[House Hearing, 110 Congress]
[From the U.S. Government Publishing Office]


 
USING TAXPAYERS' DOLLARS MOST EFFICIENTLY: PERSPECTIVES ON PERFORMANCE 
                               BUDGETING

=======================================================================

                                HEARING

                               before the

                        COMMITTEE ON THE BUDGET
                        HOUSE OF REPRESENTATIVES

                       ONE HUNDRED TENTH CONGRESS

                             FIRST SESSION

                               __________

           HEARING HELD IN WASHINGTON, DC, SEPTEMBER 20, 2007

                               __________

                           Serial No. 110-19

                               __________

           Printed for the use of the Committee on the Budget


                       Available on the Internet:
       http://www.gpoaccess.gov/congress/house/budget/index.html


                                 ______
                                     
                    U.S. GOVERNMENT PRINTING OFFICE
37-323                      WASHINGTON : 2008
_____________________________________________________________________________
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov  Phone: toll free (866) 512-1800; (202) 512ï¿½091800  
Fax: (202) 512ï¿½092104 Mail: Stop IDCC, Washington, DC 20402ï¿½090001

                        COMMITTEE ON THE BUDGET

             JOHN M. SPRATT, Jr., South Carolina, Chairman
ROSA L. DeLAURO, Connecticut,        PAUL RYAN, Wisconsin,
CHET EDWARDS, Texas                    Ranking Minority Member
JIM COOPER, Tennessee                J. GRESHAM BARRETT, South Carolina
THOMAS H. ALLEN, Maine               JO BONNER, Alabama
ALLYSON Y. SCHWARTZ, Pennsylvania    SCOTT GARRETT, New Jersey
MARCY KAPTUR, Ohio                   MARIO DIAZ-BALART, Florida
XAVIER BECERRA, California           JEB HENSARLING, Texas
LLOYD DOGGETT, Texas                 DANIEL E. LUNGREN, California
EARL BLUMENAUER, Oregon              MICHAEL K. SIMPSON, Idaho
MARION BERRY, Arkansas               PATRICK T. McHENRY, North Carolina
ALLEN BOYD, Florida                  CONNIE MACK, Florida
JAMES P. McGOVERN, Massachusetts     K. MICHAEL CONAWAY, Texas
ROBERT E. ANDREWS, New Jersey        JOHN CAMPBELL, California
ROBERT C. ``BOBBY'' SCOTT, Virginia  PATRICK J. TIBERI, Ohio
BOB ETHERIDGE, North Carolina        JON C. PORTER, Nevada
DARLENE HOOLEY, Oregon               RODNEY ALEXANDER, Louisiana
BRIAN BAIRD, Washington              ADRIAN SMITH, Nebraska
DENNIS MOORE, Kansas                 [Vacancy]
TIMOTHY H. BISHOP, New York
GWEN MOORE, Wisconsin
[Vacancy]

                           Professional Staff

            Thomas S. Kahn, Staff Director and Chief Counsel
                James T. Bates, Minority Chief of Staff


                            C O N T E N T S

                                                                   Page
Hearing held in Washington, DC, September 20, 2007...............     1

Statement of:
    Hon. John M. Spratt, Jr., Chairman, House Committee on the 
      Budget.....................................................     1
    Hon. Paul Ryan, ranking minority member, House Committee on 
      the Budget.................................................     2
    Barry Anderson, Organization for Economic Cooperation and 
      Development (OECD), prepared statement of..................     3
    Hon. Adrian Smith, a Representative in Congress from the 
      State of Nebraska, prepared statement of...................     7
    Hon. Clay Johnson III, Deputy Director for Management, Office 
      of Management and Budget (OMB).............................     8
        Prepared statement of....................................     9
    David M. Walker, Comptroller General of the United States, 
      U.S. Government Accountability Office (GAO)................    11
        Prepared statement of....................................    16
    Peter R. Orszag, Director, Congressional Budget Office (CBO).    37
        Prepared statement of....................................    40
    Paul L. Posner, director, Public Administration Program, 
      George Mason University, prepared statement of.............    78


                     USING TAXPAYERS' DOLLARS MOST
                      EFFICIENTLY: PERSPECTIVES ON
                         PERFORMANCE BUDGETING

