Executive Summary

These factors could be formidable forces on state government to do a better job, but institutional inertia and turf-protecting instincts are immense barriers to change. It is within this environment of warring pressures that California began a performance-based budgeting pilot project in 1993. While the pilot project has yet to run its course, mid-way assessments by the Legislative Analyst's Office and legislative staff have been cautious, if not negative.

The Little Hoover Commission turned its attention to the pilot project as a natural progression from other studies it has conducted on the mechanics of government. In prior reports on how the State oversees its real property assets, how the State procures goods and services and how the State manages its work force, the Commission has advocated reforms that echo private-sector practices in an effort to make government more effective and efficient. The Commission, therefore, had an interest in examining performance-based budgeting as a concept that would help the State focus on the bottom line of wisely spending scarce resources.

Study Methodology

T o carry out its study, the Commission gathered information from experts across the nation, as well as examined the experience of other states, the federal government and other nations. The Commission conducted a public hearing in Sacramento, bringing together some of those experts and the state department heads involved in the pilot project (see Appendix A for a list of hearing participants). Finally, the Commission intensively studied the experience of Oregon and Texas, the two states that appear to have made the most progress in perfecting performance-based budgeting.

The result is the following report, which focuses on the concept of performance-based budgeting rather than on an evaluation of the pilot project process in California. The Commission believes that the specifics of the pilot project implementation are less important to the eventual success of the experiment than the steps that need to be taken, as demonstrated in other jurisdictions, to build consensus and support for change.

The report begins with an Executive Summary and this introduction, followed by a background section and three separate findings with associated recommendations for each. The report ends with a conclusion, appendices and endnotes.

Background

I n an era of recession-reduced revenues and increasing demands for services, national, state and local governments are seeking ways to improve delivery of effective services at the lowest possible cost. California is no exception. In 1993, the State began a pilot program to test performance-based budgeting, a complex process of strategic planning and outcome evaluation that seeks to ensure that programs are funded according to their ability to produce positive results.

The State's first steps in this direction come at a time when many other states have taken major strides to reform their budgeting mechanisms. The federal government is embarking on a massive pilot project and foreign governments are well immersed in linking evaluations to budgetary decisions.

While California is not a leader in the movement toward performance-based budgeting, it long ago incorporated many of the elements in its own budgeting process that other states are just now adopting -- concepts such as delineating funding by program functions and providing key program indicators within the budget document. Thus, the State has not felt the same pressure for change that policy makers in other states have who are frustrated by cryptic, enigmatic budget documents. By comparison, California's budget, as introduced by governors each January, is a treasure trove of information about how much specific programs cost, how many people are employed to operate them and, often, how much service is delivered. Nonetheless, the budget falls far short of providing a picture of how successful state government is in reaching the policy goals represented by the commitment of funding.

Nationally and in California, the move to performance-based budgeting is but the latest in a long line of attempts to improve the way governments allocate funds and provide accountability. Understanding what has come before and why other reform efforts have failed is important for those who seek to make the new budgeting overhaul successful. In addition to explaining the performance-based budgeting concept, this background section provides a historical context for budget reform, summarizes the experience of other government jurisdictions and reviews the pilot project in progress in California.

A Definition

P erformance-based budgeting is an approach to allocating funds that attempts to link measured results with policy decisions. This method stresses holding departments accountable for outcomes, prioritizing spending based on a program's ability to successfully reach goals, and providing comparative data on how much value is received when dollars are allocated in different ways.

As the Legislative Analyst's Office defines it, performance-based budgeting is:

...the allocation of resources based on an expectation of performance levels, where performance is measured in specific, meaningful terms. It differs from the traditional approach to budgeting in that it focuses on outcomes rather than inputs or processes when deciding how to allocate resources. 1

For instance, rather than funding a job-training program based on the number of clients needing services and the number of caseworkers that are required to meet that need, a performance-based budget would offer policy makers a grid of expected results that would be verified over time. An investment of one level of dollars would have the potential of achieving a 10 percent reduction in unemployment, while a greater investment may bring a 20 percent reduction.

The contrast between performance-based budgeting and the more-traditional line-item budgeting is clear because the perspectives are quite different, according to a report by the National Conference of State Legislatures:

The traditional approach to government budgeting focuses on incremental changes in detailed categories of expenses called line items. More money is appropriated for specific expenses like personnel, equipment or rent as demand for an agency's service expands or new program responsibilities are added over time.

Performance budgeting differs from this traditional approach because it focuses on spending results rather than the money spent -- on what the money buys rather than the amount that is made available. 2

The report identified the following characteristics of a performance-based budget:

• A statement of the goal in allocating funds for a particular program and a set of measurable objectives.
  

• A report on past performance and the use of a common cost factor that allows direct comparisons between disparate types of programs.
  

• Flexibility for managers to redirect funds as needed, along with rewards for success and sanctions for failure.
  

• Data on program evaluation that is reliable, credible and capable of being independently verified. 3

The Governor's Budget Summary of 1993-94 provided a similar but expanded list of "essential elements" for the performance-based budgeting pilot project in California. The elements include:

• The creation of long-term strategic plans, with the State identifying missions for each program and strategies for carrying out the missions.
  

• The development of performance measures to be used in a department's program and financial planning process.
  

• The establishment of benchmarks by which to measure progress toward more efficient and effective operation.
  

• The development of annual budgetary contracts that specify the level of resources provided to a program, the level of performance expected, accountability measures and flexibility granted to managers.
  

