Performance Budgeting: Initial Agency Experiences Provide a Foundation to Assess Future Directions

Published by the Government Accountability Office on 1999-07-01.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                          United States General Accounting Office

GAO                       Testimony
                          Before the Subcommittee  on Government Management,
                          Information and Technology, Committee on Government
                          Reform, House of Representatives

For Release on Delivery
            Expected       atPERFORMANCE
2 p.m.
July 1,1999               BUDGETING
                          Initial Agency Experiences
                          Provide a Foundation to
                          Assess Future Directions
                          Statement of Paul L. Posner
                          Director, Budget Issues
                          Accounting and Information Management Division

                          Statement of Christopher J. Mihm
                          Associate Director, Federal Management
                          and Workforce Issues
                          General Government Division

                                    Accountability   * Integrity * Reliability

                                    JT'~ 71 '-
                   Mr. Chairman and Members of the Subcommittee:

                   It is a pleasure to be here today to discuss how we can advance
                   performance budgeting in the federal government. As you requested, I will
                   discuss the postponement of the performance budgeting pilots that are
                   required by the Government Performance and Results Act. I will also
                   discuss some of the challenges that confront these pilots and any effort to
                   more closely relate performance expectations and spending estimates.
                   Despite these challenges, our recent and ongoing reviews of agencies'
                   performance plans indicate that federal agencies are developing
                   approaches called for by the Results Act to link performance plans and
                   budget requests. These agency efforts deserve close attention and support,
                   not only because of their contribution to the overall implementation of the
                   act, but also because of their potential to inform our understanding of and
                   expectations for performance budgeting within the federal government.
                   But first, to set context for this discussion, I'd like to begin by briefly
                   looking at how the concept and practice of performance budgeting has
                   evolved in the federal government.

The Evolution of   The concept of performance budgeting-essentially the process of linking
                   budget levels to expected results, rather than to inputs or activities-has
                   and will likely continue to evolve. Historically, within the federal
Budgeting          government, performance budgeting has progressed from a relatively
                   straightforward efficiency concept, as evidenced in recommendations from
                   the first Hoover Commission, to the complex and mechanistic processes
                   associated with such initiatives as the Planning-Programming-Budgeting
                   System (PPBS) and Zero-Base Budgeting (ZBB).1 Similarly, foreign
                   countries and state and local governments in this country are
                   experimenting with a variety of approaches to more closely associate
                   expected performance with requested funding levels, as part of their
                   broader reform efforts to become more results-oriented. 2 These
                   governments recognize that focusing on results involves defining clear
                   missions and outcomes, measuring performance to gauge progress, and

                    Performance Budgeting: Past Initiatives Offer Insights for GPRA Implementation
                   (GAO/AIMD-9746, March 27, 1997).
                     See, for example, Budgeting for Results: Perspectives on Public Expenditure Management. prepared
                   by the Organisation for Economic Co-operation and Development, 1995, and "The State of the States:
                   Performance-Based Budgeting Requirements in 47 Out of 50," by Julia Melkers and Katherine
                   Willoughby, Public Administration Review (January/February 1998, Vol. 58, No. 1).

                   Page 1                                                                   GAOfr-AIMD/GGD-99-216
using performance information within decision processes. Performance
budgeting is the general term used to refer to the infusion of performance
information into resource allocation processes.

We have looked at the history of performance budgeting in the federal
government and the experiences of state governments and believe that two
general themes are suggested. 3

* First, although the process of budgeting is inherently an exercise of
  political choice in which performance information can be one but not
  the only factor underlying ultimate decisions, many governments have
  recognized that systematic presentation of performance information
  alongside budget amounts will improve budget decision-making. In fact,
  the Results Act is based on a premise that budget decisions should be
  more clearly informed by performance.
* Second, no single definition of, or common approach to, performance
  budgeting can encompass the range of needs or interests of
  decisionmakers, or the variety of political institutions and
  organizational arrangements of modern governments. Thus
  performance budgeting is best seen as a process of adaptation rather
  than as an adoption of a specific process.

In its overall structure, focus, and approach, the Results Act incorporates
important lessons from previous federal efforts to connect plans with
budgets. For example, past initiatives-such as ZBB in 1977-typically
devised unique and often voluminous presentation formats unconnected to
the structures used in congressional budget presentations. The Results Act
requires an agency's annual performance plan to link directly to the
presentation structures ("program activities" 4 ) used in the President's
budget submission for that agency. Also, past performance budgeting
initiatives-such as PPBS in 1965-were typically implemented
governmentwide within a single annual budget cycle. In contrast, the
Results Act defines a phased and iterative implementation process that
incorporates pilot tests and formal evaluation of key concepts, including
performance budgeting.