                              ----------                              


                      THURSDAY, SEPTEMBER 20, 2007

                          House of Representatives,
                                   Committee on the Budget,
                                                    Washington, DC.
    The committee met, pursuant to call, at 10:07 a.m. in room 
210, Cannon House Office Building, Hon. John Spratt [chairman 
of the committee] presiding.
    Present: Representatives Spratt, Edwards, Cooper, Becerra, 
Blumenauer, Berry, Boyd, Scott, Etheridge, Bishop, Ryan, 
Conaway, Porter, Alexander, and Smith.
    Chairman Spratt. Good morning and welcome to our hearing on 
performance budgeting. We have an excellent panel of government 
witnesses who have thought long and hard about this subject, 
performance budgeting. I thank them for their participation 
today.
    Given the daunting budgetary that face us, we need more 
than ever to put the taxpayers' dollars to the wisest possible 
use. Earlier this year, the committee heard from government 
witnesses about efforts to combat waste, fraud and abuse in 
health care, retirement and tax collection.
    Our budget this year made provision for additional funding 
to provide for what we call ``program integrity'' in these 
areas. And those program integrity funds are included in the 
appropriation bills that the House passed.
    Today we turn our attention to another aspect of 
stewardship. Our question today is whether or not there are 
ways that we can improve the tools we use to measure government 
performance and effectiveness to see that we are getting the 
most bang for our bucks.
    Performance budgeting is not a new concept. It has been 
undertaken by nearly every administration for the past 50 
years. The most recent comprehensive initiative was the 
Government Performance Results Act of 1993 which is aimed at 
creating a framework to align performance objectives and 
program activities.
    While the goals of the GPRA are similar to other efforts, 
it differs from its predecessors in one key respect. The basis 
for the GPRA review is statutory, explicitly linked to the 
budget process, and congressional involvement is mandated.
    During the current administration, OMB has embarked on the 
development of the rating system known as Program Assessment 
Rating Tool, or PART. While similar in some respects to the 
goals of GPRA, PART has been used mostly to assist in the 
executive branch budget formulation. There is a wide range of 
views about PART. Criticism has been raised about its 
effectiveness, the subjectivity. And some of those perspectives 
will be explored today. And from OMB, GAO and CBO we have an 
extraordinary range of expertise.
    But before turning to our witnesses, I want to yield to the 
Ranking Member, Mr. Ryan for any opening statements he may care 
to make.
    Mr. Ryan.
    Mr. Ryan. Thank you, Chairman, and I thank you for having 
this hearing. This is a very well-timed hearing and I look 
forward to our witnesses' testimony.
    Clearly, constituents want Congress to ensure that they are 
running their government and spending their tax dollars as 
effectively and as efficiently as possible. But Congress has 
long struggled to define that goal, let alone even achieve it. 
There is simply no formula on how Congress' spending decisions 
are made. Policy preferences, parochial interests, both inside 
and outside of government, and even emotional ties, all factor 
into the mix. And it is notoriously difficult to come to a 
consensus as to which programs we think are working well and 
which are not. It is therefore critical that Congress has an 
objective means for measuring agency performance.
    Today we are here to discuss several of the tools Congress 
has been provided to help us towards that end. And in 
particular, the Government Performance and Results Act and the 
administration's Program Assessment Rating Tool, or PART, are 
two good tools. Clearly there is not always going to be a 
direct correspondence between how well an agency performs and 
the amount of funding it receives.
    That said, let us look at this year's PART assessment of 
the worst performing programs that appears to show that there 
is a complete disconnect between program performance and House-
passed funding levels.
    If you could pull up chart one, please.
    Consider what has happened with our worst performing 
programs. OMB's recent PART analysis reviewed over 1,000 
programs, 3 percent of which received a rating of ineffective, 
or the worst possible rating that they provide. The President's 
budget requested lower funding levels for these programs. But 
as this chart shows, the House actually gave them considerably 
more than the President requested. Again, there is not always 
going to be a direct correspondence between performance and 
funding levels, but it does seem that Congress is failing to 
take advantage of the tools we have available to guide funding 
decisions, at least partially, on actual agency performance.
    So I think today's review of these assessment tools and how 
Congress might better utilize them is a particularly useful and 
constructive use of this committee's time. And I thank the 
Chairman for having this hearing.
    Chairman Spratt. Mr. Ryan, we will go with Mr. Johnson 
first, the Deputy Director of OBM. But before we do, we have 
got a couple of housekeeping details. First of all, we have a 
request for unanimous consent for two witnesses, the statements 
of Mr. Barry Anderson, whom we all know, head of Budgeting and 
Public Expenditures at OECD, and Paul Posner, Director of the 
Public Administration program at George Mason. I want to take 
this opportunity quickly to thank both of them for their 
contributions to the hearing. We had asked them to submit 
testimony to gain a boarder spectrum of views about the subject 
from some of those who have thought long and hard about it. 
Their observations will be part of the record and will be food 
for thought as we deal with this matter.
    [The prepared statement of Mr. Anderson follows:]

    Prepared Statement of Barry Anderson, Organization for Economic 
                   Cooperation and Development (OECD)

