Go to USC home page USC Logo PUBLICATIONS
UNIVERSITY OF SOUTH CAROLINA
COLLEGE OF ARTS AND SCIENCES | IPSPR HOME | PUBLIC POLICY & PRACTICE



HOME

ARCHIVES

CONTACT INFORMATION




USC  THIS SITE
 

he modern world is a highly complex one. It is moreover in a constant state of change. As testimony to these facts, one need only take a glance at the complexities and changes of today's demographics, economies, technologies, and environmental surroundings.

Governments are highly aware of these intricate and mutable realities and are striving, as best they can, to keep in step. Public budgeting is one area in particular that governments are giving attention to in order to respond to a changeable world. To do this, governments are attempting to provide reliable and complete information to budgeters and policy-makers alike so that substantive budget choices can be made.

Governments today are especially trying to ascertain how well public organizations and programs are doing in providing services and products to their citizenry. Governments are asking: "What kind and how many services are we getting from allocated dollars?" "Are these public services of good value?" "Are they making a difference in citizens' lives?"

To answer these questions, and other equally significant ones, governments are developing and implementing "performance-based budgeting" systems. No longer satisfied with traditional budgeting processes, new and, in some cases, renewed interest in linking planning and performance measurement to budgeting is taking hold. Governments are looking beyond inputs or line-item expenditures to make informed decisions, choices that address long-term effects or outcomes, and choices that are grounded in measurable progress or accomplishment.

In this section, performance measurement and budgeting will be discussed. Performance budgeting will be defined and its various uses will be reviewed. Next, the main characteristics or criteria associated with successful performance measurements will be examined as well as its several strategic aspects. Lastly, lessons learned from governments' prior implementation efforts will be scrutinized to determine the "dos and don’ts" of performance-based budgeting.

DEFINITIONS

There is no one single definition of performance-based budgeting (PBB). A review of the literature does, however, suggest what it means commonly. Most observers of -- and experts on -- public budgeting do agree that, generally speaking, PBB is the allocation of funds to achieve programmatic goals and objectives as well as some indication or measurement of work, efficiency, and/or effectiveness. (See Snell and Hayes, November 1993, p. 1; Garsombke and Schrad, February 1999, p. 9; Epstein, 1984 p.2). John Mikesell, for example, states that performance budgets are basically the linking of inputs or costs to program activities and goals.  He states that performance budgets may, and most often do, contain one or more of the following the elements: workload data (units of activity provided), productivity data (cost per activity), and effectiveness information (level of goal achievement) (Mikesell, 1999, pp. 185-186).

Philip Joyce, while acknowledging that "no standard definition" of PBB exists, states that it "involves a sophisticated web of relations, from inputs to outputs, to outcomes… the connecting of resources to results for budgeting purposes” (Joyce, 1999, p. 598). Similarly, Charles Dawson describes performance measurement and budgeting "as general terms applied to systemic efforts to assess government activity and enhance accountability for progress and outcomes in achieving results” (Dawson, 1995, p. 1). 

A 1994 report published by the National Conference of State Legislatures defined PBB in the following way:

Performance budgets use statements of missions, goals and objectives to explain why the money is being spent… .[It is a way to allocate] resources to achieve specific objectives based on program goals and measured results. …Performance budgeting differs from traditional approaches because it focuses on spending results rather than the money spent -- on what the money buys rather than the amount that is made available." (Carter, 1994, pp. 2-3).

As the literature implies, therefore, performance-based budgeting has four primary characteristics.  First, PBB sets a goal, or a set of goals, to which monies are "connected," i.e. allocated. From these goals, specific objectives are delineated and funds are then subdivided among them. Second, PBB provides information and data on past performance and thereby proceeds to allow for meaningful comparisons between "expected" and "actual" progress. Third, adjustments to programs are made either at this point or during a future budget preparation cycle to close any performance gaps that may exist. Fourth, as an ancillary yet important characteristic, PBB provides an opportunity for regular or special (ad hoc) program evaluations. When utilized, these evaluations are valuable in that they give independent and verifiable information to budget decision-makers and program managers alike.

FIGURE 1
Definitions of PBB Terms

INPUT MEASURES. These are the volume of resources used or total expenditures (costs) consumed to achieve a given output.

OUTPUT MEASURES. These are the quantifying of goods and services performed or delivered to customers.

EFFECTIVENESS MEASURES. These are the indices that assess how well a program achieved its goals and objectives; e.g., percent of wetlands preserved as a result of permit issuance; percent of inmates convicted of another crime after release, percent of placements successful after 30 days, etc.

EFFICIENCY MEASURES. These are indices that assess or compare how much output was achieved per unit of input (costs); e.g. cost per complaint processed, cost per license issued, cost per prisoner incarcerated, etc.

WORKLOAD MEASURES. These are indices that assess the level of effort required to carry out an activity; e.g., number of applications processed, number of inspections completed, number of miles patrolled, etc.


SOURCE: See Garsombke, H. and Schrad, Jerry. (February 1999). Performance measurement systems: results from a city and state survey. From Government Finance Review. Chicago, IL: Government Finance Officers Association.

Additionally, it is important to note that many experts in public finance believe that the cardinal aim of PBB is accountability. Performance information and data used in budgeting holds public officials, especially program managers, accountable for service quality, cost-efficiency, and program effectiveness. The focus for PBB is, once again, on results, not simply inputs. Hence, governors, legislators, service or program recipients, and the public generally can determine accountability with a degree of certainty with the use of PBB methods, where this is not possible utilizing traditional or line-item approaches. This ability to assess performance and hold managers accountable serves as a powerful incentive to ratchet up quality or positive service results.

The implications of PBB to stir agency and program officials to meet or exceed performance expectation have, of course, two sides -- one that "rewards" and the other that "penalizes." With regard to "good" or "excellent" performance, rewards can be several. For one thing, agencies or programs might be recompensed with additional or increased funding, or may be allowed to carry "saved" funds, due to efficiencies, forward to the next fiscal year. Agencies or programs could also be absolved from burdensome paperwork requirements, awarded some form of bonus, or perhaps even have agency responsibilities enhanced in some fashion. On the other side, penalties could be handed out to agencies or programs that perform poorly. This might include a reduction in funding or elimination of a program altogether. It might additionally be cause for ordering a management audit or evaluation, transferring program responsibilities to another agency, or the firing of the agency director (Snell and Hayes, 1993, p. 4).

USES

Uses of performance-based budgeting systems can be understood in two ways. One way to make sense of the use of PBB is to speak broadly to the subject matter as it relates to its primary aims. This would include PBB's twin aims of improving decision-making and enhancing service delivery. The other way to understand the use of PBB would be to examine its application or implementation among specific states. This examination would allow for grasp of PBB's use in an actual or "hands-on" way. In this section, a brief discussion of these uses will be presented with the purpose of gaining a more valuable comprehension of performance-based budgeting.