• The provision of flexibility to managers to enhance program performance through relief from control mechanisms involving procurement, personnel and other requirements.
  

• The incentive of retaining 50 percent of any savings within a department to encourage program efficiency and innovation.
  

• The commitment to quality improvement, with a focus on customer service, constant redesign of processes and employee empowerment.4

The Governor's Budget Summary emphasized the potential that performance-based budgeting holds for increasing efficiency: "The development of a performance budget requires budget writers and program managers to determine the best method to maximize program performance and enhance service delivery, given a realistic level of available funding." 5

As will be examined in more detail in Finding 1, performance-based budgeting offers policy makers the opportunity to make program evaluation part of the budget process, to provide managers with flexibility while maximizing accountability for results and to encourage cost-effective streamlining of programs. None of these are new goals, as a review of more than four decades of budgeting reforms indicates.

Historical Context

O fficials have long struggled to make dry and complex budget documents more understandable and relevant to those who must use them to make decisions. The federal Commission on the Organization of the Executive Branch of Government, more commonly known as the Hoover Commission, in 1949 recommended that the federal government reshape its budget to focus on functions, activities and projects rather than agencies and departments. The Commission used the term "program budget" to describe the new concept, which was touted as a simpler, more understandable way of presenting the budget with an emphasis on "the general character and relative importance of the work to be done" rather than on things (personnel, equipment, supplies, etc.) to be acquired by departments. The Commission's report included a nod to the State of California for already using programmatic themes in its budget.

The federal government adopted program budgeting in 1951, but this proved to be only a single change in a succession of reforms. Budgets were now formatted by programs, but expenditure categories were still driven by "things" rather than by work output.

In 1961, Secretary of Defense Robert McNamara implemented Planning-Programming-Budgeting-Systems (PPBS) in the department, which was followed by a presidential order in 1965 that all federal departments adopt the new system. This system was designed to provide macro-analysis of broad policy decisions and desired outcomes rather than focusing on the details of planning and implementation. According to one national researcher, PPBS was intended to "provide a systematic process to identify trade-offs among programs aimed at similar objectives, to analyze the performance and impact of programs and to connect these objectives to the current year's budget."6

Despite goals that were very similar to those expressed by the federal Hoover Commission, the new system was no more successful than the old in driving decisions toward proven results. In the late 1960s, witnesses told congressional committees that across the federal system there had been no apparent change in the decision-making process and the budgets being produced. 7

President Carter brought Zero-Base Budgeting to the federal government after implementing it as governor of Georgia in 1971. This approach tackles the problem of programs becoming entrenched and receiving funding long after their usefulness has diminished or their aims have become a lower priority for policy makers. Under Zero-Base Budgeting, each year's budget is supposed to be designed fresh without relying on prior years' totals. Each program is theoretically begun with a zero budget and is funded after competing with all other programs for available resources. Directors are required to identify, evaluate and rank in order of importance each function and operation to be performed by a department so that program value can be compared across all sectors of government. 8

Zero-Base Budgeting met much the same fate as PPBS: Departments quickly learned the rituals that needed to be performed, but actual decision-making remained largely unaffected by the new system, according to many observers.

By the 1980s, a national trend toward performance-based budgeting was emerging, with the California community of Sunnyvale leading the way. The 1993 publication and runaway success of Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector put the catchwords mission-driven, results-oriented and customer-focused on the agenda of every progressive policy maker. Cautious observers, however, wonder if performance-based budgeting will prove to be just one more new exercise that fails to change decision-making in any fundamental way. This perspective gained credence when the United States General Accounting Office examined performance-based budgeting in five states and concluded that the system has yet to influence state budget decisions. Instead, allocations continued to be driven by traditional budgeting practices. 9

Advocates are firm in the belief that the performance-based budgeting approach holds the key to making government efficient and responsive. And the track record of at least some of the jurisdictions using performance criteria to mold their budgets demonstrates that the system has vast potential when used well.

Success Stories

S unnyvale, a city of 120,000 people just south of San Francisco, has become internationally renowned for its performance-based budgeting. Under a federal Office of Management and Budget (OMB) grant in the mid-1970s, Sunnyvale began to experiment with an approach that it calls the Planning and Management System (PAMS). By May 1991, OMB was impressed enough with the results to tell a Senate committee that Sunnyvale's system:

...stands out as the single best example of a comprehensive approach to performance measurement that we have found in the United States...One underlying reason for the success achieved in Sunnyvale is the fact that every program manager uses the system to plan, manage and assess progress on a day-to-day basis. 10

The elements of the system include a 20-year strategic plan, a 10-year financial plan, a two-year performance budget, annual evaluations and performance-based compensation.11 Rather than formatting its budget along departmental lines, Sunnyvale matches its allocations to the seven topics in its 20-year strategic plan: transportation, community development, environmental management, public safety, socio-economics, cultural and planning and management. The budget allocates funding for tasks rather than for personnel, equipment or supplies, with quantifiable objectives that are expected to be achieved with the funding. 12

The proof of Sunnyvale's success with the system comes not just in the kudos it receives in books such as Reinventing Government and in international forums, but in the statistics it can provide. Compared to fiscal year 1984-85, Sunnyvale reported in 1993-94 that there was a 44 percent improvement in worker productivity and a 38 percent improvement (in constant dollar terms) in the cost of providing city services. 13