    Mr. Chairman and Members of the Committee: Thank you for giving me 
the opportunity to testify on the use of performance information in 
budgeting. I know this Committee has had an interest in the performance 
of Federal programs for a number of years, as I have discussed the 
subject many times with members and staff during my tenure as the 
Deputy Director of the Congressional Budget Office between 1999-2003. 
However, I am in a much different role now than I was then: I am 
currently Head of the Budgeting and Public Expenditures Division of the 
Organization of Economic Cooperation and Development (OECD) located in 
Paris. And I have been asked to submit written testimony on the 
international use of performance information in budgeting.
    OECD countries--and other countries around the world--are under 
pressure to improve public sector performance and at the same time 
contain expenditure growth. As factors such as ageing populations and 
increasing health care and pension costs add to budgetary pressures, 
citizens are demanding that governments be made more accountable for 
what they achieve with taxpayers' money. An important step in this 
process is to gather objective information about public sector 
performance.\1\ Performance information enables governments to measure 
progress towards achieving their policy and programme goals and 
provides details on what initiatives are working and those that are 
not. The central aim is to improve decision making by providing 
politicians and civil servants with better quality and more concrete 
information on the performance of agencies and programmes. In addition, 
this information can enhance transparency to the public and the 
legislature.
---------------------------------------------------------------------------
    \1\ Curristine, T., Z. Lonti and I. Joumard (2007) ``Improving 
Public Sector Efficiency: Challenges and Opportunities'', OECD Journal 
on Budgeting, Vol. 7(1), pp. 161-201
---------------------------------------------------------------------------
    The introduction of performance information into the budget 
processes across OECD countries is a widespread, long term and evolving 
process. Formal performance information is taken to refer to both 
performance measures (outputs and/or outcomes) and evaluations. 
Consider the following:
     75% of OECD countries include non-finance performance data 
in budget documentation;
     40% of countries have been working on outputs for over 10 
years;
     35% of countries introduced a new initiative in this area 
in 2005.\2\
---------------------------------------------------------------------------
    \2\ Curristine, T (2005) ``Performance Information in the Budget 
process: Results of the OECD 2005 Questionnaire'', OECD Journal on 
Budgeting, Vol. 5(2), pp. 87-131
---------------------------------------------------------------------------
    The 1990s witnessed an expansion of efforts by many governments to 
introduce performance information into their budgets as part of a 
process that seeks to move the focus of decision making in budgeting 
away from inputs (how much money can I get?) towards measurable results 
(what can I achieve with this money?). At first this movement led to 
the introduction of performance budgeting as a form of budgeting that 
directly relates funds allocated to measurable results. Performance 
budgeting has also been linked to larger reform efforts to improve 
expenditure control and/or public sector efficiency and performance. 
Performance budgeting initiatives tend to go hand in hand with 
performance management or managing for results. These reforms are often 
combined with reductions in input controls and increased flexibility 
for managers--in return for stronger accountability for the results--so 
as to enable them to decide how to best deliver public services.

                     PERFORMANCE-INFORMED BUDGETING

    The production of performance information is not an end in itself; 
quite the contrary, to make difference performance information has to 
be actually used in the decision-making process. As countries struggled 
with mechanical performance budgeting systems that some first started 
with, a major issue developed on what was the best way to use 
performance information in the budgetary process. However, whatever 
troubles countries encountered in using performance information, no 
country ever stopped using the information, although some changed their 
techniques as they learned what worked and what didn't.
    There are different models and approaches to using performance 
information. As the use of performance information evolved, the 
majority of OECD countries began to adopt performance-informed 
budgeting where performance information (performance measures and/or 
evaluations) is used along with other information on political and 
fiscal priorities to inform--but not determine--budget allocations. 
Most importantly, it is only one factor in the decision-making 
process.\3\ In performance-informed budgeting, there is no direct or 
mechanical link between measures of performance and funding. 
Performance information is used for planning and/or accountability 
purposes; that is, the Ministry of Finance (MOF) can use planned future 
performance to inform funding decisions and use performance results to 
hold agencies to account and to inform budgetary allocations. But many 
countries have found that performance-informed budgeting techniques are 
better at incorporating performance information into the budget process 
than mechanistic performance budgeting.
---------------------------------------------------------------------------
    \3\ OECD (2007) , Performance Budgeting in OECD Countries, OECD 
Publishing, Paris.
---------------------------------------------------------------------------
    Below are examples of different approaches to include performance 
information in budget and management processes taken by a few OECD 
countries.
Australia
    Australia's current performance budgeting and management framework 
has been in place since the mid-1990s. These arrangements arose from 
the budgetary reforms associated with the 1996 report of the National 
Commission of Audit and subsequent introduction of accrual-based 
outcomes and outputs policies. The current framework develops both 
performance measures and evaluations.
    At a national level, Australia operates under a devolved financial 
framework. Performance management and budgeting are generally the 
responsibility of individual ministers and their departments and 
agencies. The current system concentrates on agency-level outcomes. 
Every department and agency within the general government sector is 
required to identify comprehensive and explicit outcomes, outputs, and 
performance measures for quantity, quality, price, and effectiveness of 
their activities. They are required to report on those items and any 
major evaluations in their budget plans (portfolio budget statements) 
and their end-of-year results (in annual reports).
    Expenditure and programme reviews are a central feature of the 
Australian budget process and an area where performance information is 
used to inform budget decisions. Until recently, the lapsing budget 
measure review process was the most commonly used type of review, but 
the Australian government revised the review arrangements in October 
2006. The new arrangements give the MOF a greater role in identifying 
and managing reviews in co-operation with departments. Senior ministers 
make decisions early in the budget process regarding which major areas 
of public expenditure will be reviewed in any year. In addition, there 
can be major reviews or departmental reports, which can be short-term 
and available for the next budget or more for long-term.
Canada
    Over the years, the Canadian federal government has implemented 
numerous performance budgeting and management initiatives. The 
government currently uses performance information throughout the 
planning, measuring and assessing, and reporting phases of expenditure 
management.
    The federal government's expenditure management framework is 
complex and decentralised. All major departments produce strategic 
plans, known as reports on plans and priorities (RPPs). These are 
planning documents which are submitted to Parliament detailing the 
strategic outcomes and planned results of each department. These 
documents also include information on resource requirements over a 
three-year period. All departments report on the results in 
departmental performance reports (DPRs) which set out performance 
against commitments in the RPP. The Treasury Board (a Cabinet 
committee) produces and presents to Parliament two whole-of-government 
reports: RPP Overview guides parliamentarians through the many RPPs 
each spring, and Canada's Performance does the same for the 
departmental performance reports each autumn.
    In addition, all major departments and agencies have internal audit 
and evaluation units. In 2004-2005, evaluations covered approximately 
10% of departmental programme funding. The government has also sought 
to introduce performance information into decision making on its 750 
non-statutory transfer payment programmes that are delivered through 
third parties and must be reviewed by the Treasury Board at least once 
every five years. As the programmes come up for review, departments 
must produce evaluations. In addition, a policy on transfer payments 
(June 2000) formalised the requirement for departments to develop 
results-based management and accountability frameworks and risk-based 
audit frameworks in support of the ongoing management and renewal of 
these programmes.
    In 2005, the Treasury Board, in an effort to regain detailed 
programme-level knowledge, adopted the Management, Resources and 
Results Structure (MRRS) policy. This sets explicit requirements for 
departments on how to collect, manage and report financial and non-
financial information at a government-wide level. These results-based 
structures underpin all performance planning and reporting.
    After the 2006 election, the conservative government called for 
reform of the federal expenditure management system and greater use of 
performance information in support of resource allocation and 
reallocation decisions. In addition to the performance-based Cabinet 
reviews of transfer payment programmes already in place, current plans 
call for the commencement of a first round of strategic programme 
reviews to begin in the autumn of 2007. All available performance 
information--whether from audits, evaluations, the MRRS performance 
frameworks, or from annual assessments of departmental management 
performance--will support these reviews.
Netherlands
    In 2001, the Netherlands introduced a more policy-oriented budget 
structure. The Policy Budgets and Policy Accountability reform aimed to 
provide Parliament with a more transparent budget document and clearer 
information about the results of government actions. This reform was an 
initiative of the Lower House of the Dutch Parliament.
    In addition, since 1995 the Netherlands has developed a system of 
interdepartmental policy reviews. Policy reviews are proposed by the 
MOF and approved by the Cabinet and the respective line ministries. A 
list of proposed reviews is included in the September budget 
memorandum. These reviews are conducted by small working groups 
including representatives from the relevant line ministries, the MOF, 
and external experts. All reports are published and submitted to 
Parliament. Initially these reviews concentrated on efficiency with a 
mandatory 20% saving; however, as the economic situation improved the 
mandatory cut was abolished and the reviews focused more on 
institutional reform.
United Kingdom
    The United Kingdom first introduced the comprehensive spending 
review in 1998 and repeated the exercise in 2000, 2002, 2004 and 2007. 
This biennial approach aims to reallocate money to key priorities and 
to improve the efficiency and delivery of public services. After a 
review of existing departmental spending, each department develops a 
three-year spending plan and a public service agreement (PSA). The 
Treasury negotiates with ministries regarding their key performance 
targets for the next three-year period and these targets are included 
in their public service agreements. PSAs contain measurable targets for 
a whole range of government objectives. The current agreements mainly 
focus on outcome targets, although there are still a few output 
targets. In addition to the PSA, each department produces a technical 
note stating how the targets will be measured and a delivery plan 
explaining how it plans to achieve the targets. The technical note is 
published but the delivery plan is not.
    The PSA also states who is responsible for the delivery of the 
targets--usually the relevant secretary of state. In contrast to 
Australia, Denmark, and Sweden, this is a top-down, centrally-driven 
performance system. The development and evolution of the PSA framework 
has been led by the Treasury. All performance agreements and 
ministerial targets are agreed with the Treasury. Performance 
information is discussed as part of the spending review negotiations 
between the Treasury and ministries, although there is no automatic 
link between results and resource allocation.
    In the United Kingdom, key objectives and targets are integrated 
into the decision-making process at a high political level. There is a 
special cabinet subcommittee on public services and public expenditure, 
which is chaired by the Chancellor of the Exchequer. This committee 
discusses progress against targets and key strategic objectives and 
challenges.

