Un Pan 005913

Performance Measurement Utility in Public Budgeting: Application in State and Local Governments

by

Dr. Katherine Willoughby Andrew Young School of Policy Studies

Georgia State University University Plaza

Atlanta, GA 30303 Telephone: 404-651-4599

Email: [email protected]

Paper presented at the 14th Annual Conference of the Association for Budgeting and Financial Management a section of the American Society for Public Administration, October 10-12, 2002 in Kansas City, Missouri.

DRAFT: Cite with Permission Only

Performance Measurement Utility in Public Budgeting: Application in State and Local Governments

Introduction Budgetary reform is little different from human reform – change, when entrenched, fosters a feedback of information to support budget learning (Forrester and Adams 1997). Such knowledge presumably factors into the decisions about spending by responsible budgeters and administrators. Practitioners and academics who are “guardedly optimistic” about performance measurement applicability to public budgeting practices point out that as a reform, performance measurement has in its favor that it is not new. In fact, “building on existing budget routines, performance measures are slowly being developed, but more important, the routines for developing the measures are being routinized” (Chackerian and Mavima 2001).

In fact, over the last decade governments in the United States have been folding performance measurement and/or results oriented systems into traditional practices of management and budgeting – some more comprehensively than others. If “time allows synergistic learning and a reduction of ambiguity” (Grizzle and Pettijohn 2002) are these systems viewed as useful by those required to generate and apply measurement within the budget cycle and for budgeting purposes? This research seeks to answer this question by examining the perceptions of state and local officials and staff about performance measurement applicability to budget cycle and process in their governments. This research rests on the notion that budget structures influence budget deliberations and discussions and therefore impacts the final budget decision, an appropriation or appropriations (Ingraham and Moynihan 2001; Pettijohn and Grizzle 1997). If performance measurement can be found to be useful to government budgeters and administrators, then it can be considered successful in adding value to budgetary decision-making. This research examines the perceptions of performance measurement by those responsible regarding its utility to budgeting. And, if performance measurement is a requirement of budget documentation, where in the budget cycle is this type of information perceived as most helpful.

A quick review of the factors related to successful implementation of reform in government is provided. This is followed by a discussion of our present understanding of how useful performance measurement can be to budgetary decisions in particular at the state and local level. Then, study method and findings are presented. The conclusion summarizes results and offers suggestions for future analysis of how public budgeting decisions can be better supported.

Implementing Change The factors associated with successful implementation of any reform are dynamic. Streib and Willoughby (forthcoming) provide illustration of the characteristics that must exist for innovation in government.

For any initiative to be successfully implemented in an organization there must exist an environment for change. Certainly throughout the last decade, sub national governments in the United States were pressured externally to change management practices and service efforts, and to advance program and service accountability in terms of costs. Pressure has arisen from the public for more responsive service, as well as from professional organizations such as the GASB regarding SEA measurement, reporting innovations, templates, as well as concept statements. Currently, government officials shoulder an even heavier burden to advance and advertise the performance of their agencies and departments to weather a continually weakened economy.

Sustained leadership must support a culture of change and eventually the maintenance of an innovation or reform. A common lament of today’s public leaders is that they must promise to do more with less in order to remain in office. For example, consideration of the tax initiatives to combat budget deficits mentioned in 2002 governors’ addresses provides some indication that these chief executives are just as likely to work with what they have to carry out state business, rather than depend on the creation of new or generation of additional revenues (Willoughby 2002). Chief executives understand that new or increasing revenues are in a distant future, and in the

Knowledgeable Leadership

Slac

k R

esou

rces

Skilled Staff

SuccessfulReform

Goo

d C

omm

unic

atio

n

Stable Environment

External Pressure for

Change

Figure 1: Factors influencing government reform* *Figure adapted from: Streib, Gregory and Katherine Willoughby (forthcoming). “Local Governments as E-Governments: Meeting the Implementation Challenge.” Public Administration Quarterly.

short term muddling through necessitates focused attention to the processes for getting the work done. Related to results oriented systems, this requires their commitment to recognize the performance measurement work of agency heads and department directors in addition to simply requiring measurement as a part of budget and management reporting.