PBB Utilitarian Aims

Governments have pinpointed, according to the literature, two utilitarian aims that PBB readily strives to fulfill. These aims are (1) improved decision-making, and (2) enhanced service delivery (see Epstein, 1984, pp. 6-7; Joyce, 1999, pp. 600-601; Lee and Johnson, 1998, pp. 105-107).

Public budgeting is essentially about making choices. To make better choices, decision-makers need qualitative and complete information and data. PBB can provide this through its various components or devices; e.g., the setting of goals and objectives, the prioritizing of these ends, and the measuring of performance levels (via the indices of efficiency and effectiveness).

Pragmatically, and quite simply, measurement of performance assists government officials to assess "what" and "how well" a program is doing. For instance, what is Program X intended to do? Is Program X achieving these intended ends? Are Program X’s activities or operations cost-efficient? Asking and answering these and other similar questions will permit decision-makers to make wiser, more intelligent program policy and spending determinations.

Consider further, for example, the corresponding improvement of the decision-making process with the addition of differing levels or tiers of performance information and data. Let's assume that a state's mental health agency wants to fund a program to treat individuals with alcohol and drug addiction. The first budget tier of information might speak only to personal services, operating expenses, and assistance payments. A decision-maker (a governor, a legislator, a program manager, etc.) has little to work with here -- just broad expenditure classifications. However, add a second tier of performance information, i.e., "program objectives," and the decision-maker has more meaningful information with which to base and make a judgment. Objectives for an alcohol and addiction program might include, for example, "(1) to increase the number of community programs available for involuntary alcohol and drug services by three by June 30, 2002, and (2) to provide six scheduled treatment hours per patient per day for inpatient alcohol and drug addiction programs."

Additionally, now consider adding a third tier of performance data. With another tier the decision-maker gets a picture of the effectiveness of the program as measured from previous fiscal years. Effectiveness indices show, for example, "that the number of community programs providing involuntary services increased from three in FY 1999 to six (6) in FY 2000." Also, indices for "scheduled treatment hours per patients indicate that these have increased from 4.5 to 5.5 for the same period." Add information as to the "number of voluntary alcohol and drug admissions" and the decision-maker has workload data – a fourth tier. Add information on "cost per patient per day" and a fifth tier of efficiency data is made available. The upshot of these multiple tiers of information obviously becomes clear -- the more tiers of information available, the more informed, intelligent the decisions are going to be.

The other functional aim of PBB is, once again, enhanced service delivery. Performance-based budgeting strives to provide individuals with quality services that meet individual needs in a prompt and complete fashion. It does this through the establishment of specific objectives and measurement indices. In other words, the budgeter or decision-maker establishes performance targets and simultaneously provides for some measuring device to monitor efficiency and effectiveness of service delivery efforts. The intent or design is again enhanced service delivery.

Sometimes so-called "industrial engineering" techniques are employed to improve operations and service delivery. "Quality management" approaches are also used in many organizations to increase service performance. Particularly in the public sector, program evaluations or "performance audits" are a commonly used method to evaluate service delivery and recommend improvements.

Many authors and practitioners of public budgeting especially believe that the focus of a PBB process should center on the notion of “quality” services. The emphasis here is placed on the satisfying of customer (constituent, client, consumer or user) needs. According to the Council of State Governments, 32 states have some form of statewide quality management program in place.

Total Quality Management (TQM) and similar quality schemes are, and have been for some time, a leading philosophy and management process embraced and utilized at all levels of the public sector. TQM and other quality programs concentrate on customer satisfaction, teamwork and employee participation, performance measurement, and open or flexible organizational structures (Young, September 1998, pp. 4-5).

FIGURE 2
The Elements of Quality Management (QM)

QM KEY FACTORS

Top management leadership and support.

Focus on the customer.

Long-term commitment.

Enhanced rewards and recognition.

Commitment to training.

Employee empowerment and teamwork.

Effective and renewed communications.

Application of statistical process analysis.


SOURCE: See Federal Quality Institute. ( May 1991). Introduction to total quality management in the federal government.  Washington, DC: Federal Quality Institute.

PBB State Practices

According to the National Association of State Budgeting Officers (NASBO), states using PBB systems are on the increase. Additionally, governors and legislators, and in some cases both, are finding performance-based budgeting mechanisms useful. Currently, all 50 states, plus Puerto Rico, utilize some form of performance measures. Twenty-five states surveyed by NASBO indicate that they actively use performance measures at some stage in their budgeting and appropriating processes. An additional eight states attest that while they do not use PBB per se, they do use some "public accountability" or "goal /priority building" processes in their funding decisions. Finally, of the 50 states and Puerto Rico using some form of performance measurement, with or without linkages to budgeting practices, 38 maintain that they do regularly monitor performance (NASBO, 1999, p. 48).

FIGURE 3
Performance Measurement Use among the States

State
Performance
Measures
Measurements Monitored
Implications of
Performance
Measurement
State
Performance Measures
Measurements Monitored
Implications of Performance Measurement

Alabama

X

X

P

Nebraska

X

X

P,GP,B

Alaska

X

NA

P, GP

Nevada

X

X

P,GP,B

Arizona

X

X

P, GP, B

New Hampshire

X

 
 

Arkansas

X

X

B

New Jersey

X

Selectively

B,GP

California

X

X

P, GP, B

New Mexico

X

X

B,GP

Colorado

X

X

P, B

New York

X

NA

NA

Connecticut

X

 

P, GP

North Carolina

X

X

P,GP,B

Delaware

X

X

P, B

North Dakota

X

 
 

Florida

X

X

P, GP, B

Ohio

X

X

P,GP,B

Georgia

X

X

P, GP, B

Oklahoma

X

X

P,B

Hawaii

X

X

B

Oregon

X

X

 

Idaho

X

X

P, GP, B

Pennsylvania

X

X

P,GP,B

Illinois

X

X

P, GP

Rhode Island

X

X

P,GP,B

Indiana

X

X

B

South Carolina

X

 

P

Iowa

X

X

P, GP, B

South Dakota

X

 
 

Kansas

X

X

B

Tennessee

X

 

B

Kentucky

X

X

P

Texas

X

X

P,GP,B

Louisiana

X

X

P, GP, B

Utah

X

X

P,GP,B

Maine

X

 
 

Vermont

X

X

--

Maryland

 X

 
 

Virginia

X

X

P,GP

Massachusetts

X

X

B

Washington

X

X

P,GP,B

Michigan

X

 
 

West Virginia

X

X

GP,P

Minnesota

X

X

P, GP, B

Wisconsin

X

 

P,GP,B

Mississippi

X

X

P, GP, B

Wyoming

X

X

P,GP,B

Missouri

X

X

P, GP, B

Puerto Rico

X

X

P,GP,B

Montana

X

X

P

TOTAL

51

39

 

SOURCE: See National Association of State Budgeting Officers. (October 1999). Budget processes in the states.  Washington, DC: National Association of State Budgeting Officers.