 THE BENEFITS AND CHALLENGES OF THE USE OF PERFORMANCE INFORMATION IN 
                           THE BUDGET PROCESS

    Countries have reported a number of benefits from the use of 
performance information:
     It generates a sharper focus on results within the 
government.
     It provides more and better information on government 
goals and priorities, and on how different programmes contribute to 
achieve these goals.
     It encourages a greater emphasis on planning--especially 
when part of a medium term expenditure framework--and acts as a 
signalling device that provides key actors with details on what is 
working and what is not.
     It improves transparency by providing more and better 
information to parliaments and to the public. (Twenty-four out of 30 
OECD countries provide information on performance results to the 
public.)
     It has the potential to improve the management of 
programmes and efficiency.
    Advocates claim that the provision of objective performance 
information facilitates better decision making for the efficient use of 
resources, programme management, central resource allocation, and 
expenditure prioritisation. In sum, supporters claim that the use of 
performance information in budgetary decision-making can contribute to 
budgetary goals of improving productive efficiency, allocative 
efficiency, and even aggregate fiscal discipline.
    Despite these reported benefits, OECD countries continue to face a 
number of challenges with the development and use of performance 
information in the budget process including:
     how to improve the use of performance information in 
budgetary decision-making;
     how and if performance information should be related to 
resources;
     how to improve the measurement of activities;
     how to improve the quality of information; and
     how to get politicians to use performance information in 
decision making.
    Country experiences have shown that the existence of a procedure to 
integrate performance information into the budget process is a 
necessary but not sufficient condition to ensure its use. Other factors 
influencing use include the quality of the information, the 
institutional capacity of the MOF and the spending ministries, and the 
political and economic environment.