The City of Indianapolis for several years has produced what it calls a "popular budget" for several major departments, including parks and recreation and public safety. The budget provides mission statements, allocations by outcome objectives and comparative performance measures. An introductory statement that is specific to the parks and recreation department captures the flavor of what the city is attempting to achieve with its popular budget:

This agency is under a major transformation that represents a dramatic departure from traditional government approaches. Indy Parks hopes to communicate to the public and to its constituencies that parks and recreation are a necessity. With today's frantic lifestyles, Indy Parks will strive to provide enticing recreational opportunities so that people are healthier and more productive. With every action that Indy Parks takes, it will do so with the end customer in mind. Considering the customer's wants and needs will continue to be important at every phase of implementation. 14

The successful implementation of performance-based budgeting has not been limited to this country. The Organization for Economic Cooperation and Development, based in Paris, has promoted results-oriented approaches to governmental service for the past five years. The United Kingdom, Australia and New Zealand have been at the forefront of making government more responsible and accountable through performance measurements.15 Australia is particularly well known for its system of strategic planning, annual work plans, performance reporting and program evaluation. The government each year publishes a 19-volume set of detailed program performance statements relating funding to accomplishments. 16

In 1991, the United Kingdom began a 10-year program called the Citizen's Charter to improve public services. The program aims to make clear what standard of service people can expect and what they can do about it if service falls short. A 1994 report on the effort produced by the prime minister documented improvements in services in 24 different areas of government, ranging from health and education to the post office, defense, agriculture, planning and housing. In each area, the report delineated with specific objectives and data what had been promised, what was achieved and what future steps were planned. The United Kingdom's efforts have been widely acknowledged as successful, prompting other countries to study the model. 17

Closer to home, most efforts in the United States are too fresh for any definitive analysis of the results. But performance-based budgeting, either whole or in part, is spreading rapidly.

The National Experience

T wenty-four states have tried performance-based budgeting to some extent. Eighteen are in various stages of modifying their budget procedures: Arizona, California, Connecticut, Florida, Georgia, Idaho, Massachusetts, Minnesota, Mississippi, North Carolina, New Mexico, Oklahoma, Oregon, South Carolina, Texas, Virginia, Washington and Wyoming. Another six are emphasizing long-range strategic planning but are expected to reform budget processes as a next step: Alabama, Arkansas, Missouri, Utah, Vermont and West Virginia. Eight states may join the performance-based budgeting ranks in 1995: Alaska, Kansas, Louisiana, Maryland, Nebraska, New York, North Dakota and Pennsylvania. 18

The National Conference of State Legislatures identifies three types of performance reforms that are emerging. The broadest type rethinks what role government should play, extends beyond single budget cycles and looks at the highest levels of program outcome. Oregon, Minnesota and Utah are involved in this type of reform. "An unspoken expectation in these states is that once long-range goals are hammered out, these objectives will limit budget debate, structure program activity and establish spending priorities for years to come." 19

The second type of reform focuses on efficiency and effectiveness of current services, moving into the new budgetary system quickly and with little advance discussion. Texas fits into this category.

The third type focuses on managerial aspects, giving departments freedom from budgetary line-item control in exchange for promised levels of performance. California, Massachusetts and Mississippi are working on this concept. 20

The degree of involvement in performance-based budgeting varies. The following summarizes the experiences of several states that are considered on the leading edge of the movement:

• Texas: In a single, biennial budget cycle, Texas converted its entire funding process to a performance-based budgeting system. Beginning with a top-down strategic planning process, Texas policy makers defined a vision statement for state government and directed all agencies to develop mission statements that fit within the vision. Agencies were required to develop six-year strategic plans and performance measurements, identifying outcomes and outputs. The 1993-95 budget had more than 3,000 key performance targets that agencies were expected to be held accountable for achieving. A senior analyst told the Little Hoover Commission that, by and large, the system has worked well, especially for those agencies that have taken the process seriously. 21

• Oregon: Working from the bottom up, Oregon began conducting a series of statewide public hearings in 1988 that culminated in a strategic plan and mission statement entitled Oregon Shines. The Legislature created the Oregon Progress Board in 1989 with the mandate to develop measurements and criteria for holding government programs accountable. The result was Oregon Benchmarks: Standards for Measuring Statewide Progress and Government Performance, a document with almost 300 achievement goals that is regularly updated and subject to extensive public review. By the 1993-95 biennial budget, all state agencies were directed to use performance measurements linked to the benchmarks in developing budget requests. A state coordinator for the program estimates that only about 25 percent of the agencies have produced effective measurement systems, but others continue to make progress. 22

• Florida: Moving on a slower track, Florida has set up a gradual schedule of conversion to performance-based budgeting that is expected to encompass all agencies by 2002. Unlike many states where part of the goal is to deflect micro-management by policy makers, Florida's reform effort came from frustration with a budget that gave no program information and left agencies free to shift funds away from activities that the Legislature set as priorities. A half dozen agencies are expected to be added to the process each year.
  23

• North Carolina: Driven by a $1.2 billion shortfall in revenues for 1991-92, North Carolina created a State Government Audit Committee with legislators and private-sector representatives. In 1993, the legislature enacted many of the committee's recommendations, including a mandate to develop a performance-based budgeting system that focuses on results rather than line items and a new personnel classification system that would allow greater management flexibility and reward performance. 24