 See GAO/AIMD-97-46, March 27, 1997 and Performance Budgeting: State Experiences and Implications
for the Federal Government (GAO/AFMD-93-41, February 17, 1993).
 The term "program activity" refers to the listings of projects and activities in the Appendix portion of
the Budget of the United States Government. Subject to clearance by the Office of Management and
Budget and generally resulting from negotiations between agencies and appropriations subcommittees,
program activity structures are intended to provide a meaningful representation of the operations
financed by a specific budget account.

Page 2                                                                      GAO/r-AIMD/GGD-99-216
The need to more closely link plans and budgets is of central importance to
the Results Act. Improving agencies' performance budgeting capabilities is
critical to meet a key expectation of the act-that decisionmakers
understand what is being achieved in relation to what is being spent.
Agencies cannot credibly set performance goals without understanding
what resources are needed to achieve them. Correspondingly, these goals
will be of little value to congressional appropriations decisions without a
connection to the resources that agencies are requesting.

The Results Act actually defines two different approaches regarding
performance budgeting. First, the act requires "each agency to prepare an
annual performance plan covering each program activity set forth in the
budget of such agency." The Congress intended this provision to establish a
direct annual link between plans and budgets. To prevent voluminous
presentations, agencies are permitted to aggregate, disaggregate, or
consolidate the program activities in their budgets, so long as any major
function or operation of the agency is not omitted or minimized. The Office
of Management and Budget's (OMB) subsequent guidance regarding this
provision of the act set forth an additional criterion: plans should display,
generally by program activity, the funding level being applied to achieve
performance goals.5 In effect, OMB's guidance expected performance plans
to indicate how amounts shown for program activities in an agency's
budget request would be allocated to the performance goals displayed in
the performance plan. 6 Testifying on the Results Act prior to its passage,
the Director of OMB characterized this requirement of the act as a
"limited-but very useful-form of performance budgeting ... "7

In addition to mandating a direct link between budget requests and
performance plans, the Results Act also required that a second approach to
performance budgeting be piloted. Specifically, the Director of OMB, in
consultation with the head of each agency, was required to designate for
fiscal years 1998 and 1999 at least five agencies to prepare budgets that
"present, for one or more of the major functions and operations of the
agency, the varying levels of performance, including outcome-related

50MB Circular A-11, Sec. 220.9(e), June 23, 1997.
 Subsequently, in its guidance on fiscal year 2000 plans, OMB noted that it expected to see "significant
progress in associating funding with specific performance goals or sets of goals" in agencies' plans.
 Govemment Performance and Results Act of 1993, Committee on Governmental Affairs, United States
Senate, S. Rpt. No. 103-58, p. 19 (1993).

Page 3                                                                     GAO/I-AIMD/GGD-99-216
                           performance, that would result from different budgeted amounts."8 While
                           the act required agencies to define goals consistent with the level of
                           funding requested in the President's budget, these pilot projects would also
                           show how performance would change if the agency received more or less
                           than requested. OMB was to include the pilot performance budgets as an
                           alternative presentation in the President's Budget for fiscal year 1999.
                           Subsequently, the Director of OMB is required to report to the President
                           and to the Congress no later than March 31, 2001, on the feasibility and
                           advisability of including a performance budget as part of the President's
                           Budget. This report is also to recommend whether legislation requiring
                           performance budgets should be proposed. However, as I will discuss in this
                           testimony, the performance budgeting pilots required by the Results Act
                           have not begun.

Performance                While the Results Act's design incorporates important lessons learned from
     eting Efforts Face    previous initiatives, many challenges remain. To a large extent, these
   Budgeimg Efforts Face   challenges are inherent to a complex, political environment such as the
Many Challenges            federal government. For example, competing and at times conflicting goals,
                           the variety of service delivery approaches, and the nature of federal
                           budgetary commitments raise serious implementation concerns. Both the
                           Congress and the executive branch must continue to explore what can be
                           reasonably expected from performance budgeting.

                           Performance budgeting assumes that performance goals can be defined
                           and that valid and reliable performance measures can be developed.
                           However, as we have noted previously, reaching a reasonable level of
                           consensus on clear and precise goals will almost certainly encounter
                           political hurdles. 9 In addition, goal definition and measure development are
                           particularly challenging in the complex operating environment of the
                           federal government.