Good communication throughout an organization must confirm new behaviors and the acceptance of fresh information or system requirements. Flexibility in implementation of adopted practices can help sustain employee action toward eventual institutionalization of new practices. Of course, resources must exist to support the introduction of change. Resources may be in the form of monetary support for consultants and/or training, or for new or enhanced technologies, or additional time provided for service delivery, or greater flexibility with personnel for getting jobs accomplished. Resources may not be plentiful and in fact declining, but they can be redirected toward new practices (Berman and Wang 2000; Jordan and Hackbart 1999).

Performance Measurement Applicability to Resource Allocation Decisions

While performance development and reporting requirements in some form or fashion are identifiable as having been adopted by most state and local governments in the United States, assessment of the institutionalization of these requirements is a complex task (Melkers and Willoughby 1998; Poister and Streib 1999). Literature that focuses specifically on performance measurement implementation concurs that the above characteristics are often missed or inconsistently applied and may compromise successful application of performance measurement to management and budgeting practices (Grizzle and Pettijohn 2002, Julnes and Holzer 2001; Lee and Burns 2000; Smith 1999).

On the other hand, much research exists that has teased out some subtle and not so subtle implications of instituting performance measurement into budget systems. Franklin’s (2002) analysis of appropriation format change from 1990 to 2000 in two states finds that agencies in the reform state (that advances performance measurement), Arizona, are more likely than those in the non-reform state, Oklahoma, to have less restrictive appropriations at the end of the decade. Other variables that she finds can effect appropriation format include the restrictiveness of the format in 1990, the percent change in general fund reliance of an agency, whether the agency head is elected, and turnover of agency heads over time. As noted related to government implementation of an innovation, active and sustained leadership, good communication across the organization, and flexible resources contribute to success.

Jordan and Hackbart (1999, 84) find that “almost two-thirds of the responding states indicate that the governor’s executive budget recommendations could be impacted by [agency] achievement of performance standards” – achievement that comes to light with communication about performance measurement. These authors also find that “more than half of the respondents from survey of state executive branch budget officials answer that performance indicators are important tools for allocation decisions or budget recommendations” (1999, 78). However, they conclude that while states are reticent about basing allocation decisions strictly on performance, measurement supports making adjustments to funding to improve or maintain performance (1999, 85).

Chackerian and Mavima (2001) analyze administrative and budget reform during

Lawton Chiles administration in the State of Florida and conclude that in spite of some difficulties, “key stakeholders continue to like the idea of budget reform and performance measures. Policy conflict and ambiguity is moderate. Budget reform is seen as doable and as possibly resulting in long-term cost savings.” They conclude that:

Since implementation costs for the budget reforms are stretched over a long time, their apparent size is reduced. Agencies also reduce costs by learning from the experience of the early adopters of the phased-in budget reforms. In addition there is, of course, a great deal of experience with the existing budgeting process, which serves as a jumping-off point for the changes. Thus, while performance-based budgeting represents a departure from it, it is built upon past practice. ….The generalized legitimacy of the approach, the incremental implementation, and the existing statutory framework for budgeting and staffing have greatly reduced the implementation costs, and thereby makes full implementation more likely (2001, 353-377).

Bradley and Flowers (2001, 401) temper this enthusiasm about Florida’s application of performance measurement and reporting by noting, “extended implementation has diffused the intellectual coherence of the initiative, diluting leadership commitment. Legislative aspirations to make performance central to allocation decisions have been largely deflected in the face of persistent technical problems.” Their conclusions again point to the importance of sustained leadership (and in both branches of government) for effective reform. Research about local applications of performance measurement to budgeting is cautiously optimistic as well. Smith with Bratton (2001) considers the implementation of computerized statistics, Compstat to re-engineer police management and impact citizen safety in New York City. They are hopeful that the system will continue to support a changed culture in the police force, opening lines of communication and accountability among officers, and reorienting precincts as hubs for problem solving and performance management. Scheps (2000) illustrates a good case for sustained pressure and constant communication as successfully linking performance measures to resource allocation in Dallas County, Texas. And Steifel, Rubenstein and Schwartz (1999) generate findings that support the ongoing flexibility and enhancement of performance measures when applying to spending decisions. They find that adjusted performance measures can be used effectively to explain past resource allocation decisions. Their carefully chosen and stepped model successfully distinguishes high and low performing schools in Chicago’s Public School System, specifically illustrative of different patterns of discretionary spending in these two categories.