Codes:

 P…Public Accountability

GP…Goal/ Priority Building

B…Budgeting Decisions

NA…Not Available

As Figure 3 above shows, each state uses performance measurement in a differing or particular way. This can be ascribed to the different economic, social, cultural, and political aspects of the states. NASBO (1999) describes, for example, Minnesota, Oregon and Texas as having "broad, comprehensive PBB approaches" that have a long history. Oppositely, Virginia and California are relatively new states to the practices of PBB and are more limited in their applications. 

A more recent study of performance measurement systems among the states was published by H. Perrin Garsombke and Jerry Schard in 1999. In this study, 105 questionnaires were sent to state auditors, controllers or treasurers, in early 1998, to inquire as to their "extent of use" and "satisfaction with" performance measurement. According to their survey results based on a 60% response rate (63 questionnaires completed and returned), 67% of state governments used performance measurement systems "to some degree." Eight additional states said that while they did not use performance measurements, they were in the planning stages to do so. The Garsombke and Schard study also found that 45% of the states responding to the survey linked performance measurement, in some way, to budgeting and planning. Thirty-eight percent of these stated specifically that they linked performance measures or incorporated them into their strategic planning processes (Garsombke and Schard, 1999, pp. 9-10).

Another set of survey findings was published in the spring of 2000 by Robert Lee and Robert Burns in the Journal of Public Budgeting and Finance (see Lee and Burns, 2000). In this article, a determination of the extent of "advancement" or "backsliding" of PBB methods and uses was made for the period between 1990 and 1995. One key question of Lee and Burns was: "What changes occurred between 1990 and 1995 in the use of performance measurement in state budgeting?" Another critical question was if there was significant change, "What were the reasons for either advancement or backsliding?"

Overall, the findings were diverse. The survey results found that, between 1990 and 1995, states requiring agencies to submit program effectiveness and efficiency measures when proposing new programs dropped 16%, from 39 states requiring this information in 1990, to 31 in 1995. The survey conversely found that 6% of the states indicated that they had revised their performance measures for the same period. Additionally, effectiveness measures actually utilized in budget documents increased by 8%, from 19 to 23 states, for the same 1990-1995 period. Efficiency measures used in budget documents also increased by 8%, from 15 (in 1990) to 19 (in 1995) states (Lee and Barnes, 2000, pp. 38-32).

As the literature generally suggests, a few states have been specifically recognized as using intensively or well-developed PBB approaches. Among those frequently mentioned are Arizona, Florida, Minnesota, Oregon, Texas, Virginia and North Carolina. 

Arizona, for example, uses a budgeting system that combines strategic planning, performance measurement, and program evaluation. The system, called Program Authorization Reviews (PAR), requires all agencies to submit a one-page overview of its performance measurements for the upcoming fiscal year along with its regular detailed budget request. The recent FY 1998-1999 budget also required an extensive PAR budget submittal from 14 select agencies that included complete performance information and data on 30 programs and subprograms (Freidman, 1997, p. 17).

More specifically, PAR required these 14 agencies to answer four main questions in their budget submittals. One question addressed how programs and their objectives related to their agency mission statements. Another question asked was how efficient and effective programs were in carrying out their activities and in attaining their objectives. The two remaining questions inquired as to how well programs measured up in comparing expected to actual results and, additionally, as to the use of cost-effective alternatives (Freidman, 1997, p.17).

Arizona’s PBB approach has been applauded for not “overloading” its budget document with superfluous performance information and data. Providing decision-makers with a manageable, yet thorough, set of performance data for making good spending choices is a time-consuming and hard won endeavor. Arizona appears to have proven that this can be done.

Another state that has worked diligently to establish a working, useful PBB system is North Carolina. In 1991 North Carolina instituted a statewide agency performance-based budget system. Most recently, its FY 1998-1999 budget request process produced performance measures for more than 3,000 agency or departmental activities. These measures were “outcome focused,” with particular emphasis placed on the “effectiveness” of programs (Freidman., 1997, p. 18).

North Carolina’s PBB system includes all state and federally funded activities. These activities are grouped into ten mammoth programmatic areas that are distinguished mainly by program recipients and analogous program outcomes. These ten broad program categories include, for instance, area designations such as “human services,” “education,” “commerce,” and “justice and public safety.” By using this budget method, decision-makers are forced to make spending choices based on programs – their activities and performance – without relation necessarily to agency jurisdictions or boundaries (Freidman, 1997, p.18.).

North Carolina decision-makers are, therefore, freed – relatively speaking -- from contrived budget decisions due to agency advocacy or similar organizational influences. This is perceived as a plus, since concentration and judgments about budgets are oriented towards programs and results, not by favored or well-liked agency lobbyists and directors. 

CRITERIA

At this juncture, a few brief observations on "choosing" performance measures will be useful. In nearly all cases, performance-based literature speaks to the necessity and priority of establishing "specific criteria" to discriminate between "good" measures and "bad" ones. Based on common sense and practical experience, most observers and experts believe that in order to have a truly successful PBB system, performance measures must meet pre-established, written criteria that are set early in planning stages prior to actual PBB implementation. (See Mikesell, 1999, p. 189; Lee and Johnson, 1998, pp. 103-104; Joyce, 1999, pp. 613-615; Grizzle, 2001, pp. 357-361.)

What is meant by performance criteria? Criteria are a set of standards, guidelines or "yardsticks" by which performance measures are determined to be adequate or satisfactory. These criteria, for example, would include such standards as "relevance," "validity," "clarity," and so on.

Each criterion (of the total set of criteria) would be individually defined and, in turn, this definition would aid the budgeter or decision-maker with the task of determining if each performance measure satisfactorily met the criterion. For instance, if "relevance" is a criterion, these questions would be asked of the performance measure: "Is this information or data constituting the performance measure useful to decision-makers?" "Is there a clear, logical connection between the measure and the program objective?" 