                           * Full or ultimate program outcome is typically not under the control of a
                             single federal agency, complicating responsibility determinations and
                             resource allocation decisions. In some cases, federal activities are but
                             one--and often a small-component of total public and private sector

                           831 U.S.C. 1119(b).
                            See GAO/AIMD-97-46, March 27, 1997, and The Government Performance and Results Act: 1997
                           Governme:ntwide Imrlementation Will Be Uneven (GAO/GGD-97-109, June 2, 1997).

                           Page 4                                                               GAO/r-AIMD/GGD-99-216
  interventions in a given program area; in other cases, intended results
  cut across the activities of several agencies.l 0 In these situations,
  individual agency outcome measures could be incomplete and of limited
  value to budgetary deliberations.
* Increasingly-in program areas ranging from child welfare to
  environmental protection-state and local governments, contractors,
  and other third parties are the delivery agents for federally financed
  activities. The efforts of these nonfederal actors-and their objectives
  and concerns-are often critical factors in determining whether
  program results are achieved.
* Many federal activities, for example health and safety programs or
  research and development programs, achieve desired outcomes only
  over periods of many years. In such cases, relating these lengthy
  performance horizons to annual budget deliberations can raise special
  measurement questions.
* Finally, the predominance of entitlement spending within the federal
  budget, in which federal spending is a function of statutory eligibility
  determinations, can cloud efforts to hold agencies accountable for
  results. In these types of programs, attention is often shifted from
  outcomes (e.g., assuring a certain standard of living) to specific process
  standards (e.g., ensuring correct and prompt payments to individuals).

The high stakes involved in budgetary decisions further complicate the
development and use of outcome measures. Introducing such measures
into resource allocation processes before a reasonable level of consensus
is achieved heightens the potential for bias toward favorable results. 1
Recognizing this potential, the Results Act requires agencies to build
procedures for verifying and validating performance measures into their
plans. However, improvements in the quality of verification and validation
discussions in agencies' plans are needed if the Congress is to have needed
assurance that agencies' performance data will be credible. 12

°See for example Combating Terrorism: Opportunities to Improve Domestic Preparedness Program
Focus and Efficiency (GAO/NSIAD-99-3, November 12, 1998), Drug Treatment: Overview of Federal
Programs (GAO/HEHS-98-237R, September 3, 1998), and Homelessness: Coordination and Evaluation
of Programs Are Essential (GAO/RCED-99-49, February 26, 1999).

llProgram evaluation is critical to understanding and isolating an agency's impact on outcomes. For a
discussion of performance measurement challenges, see Managing for Results: Measuring Program
Results That Are Under Limited Federal Control (GAO/GGD-99-16, December 11, 1998), Prwgram
Evaluation: Agencies Challenged by New Demand for Information on Program Results
(GAO/GGD-98-53, April 24, 1998), and Managing for Results: Analytic Challenges in Measuring
Performance (GAO/HEHS/GGD-97-138, May 30, 1997).
  Managing for Results: An Agenda to Improve the Usefulness of Agencies' Annual Performance PlanIs
(GAO/GGD/AIMD-98-228, September 8, 1998).

Page 5                                                                   GAO/T-AIMD/GGD-99-216
In addition to developing and using nonfinancial outcome measures,
performance budgeting also requires an ability to understand how costs are
related to outcomes. Reliable cost information is essential for Results Act
implementation and was called for by the Chief Financial Officers (CFO)
Act of 1990. Cost accounting standards developed by the Federal
Accounting Standards Advisory Board (FASAB) 13 require that agencies
develop and implement cost accounting systems that can be used to relate
the full costs of various programs and activities to performance outputs.
Although these standards were originally to become effective for fiscal year
1997, the CFO Council-an interagency council of the CFOs of major
agencies-requested the effective date be delayed for 2 years due to
shortfalls in agencies' cost accounting systems. Ultimately, the effective
date was extended by 1 year, to fiscal year 1998, but with a clear
expectation that there would be no further delays.

Agencies recognize the importance of cost accounting and other financial
management systems in allocating funding to performance, but developing
the necessary tools to gather and analyze needed program and activity-level
cost information will be a substantial undertaking. For the most part,
agencies are just beginning this effort and are already experiencing
difficulties. For example, our audit of the Internal Revenue Service's (IRS)
fiscal year 1998 financial statements found that IRS does not consistently
capture cost information in accordance with cost accounting standards.i 4
Consequently, IRS was unable to reliably report cost-based performance
measures. Similarly, the fiscal year 1998 audit of the Federal Aviation
Administration's (FAA) financial statements found significant deficiencies
in FAAs cost accounting systems. FAA does not expect to have a fully
operational system until 2001.15

Finally, performance budgeting efforts will almost always disclose tensions
between budgeting and planning structures. As I mentioned earlier, the

  In October 1990, the nine member FASAB was established by the Secretary of the Treasury, the
Director of OMB, and the Comptroller General of the United States to consider and recommend
accounting standards to address the financial and budgetary information needs of the Congress,
executive agencies, and other users of federal financial information. Once FASAB recommends
accounting standards, the Secretary of theTreasury, the Director of OMB, and the Comptroller General
decide whether to adopt the recommended standards. If they are adopted, the standards are published
as Statements of Federal Financial Accounting Standards by OMB and GAO.
  Financial Audit: IRS' Fiscal Year 1998 Financial Statements (GAO/AIMD-99-75, March 1, 1999).