• Minnesota: In 1993, 20 agencies were directed to produce annual performance reports on which the 1994-95 budget could be based, with all other agencies to follow by the 1996-97 budget. The reports are required to answer three questions: What is the agency's mission? What are the expected results of agency programs? What are the actual results? The reports were to be used by the Legislature during the budget deliberation process.25 But a report by the state's auditor indicates that the first set of performance reports had very little impact on the discussion and decisions by executive and legislative branches for the 1994-95 budget. 26

• Pennsylvania: Focusing on the development of a solid measurement system first, Pennsylvania requires all departments to form five-year strategic plans and annual performance plans. Based on three years' experience with projections and actual results, the Budget Office is expected to recommend the best way to implement a performance-based budgeting system by late 1998. 27

• Washington: The Washington Performance Partnership is a multi-year plan by the state's leaders to refocus state government programs on meeting the needs of citizens. Using focus groups, citizen input and surveys, the state's leaders have put together a strategic statement for Washington government and is directing the implementation of performance measurements throughout state agencies. The biennial 1997-99 budget is expected to be based on performance measures and strategic visions. 28
  

• Virginia: The state's 1992 budget act required all new programs to be set up with performance measures. Existing programs were to be moved into performance-based budgeting on a pilot basis in future years.29

The federal government is also moving in the direction of performance-based budgeting. In 1993, the National Performance Review recommended that future budgets be based on outcomes, finding that the current budgeting process is both messy and costly. The panel's report described the lengthy negotiations and their effect:

Until the end, agency officials troop back and forth to [the Office of Management and Budget] and to the Hill to make their case. States and localities, organizations and advocates seek time to argue their cause. Budget staffs work non-stop, preparing estimates and projections on how this or that change will affect revenues or spending. All this work is focused on making a budget -- not planning or delivering programs....

The process is devoid of the most useful information. We do not know what last year's money, or that of the year before, actually accomplished. Agency officialsdevise their funding requests based on what they got before, not on whether it produced results.

In sum, the budget process is characterized by fictional requests and promises, an obsession with inputs rather than outcomes, and a shortage of debate about critical national needs. 30

The same year the report was issued, the Government Performance and Results Act was enacted. It required at least 10 federal pilot projects to test strategic planning, annual performance plans and measurable outcomes. To date, more than 75 agencies are involved. The same act required all agencies beginning in 1997 to submit five-year strategic plans, to be updated each three years, and annual performance plans. 31

The fact that all levels of government are experimenting with budgeting techniques reflects the shared need to improve and streamline programs. With widespread interest both in the United States and around the world in reforming budgeting techniques, it is not surprising that California would become involved. What is notable, however, is that the State -- with a history of being an aggressive leader, if not innovator, in reform movements -- has taken a cautious, small-step-at-a-time approach.

California

E nacted by both Executive Order of the Governor and statute in 1993, California has a performance-based budget pilot project that currently involves four departments: the Department of General Services, the Department of Parks and Recreation, the Department of Consumer Affairs and the California Conservation Corps. The Department of Finance provides oversight for the pilot project and is required to report to the Legislature on January 1, 1996 on the extent to which performance-based budgeting results in more cost-effective and innovative programs and services.

The departments in the pilot project are diverse in many respects, including size, areas of responsibility and types of programs. For instance, the California Conservation Corps, with 415 employees and a $56.7 million budget, is tiny compared to the Department of General Services, with 3,740 employees and a $503.1 million budget. The Department of Consumer Affairs deals extensively with "clients" outside of state government (the people it regulates and the State's consumers), while the Department of General Services' activities are directed toward other state agencies. The Department of Parks and Recreation manages natural resources, while the California Conservation Corps focuses on training youths for productive lives.

The goal of the pilot project is to find more cost-effective ways to deliver government services through strategic planning, performance measurement and the budgetary decision-making process. The statute creating the pilot project lists nine elements that in general outline a program where departments will be given flexibility but in turn will be held accountable for results that are delineated in a contract each year between the Legislature and the department.

Each department has followed a fairly similar path of strategic planning (developing mission and goal statements), identifying measurable objectives and linking budgetary needs to program activities. As of mid-1995, each department had a strategic plan in place, as well as mechanisms for updating and revising the plans; each had established performance measurements and baseline data, although some were still being revised; and each had completed a budget contract with the Legislature. The contracts, however, focused on the flexibility that managers would be allowed and the achievements that would be pursued. The actual budgets adopted for each department for the 1995-96 fiscal year continued to reflect the line-item format used for all state departments.

Managers told the Little Hoover Commission that a lack of guidance from the Department of Finance, particularly in terms of standardized processes and explicit expectations, has forced each department to struggle with a steep learning curve. The Legislative Analyst's Office particularly noted that there were no guidelines to help departments determine how to conduct budget negotiations with the Legislature, and that no common reporting formats or computer application standards had been set to facilitate the shift to a performance-based budget document. 32

The Department of Finance, however, has maintained that since each department is so different an important part of the pilot project process is to allow each to develop separately with little constraint. In sharp contrast to enthusiastic advocates of performance-based budgeting who testified to the Commission and the leaders of state departments in the pilot project who uniformly said they saw positive changes in their operations, the Department of Finance sees performance-based budgeting as one in a long line of attempted reforms that has only limited applicability to state government. The Chief Deputy Director of the Department of Finance told the Commission:

We continue to believe that not all agencies are well suited to performance-based budgeting. Our focus should remain on those agencies whose services mirror the private sector and which have identifiable measures of performance. Some agencies administer programs for the federal government; they operate under rules that we do not control in California. Other agencies have mandated responsibilities that are not amenable to the level of discretion necessary for performance-based budgeting to succeed. 33