Kluvers (2001) finds that program budgeting including the use of performance indicators does not foster reallocations, but does provide useful information that allows for the sorting of direct and allocated costs and fosters a changed attitude toward planning. The institution of planning programming budgeting did not assist in eliminating duplicative activities or change objectives based on performance indicators. He cannot claim that better budgeting decisions are being made, but suggests that the

subtle effects of measurement application could influence spending over time. His results would square with those of Ammons (1995) in that performance measurement has had more utility to managers internal to governments rather than to citizens external to them.

Poister and Streib (1999) review the inconsistency of use of performance measurement among local governments, although they do find that larger local governments and those with a council-manager form of government are more likely to make use of measurement. Not surprisingly and particularly at the local level, workload and output measures are a measurement mainstay. And local officials are more often pressed to use such measures to track costs and improve accountability to the local government manager or chief administrator. These authors find that in governments with centralized systems, performance measurement is used in a meaningful way for planning and is considered very important for budgeting purposes. Most officials (80 percent) with centralized systems of performance measurement indicated that measurement improves decisions in their jurisdictions, at least to a moderate degree. Of this group, “more than 60 percent report moderate or substantial changes in budget allocations as an impact of their performance measures” (1999, 331).

Berman and Wang (2000) concur that existence and meaningful use of performance measurement are separate issues and that overstatement of use often masks the inadequacies of the organization in supporting such reform. Their research provides good corroboration of the work by Jordan and Hackbart (1999), which finds that organizational capacity is vital to the usefulness of performance measurement for budgeting. Jordan and Hackbart (1999, 85) suggest that governments can best support their performance initiative by empowering the central budget office given its influence on budget deliberations later on in the budget cycle. As the nexus between policy development and its administration, these offices provide an appropriate sieve through which information flows about what agencies do and what they need. Depending upon orientation (capacity as a control versus policy office) these offices house examiners who understand the usefulness of performance measurement to spending decisions, and are in a strong position regarding when and where to use such information for “success” in terms of gaining support for spending by agencies from the budget director, the chief executive, and eventually the legislature (Thurmaier and Willoughby 2001).

The body of research above illustrates that the application of performance measurement to decisions by government officials is widespread although highly variable among sub national governments in the United States. Those responding to past studies of performance measurement use indicate that such data is most helpful for managerial and communication purposes. The influences of performance measurement to budgetary decision making and traditional practices however, are faint in terms of directly linking this information to budget allocation changes.

Research Focus and Methodology This research addresses the following questions to further our understanding of performance measurement adoption and use in state and local governments:

1. Do state and local government officials perceive performance measurement as

effective in general? 2. For what types of decisions and processes is performance measurement perceived

as most effective? 3. What is the utility of performance measurement for budgeting purposes

specifically? 4. Are there distinctions across levels of government among the budgeters regarding

their perceptions of performance measurement utility for budgetary purposes?

Survey protocol, variable measurement and response rate This study is part of more extensive research about performance measurement that has been funded in part by the Alfred P. Sloan Foundation. A mailed survey to state and local government budgeters and administrators in the summer of 2000 was a follow-up to approximately two dozen case analyses conducted by the Governmental Accounting Standards Board (GASB) in selected governments regarding the success of government performance measurement initiatives. This work stems from the GASB’s Service Efforts and Accomplishments project that supports government experimentation with performance measurement and reporting. In the present study, similar, though tailored versions of the instrument were mailed to the executive and legislative branches of the different levels of government as well as to departments and agencies in these governments.

Those contacted for this survey are in the business of performance measurement application in their governments and includes budget and fiscal officers and staff from executive and legislative budget offices of state government, state agency directors, division directors, program managers and staff, and local government managers, department and division directors, program managers and budget, accounting and audit staff. Questionnaires were sent to 121 officers in executive and legislative budget offices in all 50 states. Also, 434 questionnaires were sent to the heads of selected agencies in states, including those in Corrections, Education, Economic Development/Welfare, and Transportation. Finally, 735 questionnaires were sent to those in similar positions in a sample of local governments across the nation.