15Federal Aviation Administration: Financial Management Isues   (GAO/T-AIMD-99-122,
March 18, 1999).

Page 6                                                                   GAO/r-AIMD/GGD-99-216
                      Results Act requires agencies to link performance goals to their program
                      activity structures, which form the basis for their budget requests. This
                      requirement is aimed at assuring a simple, straightforward connection
                      among goals, budgets, and performance information. However, achieving
                      this link is dependent on the capacity of agencies' program activity
                      structures to meet dual needs. These budget structures have evolved to
                      help the Congress control and monitor agency activities and spending and,
                      as such, are geared more to fostering accountability for inputs and outputs
                      within the control of agencies. 16 On the other hand, performance plans
                      need to be broad and wide-ranging if they are to articulate the missions and
                      outcomes agencies seek to influence. Strategies for bringing budgeting and
                      planning structures together must balance both sets of needs and values.

                      For example, planning structures and presentations that bore no
                      connection to budget structures and presentations hampered performance
                      budgeting initiatives prior to the Results Act. In the fiscal year 1999
                      performance plans, agencies attempted to bring these structures and
                      presentations together by (1) changing budget structures to more closely
                      align them with goals in the performance plan, (2) using the budget
                      justification to provide more details on goals contained in their
                      performance plans, or (3) using crosswalks or tables to show relationships
                      between planning and budgeting structures.

The Postponement of   Many of the challenges I have just described were evident in OMB's
Performance           decision to delay the performance budgeting pilots required by the Results
                      Act. The performance budgeting pilots were scheduled to start in fiscal
Budgeting Pilots      year 1998-four years after initiation of theact's pilot projects for
                      performance plans and reports-"so that they would begin only after
                      agencies had sufficient experience in preparing strategic and performance
                      plans, and several years of collecting performance data."17 In this context,
                      and indicating the importance of concentrating on governmentwide
                      implementation in fiscal year 1998, the Director of OMB in his statutorily
                      required May 1997 report on Results Act implementation announced that
                      the pilots would be delayed for at least a year. The Director stated that the
                      performance budgeting pilots would require the ability to calculate the

                      16Budget Account Structure: A Descriptive Overview (GAO/AIMD-95-179, September 18,1995).

                      17S. Rpt No. 103-58, p. 38.

                      Page 7                                                               GAOIT-AIMD/GGD-99-216
effects on performance of marginal changes in cost and funding. According
to OMB, very few agencies had this capability, and the delay would give
time for its development.

Subsequently in September 1998, OMB suggested possible formats and
time frames for the pilots in a discussion paper sent to federal agencies. In
this document, OMB noted that pilot projects would not be designated
unless they could "fairly test the [Results Act's] concept of performance
budgeting," which the discussion paper described as "the application of
multi-variate or optimization analysis to budgeting." The paper described
three analytical alternatives that could be tested, involving performance
tradeoffs (1) in the same program with changes in program funding,
(2) in the same program with no change in total program funding, or
(3) in several programs with shifts in intra-agency funding between these
programs. OMB solicited agencies' comments on the discussion paper and
on their capability to produce the alternative budgets suggested in the
committee report accompanying the Results Act. However, according to
OMB, no agency volunteered to participate. In its discussion paper, OMB
stated that "the absence of designated pilots or having fewer designations
than required would be an indication of agency readiness to do
performance budgeting, and would be discussed in the OMB report to
Congress," which is required on March 31, 2001.

Agencies' reaction to the performance budgeting pilots reaffirms the
challenges and tensions that performance budgeting will face within the
federal government. Whether due to imprecise goal definitions, the absence
of valid and reliable financial and nonfinancial performance information,
uncertainty concerning the relationship between agency activities and
desired outcomes, or a lack of priority attention, no federal agency was
prepared to associate itself with pilot projects that appeared to
mechanically link resources to results. Although not required to do so,
OMB has not publicly communicated agencies' reactions to the discussion
paper or agencies' reasons for declining to participate in the pilots. As a
result, it has been over 2 years since OMB reported to the Congress on
challenges facing these pilots, and more information is needed to
determine not only the viability of the pilots but also the direction that
federal performance budgeting efforts can and should take. In effect, an
opportunity to better understand the specific challenges facing the Results
Act's performance budgeting pilots has been missed.