An expert from Texas, however, told the Commission that his state's overnight conversion to performance-based budgeting in all agencies has worked well for the most part. He said the factor that seems to determine whether the method will work for an agency is not the agency's mission or program constraints, but the degree of seriousness with which the agency approaches the process of strategic planning and measurement development. 34

Similarly, the U.S. Senate counsel most closely connected to the federal experiment (and a former mayor of Sunnyvale) said that performance-based budgeting can work with any level of government and any type of program. He said an appropriate analogy is the difference between a 747 jet and a small single-engine plane; both fly under the same general principles of aerodynamics but each is designed and constructed differently. 35

Both the Department of Finance and the Legislative Analyst's Office agree that the State has seen little tangible results from the pilot project so far, although departments have reported improved customer sensitivity. Departments have spent more than $5 million on the process of creating strategic plans and identifying performance measures, although some of that cost involves staff time that would have been spent on budgeting activities regardless of the pilot project. Cost savings are not expected to accrue until departments have been able to implement the program fully -- and that is not expected to be before the January 1, 1996 reporting deadline.

While the results of the pilot project have yet to be assessed, policy makers have taken several actions to set the stage for the method's eventual expansion to other agencies.

In 1994, the State Government Strategic Planning and Performance Review Act (AB 2711, Chapter 799, Statutes of 1994) said that "strategic planning is a prerequisite for effective performance reviews and effective performance budgeting." The act requires the Department of Finance to survey annually state agencies regarding the status of their strategic plans and to recommend which ones should develop or update a plan. The act also requires the Controller, the Department of Finance and the Bureau of State Audits, in consultation with the Legislative Analyst, to develop a plan for conducting performance reviews of all state agencies.

In addition, executive orders and statutes have focused on total quality management programs and results-oriented processes. A total of 27 "Pioneer Projects" have brought total quality management techniques to state agencies under a 1993 executive order. A 1993 statute (SB 1082; Chapter 418, Statutes of 1993) requires Cal-EPA to develop a model quality program and to begin in 1998 submitting a yearly progress report on the achievement of performance objectives as part of the budget process.

Regardless of the eventual assessment of the performance-based budgeting pilot project, California clearly is moving away from traditional top-down, command-and-control mechanisms for delivering state services. The following three chapters of findings examine the experience of those most familiar with the necessary steps to re-engineer government effectively and provide recommendations designed to speed the process and avoid missteps in California.

Flawed Process

Finding 1: The current process for allocating funds and setting program priorities is not a framework that encourages the best policy decisions, especially in times of economic contraction.

T he traditional line-item budget invites policy makers to add funds automatically to existing programs each year to take care of caseload growth. In years when resources are growing, such reflexive action is possible even as new layers of programs are added. But when resources fail to keep pace with demands, policy makers would be better served by a system that helps them make rational choices. Such a system would quantify outcomes that will be achieved by various levels of spending. Informed decisions about how to get the most value out of limited resources to meet competing needs could then be made. The focus could shift to reaching consensus about priorities rather than on battling to protect existing programs, regardless of performance.

Government budgeting experts agree that no system can make difficult choices on behalf of policy makers. Decisions must be made about how to meet the competing demands on state government, regardless of how data is presented and tracked. No system provides an "automatic pilot" for arriving at the proper balance of public expenditures. But it is widely acknowledged that some systems enhance the ability of policy makers to focus on pertinent data and make sound comparisons between programs, including their effectiveness and efficiency.

Most governmental entities use a budgeting system referred to interchangeably as line-item, baseline, current services or workload. The budget for a department under this system lists personnel, equipment, supplies, travel, training and other well-defined categories for how money is to be spent. An unrefined version of this type of budget makes no reference to the separate programs that may be operated by a department, the source of funding, the operational goals, the number of clients served or other information that sheds light on how money is used in a broad sense. Instead, it merely describes how money is spent.

California's Budget

C alifornia's budget has a long history of being more informative than a straightforward line-item budget. The 1949 federal Hoover Commission report acknowledged California as a leader in describing expenditures in terms of programs rather than departments.³⁶ This allows people to tell how much is being spent on specific groups of services rather than lumping the many operations of a single department together.

In the late 1970s, the push by "tax revolt" proponents to cut government waste prompted the Legislature to improve budget oversight and require more information on which to base decisions. One major bill, Chapter 1284 of the Statutes of 1978, put into place most of the recommendations from a study by Deloitte Haskins & Sells on the State's fiscal management. Among other things, the law required:

• The Department of Finance to operate an accounting system with expenditures by program, item, organization and fund source.
  

• The Governor to propose his budget in terms of programs rather than simply by department.
  

• Federal allocations to be incorporated in the budget as a source of funds by program.
  

• The Department of Finance to develop an online California Fiscal Information System to monitor the budget, forecast revenues and allow comparison of the affect of different spending choices -- and to provide access to the information to both the Legislature and the Administration.
  

• The Department of Finance to develop departmental performance measures for each state agency.³⁷

Although several of the requirements were later modified, the statute substantially affected how the State's annual budget was presented. The box on the next page, which contains in abridged form all of the elements in the 1995-96 Governor's Budget for the California Conservation Corps, shows the range of information in the budget.

The elements in the budget include:

• An introductory paragraph about the department, followed by the statutory citation that provides authority for the program.
  