A total of 1,311 questionnaires were mailed and 491 received, with 489 useable instruments. The overall response rate for the survey was 37 percent and includes 60 state government budget officers or staff, 152 state agency personnel, and 277 local government officials, half who are identified as budget officers or staff and half as agency administrators or staff. Representation by government includes responses from 37 states, 47 counties and 168 cities. The inclusion of those responsible for and knowledgeable about performance measurement across government branches and by level helps avert threats to the validity of this research. And as is typical of this type of research, most questions were accommodated to Likert-scale items in which responses

fall along a continuum regarding level of agreement, evaluation of effectiveness, or frequency of use. The scales used range in value from 1 (the lowest) to 4 or 5 (the highest). Research Findings Measurement utility in government Survey results indicate significant difference in the comprehensiveness of application of performance measurement by level of government and type of measure. According to Table 1, higher proportions of local respondents compared with state respondents claim performance measurement use (of all types) throughout all agencies and departments.1

===Table 1 about here=== Considering types of measures used helps to explain the seemingly broader application at the local level, however. That is, while strong majorities of officials from both levels of government admit use in over 50 percent of agencies of input, process and output types of measures, perception of application across all agencies drops rather dramatically as measurement becomes more difficult or complex – beginning with outcome measures, and continuing a decline in perceived comprehensive use of quality, explanatory and benchmark measures. It seems that performance measurement application is still quite selective; the more easily measurable activities are tracked most often by using rather elementary data. Also, performance measurement seems to realize more comprehensive use at the local level of government. This coincides with the fact that many local government activities are quite amenable to measurement. For example, it is easy to see why local governments might develop quality measures by surveying customers about service delivery timeliness and quality, given the more direct relationship between citizens with local service delivery. Respondents perceptions of performance measurement effectiveness Table 2 shows percent of responses by level of government affirming performance measurement effectiveness for a number of processes and interactions within and external to governments.

===Table 2 about here=== Those surveyed were asked to respond if performance measures were very effective to not at all effective regarding the items listed. Generally speaking, state level respondents

1 As is common for nominal level data like this, contingency tables and the Chi Square statistics are calculated using level of government of respondent as the independent (column) variable and the extent of use of the measure (as scaled from all to no departments) as the dependent (row) variable. The χ2 values for each of nine tests of independence of extent of use of each measure by level of government is significant at probability values ranging from less than 0.05 to 0.001 (See Agresti and Agresti 1979 for a discussion of the use of the Chi Square test statistic).

are more likely than their local counterparts to find performance measurement effective, particularly for promoting a culture that focuses on results, for influencing management changes to achieve desired results, and for improving communication both horizontally and vertically throughout government. It is interesting that the χ2 values found to be significant (p<0.05) are associated with just two items, Improving communication between departments and programs (state officials are 11 percent more likely than local ones to find performance measurement effective or very effective for this); and Changing the substance or tone of discussion among legislative members in budgeting or oversight. State officials seem to recognize greater effectiveness of performance measurement in adding value to the budgeting deliberations of legislative branch member than local officials. Perhaps because many of the performance budgeting and measurement initiatives by states in the last decade were legislatively initiated (Melkers and Willoughby 1998), legislators may feel some pressure to take advantage of the information they are requiring of executive branch officials and staff. Or, given natural legislative turnover, new, younger and perhaps more technically savvy legislators just feel more comfortable incorporating this type of data into their decision making. In general, these results affirm that performance measurement can support advancements to management decisions and foci, as well as to communication within governments. At least a third or more of both the state and local level respondents consider that performance measurement is effective or very effective for the first nine items. These administrators and budgeters consider that performance measurement is most effectively used to focus the attention of government officials on program results and the factors contributing to those results. They agree that communication across departments, with the executive budget office and with legislative branch members are enhanced by using performance measures. Finally, program and service quality are advanced given the institution of performance measurement application in the governments represented by these respondents. The processes and interactions in which performance measurement is less likely to prove effective include changes to program costs or their appropriations, advancing coordination among agencies and programs, saving money, reducing duplicative services, or enhancing external relationships or communications. Relatively small proportions of both groups (a fifth or less of state and local respondents) in fact find performance measurement effective for reducing duplicative services, changing appropriations levels, eliminating ineffective services programs, affecting costs savings, or improving external support to government. It is interesting that a smaller proportion of local respondents than state ones find performance measurement effective for improving external government cooperation or coordination. Budgeters’ views of measurement utility to budget reporting and cycles The following teases out the perceptions of the budgeters only. Central budget office staff have a more global view of the application of performance measurement to budgeting practices than do agency administrators and staff (see Thurmaier and Willoughby 2001 for a full examination of the orientations of central budget offices and the budget analysts employed therein). Sections of the survey related to performance measurement applicability to budgetary processes and cycles were included for state

budgeters, but not the administrators, and included for all local budgeters and administrators. To enhance reliability, the perceptions of state budgeters are compared to those of the local budgeters only. The tables below take advantage of responses from 60 state budgeters and 140 local budgeters. Budgeters are defined as directors, deputy directors, or analysts working within a central budget office (either executive of legislative) or staff working in the capacity of a government wide auditor. Table 3 shows responses of budgeters by level of government acknowledging the appearance of output or outcome measures in budgeting documentation. The difference