Page 8                                                  GAO/Tr-AIMD/GGD-99-216
OMB Can Take Steps to   Despite the postponement of the performance budgeting pilots, our review
Achieve the Intent of   of agencies' performance plans shows that some agencies have been able
                        to develop approaches to make perhaps a more basic, but still useful,
the Pilot Projects      connection between proposed spending and expected performance. The
                        experience of these agencies during the first two performance planning
                        cycles under the Results Act can provide a valuable foundation for future
                        efforts to more closely demonstrate the performance consequences of
                        budgetary decisions.

                        In summary, we found that 30 of the 35 fiscal year 1999 agency performance
                        plans we reviewed defined some relationship between program activities
                        and performance goals, as called for by the Results Act.18 However, only 14
                        of these plans translated this relationship into budgetary terms by (1)
                        identifying the proposed funding level needed to achieve a discrete set of
                        performance goals and (2) describing how that funding had been derived
                        from the program activities in the agencies' budget requests. Figure 1
                        illustrates the relationship between resources and results that was
                        expressed by 1 of the 14 agencies-the Nuclear Regulatory Commission
                        (NRC). As shown in this figure, NRC not only indicated that it would need
                        $211 million to achieve the nuclear reactor safety performance targets
                        described in its plan but also explained that the $211 million had been
                        derived from a single program activity in its budget request. As a result,
                        NRC's plan not only indicates the estimated cost of a given level of
                        performance but also shows which goals would be primarily affected by
                        changing the level of program activity funding from NRC's proposal.

                        '8See GAO/AIMD/GGD-99-67, April 12, 1999. We could not determine linkages between program
                        activities and performance goals for five agencies from the information provided in their performance

                        Page 9                                                                    GAO/T-AIMD/GGD-99-216
Figure 1: The Nuclear Regulatory Commission Aligns Budget and Planning
Structures to Create a Simple Relationship Between Program Activities and
Performance Goals

    NRC budget accounts and                                              NRC strategic goals and
       program activities                                                  performance goals

   Salaries and expenses account                                  at e    goal: nuclear reactor safety (211)
     1.i Nuclear reactor safety   211                     Performance goals:
                                                                , Goal l.A -zero civilian nuclear reactor accidents
     2.1Nuclear materials safety ($49) a                        * Goal I.A.1- maintain low frequency of events which
                                         ~3.    wastesafety             lead to a severe accident
                                                                * Goal l.B- zero deaths due to radiation or
     4. Common defense and security and                         radioactivity releases from civilian nuclear reactors
         international involvement                                 Goal I.B.1- zero significant radiation exposures due
                                                                to civilian nuclear reactors
     5. Protecting the environment
                                                          Strategic goal: nuclear materials safety ($49) I
     6. Management and support
                                                          Performance goals:
                                                                * Goal IIA- zero radiation-related deaths due to
                                                                civilian use of source, byproduct, and special nuclear
                                                                * Goal II.A.1- no increase in the number of significant
                                                                radiation exposures due to loss or use of source,
                                                                byproduct, and special nuclear materials
                                                                * Goal lI.A.l.a- no increase in the number of losses
                                                                of licensed material as reported to Congress annually
                                                                * Goal II.A.1 .b- no accidental criticality involving
                                                                licensed material.
                                                                * Goal II.A.2- no increase in the number of
                                                                misadministration events which cause significant
                                                                radiation exposures

Note: Dollars in millions.
Source: GAO analysis based on NRCs fiscal year 1999 performance plan and Budget of the United
States Government Fiscal Year 1999-Appendix.

We also looked across these 35 plans to determine whether agencies and
their plans shared common characteristics. We found that three
approaches, either alone or in combination, were used more frequently by
the 14 agencies that were able to relate resources to results. 19 These
agencies more often (1) showed simple, clear relationships between
program activities and performance goals, (2) fully integrated performance
plans into congressional budget justifications, or (3) changed their budget
program activity structures to reflect the goal structures in their
performance plans.

191n this review, we did not assess the quality of the goals presented in the plans or independently verify
the funding: levels associated with the goals.

