budgeting in government. First, it is critical to recognize that efforts to measure performance must confront the issue of the appropriate combination of executive and legislative branch action. Then, considering that such endeavor is complex, the use of a deliberate approach may prove most fruitful. Finally, it is important to grasp how performance measures might influence the budget process, which requires understanding the limitations of this type of change in influencing decision-making policy. Appropriate Combination of Legislative and Executive Activity In order to increase the potential for success in linking performance measurement to either management or budgeting, the appropriate combination of legislative and executive activity must be applied. This requires understanding how each is limited in bringing about lasting changes. States, due in part to the nature of their government organization, have had more difficulty than local governments in interesting legislative bodies in performance measurement. Indeed, the legislative and executive branches at both the state and federal levels are both more separated and more fragmented than those of local governments. Second, legislating performance-based budgeting will not gain widespread acceptance without commitment from the executive branch and such effort might be doomed to fail. But, it important to recognize that legislation has other limitations. Hence, improving the use of performance measures for financial management, financial reporting, or budgeting implies commitment to changing behavior in the two branches of government. the big picture, the results obtained from public programs. A Deliberate Approach May Be Most Desirable Performance measurement might face obstacles in bringing changes, which are beyond the commitment of the two branches of government; however, these challenges derive simply from the difficulty of measuring government performance itself. Designing systems that appropriately link the goals of programs with their results, and that link results to budgeting and financial reporting, is an extraordinarily complex task for all levels of government. Such behavioral change would involve a switch from focusing on micro-level inputs to emphasizing

16

Difficulties are compounded at the federal level, where the success of so many programs is influenced by other actors, including state and local governments, private businesses, and individuals. While it is hard to disagree with the goal of improving the measurement of government performance, acting precipitously by implementing far-reaching reforms without a fairly complete understanding of their effects can be counterproductive. It is also important to indicate that performance measures should not be viewed as ends in themselves and that there are concerns regarding the relationship between performance measurement and program evaluation. In most cases, performance measures can only offer clues about how well a program is achieving its results. In many cases, a comprehensive evaluation of a program is necessary in order to determine whether it is operating successfully. Hence, it is necessary to understand the limitations of performance measures and interpret them accordingly (CBO, 1993). Limits on Using Performance Measurement for Budgeting It is difficult to disagree with the concept of budgeting based on performance. But the experience of other levels of government suggests limited success. There is little evidence that policy-makers use performance measures to help them make large changes in the allocation of resources, or that they receive detailed information on the relationship between resources and outcomes. Where performance measurement has taken hold in the budget process, agency managers tend to use measurements to manage their budgets. Thus, performance measures are used much more extensively in carrying out the budget than in preparing it. However, the fact that performance-based budgeting has not gained widespread acceptance is not reason enough to discount its potential. government budgeting, there would still be obstacles. Even if both legislative and executive branches were committed to it, and a genuine effort was undertaken to transform It is not clear how performance Finally, measures should be used to allocate resources. Furthermore, performance measures seldom make the task of choosing between different uses of public resources easier. budgeting based on performance flies in the face of existing budgeting practice. A system that affords less control over individual line items in order to hold agencies solely accountable for results would be a fundamental change from the current system. Such a system could not