===Table 3 about here=== between state and local budgeters responses regarding performance measurement appearance in department budget requests is significant (χ2=12.83, p<0.01) as is the difference in responses regarding appearance of measures in quarterly reports, albeit just barely (χ2=6.96, p<0.10). Generally, the results confirm that performance measurement gets more attention during budget development than in other phases of the budget cycle. The traditionally strong budget guidelines of governors to agencies as they prepare their requests undoubtedly calls for information about program performance and often comparison of measures with those of the current and past fiscal years. The fewer local budgeters indicating the existence of these measures in their department requests could attest to the more relaxed process that may exist at the city or county level, or be related to the fact that departments at this level still cling to input and activity based measures over more advanced measures of output or outcome. It is interesting in fact, that the output and outcome measures do not seem to have stronger application during budget execution, as noted by the fact that very strong majorities indicate a lack of appearance of these measures in quarterly reports in any departments. Finally, the appearance of these measures in annual reports does pick up a bit with over half of state budgeters indicating appearance in 50 percent to all departments, while just one third of the local budgeters indicate such appearance. The requirement for and appearance of performance measures in various budgetary documents is important as an information source and part of the compendium of materials that will support budgetary deliberations through their sheer publication. Yet, how important is this information to the process of making spending decisions. State and local budgeters were asked to rate performance measurement existence throughout departmental and government wide budget cycles on a scale from not important at all to very important. Table 4 presents the responses for those indicating that the existence of this information is important or very important to that segment of the budget cycle so noted. In general, from one quarter to a third of these budgeters consider

===Table 4 about here=== this information important or very important to departmental budgeting processes, with virtually similar proportions of budgeters by level considering such information important to department budget development, and somewhat fewer considering these measures important to budget appropriations, execution and assessment of results. There

are no significant differences between state and local budgeters regarding performance measurement importance to departmental budget processes. There are more pronounced distinctions by level of government of these budgeters’ perceptions of performance measurement importance to government wide budget processes, however. For every stage listed, a greater proportion of local budgeters find the information to be important or very important. It is interesting that according to Table 3, fewer local budgeters (55 percent) indicate the appearance of output or outcome measures in the budget requests of 50 percent or more of their departments when compared to state budgeters (75 percent). Yet, local budgeters are more likely than state ones to consider this information important to government wide budget development. The test of independence is significant only for the difference in proportions of budgeters’ perceptions of the importance of performance measurement to government wide final appropriation determinations (χ2=9.57, p<0.01). Local budgeters are significantly more likely than their state counterparts to consider output or outcome measures as important or very important to final spending decisions. Discussion and Conclusion Allen Schick ponders that “[t]he problem [with performance measurement] is not in how we measure performance but in how we use the results” (2001). While the influences of performance measurement to advance management practice are more easily documented, gauging measurement applicability to budget practice is a tougher assessment to make. There certainly is little information confirming Schick’s strict definition of performance budgeting that requires linkage of each increment in budget resources to an increment in outputs or outcomes. The results of this research actually provide some support to skeptics regarding the applicability of performance measurement to budgeting, however, there is also support for optimists – encouragement to those who hold out hope that such information will continue to seep into budgetary practice as a reliable resource for important budget actors. Results supporting skeptics include:

• Performance measurement is perceived as having its greatest effectiveness related to management decision-making and communication enhancement and not directly to resource allocation decisions.

• There remains a strong attachment to more easily measured aspects of

government operation, including input, activity and output measures. There is less use of outcome measures, and very slow advancement of more complex measures of quality, explanation or benchmarking. Local governments are however, stronger users of quality measures than state governments.

• Strong majorities of state and local budgeters indicate the appearance of

output or outcome measures in the early stages of the budgeting cycle, in agency budget requests, in the executive budget report and in annual operating budgets. Appearance of these measures drops rather dramatically for both levels of government in quarterly reports, picking up

slightly in annual reports where we would more likely expect to find summary measures of performance.

Results supporting optimists include:

• Strong majorities of state and local officials indicate performance measurement use in 50 percent to all of their departments.