Page 10                                                                                 GAOtr-AIMD/GGD-99-216
NRC adopted each of these approaches. As figure 1 shows, NRC presented
a simple one-to-many relationship between its program activities and its
performance goals in its fiscal year 1999 performance plan.2 0 The
allocation of funding to performance goals in the NRC plan was essentially
automatic because each of the agency's program activities generally aligns
with a strategic goal and its supporting performance goals. This alignment
was facilitated by NRC's decision to change its budget structure to align
with its strategic goals, as shown in figure 1.21 For the fiscal year 2000
performance plan, NRC maintains the relationships shown in figure 1 while
fully integrating its performance plan and budget justification. 22
Information traditionally contained in a budget justification, such as
descriptions of accounts and their funding, was combined with
performance information such that the NRC budget justification and plan
could not be separated. In contrast, other agencies, such as the Department
of Veterans Affairs (VA), did not identify the funding levels needed to
achieve performance goals and associated numerous program activities
with numerous performance goals in their fiscal year 1999 performance
plans-a "many-to-many" presentation that did not indicate the
performance consequences of the agencies' budget requests. Like some
other agencies, VA has noted that it is considering changes to its budget
structure to improve its ability to relate resources and results.

Although they used some common approaches, the 14 agencies that
connected budgetary resources to results represented a range of federal
missions, including agencies providing services directly to the public
(e.g., IRS), those principally involved in grant or loan making (e.g., the
U.S. Agency for International Development), and some with principally a
regulatory mission (e.g., NRC). Similarly, these agencies achieved linkages
despite varying planning and budgeting structures. The relative complexity
of these structures-measured in terms of the number and layers of goal
structures, and the number of budget accounts and program activities and
concentration of funding within those accounts-did not appear to be a
significant factor in an agency's ability to relate proposed resources to
expected results.

     Figure 1.2 in appendix I provides another illustration of a simple one-to-many relationship.

21As shown in figure 1.3 in appendix I, the Environmental Protection Agency proposed an alignment of
its program activities and strategic goals.
  Figure 1.4 in appendix I provides an illustration of another agency that fully integrated its
performance plan and budget justification.

Page 11                                                                        GAO/I-AIMD/GGD-99-216
During our preliminary review of the fiscal year 2000 plans for the same
35 agencies, we have noted little change in the overall number of agencies
clearly relating resources to results. But, it does appear that some progress
is being made in presenting the performance consequences of budgetary
decisions. For example, more fiscal year 2000 plans associated funding
levels with specific performance goals, although in many cases these
funding levels were not linked back to the program activities in agencies'
budget requests.

The extent of agencies' progress in linking plans and budgets is not
surprising; translating the use of agency resources into concrete and
measurable results will be a continual challenge that will require both time
and effort. However, based on our review of the first two cycles of
performance planning under the Results Act, we believe that the
approaches being developed by some agencies provide a valuable
foundation for further experimentation in identifying useful methods to
connect planning and budgeting structures. Some of these approaches-
such as those used by NRC, the Administration for Children and Families,
the Health Resources and Services Administration, and IRS-are illustrated
in figure 1 and appendix I to this testimony.

In fact, agency efforts to link performance goals and program activity
funding essentially constitute a first step toward achieving the intent of the
performance budgeting pilots. As defined by the Congress, the original
intent for the act's pilot projects was twofold: to allow OMB and agencies
to develop experience and capabilities towards realizing the potential of
performance budgeting, and to provide OMB with a basis for reporting to
the Congress on next steps and needed changes. In addition to providing
some practical experience with the concept of performance budgeting,
agencies' fiscal year 1999 and 2000 performance plans also provide a
baseline from which OMB could assess progress and determine what
changes, if any, may be needed to the act and federal budget processes.

OMB is the lead agency for overseeing a framework of recently enacted
reforms designed to improve the effectiveness and responsiveness of
federal agencies. 2 3 Thus, OMB should be well-situated to assess (1) the
practicality of performance budgeting pilots as currently defined in the law,

  The Results Act: Observations on the Office of Management and Budget's July 1997 Draft Strategic
Plan (GAO/AIMD/GGD-97-169R, August 21, 1997).

Page 12                                                                 GAO/T-AIMD/GGD-99-216
(2) agency approaches and continuing challenges to linking budgetary
resources and performance goals, and (3) options to encourage progress in
subsequent planning and budgeting cycles.

In light of the delay of the performance budgeting pilots required by the
Results Act and the experiences of agencies during the fiscal year 1999
performance planning and budgeting cycle, we recommended in April 1999
that the Director of OMB assess the approaches agencies used to link
performance goals and program activities in the fiscal year 2000
performance plans. OMB's analysis, building on our review of fiscal year
1999 performance plans, should develop a better understanding of
promising approaches and remaining challenges with respect to the
concept of performance budgeting within the federal government.
OMB's analysis should address, for example,

* the extent of agencies' progress in associating funding with specific or
  sets of performance goals,
* how linkages between budgetary resources and results can be made
  more useful to the Congress and to OMB,
* what types of pilot projects might be practical and beneficial, and
* when and how those pilot projects would take place.