• A summary that tells how many people work for the department, how much money is spent and where the money comes from. All figures in the budget -- expressed in thousands of dollars -- are compared across three years: the actual prior year, the anticipated current year and the proposed next year.
  

• A statement about the program's objective and a description of how it operates.
  

• A listing of the major budget adjustments for the current year and those that would affect the proposed year.
  

• A detailed budget showing the total funding coming from separate sources and the different programs within the department that funding will be spent on, by source.
  

• A summary for the whole department showing how much will be spent on personnel salaries and benefits, operating expenses and equipment.
  

• A reconciliation of the appropriations by funding source.
  

California's budget, then, displays a broad range of information. But all of it is information relating to what a program is expected to do -- not what it has actually accomplished. It is full of "inputs" (number of personnel, amounts to be spent on salaries, benefits, equipment, etc.) and not "outcomes."

In addition, like all line-item budgets, it encourages policy makers to make automatic cuts or increases -- rather than focusing on the quality or level of service being funded -- since there is no information that would allow such a judgment. The Governor's Budget Summary for 1993-94 described the exercise the administration goes through to propose the next year's budget:

The Executive Branch phase of this process begins by developing a workload budget for each program for the upcoming fiscal year. The workload budget is calculated by adjusting the program's authorized level of spending in the current fiscal year for one-time costs, increases in the cost of operations, implementation of new legislation and for additional workload anticipated under current law.³⁸

None of this exercise involves determining how well a program has performed and making some assessment of whether the program should continue. In writing about the problems with traditional budgeting, the National Conference of State Legislatures agreed that the focus on line items leads to an incremental approach.

Incremental budgeting assumes that past decisions remain appropriate and would not typically be subject to major revision. Incremental budgeting does not easily allow a state to shift large sums of money from existing programs to more pressing needs.³⁹

In testimony to the Commission, the Department of Finance said that such an approach to budgeting works when revenues are growing but:

...it fails when revenues are shrinking because it builds false expectations by calculating a program's "entitlement" and then failing to fund the program to that level. The result is to create "funding gaps" which pit one program against another to see which program can fill more of its gap. The losers are the people of California. In the environment of today's realities, we need a new budget process that shifts from a focus on expenditures to a focus on available revenues. In this new budget paradigm, the most important question becomes, "What is the best program I can deliver for the funds that are available?"⁴⁰

Because the current process builds on prior budgets, new and innovative programs often get squeezed out with little consideration. The revenues that exist beyond that needed to cover workload increases in existing programs are the only source of funding for new concepts. When resources are not expanding, earmarking funding for one program, either new or old, often means cutting another. Competition among programs for limited dollars creates countervailing pressures on policy makers that are easiest to avoid by across-the-board cuts that ignore issues like performance, quality and need.

Summing up the problems with the workload budget concept, the Governor's Budget Summary for 1993-94 said:

The process of developing a workload budget has created unrealistic expectations among department managers and employees, elected officials and program recipients that the current level of service is an entitlement that can be continued without regard to the State's fiscal condition. It makes no allowance at all for productivity or quality improvements. It places too much emphasis on the level of funding promised to each program, and too little emphasis on program performance. Finally, it has relegated the development of the budget to a technical calculation of desired levels of funding, when the budget should instead be a tool for improving program management.⁴¹

Frustration with the current system goes beyond the state bureaucrats and policy makers struggling to make it work. The California Business-Higher Education Forum, the California Taxpayer's Association and the California Policy Seminar have all issued statements backing the performance-based budgeting concepts.⁴²

In addition, experts from other governmental jurisdictions that are at various stages of making performance-based budgeting work uniformly praise it as a much-needed improvement over traditional budgeting. The following sections provide their observations and advice, as well as samples of their budgeting techniques.

Sunnyvale

S unnyvale's budget document is rich in information, ranging from past trends in performance to current statistics on the unit cost of different activities and targets for future performance. Organized by function rather than department (for instance, solid waste disposal rather than the Department of Public Works), the budget reflects increased productivity over time and captures the return on innovative program management.

Perhaps the most striking aspect of the city's budget is the long list of performance indicators -- the measurements that tell what is being achieved by the city's programs and services. The ones in the box on the next page deal with solid waste disposal, but there are many other examples: having fewer than 10 complaints about telephone information services; repairing damage from vandalism within three days; successfully assisting on library information requests 95 percent of the time; responding to police emergency calls in 5.6 minutes or less 90 percent of the time; maintaining files so they can be retrieved within five minutes 85 percent of the time; and getting positive survey results from recreational customers 90 percent of the time.⁴³

If such measurements are rare in most governments, they are the heart of Sunnyvale's intensively monitored system. It is a system that has allowed the city to focus on its goals even in times of declining revenues. The Sunnyvale city manager says that performance-based budgeting allows government to accomplish three things:

• A clear articulation of the purpose for which funds are appropriated, which in turn assures that the funds are being spent for the "right thing."
  

• The enhanced ability to provide oversight and proper management of services.
  

• An improvement in the quality and efficiency of service delivery.