Criteria to determine the sufficiency and adequacy of performance measures have been identified fairly well given the history, relatively speaking, of PBB. All governments, however, decide which set of criteria most benefits their unique circumstances and preferences.  Governments also usually place emphasis on some individual criterion or criteria more than others. Thusly, one state's PBB system may have eight criteria, another a dozen or more. In one state, "validity," clarity," and "quantification" may be much more valued or weighted greater than other criteria, while in another state these may be important but other criteria such as "reliability" and "accuracy" may rank much higher in terms of significance to decision-makers.

FIGURE 4
Performance-based Budgeting Criteria
for Selecting "Good" Measures

CRITERIA

Relevance -- Is this information useful to decision-makers? Is there a clear, logical relationship between the measure and the program objective?

Validity -- Does it accurately gauge what is supposed to be measured?

Significance -- Is it important? Is it worth measuring?

Uniqueness -- Does it contain information not supplied by any other measure?

Clarity -- Will decision-makers understand it? Is it lucid, distinct and intelligible?

Timeliness -- Will the information be collected in time for decision-makers to use it for making policy?

Reliability -- Will the data be collected from the same source in the same way every time?

Quantification -- Can the measure be expressed in the form of a numerical value? Does it measure quantity and/or quality of service provided?

Practicality -- Can the data be collected and processed on a regular basis, at a reasonable cost, without undue strain on staff resources?

Completeness -- Does the measure(s) cover all major elements of the program objective(s)?

Control -- Is the activity or objective being measured within the control, authority or power of the agency?


SOURCE: See State Reorganization Commission. (1984). Program Performance Workshop Manual. Columbia, SC: State Reorganization Commission.

One study of performance measurement criteria consisted of a review of 24 books and articles to determine which criteria were most frequently used and how they were generally defined. The criterion most cited, and presumably most used, was "validity." It was cited 15 times and was defined commonly as a "measure that logically represented the concept or construct to be measured" (Grizzle, 2001, pp. 358-359). 

Other criteria most frequently cited were "clarity" (14), "reliability" (13), "relevance to objectives or decisions" (11), "accuracy" (10), and "sensitivity" (8). The criterion sensitivity was defined as "the distinguishing power of the measurement procedure or operation that is ample enough to capture the change and diversity that occurs in the object, event, or situation being measured" (Grizzle, 2001, pp. 358-359).

This study further identified a total of 22 separate criteria used to determine the sufficiency of performance measures. Of the varying criteria, the study's author organized them into broad categories that captured their group-type similarities. For example, "practicality" was a category used to capture the sense of both "cost" and "ease of data collection." Another category was "utility-user independent" which encompassed the criteria of "comparability," "sensitivity," and "clarity." Alternatively, the category of "utility-user dependent" included "relevance," "timeliness," and "controllability" (Grizzle, 2001, pp. 358-359).

STRATEGIC ASPECTS          

What is strategic planning? Drawing on Olsen and Edie (1982), I define strategic planning as a disciplined effort to produce fundamental decisions and actions that shape and guide what an organization is, what it does, and why it does it. To deliver the best results, strategic planning requires broad yet effective information gathering, development and exploration of strategic alternatives, and an emphasis on future implications of present decisions. Strategic planning can help facilitate communication and participation, accommodate divergent interests and values, foster wise and reasonably analytic decision making, and promote successful implementation. In short, at its best strategic planning can prompt in organizations the kind of imagination--and commitment--that psychotherapist and theologian Thomas More thinks are necessary to deal with individuals' life conundrums.

John M. Bryson, 1995

Performance-based budgeting, when linked to strategic planning methodologies, is a powerful and advantageous decision-making tool. Today, many states are utilizing PBB systems along with strategic planning, recognizing that the two systems taken or applied together are a logical and practical fit. In fact, according to the survey conducted in 1998 by Garsombke and Schrad, 38% of states using PBB systems were also combining such efforts with formal strategic planning practices (Garsombke and Schrad, 1999, p. 10.)

Strategic planning is a process of developing a long-term plan to guide an organization, for example, a state agency, department or commission, towards a clearly articulated mission, goals and objectives.  It is a process of assessing where an organization is presently, ascertaining the challenges and opportunities that present themselves, and determining what destination is most desirable and how to get there. PBB adds or emphasizes the critical, additional step of measuring progress (see Southern Growth Policies Board, 1996, pp. 6-7).

A review of recent literature suggests that states could benefit from the strategic planning process mainly for the reason that the development of multi-year policy plans could link present situations or circumstances with a more meaningful vision of the future. In other words, a strategic planning process would enable, let's say, the governor and the legislature, to understand more clearly where their state is now and where they would like it to be in the future. Basically, a strategic plan would indicate to state leaders -- more lucidly -- what is state government’s (or more particularly an agency's) overall mission, its goals and objectives, its strategic or programmatic activities, and its resources (people, monies, technologies, facilities, etc.). This process would further allow state officials to have a solid grasp of the state’s on-going performance and what results are actually being achieved (Young, 1998, p. 13). More specifically, the benefits of a statewide strategic planning process would be:

  • The establishment of a long-range, unified and broad direction – a “plan” – for state government in the policy areas of education, health and human services, transportation, public safety, commerce, natural resources, and criminal justice.
  • The facilitation of the governor and legislature in being more responsive and accountable to the current and emerging needs of their state.
  • The allocation of limited resources, via the state’s budgetary process, in a more rational, and “results-producing” way.
  • The improvement of communication among all state leaders and better coordination of the “omnibus” policy/fiscal decision-making process.
  • The measurement of the progress of statewide strategic efforts, by all planning participants, and the updating or revision of these efforts as warranted (Young, 1998, p.13).

What are the key elements or steps of the strategic planning process? Strategic planning is simply a formal yet flexible process to determine where an organization is currently and where it should be in the future. There is agreement, as evidenced in recent literature, in both theory and practice, on the general steps that are involved in a strategic planning process. By and large, there are six steps that can be summarized as follows:

  1. An “environmental scan” or a situational analysis of the strengths and weaknesses of one’s organization, including an analysis of external threats and opportunities;
  2. The formation of or the “putting into words” of a vision for the future and an accompanying mission statement which defines the fundamental purpose of an organization, its values, and its boundaries;
  3. The development of general goals, specific targets or objectives, and performance measurements to gauge organizational progress;
  4. A set of strategies to indicate what will be done to accomplish its goals and objectives;
  5. The implementation of detailed operational or tactical plans that provide for staff assignments and schedules; and finally,
  6. An evaluation component to monitor and revise the overall strategic approach as it unfolds (Southern Growth Policies Board, 1996, pp. 6-7).