On the basis of this analysis, we recommended that OMB work with
agencies and the Congress to develop a constructive and practical agenda
to further clarify the relationship between budgetary resources and results,
beginning with specific guidance for the preparation of agencies' fiscal year
2001 plans. We further recommended that this analysis and the resulting
agenda become the foundation for OMB's report to the Congress in March
2001, as currently required by the Results Act, on the feasibility and
advisability of including a performance budget as part of the President's
Budget and on any other needed changes to the requirements of the act.

In summary, Mr. Chairman, much can be learned from the initial efforts of
some agencies to demonstrate the performance consequences of budget
requests. Given the importance of performance budgeting to achieving the
Results Act's full potential and the delay of performance budgeting pilots
called for by the act, it is critical that promising approaches be explored
and encouraged. The performance plans being developed under the Results
Act show potential to inform the budget process and change the nature of
its dialogue by more routinely introducing performance information into
budgetary decision-making. To be sure, many challenges will remain-from
defining outcome goals to developing effective performance measures and
reliable cost information.

Page 13                                                GAOfr-AIMD/GGD-99-216
                    At the same time, the Results Act and, in fact, any performance budgeting
                    initiative cannot be expected to eliminate conflict inherent in the political
                    process of resource allocation. The linkage of performance plans and
                    budget requests does not guarantee that decisions will be made solely on
                    the grounds of performance-nor should they be, there are other important
                    criteria. However, the absence of meaningful links can inhibit the
                    usefulness of performance information for resource allocation decisions.
                    Only through continued experimentation and the mutual efforts of the
                    Congress and the executive branch will the potential, and limits, for
                    performance budgeting within the federal government be determined.

                    Mr. Chairman, this concludes my statement this morning. I would be
                    pleased to respond to any questions you or other Members of the
                    Subcommittee may have.

Contact and         For information about this testimony, please contact Paul Posner at
 Ackn1owledgments   (202) 51"2-9573 or by e-mail at posnerp.aimd@gao.gov,or J. Christopher
                    Mihm at (202) 512-8676 or by e-mail at mihmj.ggd@gao.gov Individuals
                    making key contributions to this testimony included Michael J. Curro and
                    Laura E. Castro.

                    Page 14                                                GAO/T-AIMD/GGD-99-216
Page 15   GAO/r-AIMD/GGD-99-216
Appendix I

Illustrations of Approaches Used to Connect
Resources to Results in Agencies' Fiscal Year
1999 Performance Plans

               Figure 1.1: The Administration for Children and Families Crosswalks Program
               Activities to Performance Goals

                          ACF budget accounts and                             ACF strategic goals, strategic
                             program activities                             objectives, and performance goals

                   Family support payments to states                       Strategic goal:
                   account                                                 Increase economic independence and
                         1 State child support administrative              productivity for families
                          sts ($2,749)
                         2. Federal incentive/hold harmless                   Strategic objective:
                             paynents   to states ($469)           1       BIncrease parental      responsibility ($3,257)
                         3 Access and visitation grants ($10)
                         4. Payments to territories                           Performance Goals:
                                                                                * Increase the paternity establishment percentage
                         5. Repatriation                                        among children born out-of-wedlockto 96 percent
                         6.Aid to families with dependent children               · Increase the percentage of cases having child
                         benefit payments                                       support orders to 74 percent
                         7. Emergency assistance                                * Increase the collection rate for current support
                                                                                to 70 percent
                  Children's research and technical                             · Increase the percentage of paying cases among
                  assistance account                                            arrearage cases to 46 percent
                              iFe2deral parent locator service ($30)   |          Increase
                                                                                   . .· the cost-effectiveness ratio (total dollars
                                                                                collected per dollar of expenditures) to $5.00
                         2. Training and technical assistance
                         3. Child welfare study
                         4. Welfare research
                         5. Evaluation of welfare to work
                         6. Evaluation of abstinence education

               Note: Dollars in millions. Numbers may not add due to rounding.
               Source: GAO analysis based on the Administration for Children and Families' fiscal year 1999
               performance plan and Budget of the United States Government Fiscal Year 1999-Appendix.