17

resolve the issue of how much money goes to one agency budget and how much goes to other spending solely by using performance measures. It should not be forgotten that budgeting, in a democratic society, is inherently political. Nevertheless, the limited potential of performance measures to influence budget outcomes directly, however, does not mean that they have no place in the budget process. There are at least two other ways that performance measures can be tied to the management of resources. Performance measures can be used to assist agencies in the management of a relatively fixed level of resources. Performance measures can also be used to present information on the relationship between inputs and outcomes. That would define for policymakers the relationship between given levels of inputs and the results expected from them. For example, the Internal Revenue Service might present information on how much faster it could process income tax refunds for each of several different levels of funding. But it is one thing to present such information to decision-makers; it is quite another for them to use it. If performance measures are to be used to influence the allocation of resources, the change is not likely to happen suddenly. Rather, it may be the result of a change in culture that starts with the development of better, valid performance information at the agency level and with the reporting of that information for non-budgetary purposes. As more attempts are made to tie performance measures to the budget, using the right measurements and collecting accurate information become more important. But the higher the stakes are, the greater the incentives are for people to identify self-serving measures and report misleading data. Without a process of ensuring that the right measures are chosen and reported accurately, performance measurement will never deliver on its promise. The limited potential, in the short run, for performance measures to influence the allocation of resources should not discourage governments from continuing to concentrate more resolutely on the results of public programs. In fact, the greatest reward to be gained from the use of performance measures may have less to do with government wide budgeting than with the task of using existing resources to improve performance. But encouraging managers and employees to think in terms of outcomes rather than inputs or outputs will produce desirable results. Because the measurements will not be right at first, one must be realistic about how much they can be used to influence budgeting in the near term. In short,

18

improving agency performance measurement should precede using those measures for budgeting (CBO, 1993). VIII. Conclusion If mandating performance measures is expected by itself to reform [federal] budgeting and management, such endeavor may prove disappointing. Executive branch commitment is probably a more important factor than legislation, although legislation aimed at reducing impediments to managers might be helpful as well. Further, effort to identify agency missions and goals should precede the effort to arrive at measures of performance, and the development of performance measures should precede requiring their use for budgeting. In the final analysis, the budget process is unlikely to be changed substantially until and unless decision-makers demand and use information on program performance when making decisions about allocating resources. Having this information, if good measures of program success can be developed, is a necessary; but, it is not sufficient or prerequisite to changing the policy process.

VIII. Selected Annotated Bibliography 1. Ammons, D. (1996). Municipal Benchmarks: Assessing Local Performance and Establishing Community Standards, Thousands Oaks, CA: Sage Publications.

19

This book is a collection of benchmarks, tools that are useful in gauging and improving the service delivery performance of local governments. The author indicates that the collection of indicators was gleaned from the documents of more than 100 local governments of various sizes scattered across the nation. With these benchmarks, municipal officials can make comparisons that will provide them an internal gauge as well as an external gauge, in having the performance of their organization stack against other jurisdictions. Ammons excluded workload measures because of their limited usefulness in crossjurisdictional comparisons and their inability to answer the how well and how efficiently questions. He also excluded unit costs because of their extreme vulnerability to inflation, economic differentials, and accounting vagaries. He cautions that benchmarks are most effective in the hands of peoples who know how to use them and understand their limitations.
2.

Berman, E. (1997). Dealing with cynical citizens, Public Administration Review 57 (2): 105-112.

3. Blanche-Kapplar, S. and L. Lissman (1996). Development and Use of Outcome Information in Government: Oregon Adult and Family Service Division. Washington, D.C.: American Society for Public Administration. The Oregon Adult and Family service Division, serving as part of pilot test under the Oregon Option for new forms of intergovernmental cooperation with the federal government based on managing for results, developed and used performance measures as a core tool in the redesign of its culture and management. Rather than focusing on process or input measures, it placed the emphasis on outcomes, which would provide quantifiable measures that the legislature, staff, and the public could use to gauge its progress. It moved from a long held mission of determining eligibility to a mission of promoting client selfsufficiency. It tracked performance measures at various organizational levels and sent performance reports to its partners to stimulate a team approach to achieve the desired results. Through this process, it learned about factors affecting outcomes, its service cycles, and outcome-based systems.
4.

Broom, C. A. and L. A. McGuire. (1995). Performance-Based Government Models: Building a Track Record, Public Budgeting and Finance 15 (4): 3-17.

These authors review performance-based government efforts undertaken by the states of Texas, Oregon, Minnesota, Virginia, and Florida in terms of the stated need, the approach, and the results to date. They discuss the elements of success and the lessons learned from these performance-based government pioneers. Such study was undertaken to answer broad questions of whether the performance-based programs implemented by these states could be sustained over long period.
5.