  OECD GUIDELINES ON DESIGING AND DEVELOPING BUDGET SYSTEMS THAT USE 
                        PERFORMANCE INFORMATION

    Despite these challenges, countries are not discarding the use of 
performance information, but continue to evolve new approaches to using 
the information. The OECD has developed general guidelines for 
countries as they adopt and evolve initiatives to improve the use of 
performance information in budgeting processes. Some important factors 
to consider in this respect are:
     There is no one model of performance budgeting; countries 
need to adapt their approach to the relevant political and 
institutional context.
     A whole-of-government planning and reporting framework is 
important.
     Performance information should be integrated into the 
budget process.
     Designing government-wide systems that automatically link 
performance results to resource allocation should be avoided. Such 
automatic linkages distort incentives, ignore the underlying causes of 
poor performance, and require a very high quality of performance 
informaiton that is rarely available.
     Meaningful and accountable performance information 
requires reliable output and outcome data that are continuously 
updated.
     Timely and straightforward assessments of performance 
information should be carried out independently of the spending 
departments and be supported by external expertise.
     The support of political and administrative leaders is 
vital for implementation.
     The staff and resource capacity of the ministry of finance 
and spending ministries is critical.
     Reform approaches need to be adapted to evolving 
circumstances.
     It is important to develop incentives to motivate 
politicians and civil servants to change their behaviour. Regardless of 
the type of performance information (evaluations or performance 
measures), consideration needs to be given to whether and how it will 
be used by decision makers. It is a matter not just of process, but 
also of having the right incentives to motivate decision makers to use 
the information.

                              CONCLUSIONS

    Previous incarnations of performance budgeting in many countries 
began with expectations that were unrealistic, thus ensuring 
disillusionment when the predicted results failed to materialise. It is 
important from the outset to manage expectations in terms of the length 
of time it takes for the reforms to produce results. There are no quick 
fixes. A long-term approach and patience are necessary as countries go 
down this road. Some countries estimated that it took 3-5 years to 
establish a government-wide performance measurement framework. There 
can be expectations that performance budgeting will create an 
environment of rational decision-making and will enable governments to 
financially reward good performance and punish bad. While this may be a 
simple and appealing idea, it does not take account of the fact that 
budgetary decision-making takes place in a political context, or that 
the issues and context surrounding budget decisions are complex. In 
most cases such an approach is not desirable. The more realistic 
expectation is that countries will engage in performance-informed 
budgeting.
    In the governments of OECD member countries, a great deal of the 
annual budget process remains incremental and inputs still play a key 
role. Results information will never completely replace inputs. There 
is a trend towards reforms that are, however, slowly shifting the 
thinking of decision makers at all levels--politicians, the MOF, 
spending departments and agencies, and the general public--towards a 
greater focus on results, and there is a clearer understanding of the 
need to see public policy and government actions in terms of achieving 
results. Despite the challenges encountered, countries are continuing 
to move forward with reforms to improve the use of performance 
information in budgetary decision-making. As long as citizens demand 
results from their governments for their tax dollars, there will be a 
continuing need for performance information.

    In addition, before we proceed to testimony, we welcome--I 
have got it, I would like to ask unanimous consent that all 
members be allowed to submit an opening statement for the 
record at this point.
    Without objection, so ordered.
    [The prepared statement of Mr. Smith follows:]

 Prepared Statement of Hon. Adrian Smith, a Representative in Congress 
                       From the State of Nebraska

    Good morning and thank you, Chairman. I am so pleased we are 
holding this hearing today.
    We have an obligation to exercise accountability and fiscal 
responsibility in government spending. As this Committee plans for 
future federal budgets, the ideas and information expressed to us today 
are of utmost importance.
    While we address the many fiscal challenges we face, we must assess 
the value of the information available to us. This hearing will give us 
the opportunity to explore ``performance-based budgeting.'' The models 
and results of ``performance-based budgeting'' could serve as a 
valuable tool for setting our spending priorities. Too often in 
Congress, we debate, vote on, and pass--and in some cases regrettably 
so--legislation about which we have no measure of its efficacy. And Mr. 
Chairman, that just has to change.
    Financial discipline is one of my greatest concerns. By serving on 
this Committee, my goal is to add fiscal restraint to the budget 
process. Congress must promote policies which will make government 
programs better, more efficient, and more sustainable for the long-
term. Commonsense tells us this can only be achieved with an 
accountability system in place. Adding accountability to the budget 
process will combat waste, fraud, and abuse while we work to control 
the growth in government and restrain spending.
    I appreciate the Committee for holding this hearing as an important 
step to meeting our goals.
    Chairman, I look forward to continuing to work with you, and I 
thank you for your time.

    Chairman Spratt. Let me welcome my witnesses this morning 
and say that if you have prepared written statements, we will 
treat your witness statements as made part of the record. You 
can summarize them as you please, but the floor is yours.

STATEMENTS OF CLAY JOHNSON III, DEPUTY DIRECTOR FOR MANAGEMENT, 
 OFFICE OF MANAGEMENT AND BUDGET; DAVID M. WALKER, COMPTROLLER 
 GENERAL OF THE UNITED STATES, U.S. GOVERNMENT ACCOUNTABILITY 
  OFFICE; AND PETER R. ORSZAG, DIRECTOR, CONGRESSIONAL BUDGET 
                             OFFICE

    Chairman Spratt. We will proceed first with Mr. Johnson. 
Thank you very much for coming. We look forward to your 
testimony.