The Council of State Governments published a paper, in 1997, examining state trends and models of state strategic planning and "benchmarking" (see Council of State Governments, April 1997). Several statewide planning initiatives were highlighted including those in Utah, Oregon, Minnesota, Florida, Texas, Connecticut, Pennsylvania, Kentucky and Michigan. The Council found that each state’s strategic planning process contained unique characteristics. Most states did, however, attempt to set into place the key steps that constitute, by generally accepted practices, a strategic plan.

FIGURE 5
Strategic Planning Process Model

Where Are We Now?
Internal/External Assessment
Mission and Principles
  • Situation Inventory/ Environmental Scan
  • Customer Analysis
  • Broad, Comprehensive Statement of the Organization’s Purpose
  • Core Values, Actions to Achieve Mission
Where Do We Want To Be?
Vision Goals and Objectives
  • Compelling Image of Desired Future
  • The Desired Result After 3 or More YearsSpecific & Measurable Targets for Accomplishment
How Do We Get There?
Action Plans
  • Strategies Used to Accomplish Goals and
  • Objectives Detailed Work Plans
How Do We Measure
Our Progress?
Performance Measures Monitoring
and Tracking
  • Methods to Measure Results
  • Systems to Monitor Progress & Compile Management Information

SOURCE: See Office of Strategic Planning and Budgeting. (1995). Managing for results: strategic planning and performance measurement handbook. Phoenix, AR: Office of Strategic Planning and Budgeting.

Two states that are particularly noted for their long-run and exceptional efforts at linking PBB with strategic planning methods are Oregon and Minnesota. A few comments on these two state's experiences will be worthwhile.

Oregon is recognized, arguably, as the most sophisticated or highly evolved state in terms of model strategic planning and PBB initiatives. Called "Oregon Benchmarks" –   and alternately "Oregon Shines" – the model system was introduced in 1989 when over a hundred citizens and policy-makers came together to develop a multi-year strategic plan for the state.  The state legislature also created that year the Oregon Progress Board to maintain, revise and oversee the implementation of the comprehensive state's strategic plan "well into the twenty-first century" (retrieved January 3, 2003 from http://www.econ.state.or.us/opb/jttestim.htm).

In 1991, with plentiful input from all levels of government and the people of Oregon, the Progress Board adopted 158 indices or "benchmarks" that they considered of the greatest priority to the progress of the state. These measures were oriented to performance and not effort. The Progress Board was interested, for example, not in measuring or monitoring school expenditures to assess school performance, but rather, in measuring student achievement as predicated on standardized testing (retrieved January 3, 2003 from http://www.econ.state.or.us/opb/jttestim.htm).

In 1994, the Progress Board implemented a program to facilitate performance by restructuring many of the state's intergovernmental and programmatic relationships. For instance, it managed to relax federal guidelines and restrictions to implement more efficiently and effectively programs dealing with child services, disabled employees, wildlife preservation, juvenile justice, and welfare recipients. As of 1997, 32 agencies were participating in the Progress Board's "restructuring" program (retrieved January 3, 2003 from http://www.econ.state.or.us/opb/jttestim.htm).

In 1997, Oregon's legislature mandated that the Progress Board's strategic planning/PBB process be a permanent fixture of the state's government. The law required that the Progress Board report to the state legislature as to the general status of efforts in strategic planning and PBB among Oregon's agencies. A detailed and "complete update" of Oregon Benchmarks is to be completed and reported to the legislature every six years (retrieved January 3, 2003 from http://www.econ.state.or.us/opb/jttestim.htm).

Today, agencies in Oregon's state government are required to develop "results-oriented" performance measures that are tied directly to both agency strategic plans and budgets. Input is encouraged not only from internal agency personnel but also from other state agencies, elected officials, service delivery clients, interest groups, and the public at-large. Participants and observers alike believe that this input is invaluable to the planning and budgeting process and ultimately reflects the values and priorities of all Oregonians  (retrieved January 3, 2003 from http://www.econ.state.or.us/opb/jttestim.htm).

Minnesota is another state that has received much attention with regard to its strategic planning and performance measurement efforts. "Minnesota Milestones" was begun in 1991 as a planning/performance measurement system that established some 20 goals and 79 milestones to measure progress. Termed a "citizen-based planning process," literally thousands of Minnesotans came together in a series of meetings across the state, and through participation in mail-in surveys, to contribute to the first Minnesota Milestones document. The planning document, containing dozens of measurement indicators, was extended to a 30-year time frame with designated yearly milestones to indicate progress. According to the Minnesota Planning Division, the planning document was and continues to be a centerpiece, for developing the state's budget recommendations (retrieved January 3, 2003 from http://www.mnplan.state.mn.us/pdf/mm98-2.pdf).

In 1997 and 1998, the public came together once again to review, update and adjust the Minnesota Milestones' master plan. The result was the determination of 19 major goals and 70 measurement indicators. According to Minnesota's planning division, from 1990 to 1998, the state has achieved marked success on seven goals, relapsed slightly on two, and had "mixed" results on five goals  (retrieved January 3, 2003 from http://www.mnplan.state.mn.us/pdf/mm98-2.pdf).

FIGURE 6
Assessing Progress of the 1990's
Goal Categorical AreasThe State of Minnesota

Goals showing progress

Goals with stable or mixed results

Academic achievement

Safe, caring communities

Health

Decent, affordable housing

Support for independent living

Viable rural and urban economies

Sustainable economic growth

Conservation of natural resources

Skilled work force

Healthy ecosystems

Standard of living

 

Outdoor recreation

Goals with inadequate 1990s data

 

Child poverty

Goals showing decline

School readiness

Stable, supportive families

Inclusive communities

Participation in democracy

Responsive, efficient government

 

Quality of the environment

SOURCE: See http://www.mnplan.state.mn.us/pdf/mm98-2.pdf.

And what has Minnesota learned from its decade-long effort at implementing a strategic planning and performance-based budget system? The Minnesota Planning Office says that on several fronts it is still too early to tell. What they do know, however, is two things with certainty. One is that forming a mission or vision with broad public participation works for them. Policy experts, public administrators, public officials and the citizenry have worked well together and have been able to reach agreement on milestones. On the down side, the Planning Office admits that some legislators are not actively involved in using the planning/measurement process. Though some legislators were involved in the planning series of meetings, other legislators are hesitant or simply lax in using the Minnesota Milestones in budget decision-making. Many believe that this hesitancy is based on partisan politics. Nevertheless, many involved in the budget process are optimistic that political wrangling can be minimized in certain situations where good performance data are available, and cannot be ignored  (retrieved January 3, 2003 from http://www.mnplan.state.mn.us/pdf/mm98-2.pdf).