• Strong majorities of state and local budgeters indicate the appearance of

output or outcome measures in the early stages of the budgeting cycle, in agency budget requests, in the executive budget report and in annual operating budgets. Appearance of these measures drops rather dramatically in both levels in quarterly reports, picking up slightly in annual reports in which we would more likely expect to find summary measures of performance.

• Performance measurement acceptance as a resource is seeping into the

allocation deliberations of legislative branch members. Executive branch members have been working for years to develop, redress, and report about the performance of government activities. It is important for executive branch members to understand if and how legislative branch members use such information to provide legitimacy to performance measurement system requirements.

Notice that the second point regarding appearance of performance measures in the budget cycle can support skeptics or optimists. Skeptics might find solace in the fact that performance measures are less apparent in latter stages of the typical budget cycle, somewhat restricted to a supportive role during the budgetary development stage. Optimists can consider this positively, in that the greatest gains for influencing appropriations can occur for agencies at the development stage – when they must make their case to executive budgeters for their spending needs.

Finally, results show that there is little distinction by level of government of budgeters’ perceptions of the importance of output or outcome measures to departmental budget cycle. However, larger proportions of local budgeters view this information as important to government wide budget processes, and these budgeters are significantly more likely than state budgeters to view performance measurement as important or very important to final appropriation determinations. Such results might speak to the difficulty of gaining a comprehensive view of state government performance, when compared to local government performance, on a government-by-government basis. In any event, it remains important to find out what types of performance measures are most useful and where in a public budgeting cycle and/or within which decision processes it holds its greatest likelihood for effectiveness. This research has provided some enlightenment about the usefulness of this information. And, there should be little threat that so called rational decision aids will supplant the politics of public budgeting. Rather, in the words of David M. Walker, U.S. GAO Comptroller General (2002, 2),

Linking performance information to budgeting carries great potential to improve the budget debate by changing the kinds of questions and information available to decision makers. However, performance information will not provide mechanistic answers for budget decisions; nor can performance data eliminate the need for considered judgment and political choice.

References Agresti, Alan, and Barbara Finlay Agresti. 1979. Statistical Methods for the Social

Sciences. San Francisco, CA: Dellen Publishing Company. Ammons, David N. 1995. “Overcoming the Inadequacies of Performance Measurement

in Local Government: The Case of Libraries and Leisure Services.” Public Administration Review (January) 55, 1: 37-47.

Berman, Evan, and XiaoHu Wang. 2000. “Performance Measurement in U.S. Counties:

Capacity for Reform.” Public Administration Review (September/October) 60, 5: 409-420.

Bradley, Robert B., and Geraldo Flowers. 2001. “Getting to Results in Florida,” Chapter

14 in Quicker, Better Cheaper: Managing Performance in American Government, edited by Dall W. Forsythe (NY: The Rockefeller Institute Press): 365-416.

Caiden, Naomi. 1998. “Public Service Professionalism for Performance Measurement

and Evaluation.” Public Budgeting and Finance (Summer) 18, 2: 35-52. Chackerian, Richard, and Paul Mavima. 2001. “Comprehensive Administration Reform

Implementation: Moving Beyond Single Issue Implementation Research.” Journal of Public Administration Research and Theory (July) 11, 3: 353-377.

Forrester, John P., and Guy B. Adams. 1997. “Budgetary Reform Through

Organizational Learning: Toward an Organizational Theory of Budgeting.” Administration and Society (February) 28, 4: 466-488.

Franklin, Aimee. 2002. “An Examination of the Impact of Budget Reform on Arizona

and Oklahoma Appropriations.” Public Budgeting and Finance (Fall) 22, 3: 26- 45.

Garsombke, H. Perrin, and Jerry Schrad. 1999. “Performance Measurement Systems: Results from a City and State Survey.” Government Finance Review (February)

15, 1: 9-12. GASB 2002. Performance Measurement Survey: Performance Measurement at the State

and Local Levels: A Summary of Survey Results. http://www.seagov.org/index.html. GASB, 2000. State and Local Government Case Studies: The Use and the Effects of

Using Performance Measures for Budgeting, Management, and Reporting. (April). http://www.rutgers.edu/Accounting/raw/seagov/pmg/index.html.

Grizzle, Gloria A., and Carole D. Pettijohn. 2002. “Implementing Performance-Based Program Budgeting: A System-Dynamics Perspective.” Public Administration Review (January/February) 62, 1: 51-62.