                 STATEMENT OF CLAY JOHNSON III

    Mr. Johnson. Chairman Spratt, Ranking Member Ryan, members 
of the committee, thank you for having me. We all want the 
Federal Government to be effective. We want to spend money on 
programs that work. We want not to spend money or waste money 
on programs that don't work or can't be made to work. And we 
want to get what we pay for. We all want to be held accountable 
for this. We are here to do this work. In fact, I think we 
should all publicly declare that we want to be held accountable 
for this.
    To do this, to deliver on this, for every Federal program 
we must have clear outcome goals and we must know how we are 
performing relative to those goals. We can't manage programs 
unless we know what we are trying to manage to.
    Today, every Federal program, and therefore every Federal 
agency, has better outcome goals and more information about its 
performance than ever before. It is very good information. It 
is not perfect, nor will it ever be perfect. Performance 
information, by its very nature, we will get smarter and 
smarter about--as we learn more and more about programs and 
develop ways to capture relevant performance information. But 
this information is very, very good. It is subjective, it is 
consistent with the information we have for like programs. It 
is appealable by those who disagree with it, and it is reviewed 
for accuracy and quality on a regular basis.
    And by the way, Chairman Spratt, let me add that it is not 
separate and apart from GPRA. It is very consistent with GPRA. 
It is a more refined way of dealing with--in our opinion, 
dealing with program performance. It is more information with 
which to comment and report on how programs are performing. But 
it is not something totally different. Because we have this 
information, we can now, and only now, link managers' 
evaluations to the performance of the programs. We can more 
formally focus every manager and employee on the desired 
outcomes of the program they work on, which means we can more 
purposefully and intelligently pursue the goal of greater 
effectiveness.
    In fact, we are going to talk here about the use of this 
information in the budgeting process. But I believe the most 
important, most valuable use of this quality goal and 
performance information is to help managers and employees cause 
programs to be more effective.
    But in addition to helping programs be more effective, we 
can also use this information to make smarter budget 
appropriation decisions, which is the subject of this hearing. 
We can, for instance, invest more in programs known to be 
effective and less in those that are known not to work. We can 
decide not to create and fund new programs that duplicate 
existing programs, especially if they are not known to work. We 
can decide to increase funding for a less effective program 
only if its managers have a plan to fix it. And we can look at 
our spending by key indicator or strategic goal or desired 
outcome, across all agencies, to more intelligently debate 
relative spending levels and how to invest effectively and 
efficiently to accomplish our goal.
    None of this is possible without good goals and good 
information about how we are performing relative to those 
goals. This is not something that is nice to have; this is 
something that we must have for you all to do your work and for 
us to do our work to drive behavior and performance in the 
executive branch.
    I look forward to working with you and answering any 
questions you have at the end of this hearing. Thank you very 
much for having me here.
    Chairman Spratt. Thank you, Mr. Johnson.
    [The prepared statement of Clay Johnson follows:]

   Prepared Statement of Hon. Clay Johnson III, Deputy Director for 
           Management, Office of Management and Budget (OMB)