Campbell, M. D. (1997). Outcome and Performance Measurement Systems: An Overview, a WWW document available from the Alliance for Redesigning Government at URL: http://www.alliance.napawash.org/ALLIANCE/Picases.nsf/.

20

Michael Campbell discusses and defines the concept of performance accountability. He describes a performance accountability system that comprises three components, which are defining outcomes, measuring performance, and reporting results. In his overview, the performance accountability system is called outcome and performance measurement system.
6.

Cothran, D. A. (1993). Entrepreneurial Budgeting: An Emerging Reform?, Public Administration Review 53 (Sept.-Oct): 445-454.

Dan Cothran examines three new approaches to budgeting that use decentralization in order to stimulate entrepreneurial behavior by public administrators. These approaches include a method of expenditure control budgeting used by some municipalities in the United States, the second method of budgeting for results used by several industrialized democratic governments, and the third method of mission budgeting proposed for use to improve US defense budgeting at the Pentagon. The author claims that, after top policy makers have established their overall control through general guidelines, these approaches decentralize actual management decisions the same way as the market by providing agency managers with incentives to be more efficient and effective. Furthermore, they aim to make administrators more accountable for the results of their decisions by monitoring actual performance more than in the past.
7.

Gianakis, G. (1996). Integrating Performance Measurement and Budgeting. In A. Halachmi and G. Bouckaert (eds.), Organizational Performance and Measurement in Public Sector. Toward Service, Effort and Accomplishment Reporting. Westport: Quorum Books.

Gerasimos Gianakis examines the possible benefits, the potential problems, and the salient issues associated with the use of performance data in the typical municipal budget process. He first outlines the history of budget reform to provide perspectives on the program performance measurement issue. Second, he describes how performance data might be used in conjunction with a variety of budget formats and examines the ramifications of the use of performance measures in each format. Finally, he summarizes the prospects for integrating performance measurement and budgeting. He asserts that the successful implementation of the GASB measurement project will not directly affect the resource allocation processes of municipal governments. But, he contends that a program measurement system tied to the resource allocation process can help enhance communication in the policy-making process. He claims that citizen surveys can enhance communication and participation and, hence, the political rationality of the resource allocation process. Finally, he indicates that, in order to realize the allocative efficiencies promised by the integration of performance measurement and budgeting, the resource allocation decision-making process should use outcome or resultsoriented measures of program effectiveness. 8. Governmental Accounting Standards Board of the United States. (1994). Concept Statement No.2: Service Efforts and Accomplishments Reporting. Norwalk, CT: Governmental Accounting Standard Board.

21

The Board elaborates its views about the importance of reporting performance measures in annual financial reports. It states that service efforts and accomplishments information, both financial and non-financial performance measures, is an essential aspect of the measurement of governmental performance and is necessary for assessing accountability and in making informed decisions in the context of the public administration system and budgetary cycle. It holds that an ideal governmental entity should establish and communicate clear, relevant goals and objectives, set measurable targets for accomplishment, and develop and report indicators that measure its progress in achieving those goals and objectives.
9.

Grifel, S. S. (1993). Performance Measurement and Budgeting Decision Making, Public Productivity and Management Review 16 (4): 403-407. Grizzle, G. (1987). Linking Performance to Funding Decisions: What is the Budgeters Role? Public Productivity Review 41 (Spring): 33-44. Gross, B. M. (1969). The New Systems Budgeting, Public Administration Review 29: 113-137.

10.

11.

12. Hatry, H. and J. Fountain. (1990). Service Efforts and Accomplishments Reporting Its Time Has Come: An Overview. Norwalk, CT: Governmental Accounting Standards Board In these series of research reports, the authors analyze the state of the art in service efforts and accomplishments measurement and reporting for twelve service areas of state and local government. These reports include a set of performance measures for each service areas, which are intended to serve as a starting point for those governments wishing to establish their own performance measures.
13.