By focusing on outcomes and desirable objectives, policy makers not only give better guidance to those who are managing the programs but they also free themselves from being concerned about issues that are irrelevant to their goals. The city manager told the Commission:

Traditional approaches to governmental budgeting often are criticized for being micro-managed, with policy leaders involved in day-to-day administration, establishing numerous prescriptive rules as to how to get the job done. In the case of Sunnyvale, those deterrents to the provision of economic and quality services have largely been done away with. We have found that when policy leaders know that they are responsible for the clear articulation of the level and quality of services to be provided and when they receive frequent feedback on what was actually accomplished, that they no longer feel the necessity of micromanaging, and as a result responsible management and line personnel have greater freedom to focus on results.⁴⁴

One key aspect of the Sunnyvale system that is atypical of government is that management is on a pay-for-performance system. Managers are evaluated based on their ability to meet the goals set in the budget document. Those who succeed earn bonuses of up to 10 percent while those who fail may face a reduction of 5 percent in their pay (and this sanction has been used, according to the city manager, although managers with a pattern of under-performing usually do not stay with the city). Because Sunnyvale emphasizes achieving specified levels of service more efficiently, managers are rewarded more for meeting the objective at a lower cost than for exceeding the objective at the budgeted amount.

Another aspect of the Sunnyvale system is the constant accountability. Managers receive monthly reports on performance in relationship to goals. The city manager told the Commission that such constant reminders that performance is expected is alien to many people in public service and that some have a tough time adjusting. But many flower in an atmosphere that not only allows them to be creative but also rewards them for innovation and success. Because of the city's commitment to training and shifting employees to new areas rather than laying them off when changes are made, employees and their unions in general are supportive of the system.

Even with a system that is being lauded and studied internationally, Sunnyvale is not content to rest on its laurels. The city manager told the Commission that the city is now evolving into a system that focuses on "high-level outcomes." For instance, instead of measuring the time it takes to respond to police calls, the city is exploring ways to objectively measure activities that lead to reduction in crime.

There seems to be a trend emerging at higher levels of government to focus on overall outcomes as opposed to the details of defining specific services, which are often output based. The City of Sunnyvale, recognizing the value of high-level, but still measurable, outcome statements, is now moving to a major refinement of its historical performance budgeting approach to the development of high-level outcomes. In spite of the success we have enjoyed under the past framework, we are finding new opportunities to better focus scarce resources, as well as to clarify what results government should be attempting to accomplish.⁴⁵

Thus, Sunnyvale is committed to continuing to refine its process and, if necessary, move in new directions -- even after more than 15 years of performance-based budgeting. The city manager said the one lesson the city has learned is that change takes time, but patience as the system evolves is critical.

Performance-based budgeting and management takes time. It is not a quick fix. Its results evolve over time. It should be viewed as a framework and theoretical construct rather than a new fad, and therefore requires perseverance at both the political and executive level so that its full capabilities are allowed to grow and emerge rather than any expectation that it will be judged an instantaneous success or failure.⁴⁶

The need for patience has also been stressed by others with experience in performance-based budgeting, including those who are running the system in a governmental jurisdiction that far surpasses Sunnyvale in size and complexity: Texas.

Texas

U nlike California, which is edging into performance-based budgeting with a largely unstructured pilot project, Texas has whole-heartedly embraced the new system, taking a top-down approach. While Sunnyvale's emphasis is on performance indicators and well-defined unit costs, Texas has concentrated on strategic planning. In 1992, the then-Governor published Texas Tomorrow, laying out the statewide vision, philosophy andmission that would become the foundation for the marching orders to individual state agencies to define their own missions, philosophies, goals, objectives and strategies.

The state's vision, philosophy and mission read:

We envision a Texas where all people have the skills and opportunities they need to achieve their individual dreams; a Texas where people enjoy good health, are safe and secure from harm and share a quality standard of living; a Texas where we and future generations can enjoy our bountiful natural beauty and resources.

Public service is a public trust. As public servants we take pride in the service we provide for our fellow citizens. We will be open, ethical, responsive, accountable and dedicated to the public we serve -- providing legendary customer service. We will foster a working environment free of bias and respectful of the individual. We will operate efficiently and spend the public's money wisely.

The mission of Texas state government is:

• to provide educational opportunities for all its people;

• to protect and enhance the health, well-being and productivity of all Texans;

• to preserve the state's environment, and ensure wise, productive use of the state's natural resources;

• to build a solid foundation for social and economic prosperity; and

• to ensure the safety of our communities.⁴⁷

Each of the mission statements has several goals, each of which is accompanied by a list of key indicators of success. For instance, one of the goals for the mission of building a solid foundation for social and economic prosperity is: "The Texas economy will be diversified and healthy, creating and retaining the jobs needed for a prosperous Texas." Key indicators for success under this goal include: the Texas employment rate, median household income, net number of new jobs created, net number of start-up businesses, and economic diversity as measured by the percentage of jobs outside of the state's five largest industries.

Another goal under the same mission is: "Our communities will be socially prosperous." Key indicators of success include: percentage of Texans with access to affordable housing, home and auto insurance cost as a percentage of median household income, percentage of Texans with access to public transportation, number of civil rights/harassment violations per 10,000 Texans, and percentage of state professional licensee population without documented complaints.⁴⁸

Working within these state-set missions and goals, each Texas department determines a set of objectives and outcome measurements and then puts in place a strategy to meet the measurement levels specified in the budget. For instance,the Texas Commission for the Blind has several objectives, including increasing by 6 percent by 1995 the number of blind people who meet their independent living goals and to increase by 7 percent the number of children who reach their habilitative goals. Related strategies are to provide a statewide program for developing independent living skills and to provide services to all blind and visually impaired children.