    Since and even before enactment of the Government Performance and 
Results Act (GPRA) in 1993, Congress and the Executive Branch have been 
trying to discern just what we're getting for the tax dollars we spend 
and how to get even more. The first step is to have good information 
about what works and what doesn't.
    GPRA implementation in the 90's laid a strong foundation for 
strategic and performance planning in the Federal Government. It 
required agencies to set goals and report on whether they were 
achieving them. This was the beginning of a candid assessment of 
government's performance, but any connection with the budget was 
remote. At the time, the Government Accountability Office (GAO) found 
that, although most agencies (30 of 35 reviewed) defined some type of 
relationship between the program activities of their proposed budgets 
and the performance goals in their annual performance plans, most plans 
did not explain how spending on the program would help it achieve 
performance goals.\1\ More importantly, managers reported to GAO that 
their agencies were not well-positioned to answer the question of 
whether their programs produced real results.\2\
---------------------------------------------------------------------------
    \1\ Government Accountability Office, ``Performance Budgeting: 
Initial Experiences Under the Results Act in Linking Plans with 
Budgets.'' GAO/AIMD/GGD-99-67 (April 1999).
    \2\ Government Accountability Office, ``The Government Performance 
and Results Act: 1997 Governmentwide Implementation Will be Uneven.'' 
GAO/GGD-97-109 (June 1997).
---------------------------------------------------------------------------
    With that constructive criticism in mind, the Administration 
developed a tool--the Program Assessment Rating Tool--to assess 
performance at the program level and use the information to identify 
ways to help programs improve. Using the performance goals and 
information developed to implement GPRA, agencies and OMB assessed the 
available evidence and asked of every program:
     Is its purpose clear and is it well-designed to achieve 
that purpose?
     Does it have clear, outcome-oriented long-term and annual 
goals with aggressive targets?
     Is it well managed?
     Does it achieve its goals?
    Over the last six years, agencies and OMB have assessed more than 
1000 programs that total $2.6 trillion in spending. These assessments 
show that 78 percent of the programs assessed are considered 
``performing.'' Fifty percent of programs received the impressive 
ratings of Effective or Moderately Effective. Only 3 percent were 
demonstrably Ineffective.
    Is the information valid and reliable? Yes. The questions asked are 
those a reasonable person would ask to assess a program's performance. 
There are clear, written rules that guide what evidence is required to 
get a ``yes'' answer to a PART question.\3\ Agencies and OMB agree on 
the right answer based on the available evidence. Then all PART 
questions are reviewed by experts in the PART process to ensure they 
are consistent with the rules. Agencies then have an opportunity to 
appeal questions to a high-level appeals board of a selection of agency 
chief operating officers. All of these factors contribute to the 
validity and reliability of PART information. We will continue every 
year to improve the quality of program goals, the consistency of PART 
answers, and the accuracy of performance information.
---------------------------------------------------------------------------
    \3\ Office of Management and Budget, ``Guidance for Completing 2007 
PARTs.'' Program Assessment Rating Tool Guidance No. 2007-02 
(www.omb.gov/part/fy2007/2007--guidance--final.pdf) (January 29, 2007).
---------------------------------------------------------------------------
    As part of our ongoing effort to ensure PARTs reflect the most 
accurate picture of a program's performance, agencies and OMB over the 
next several months will systematically review all of the assessments 
to ensure the program goals are clear and outcome-oriented; the 
performance information is valid; and the program improvement plans are 
aggressive.
    GPRA required public planning and reporting. PART results, the 
evidence on which they are based, the results each program achieves, 
and the amount of money spent and requested for each program, are all 
posted on ExpectMore.gov. We must continue to be candid about how well 
(or not) programs are performing and what we are doing to improve if we 
want to hold programs and agencies accountable for achieving outcomes 
that matter to the American people.
    We want to incorporate performance information into budget 
justifications so Congress finds that it is a valuable factor in the 
decision-making process. Finding the best way to do this remains a 
challenge. The report accompanying the FY 2008 Financial Services and 
General Government Appropriations Bill criticized agency budget 
justifications for being ``filled with references to the [PART], 
drowning in pleonasm, and yet still devoid of useful information.'' \4\ 
[I'm sure the Members of the Committee know this, but I had to consult 
the dictionary to find that pleonasm means ``the use of more words than 
are necessary to express an idea; redundancy.''] But the Labor, Health 
and Human Services Appropriations Bill cites the Department of 
Education's Budget Justification as a model.\5\ Education's budget 
justification integrates performance information in three different 
ways:
---------------------------------------------------------------------------
    \4\ United States House of Representatives Committee on 
Appropriations, Report Accompanying the FY 2008 Financial Services and 
General Government Appropriations Bill, House Report 110-207 (June 
2007).
    \5\ The Departments of Labor, Health and Human Services, and 
Education FY Appropriations Act for FY 2006, Public Law 109-149, 
Section 107.
---------------------------------------------------------------------------
     A separate 25-page Performance Budget tab displaying 
budgets, programs, and key performance measures by strategic plan goal;
     A detailed discussion of performance information--both 
GPRA and evaluation--in each program justification; and
     References to performance information and PART in the 
program account narratives.
    Education's budget staff has briefed every agency on its 
justification, and OMB has encouraged others to emulate it, provided 
they consult with appropriators first. In an effort to encourage 
greater use of performance information in Congressional decision-making 
and address a helpful recommendation from GAO, agencies now also brief 
their House and Senate appropriations and authorizing committees each 
year on completed and planned PART assessments.
    Of course, it was never our goal, nor should it ever be, to make 
decisions mechanically based on performance information. It is not true 
that a program's budget should be cut just because it is not currently 
achieving results. Such a program may require additional resources in 
order to improve. Likewise, just because a program is achieving results 
does not mean it warrants an increase in funding. Such a program may 
have outlived its usefulness.
    In some cases, poor performance has been the basis for a request 
for additional resources. For instance, the National Park Service 
Facilities Maintenance was unable to demonstrate results when it was 
first assessed in 2002. The program then began using a facility 
condition index to assess the state of its facilities and prioritize 
investment decisions for addressing identified deficiencies. Use of the 
index also helped the program provide a better justification for 
additional resources. This program has received funding increases and 
is steadily improving the quality of its facilities. Additional funding 
for the Earned Income Tax Compliance Program was proposed to improve 
its efforts to reduce improper payments. Improper Payments in this 
program are estimated at between 23 and 28 percent of total program 
payments.
    Our ultimate goal is to make programs work better. We believe that 
making information about a program's performance more of a factor in 
all deliberations about the program will help it and other programs 
become more effective.
    We have more information about what works and what doesn't today 
than ever before and we are always working diligently to improve its 
quality. Performance information for both agencies and programs is 
available to the public on ExpectMore.gov. We can and must make greater 
use of this information in our deliberations about programs, including 
funding and other issues affecting program performance. I look forward 
to working with Congress, especially Members of this Committee, to find 
ways we can be ever smarter about spending the taxpayers' money.

    Chairman Spratt. Now, General David Walker, the Comptroller 
General of the United States.