Hendrick, R. and J. P. Forrester. (1999). Budget Implementation. In R. T. Meyers (ed.), Handbook of Government Budgeting. San Francisco: Jossey-Bass.

The authors review concepts and procedures of budget implementation associated with input budgeting and outcome or results-oriented budgeting systems. They indicate that decisions focus on process, procedures, and inputs in traditional line-item budgeting and that the budget is implemented to uphold conventional views of accountability through compliance and control. They assert that, on the other hand, decisions focus on programs, performance, and outcomes in outcome budgeting. In this system of budgeting, they further note that the bounds of accountability are broadened to closely parallel the notion of accountability used by nonprofit and private sector organizations. They thus claim that accountability is broadened to encompass effectiveness, efficiency, and quality in addition to the outcome monitoring and reporting of input levels.
14.

Key, V. O., Jr. (1940). The Lack of Budgetary Theory, American Political Science Review, 34 (December): 1137-1140.

22

15.

Kravchuck, R. and R. Schack. (1996). Designing Effective Performance Measurement Systems Under the Government Performance and Results Act of 1993, Public Administration Review 56: 348-358.

These authors assess the challenges posed for public sector decision makers by the Government Performance and Results Act of 1993 (GPRA). They contend that increasing reliance on formal measurement approaches, at the expense of "hands-on" management and evaluation, will present decision-making problems for senior public officials. Paradoxically, these authors also assert that performance measures may misinform as much as they inform, if users are unaware of the subtle limitations of measurement systems. The authors argue that the rigors of general systems theory can contribute to effective compliance with GPRA. Hence, they present ten general design principles for effective performance measurement systems.
16.

Lee, Jr., R. D. (1991). Developments in State Budgeting Trends in Two Decades, Public Administration Review 51 (3): 254-262.

Robert Lee reviews survey data from state budget offices and finds that considerable change took place between 1970 and 1990, in regard to state budgetary processes. He asserts that functions that were relatively uncommon in 1970 were now common in 1990. These functions include: 1). dollar-level ceilings and policy ceilings issued by the executive to agencies for preparing budget requests; 2). computer usage in agency budget preparation, analysis of agency requests by the budget office, and preparation of the governor's budget; and 3). use of analysis in decision making by the executive and legislative branches. Undoubtedly, the typical budget system of 1990 was noticeably different from that of 1970. However, he notices that a few characteristics remained relatively unchanged. But, a case can be made that some trends are largely irreversible and will continue. Such is the trend of hiring budget office personnel with master's degree is expected to continue. Nevertheless, he concludes that changes among states in the future will likely be more limited and less dramatic.
17.

Lewis, V. (1952). Toward a Theory of Budgeting, Public Administration Review 12 (1): 43-54. Lindblom, C. E. (1979). Still Muddling, Not Yet Through, Public Administration Review 39: 517-526. Lynch, D. T. and S. E. Day. (1996). Public Sector Performance Measurement, Public Administration Review 19 (4): 404-419.

18.

19.

Thomas Lynch and Susan Day review the evolution of the governmental performance measure concept and comment on its contemporary development in terms of history, state of the art, theoretical considerations of performance measurement. They also suggest the implications of the current developments in terms of contemporary and future public administration.

23

20.

Melkers, J. and K. Willoughby. (1990). The State of the States: Performance-Based Budgeting Requirements in 47 out of 50, Public Administration Review 58 (1): 66-73.

The authors examine the trend toward improving performance in government by analyzing the foundations for conducting performance-based budgeting in the United States. The survey of 50 states regarding existing or planned legislation related to PBB as well as administrative requirements allowed them to review legislation and budget guidelines to determine their scope and focus. Their findings show that all but 3 states have performancebased budgeting requirements, and most have established these requirements within the last few years. Thirty-one states have legislated performance-based budgeting to be conducted, while 16 states have initiated this reform through budget guidelines or instructions. Their research serves as a stepping-stone to determining effective methods to create and sustain this budget system in the states.
21.