LESSONS--DOS AND DON'TS

As discussed earlier, performance-based budgeting has been around for some time. Indeed, the late 1980s and early 1990s saw a renewed interest in PBB processes at all levels of government. The experiences of cities, states and the federal government have resulted in an amassed compilation of literature on lessons learned from devising and implementing PBB systems, the "dos and don'ts" so to speak. In this short analysis, a few of these lessons are conveyed for those who wish to understand or perhaps profit from these past hard-won trials and experiences.

The National Performance Review

In 1993, President Clinton and Vice President Gore began a government-wide review process to overhaul federal government administrative and management practices. The "National Performance Review" (NPR) was aimed at "reinventing government" as based on the premises discussed in the much celebrated David Osborne and Ted Gaebler book published in 1992 (see Osborne and Gaebler, 1992). NPR hoped to foster the Osborne and Gaebler reinvention themes of cooperative and systemic benchmarking, identifying best practices, adapting them, and replicating them where and when possible.

Four years later, in early 1997, NPR published a study on best practices documenting problems with the bureaucracy and how to go about fixing them. In this comprehensive study, entitled "Benchmarking: Best Practices in Customer Driven Strategic Planning," was a mini-report on best practices in performance measurement. This mini-report summarized best practices across the U.S., discussed a model methodology for establishing and using performance measurements, and offered strategies on how to implement a successful PBB process. (See  http://www.npr.gov/library/papers/benchmkr/nprbook.html).

NPR identified nine findings -- "lessons learned" -- that were critical in its study of performance measurement. The following summary touches upon each of these findings.

  1. One finding or lesson is that "leadership" is of the utmost importance in developing and implementing PBB systems. Agency heads, senior managers, line managers, supervisors, and other management personnel, including commission or board members if applicable, should be actively involved in PBB efforts. Agency leadership should be clearly present at each stage of developing and executing the organization's vision, mission, objectives, strategies, and performance indices.
  2. An organized, comprehensible "framework" for PBB should be established, according to NPR. As such, governmental entities should put together a written, easily understood document that shows how the PBB process works and a calendar that clearly states when milestones, target dates and so forth, should occur. This framework should link logically with missions, goals and objectives.
  3. NPR found that those governmental entities which foster communications, both externally and internally, are the most successful ones. Agency and departmental employees should regularly and frequently communicate with one another regardless of specific duties and responsibilities within the organization. Agencies should also take proactive steps to communicate with stakeholders and agency clients -- inquiring as to needs, satisfaction with services, etc.
  4. It was found that accountability is a primary factor in PBB initiatives. Government employees must not only be held accountable for "actions" but also for "results." PBB is crucial for determining accountability.
  5. PBB, it was discovered, is about "intelligence gathering," not simply data collection. PBB indices must be substantive, informative and meaningful to all participants and stakeholders.
  6. NPR found additionally that rewards, whether in the form of compensation or recognition, should be related to performance. If employees are held accountable for achieving certain measures, successful attainment should result clearly or unambiguously in the receipt of any rewards.
  7. Similarly, and without contradiction, NPR discovered that PBB systems should not be punitive in nature. Non-attainment of agency performance goals and objectives should be constructively reviewed, problems should be identified, adjustments should be made, and re-implementation should be instituted, as necessary.
  8. Progress and the accomplishment of results should especially be communicated to all participants and to the public at-large. Governmental entities should report the outcomes and advances made through all media, including the Internet.
  9. The final general finding of the NPR study is that PBB systems are not a panacea. PBB systems are useful to managers and decision-makers for accumulating the best information and data possible, accurately measuring performance, and readjusting and fine-tuning programs and agency operations and outcomes. PBB processes are not a cure-all to make all things happen in either a precise way or a special elixir to achieve complete and flawless results. PBB is a process that requires continuous upkeep and adaptation. It will never deliver perfection, only improvements based on the efforts of all participants (retrieved January 2, 2003 from http://www.npr.gov/library/papers/benchmkr/nprbook.html).
FIGURE 7
Why is Performance Budgeting Important?

PROVIDES ACCOUNTABILITY TO THE PUBLIC.
In the public sector, resources are borrowed from the shareholders. As stewards of the resources ,governments are required to deliver some product or result.

DRIVES REDESIGN OF PROGRAM (FOCUSES ON IMPROVEMENT).
Performance supported budgeting can be a driving force in the redesign of programs and the driving force in integration within agencies, and across agencies. If it is focused on improving, there will be a more effective overall plan.
HELPS RATIONALIZE BUDGET ALLOCATIONS (USES PERFORMANCE INFORMATION
AS A BASIS OF EVIDENCE).

Performance budgeting ensures that performance information is part of the budget and resource allocation debate. Performance information needs to be, in some manner, part of the resource allocation.
IMPROVES UNDERSTANDING OF CROSSCUTTING PROGRAMS IN GOVERNMENT.
It is possible to better understand total costs and benefits of comparative crosscutting programs if they can indeed frame them to be understandable to each other, outsiders, and stakeholders.
HELPS AGENCIES LINK THEIR DAILY ACTIVITIES TO OVERALL GOVERNMENT OUTCOMES AND SIMILAR ACTIVITIES OF OTHER AGENCIES.
Tracking costs and performance-based budgets against goals helps to understand roles in achieving government goals. Process outputs need to be identified and measured at each step, then tracked against the outputs. Also, each agency has to reassess its outcomes.
HELPS ALIGN GOVERNMENT SPENDING WITH OVERALL GOALS.
There needs to be an assurance that the resources provided by the public are spent sensibly for public purposes. Performance has to be an integral part of that equation of getting it done right.
COMPARES COST EFFECTIVENESS BETWEEN PROGRAMS.
More can be achieved if we know which activities are most effective. It is a cost-effectiveness argument where there are similar measures for different programs.

 
SOURCE: See  http://www.performance.napawash.org/mainfiles/TTbudgets.htm.

American Society for Public Administration

Kathryn Newcomer and Sharon Caudle presented a research paper, in 1999, at the American Society for Public Administration's (ASPA) 60th National Conference in Orlando, Florida. The paper dealt with the subject of support ("technical" and "cultural") for performance-based measurement and management systems and, additionally, summarized lessons learned as relates to "what is currently happening at various levels of government in the United States" (see Newcomer and Caudle, 1999, pp. 1-15). 