A senior budget analyst for Texas told the Commission the system is holding up well as the state moves into the second two-year budget cycle using performance measures. Policy makers who scoffed at the concept that the new system could make a difference are now converts to the process, he said. Among the lessons he identified that Texas has learned and can share with others are:

Indianapolis Parks and Recreation
Indy Parks has three missions: providing quality programs, providing leisure opportunities and developing park resources. Excerpts from their budget:
Budget summary by external outcome
Provide quality programs		$12,160,189
Provide leisure opportunities		6,124,965
Develop park resources		4,733,861
TOTAL for whole department		$23,019,015
Policy Goal: Strong neighborhoods
External Outcome: Provide quality park facilities and areas
The quality of Indianapolis neighborhoods is enhanced by well-built and maintained parks and facilities. A pleasant environment increases the attendance and adds to overall positive experiences.
Regional Attractions		$4,282,691
	Non-tax revenue per $1 expended	.80
	Percent prime-time use	75%
	Percent non-prime-time use	50%
	Percent of surveys showing satisfaction or better	86%
	Percent of critical indicators met	90%
	Number of active partnerships	20
	Percent reduction in vandalism incidents	25%
	Average time in minutes for ranger response	15
Magnet Parks		$4,217,832
Neighborhood Parks		$3,659,666
TOTAL for providing quality programs		$12,160,189
Operating Budget for Providing Quality Parks
	Personal Services	$6,708,491
	Supplies	926,327
	Other	4,381,452
	Capital	143,919
TOTAL for providing quality programs		$12,160,189

• Involve everyone, instilling a sense of ownership through orientation, training, information sharing and active solicitation of ideas. "Everyone" includes elected officials, department heads, managers, line employees, service populations, non-profits, community groups, local units of government, business, labor and the media.
  

• Be flexible and patient. "The job is never done. This is a long-term (10 years?) iterative process centered around continuous learning and continuous improvement."
  

• Create incentives, not just disincentives. Reward genuine participation by departments with additional operational flexibility, public recognition and relief from reporting requirements. In the long run, rewards should include increased appropriations and pay for performance.
  

• Report, regularly and publicly, about the results with easy-to-read formats.
  

• "Think of this as not just another system or a new process, but rather as a revolution in thinking, attitudes and emphasis."⁴⁹

Indianapolis, Palo Alto

E xamples of budgets on this page and the next from Indianapolis and Palo Alto demonstrate again the level of information that a performance-based system provides. The Indianapolis budget -- a document that the mid-western city uses more for public information sharing than for policy decisions -- concentrates on missions, policy goals and external outcomes that demonstrate the goals are being met. Appropriations are delineated according to both mission and operating expenses.

Palo Alto Public Works
Excerpts from the City of Palo Alto's new mission-drive budget for the Department of Public Works:
Department Summary
	Streets	$3,198,697
	Sidewalks	496,047
	Trees	1,520,010
	Structures & Grounds	3,691,203
	Private Development	406,764
	TOTAL EXPENDITURES	$9,312,720
	FULL-TIME POSITIONS	76.88
	TOTAL DEPARTMENT REVENUE	$1,867,276
Major Activity:Streets -- In-House Maintenance
To maximize the useful life of streets through effective repairs
	Salaries, benefits	$366,680
	Non-salaries	230,514
	Allocated expenses	758,519
	TOTAL EXPENDITURES	$1,355,713
	FULL-TIME POSITIONS	6.43
Impact measures
	Repair potholes within 5 working days	85%
	Respond to bicycle path maintenance requests within 5 days	95%
Major Activity: Trees -- In-House Maintenance
To provide for healthy trees and public safety through quality and specialty activities, including ornamental trimming, watering, etc.
	Salaries, benefits	$407,368
	Non-salaries	135,800
	Allocated expenses	225,219
	TOTAL EXPENDITURES	$768,387
	FULL-TIME POSITIONS	8.00
Impact measures
	Annually trim, prune City trees	10%
	Respond to emergency requests within 8 hours	80%
	Contain controllable infestations within 90 days	80%

Palo Alto's new budget, which is an initial, untested system for the San Francisco Bay Area city, divides up departments by function and describes the major activities related to each function. The impact measures set the goals for performance and, once baseline data is established, will demonstrate the trends over time. Other portions of the budget not shown here detail the numbers and type of personnel involved in each activity and provide a summary of the reasons for proposed funding changes.

Benefits and Pitfalls

T he message from those who have studied performance-based budgeting and those who are using it is that it is particularly useful in at least three ways for policy makers:

• It provides data in a format that makes it easier to re-assess priorities as conditions change instead of merely duplicating the past.
  

• It is a system that focuses on available revenue and how to divide those funds rather than on "funding gaps" that are created when allowing automatic program growth causes demands to outstrip expected resources.
  

• It makes performance a key factor in funding decisions and provides a clear statement of program expectations.
  

Converting to a performance-based budgeting system, however, is not easy. Among the potential pitfalls outlined by experts are:

• Delayed gratification: As noted by Commission witnesses from Sunnyvale and Texas, perfecting a performance-based budgeting system takes time and missteps are likely to occur along the way. The department managers in California's pilot project are already aware of the pressures they face to produce results. The Department of Parks and Recreation told the Commission, "Results have been expected on a time frame which is difficult to produce considering the rather monumental undertaking of changing California's budgetary process. It will not be the failure of the pilot, but the failure of government to accept or allow the pilot's success that will be the ultimate barrier to the project."⁵⁰

• Silver-bullet syndrome: Those who believe a new system w