Martin, L. L. (1997). Outcome Budgeting: A New Entrepreneurial Approach To Budgeting, Journal of Public Budgeting, Accounting and Financial Management 9(1): 108-126.

Lawrence Martin attempts to shed some initial lights on the topic of outcome budgeting, which is receiving increased attention in government today. He tries to answer the questions of what outcome budgeting is and whether such concept is really something new or an old one revisited. He provides in his discussion a conceptual framework for thinking about outcome budgeting along with two examples of outcome budgeting drawn from the states of Florida and Oregon. 22. Oregon Progress Board. (undated). Oregon Benchmarks: Standards for Measuring Statewide Progress and Institutional Performance. Salem, OR: Author. Oregon is considered as one of the states in the nation to be at the forefront of current performance measure efforts and has developed benchmarks, useful tool for gauging and improving performance in state government. 23. Osborne, D. and T. Gaebler. (1992). Reinventing Government: How the Entrepreneurial Spirit Is Transforming the Public Sector. Reading, MA: Addison-Wesley. This book identifies ten principles of entrepreneurial characteristics necessary to achieve government reinvention and indicates that flexibility, creativity, and responsiveness of public organizations are catalysts for government change into a decentralized entrepreneurial model. Gaebler and Osborne further assert that the presence of certain characteristics such as mission-driven; results-oriented; enterprising; customer-driven; community-owned; decentralized and market oriented are critical for an entrepreneurial government to exist within the reinvention framework. While these principles of government have been identified with government reinvention, these authors assert that crises and certain conditions can quicken changes in government' s approach to governing. These conditions include fiscal stress, squeeze on administrative expenditure, decentralization of previously centralized controls, threat of privatization of certain government functions. Gaebler and Osborne also

24

observe that factors such as strong leadership, continuity of leadership, a shared vision, trust, and outside resources are potent motivators toward important change in government.
24.

Pilegge, J. C. (1990). Budget Reforms. In J. Rabin (ed.), Handbook of Public Budgeting. New York, NY: Marcel Decker, Inc.

The author examines the aspirations and complications as well as the successes and failures of budgetary reform. He assesses the legacy of such phenomenon by reviewing the literature on budgeting and financial management.
25.

Poister, T. H., and R. P. McGowan. (1984). The Use of Management Tools in Municipal Government: A National Survey, Public Administration Review 44 (3): 215-223. Rubin, I. (1990). Budget Theory and Budget Practice: How Good the Fit? Public Administration Review 50 (2): 179-188.

26.

Rubin contends that there exist fewer disparities between normative and descriptive budget theory than are apparent. She asserts that normative theory prescribes based on values and suggests a planned approach to budgeting. She indicates that budget reform such as program budgeting, zero-based budgeting, management by objectives as well as performance budgeting constitute attempts at normative budgeting. On the other hand, descriptive budget theory, which describes and infers based on observations, sees budget reform as a function of incrementalism precluding the policy-making role at the state and local levels. When greater clarity is brought to bear in the budget format and process, policies and plans will improve tremendously.
27.

Rubin, I. (1996). Budgeting for Accountability: Municipal Budgeting for the 1990s, Public Budgeting and Finance 16 (Summer): 112-132.

Rubin describes changes that are taking place and suggest additional changes to make local budgeting more accountable. The author indicates that in the 1990s, the increased focus on accountability, due to widespread public belief that government was uncontrollable and responsible only to special interests, has lead local budgeters to embark on efforts to improve the budget accountability to the public.
28.

Schick, A. (1990). Budgeting for Results: Recent Developments in Five Industrialized Countries, Public Administration Review 50 (1): 26-34.