Authors Newcomer and Caudle identified ten lessons learned about performance-based measurement and management systems. Some of these lessons included those found by NPR such as do communicate to all participants and stakeholders, do provide incentives or rewards for using PBB processes correctly, and do ensure committed leadership at all phases of PBB. Other lessons disclosed in the ASPA presentation, and not addressed fully or only in a limited fashion in the NPR study, include:

  • Make strategic planning the PBB point of departure. Strategic planning embraces all the requisite mechanisms that are essential to initiating successfully a PBB approach -- vision and mission statements, objectives, environmental scans, action strategies, etc.
  • Form genuine "partnerships" with stakeholders. To make PBB work, partnering with customers and other stakeholders is important. Stakeholders, therefore, must be real participants and not given just lip service. PBB cannot be viewed as an "insiders' game," it must be an open process where program recipients or clients, interest groups, and the public generally participate, in some way, in goal setting, quality measurement, and improvement strategies.
  • Determine responsibilities, especially if these are split among differing levels of government. If responsibilities for performance activities and results are divided among federal, state and local authorities, sort out early in the PBB developmental stages who will be accountable for what. If changes in activities occur during implementation, adjust responsibilities accordingly.
  • Focus on the "right" performance objectives and measures. PBB systems should focus on only vital goals, targets, objectives, and measures. Too much or superfluous information and data will "overload" the system and ultimately "turn off" decision-makers.
  • Think long-term. First, PBB systems take time to implement. A minimum of three to four years should be considered to get a system up and running reasonably well. Second, performance goals and objectives cannot always be achieved overnight. Depending on the complexity of the organizational or agency program, its clientele, funding levels and so on, it may take several years to reach the goals envisioned. Long-term and strategic planning methods will enhance an agency's or department's chances on achieving future successes.
  • Use a centrally directed approach and provide sufficient resources. PBB systems are most successful when they are coordinated and assisted by a government's central administrative unit. This approach lends itself to consistency, coherence and uniformity of PBB processes among several and variant governmental entities. It also facilitates in providing on-demand, or "timely," technical expertise or assistance to individual agencies at critical stages during PBB implementation. Adequate resources (sufficient staff, equipment, and funds) are essential to PBB success, second only perhaps to the requirement of "good and sustained leadership" (Newcomer and Caudle, 1999, pp. 1-15).

National Conference of State Legislatures

Though published in 1993, Dianna Gordon's short article for the National Conference of State Legislature's (NCSL) magazine, State Legislatures, is recognized among many state budget practitioners as an excellent overview of both the dos and don'ts of budget reform. (Gordon, October 1993, p. 17). Based on interviews with the top budget officials from the states of Arizona, Texas, and Mississippi, Gordon deciphered what appeared to be working as well as what seemed not to be working for these three states as they reformed their respective budget systems in the early 1990s.

The "dos" are relegated to nine brief observations, lessons or statements by Gordon. For example, states should adopt realistic expectations when introducing budget reforms. Some states expect too much too soon, and are quickly disappointed when things fail to meet anticipated outcomes. Also, Gordon culled from the three states that it is imperative to study other state experiences in undertaking PBB and budget reform. Such a review will have all the usual benefits, i.e., save money, time and effort. Additionally, using a pilot approach to budget reform will provide a testing ground for all aspects of the budget reform initiative. It is a more manageable and reliable way to get sophisticated PBB systems, for example, off to a good start. Other dos include: Consider the application of new reform ideas or concepts on an individual basis; provide training related to budget reform to all participants, especially lawmakers; and be prepared to develop more reliable data and information as the budget reform effort evolves (Gordon, 1993, p.17).

The "don'ts" stated by Gordon are, in their own fashion, sound advice of what to avoid or not do. For instance, states don't want to repeat the blunders, errors and oversights that other states have made. "To err is human, but to repeat the mistakes of other states is foolish and wasteful." Moreover Gordon says don't go "overboard" with complex and "impressive sounding" reform schemes. Focus rather on budget reforms, such as PBB and its various components, which seem right for the needs and wants of your state's particular set of circumstances and players. "Things that work in some other states may not be right for your state." Also, don't give up on budget reform mechanisms that are working. If they work, then "build on them." States such as Texas and Arizona agree that persistence with and expansion of those things that work pays off in the short and long runs. And finally, Gordon re-emphasizes a recurrent admonition among budget reform practitioners: avoid or don't attempt a "massive overhaul" of the current system. Stalinist-type schemes and directives to reform complete budget systems within a given timeframe are doomed to failure. Realistic and well-planned approaches to budget reform are a more sensible and ultimately workable design for governments (Gordon, 1993, p.17).

The California Study

In 1998, the California Legislative Analyst's Office (LAO) published a report on the state's performance budgeting pilot project. California's governor initiated the pilot project among select agencies in January 1993 after he had declared the state's budget process as "dysfunctional." The legislature followed the governor's remarks and actions by establishing in law a PBB system and providing for periodic review. Hence, the LAO study comes five years after the beginning of the pilot project. (Cornett, 1998, pp. 1-4).

The LAO findings speak to the lessons learned from 1993 through 1998. Many of the findings are duplicative of the NPR, ASPA, and NCSL conclusions. These include, by way of illustration, the previous findings "that focusing on missions, goals and objectives and linking these to performance measurement is a time-consuming and difficult task that requires ample agency resources." The LAO, however, did emphasize some conclusions not fully touched upon or discussed in the other studies summarized above. For example, California's pilot agencies have observed a heightened sense of enthusiasm among their employees. This is due, according to the LAO, to their ability to clearly recognize what they are aiming for in terms of program goals and objectives, and their new ability to gauge their performance in achieving them. Pilot participants have likewise welcomed the new managerial flexibility resulting from relaxed administrative regulations. Employees are able now to administer their programs with greater ease and concentrate on programmatic results rather than excessive, redundant and tedious compliance requirements (Cornett, 1998).

The LAO also found that, on the downside, agencies have not been able to "redirect savings" due to PBB techniques and implementation. The pilot project designers had anticipated that any savings due to PBB analyses would be shifted to other agency needs. It is not clear that this has, in fact, occurred. Additionally, the LAO concluded that that – overall – many remnants of the traditional budgeting approach have remained despite the best efforts to get decision-makers to use PBB-generated information and data. The LAO believes that this will require more time and some targeted training to special users (Cornett, 1998).

The North Dakota Study

The North Dakota Legislative Council compiled a listing of key lessons learned for presentation to that state’s Budget Committee on Government Finance. Presented in June of 1998, the list identifies a number of findings other states have discovered about using PBB systems.

The Legislative Council, for instance, discovered that the fundamental or primary aim of performance-based budgeting is "not simply to measure but to improve agency services." The Council found that many states and their organizational units strive to put measures in place and that this becomes their focus rather than actually improving program activities and results. In other words, some states have become so caught up in the conceptual framework, methods, and technicalities of PBB that the process itself, and its "machinations," take priority over producing results or improving services (North Dakota Legislative Council, 1998).