Grizzle, Gloria A. 1986. “Does Budget Format Govern Actions of Budgetmakers?” Public Budgeting and Finance (Spring) 6, 1: 60-70 Harris, Jody L. 2000. “Best Value and Performance Management: Lessons Learned from the United Kingdom.” Government Finance Review (August) 16, 4: 27-32. Hatry, Harry. 1999. Performance Measurement: Getting Results. Washington, D.C.: The Urban Institute Press. Ingraham, Patricia W., and Donald P. Moynihan. 2001. “Beyond Measurement: Managing for Results in State Government, Chapter 12 in Quicker, Better Cheaper: Managing Performance in American Government, edited by Dall W. Forsythe (NY: The Rockefeller Institute Press): 309-334. Jordan, Meagan M., and Merl M. Hackbart. 1999. “Performance Budgeting and Performance Funding in the States: A Status Assessment.” Public Budgeting and Finance (Spring) 19, 1: 68-88. Julnes, Patria de Lancer, and Marc Holzer. 2001. “Promoting the Utilization of Performance Measures in Public Organizations: An Empirical Study of Factors Affecting Adoption and Implementation.” Public Administration Review (November/December) 61, 6: 693-708. Kluvers, Ron. 2001. “An Analysis of Introducing Program Budgeting in Local Government.” Public Budgeting and Finance (Summer) 21, 2: 29-45. Lee, Robert D., and Robert C. Burns. 2000. “Performance Measurement in State Budgeting: Advancement and Backsliding from 1990 to 1995.” Public Budgeting and Finance (Spring) 20, 1: 38-54. MacReynold, Merlin, and Kevin Fuhrer. 2002. “Want to Balance Your Budget? Constraint Budgeting Works.” Public Management (May) 84, 4: 19-21. Melkers, Julia, and Katherine Willoughby. 2001. “Budgeters Views of State Performance Budgeting Systems.” Public Administration Review (January/February) 61, 1: 54- 64. ________. 1998. “The State of the States: Performance-Based Budgeting Requirements

in 47 Out of 50.” Public Administration Review (January/February) 58, 1: 66-73. Pettijohn, Carole D., and Gloria A. Grizzle. 1997. “Structural Budget Reform: Does It

Affect Budget Deliberations?” Journal of Public Budgeting, Accounting and Financial Management (Spring) 9, 1: 26-45.

Poister, Theodore H., and Gregory Streib. 1999. “Performance Measurement in Municipal Government: Assessing the State of the Practice.” Public Administration Review (July/August) 59, 4: 325-335.

Scheps, Philip. 2000. “Linking Performance Measures to Resource Allocation.”

Government Finance Review (June) 16, 3: 11-15. Schick, Allen. 2001. “Getting Performance Measures to Measure Up.” Chapter 3 in

Quicker, Better, Cheaper,: Managing Performance in American Government, Dall W. Forsythe, Editor (Albany, NY: The Rockefeller Institute Press): 39-60.

Smith, James Fielding. 1999. “The Benefits and Threats of PBB: An Assessment of

Modern Reform.” Public Budgeting and Finance (Fall) 19, 3: 3-15. Stiefel, Leanna, Ross Rubenstein, and Amy Ellen Schwartz. 1999. “Using Adjusted

Performance Measures for evaluating Resource Use.” Public Budgeting and Finance (Fall) 19, 3: 67-87.

Streib, Gregory and Katherine Willoughby (forthcoming). “Local Governments as E- Governments: Meeting the Implementation Challenge.” Public Administration

Quarterly. Thurmaier, Kurt, and Katherine Willoughby. 2001. Policy and Politics in State

Budgeting. M.E. Sharpe Publishers. Walker, David M. 2002. “Performance Budgeting: Opportunities and Challenges.”

Statement of the Comptroller of GAO Before the Government Reform Subcommittee on Government Efficiency, Financial Management and Intergovernmental Relations, Rules Subcommittee on Legislative and Budget Process, House of Representatives. GAO-02-1106T

Wang, XiaoHu. 2000. “Performance Measurement in Budgeting: A Study in County

Government.” Public Budgeting and Finance (Fall) 20, 3: 102-118. Willoughby, Katherine G. 2002. “State Revenue Choices in Slow Times.” Spectrum: The

Journal of State Government (Spring) 75, 2: 9-11.