Schick describes findings concerning the experience of five industrialized countries with results-oriented management reforms. He assessed the efforts that Australia, Canada, Denmark, Sweden, and the UK have undertaken to budget for results. He attempts to grasp why the old problem of the use of performance measures in allocating resources has received renewed attention in these democratic countries. He explains that the interest in productivity is clearly related to fiscal stress and to the squeeze imposed on administrative expenditure.

25

Furthermore, delegation of previously centralized controls has stirred a rethinking of the role of the central budget organization. He asserts that the primary contemporary emphasis is not so much on generating new measures as on fostering a managerial environment that is attentive to performance when funds are parceled out. He concludes that in practice, none of the governments has forged a tight relationship between resources and results.
29.

Schick, A. (1992). The Road to PPB: The Stages of Budget Reform. In A. C. Hyde (ed.), Government Budgeting: Theory, Process, Politics. Belmont, CA: Wadsworth, Inc., pp. 46-60.

Schick citing PPB as an improvement on traditional budgeting, highlights the fact that in spite of budgetary reform, political issues are what drive the budget, not management related issues. He points to three processes that serve the budget, which are: strategic planning that links organizational objectives to program priorities and results; management control whereby management ensures that given resources are used to obtain objectives identified during the planning process; and operational control, which entails realizing objectives with given resources. He asserts that, although PPB incorporates a planning function and moves budgeting along budgetary reform continuum, legislators and managers cannot dictate the budget. They have no power of the purse, which only the executive branch controls.
30.

Schick, A. and H. Hatry. (1982). Zero Base Budgeting: The Manager Budget, Public Budgeting and Finance 2 (1): 72-87.

31. Southern Growth Policies Board. (1996). Results-Oriented Government. A Guide to Strategic Planning and Performance Measurement in the Public Sector. Research Triangle Park, NC: Southern Growth Policies Board. The Board recognized that the lack of widespread familiarity with techniques that have proven effective in both public and private sector applications constitutes a major stumbling block to implementing effective planning and performance measurement. It thus provides a guide for a unique and flexible training program in goal-setting and performance measurement. The training program is designed to provide a comprehensive curriculum in strategic planning, benchmarking best practices, performance measurement, using performance results for project management, performance-based budgeting and contracting, and in creating environment that supports results-oriented management.
32.

Thompson, F. (1991). Management Control and the Pentagon: The Organizational Strategy-Structure Mismatch, Public Administration Review 51 (Jan.- Feb): 52-66.

In the debate over whether the Pentagon is capable of managing the cutbacks and reductions it faces in the post-Cold War era, it has been argued that an historical mismatch between strategy and structure at the Department of Defense (DOD) will make the process difficult. Thompson indicates that the solution to this problem can be found in the recommendations made by Mosher in 1954 and by Anthony in the 1960s. Indeed, Mosher and Anthony recommended a broad organizational template, namely a matrix structure with

26

mission and support centers linked by transfer prices, responsibility budgeting, and accrual accounting. Hence, he contends that the reform of the DOD depends on aligning strategy with structure, decentralizing, and establishing an account structure that is grounded in the principles of responsibility budgeting and accounting. He argues that the recommended broad organizational template should serve well the national interest under almost any set of circumstances. 33. U.S. Congressional Budget Office (CBO). (1993). Using Performance Measures in the Federal Budget Process. Washington, D.C.: Government Printing Office. The Office analyzes the use of performance measures in the budget process. Its study attempts to review the issues raised by performance budgeting in the context of past and current efforts to link performance measures and budgeting. The study focuses on two points, which are the status of the current federal performance measurement efforts and specific observations designed to inject a note of caution into the ongoing debate about performance measurement and budgeting.
34.

Wildavsky, A. (1966). The Politics of Economic Efficiency: Cost Benefit Analysis, Systems Analysis, and Program Budgeting, Public Administration Review 26: 292310. Wildavsky, A. (1969). Rescuing Policy Analysis From PPBS, Public Administration Review 29: 189-202. Wildavsky, A. (1979). Little, Brown. The Politics of the Budgetary Processes. (3rd ed.) Boston, MA:

35.

36.