The Council also found that a few states have published their performance measures before anyone had been instructed on how to use them. This can be catastrophic and bring the PBB process to a grinding halt in terms of meaningful usage among decision-makers. States must give attention when planning and implementing PBB systems to the "educational process" that is required to bring about maximum utilization of performance measures (North Dakota Legislative Council, 1998).

Reward and penalty clauses establishing PBB systems may cause the inaccurate reporting of performance or progress, according to the Legislative Council. This is so because state agencies are tempted to present performance in a favorable light to avoid punishment. Agencies can do this in several ways including publishing meaningless workload data, "low-balling" performance targets or objectives, or presenting data and information that may be "interesting" but speaks little to efficiency or effectiveness (North Dakota Legislative Council, 1998.). Given these and other findings, the North Dakota Legislative Council recommends seven important considerations for those states contemplating the use of PBB:

  1. Confine measures to only the most significant ones.
  2. Tie measures, to the fullest extent possible, to the budget directly.
  3. Develop baseline data and information for comparison purposes when assessing data reported.
  4. Assist agencies with mediocre or poor performance rather than punish them.
  5. Ensure that measures are understandable to everyone involved, especially the governor, legislators and the public.
  6. Provide training to all participants in the PBB process.
  7. Review the correctness and accuracy of the measures on a regular basis (North Dakota Legislative Council, 1998).

REFERENCES

Bryson, J. M. (1995). Strategic planning for public and nonprofit organizations: A guide to strengthening and sustaining organizational achievement. San Francisco, CA: Jossey-Bass Publishers.

Carter, K. (1994). The performance budget revisited: A report on state budget reform. Legislative Finance Paper No. 91. Denver, CO: National Conference of State Legislatures.

Cornett, C. (1998). California's performance budgeting pilot project: A view from the legislature. Cal-Tax Digest, February.

Council of State Governments. (1997l). Managing for success: A profile of state government for the 21st century. Lexington, KY: Council of State Governments.

Dawson, C. S. (1995). Performance measurement and budgeting: Relearning old truths.  Albany, NY: Legislative Commission on Government Administration.

Epstein, P. D. (1984). Using performance measurement in local government: A guide to improving decisions, performance, and accountability. New York, NY: Van Nostrand Reinhold Company, Inc.

Freidman, M. (1997). A guide to developing and using performance measures in results-based budgeting. Baltimore, MD: Fiscal Studies Institute.

Garsombke, H. P. & Schrad, J. (1999). Performance measurement systems: Results from a city and state survey. Government Finance Review, February.

General Accounting Office. (1997). Performance budgeting: Past initiatives offer insights for GPRA implementation. Washington, DC: General Accounting Office.

Gordon, D. (1993). The dos and don’ts of budget reform. State Legislatures, October.

Grizzle, G. A. (2001). Performance measures for budget justifications: Developing a selection strategy."  In.G. J. Miller, W. B. Hildreth, & J. Rabin (Eds.),. Performance-based budgeting Boulder, CO: Westview Press.

Joyce, P. G. (1999). Performance-based budgeting. In R. T. Meyers (Ed.), Handbook of government budgeting. San Francisco, CA: Jossey-Bass Publishers.

Lee Jr., R.D. and Burns, R. C. (2000). Performance measurement in state budgeting: Advancement and backsliding from 1990 to 1995. Public Budgeting and Finance, Spring.

Lee Jr., R. D. and Johnson, R. W. (1998). Public budgeting systems. (6th ed.). Gaithersburg, MD: Aspen Publishers, Inc.

Mikesell, J. (1999). Fiscal administration: Analysis and applications for the public sector. (5th ed.). Fort Worth, TX: Harcourt Brace College Publishers.

Miller, G. J., Hildreth, W. B., and Rabin, J. (2001). Performance-based budgeting. Boulder, CO: Westview Press.

National Association of State Budgeting Officers. (1999). Budget processes in the states. Washington, DC: Author.

National Conference of State Legislatures. (1999). Legislative budget procedures: A guide to appropriations and budget processes in the states, commonwealths and territories. Denver, CO: Author.

Newcomer, K. & Caudle, S. (1999, April). Structural support for performance-based management: Cracks in the foundation. Paper presented at the American Society for Public Administration 60th National Conference, Orlando, FL.

North Dakota Legislative Council. (1998). Performance budget information. Bismarck, ND: Author.

Osborne, D. & Gaebler, T. (1993). Reinventing government: How the entrepreneurial spirit is transforming the public sector. New York: Dutton/Plume.

Southern Growth Policies Board. (1996). Results-oriented government: A guide to strategic planning and performance measurement in the public sector. Research Triangle, NC: Author.

Snell, R. & Hayes, K. (1993, November). Performance budgeting and the states. Presented to the Nebraska Legislative Appropriations Committee, Lincoln, NB.

Young, R. D. (1998). A statewide strategic planning process: The need for the state's leadership in South Carolina to plan ahead. Unpublished manuscript. University of South Carolina.

ABOUT THE AUTHOR

Richard D. Young has been a senior research associate with the Institute for Public Service and Policy Research at the University of South Carolina since 1998. He conducts research on a myriad of public policy and public administration topics relating to state and local governments. Mr. Young previously worked with the Senate of South Carolina and the State Reorganization Commission in various positions of research. Prior to this, Mr. Young taught at the University of Louisville, Hanover College, Indiana University Southeast, and the University of Kentucky Campus in Louisville, Kentucky. Mr. Young has a B.A. (1973) and M.A. (1975) from the University of Louisville. Mr. Young has written and published several papers and reports on public policy issues and public management theory, including A Brief Guide to State Government in South Carolina (1999), A Guide to the General Assembly of South Carolina (2000), and  Perspectives on Public Budgeting: Budgets, Reforms, Performance-based Systems, Politics and Selected State Experiences (2001). Mr. Young has recently published State Reorganization in South Carolina: Theories, History, Practices, and Further Implications (2002). Along with Dr. Luther F. Carter, President of Francis Marion University, he is author of The Governor: Powers, Practices, Roles and the South Carolina Experience, scheduled for publication in 2003.  Mr. Young heads up the Institute’s Project for Excellence in Government, a project to foster and promote scholarly research on Baldrige concepts and criteria as they relate to state and local governments. Mr. Young can be contacted at YOUNG-RICHARD@sc.edu .


CONTACT:

Richard D. Young, Editor in Chief
Public Policy & Practice
Institute for Public Service and
Policy Research
University of South Carolina
Columbia, SC 29208
Phone: (803) 777-0453
Fax: (803) 777-4575
e-mail: young-richard@sc.edu
College of Liberal Arts: Learning that lasts a lifetime
RETURN TO TOP
USC LINKS: DIRECTORY MAP EVENTS VIP
SITE INFORMATION