Willoughby, Katherine G., and Julia E. Melkers. 2001. “Performance Budgeting in the States.” Chapter 14 in Quicker, Better, Cheaper? Managing Performance in American Government, edited by Dall Forsythe (The Rockefeller Institute Press): 335-364.

________. 2000. “Implementing PBB: Conflicting Views of Success.” Public Budgeting and Finance (Spring) 20, 1: 105-120.

Appended Tables

All Departments >50% of Departments <50% of Departments Few or No Departments

Used by: Types of Measures

(# state respondents/# local respondents) State Local State Local State Local State Local

Performance Measures (191/245)

38% 48% 36% 20% 11% 6% 15% 26%

Input Measures (177/238)

42 52 26 19 11 5 22 25

Activity/Process Measures (181/237)

33 44 33 26 13 7 21 23

Output Measures (195/253)

31 42 35 25 16 12 17 21

Outcome Measures (188/243)

20 24 31 23 26 17 24 36

Cost/Efficiency Measures (186/238)

10 21 26 23 28 19 36 37

Customer Satisfaction Measures (179/241)

6 20 18 23 26 18 50 39

Explanatory Measures (152/186)

9 12 15 15 25 13 51 60

Benchmarks (167/226)

5 11 22 13 20 19 53 58

Table 1: Comprehensiveness of Performance Measurement Application in Governments, Responses by State and Local Administrators and Budgeters in Percents

Development and use of performance measures has been effective or very effective in our agencies and departments for:

State Level Respondents

%

responding (n)

Local Level Respondents

%

responding (n)

1) Increasing awareness of, and focus on, results 53% (189)

44% (239)

2) Increasing awareness of factors that affect performance results 50% (187)

41% (239)

3) Changing strategies to achieve desired results 42% (185)

33% (238)

4) Improving communication between departments and programs 41% (189)

30% (238)

5) Improving communication with the executive budget office 40% (185)

40% (239)

6) Improving responsiveness to customers 40% (185)

39% (244)

7) Improving communication with the legislature/council/commission and/or their staff

40% (189)

34% (241)

8) Improving effectiveness of department programs 38% (188)

33% (242)

9) Improving programs/service quality 35% (186)

36% (242)

10) Changing the substance or tone of discussion among legislative members in budgeting or oversight

32% (184)

22% (234)

11) Communicating with the public about performance 31% (187)

34% (242)

12) Changing the questions legislators or their staff ask of government managers or executives

27% (185)

27% (237)

13) Improving cross department cooperation/coordination 25% (186)

22% (241)

14) Reducing duplicative services 21% (183)

17% (239)

15) Changing appropriation levels 20% (185)

17% (236)

16) Reducing/eliminating ineffective services/programs 20% (185)

16% (241)

17) Improving external government cooperation/coordination 20% (183)

15% (239)

18) Affecting cost savings 17% (184)

21% (240)

Table 2: Responses by Level of Government Affirming Performance Measurement Effectiveness

All Departments >50% of Departments <50% of Departments Few or No Departments

Extent that Output or Outcome Performance Measures Appear in

(# state/# local) State Local State Local State Local State Local

Department budget requests (56/138)

45% 32% 30% 23% 16% 12% 9% 34%

The 1999-2000 executive budget report or documents

(54/134)

37

40

30

19

6

8

28

34

Annual operating budgets (47/138)

30 43 21 20 6 8 43 30

Quarterly reports (40/121)

15 12 13 7 15 5 58 76

Annual reports (39/133)

21 19 23 14 18 9 39 59

Table 3: Responses of Budgeters by Level of Government Acknowledging the Appearance of Output or Outcome Measures in Budget Documents

State

% (n)

Local

% (n)

Department budget development 35 (57) 36 (140) Department budget appropriations 26 (57) 31 (140) Department budget execution 28 (57) 29 (139)

Dep

artm

ent

Department audit/assessment of results 30 (54) 31 (138)

Government wide budget development 30 (54) 39 (140) Chief executive deliberations 27 (56) 31 (140) Legislative/council/commission deliberations 21 (56) 29 (138) Government wide budget appropriation

determinations 13 (56) 35 (139)

Government wide budget execution 16 (56) 26 (140) Gov

ernm

ent w

ide

Government wide audit/assessment of results 18 (56) 27 (139)

Table 4: Responses of Budgeters by Level of Government Claiming Output or Outcome Measures as Very Important or Important in the Budgetary Process