               Page 16                                                                           GAO/r-AIMD/GGD-99-216
Appendix I
Illustrations of Approaches Used to Connect
Resources to Results in Agencies' Fiscal Year
1999 Performance Plans

Figure 1.2: The Health Resources and Services Administration Creates a Simple
Relationship Between Program Activities and Performance Goals

          HRSA budget accounts                             HRSA strategic goals and
          and program activities                              performance goals

      Health resources and                          Strategic goals:
      services account                               1. Eliminate health care disparities
                                                     2. Eliminate barriers to care
                                                     3. Assure quality

      20 program activities including:
      HIV/AIDS                                      Performance goals:
                                                            Increase the number of clients served
              Activity 1: AIDS emergency                  by Title I grant programs by 7.5 percent
                                                                    (3          goals

                                                            ncrease the number of clients receiving
              Activity 2 H V care grants to               anti-retro viral therapy to 57,500
              states ($670)
                                                          *5 other performance goals

                                                          * ncrease by 5 percent the number of
              intervention services ($86)                  4 other pe            goals

              4 Other activities

Note: Dollars in millions.
Source: GAO analysis based on the Health Resources and Services Administration's fiscal year 1999
performance plan and Budget of the United States Government Fiscal Year 1999-Appendix.

Page 17                                                                     GAO/T-AIMD/GGD-99-216
Appendix ][
Illustrations of Approaches Used to Connect
Resources to Results in Agencies' Fiscal Year
1999 Performance Plans

Figure 1.3: The Environmental Protection Agency Proposed Aligning Budget and
Planning Structures

           EPA budget accounts and                               EPA strategic goals,
              program activities                               strategic objectives, and
                                                                  perfonmance goals
     Science and technogyW account
                 1.1 Qeanair~($1~374~                        Strategic goal: clean air
            1.1Ceana ir( S137)d                   $4 --
            2. Clean water
            3. Sae food                                      Strategic ojective: acd rain $22)
            4. Preventing pollution
            5. Waste management                              Perfomancegoals:
            6. Global and oss border                              6.Maintain
                                                                             brder tons ofsulfur
            7. Righttoow
            7. Right tosono                                       dioxide reductions from utility sources
            8. Soundscience
            9. Credible deterrent                                 . Maintain 300,000 tons of nitrogen
                                                                  oxides reductions from coal-fired utility
     Envirrnmental programs and                                   sources
     mananaeen       t account                                    * Launch the nitrogen oxides Emissions

            1.1Cleanair($169) I--                  $13             andA loanceTracldngSytemforthe
                                                                  t1one Transport Region
            *Other program activities corresponding to
            EPA's other strategic goals (similar to above)

     State and tribal assistance grants
            1. Cleanair($201)                  $5
            * Other program activities corresponding to
            EPA's other strategic goals (similar to above)

Note: Dollars in millions.
Source: GAO analysis based on the Environmental Protection Agency's fiscal year 1999 performance
plan and Budget of the United States Government Fiscal Year 1999-Appendix.

Page 18                                                                     GAO/'r-AIMD/GGD-99-216
           Appendix I
           Illustrations of Approaches Used to Connect
           Resources to Results in Agencies' Fiscal Year
           1999 Performance Plans

           Figure 1.4: IRS Integrates Its Budget Justification and Performance Plan

                                                Processing, Assistance, and
                                               Management Account ($3,162)

                     Program activity: Functions: This activity provides for the salaries, benefits, and related
                submission processing costs to process tax returns and supplemental documents, account for tax
                                ($888) revenues, issue refunds and tax notices, develop and print tax returns and
                                       publications ... Also included are resources to: process information
                                       returns such as wage, dividend, and interest statements; provide for
                                       payment of refunds ... identification of possible non-filers for investigation;
                                       and, provide tax returns for audits...

                    Performance goal #1:               Performance goal #2:           Performance goal #3:
                  Improve customer service             Increase compliance            Increased productivity

                   Performance measure:               Performance measure:             Performance measure:
                   Number of individual               211.8 million primary returns    19.5 percent of individual
                   refunds issued will equal          processeda                       returns filed electronically
                   93.3 millions
                                                                                       Performance measure:
                   Performance measure:
                   Refund timeliness--paper                                            78.2
                                                                                       78.2 percent
                                                                                            percent of
                                                                                                    of dollars
                   40 days                                                             received electronically

                   5 other performance                                                 Performance measure:
                   measuresrfrmnc                                                      70.9 percent dollars
                                                                                       received via third party

           alIRS noted that this is a projection for budget purposes and is not used in the agency's business
           Note: Dollars in millions.
           Source: GAO analysis based on Internal Revenue Service's fiscal year 1999 performance plan.

(935323)   Page 19                                                                          GAO/T-AIMD/GGD-99-216