                  STATEMENT OF DAVID M. WALKER

    Mr. Walker. Thank you, Chairman Spratt, Ranking Member 
Ryan. I am pleased to be here today to discuss performance 
budgeting as a way to help the government meet the pressing 
challenges of the 21st century by prompting a much needed and 
long overdue review of Federal activities and programs.
    The Federal Government is in a period of profound 
transition. It faces a range of challenges and opportunities 
that can enhance performance, ensure accountability and better 
position the Nation for the future. A number of overarching 
trends, including the Nation's long-term fiscal imbalance, 
drive the need to reexamine what the Federal Government does, 
how the Federal Government does business, and who does the 
Federal Government's business.
    The term ``performance budgeting'' encompasses a range of 
approaches, activities and processes, but they all have one 
idea in common and that is more explicitly linking resources to 
results. And in all candor, the Federal Government does an 
extremely poor job of that. As it holds promise, this is a 
means for facilitating reexamination of the government.
    Through the President's management agenda and its related 
initiatives with the Office of Management Budget's Program 
Assessment Rating Tool, the administration has taken steps in 
the right direction by calling attention to successes and 
needed improvements in Federal management and performance.
    As we previously noted, PART itself has certain strengths 
and weaknesses. The weaknesses need to be addressed and the 
strengths need to be capitalized on. Whatever approach is taken 
in the future, however, in our view it will be important not 
only to look at programs on activities that run through the 
spending side of the budget, but also to look at those policies 
that run through the tax side of the budget. Any reexamination 
or performance budgeting effort that fails to include tax 
expenditures in the review of Federal activities and policies 
and whether or not they are achieving their intended goals will 
fall short of its full potential and will fall short of our 
Nation's needs.
    As I have previously testified before this committee, known 
demographic trends and rising health care costs are major 
drivers to our Nation's large and growing structural deficits. 
And I am going to take you through three quick graphics, two of 
which are in your material. The first one is not.
    First graphic, please.

    
    
    This represents the long-run fiscal situation for the 
Federal Government based upon CBO's baseline extended, or, in 
my view, a more realistic alternative simulation which assumes 
no reform of Social Security and Medicare; discretionary 
spending growing by the rate of the economy; and historical tax 
levels over the long term. Reality may be somewhat in between, 
but either one is unacceptable.
    Next please.

    
    
    The next one is on page 11 of my testimony. It shows you 
what has happened since 1982 in constant dollars with regard to 
mandatory spending, which is the green line, which is growing 
out of control, represents 62 percent of the Federal budget 
last year. That is on auto pilot. The blue line, which is 
discretionary spending, which Congress has responsibility for 
dealing with every year, and the red line which represents the 
sum of tax expenditures or the revenue losses associated with 
them. And as you will see, the tax expenditures in some years 
exceed discretionary spending and yet they are not in the 
financial statements, they are not part of the budget process, 
they are not part of the appropriations process. They are off 
the radar screen. And it is important that they be on the radar 
screen because they cost money and they may or may not be 
working.
    Next, please.

    
    
    It is important to keep in mind that these involve $800-
plus billion in the aggregate. Health care alone is close to 
$200 billion for 2006. Now, these are based upon Treasury 
estimates after the end of 2006. And in this particular case, 
we unbundled the defined contribution plans from the defined 
benefit plans. So If you add those together, you will see that 
the pension tax preferences would be number two to health care 
on that basis. But the bottom line is these numbers are just 
too big to be off the radar screen.
    Accordingly, to reexamine the base of all major Federal 
spending and tax programs, policies, and activities by 
reviewing the results and testing their continued relevance and 
relative priority for our changing society is important, and 
the sooner we start doing it, the better. Reexamination is not 
a one-time activity. It must be a continuing process and we 
should expect even the initial round to take several years.
    We could accomplish more if the intent of the Government 
Performance and Result Act, which was to develop a 
comprehensive governmentwide performance plan, was fully 
implemented. We think more needs to be done in that regard. But 
in doing it, we need to move beyond programs by programs and 
agency by agency to looking more horizontally and to also 
considering tax preferences.
    I think it is also important to keep in mind that the 
United States is one of few major industrialized nations 
without a set of key national outcome-based indicators. What do 
I mean by that? Economic, safety, security, social, and 
environmental outcome-based indicators in order to be able to 
assess which programs and policies are working and which aren't 
working. How do we stand as a Nation? How are we trending? And 
how do we compare to others?
    And when you look at some statistics published by the OECD, 
of which the United States is one of 30 member countries, on a 
portfolio of indicators, the United States ranks 16 out of 28 
on outcome-based indicators. As a certified public accountant, 
I can tell you that is below average.
    We can and we must do better. And I think performance 
budgeting is a way that can help us see the way forward and 
make some very tough choices.
    In summary, performance budgeting is one means that can 
help us be able to put our Nation on a more prudent, 
sustainable fiscal path, to separate the wheat from the chaff 
with regard to which programs and policies are working and 
which aren't, and make sure we are targeting our resources and 
preferences to accomplish the best greater good in a way that 
does not undercut the economy. But I cannot stress enough that 
performance budgeting cannot be merely an executive branch 
action. The legislative branch has to be a key partner in this 
effort. And for performance budgeting to work, the information 
must be not only useful, it must be used. And if it is not, it 
is a waste of time.
    Thank you, Mr. Chairman.
    Chairman Spratt. Thank you, General Walker.
    [The prepared statement of David M. Walker follows:]

    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    Chairman Spratt. And now the Director of the Congressional 
Budget Office, Peter Orszag.

                  STATEMENT OF PETER R. ORSZAG

    Mr. Orszag. Thank you very much, Mr. Spratt, Mr. Ryan and 
other members of the committee. Both economic and common sense 
suggest that a program's budget should be linked to its 
effectiveness in achieving its objectives. The concept of 
performance-based budgeting, though, has been applied mostly to 
discretionary spending, which represents less than 40 percent 
of overall Federal expenditures. So, like Mr. Walker, my 
written testimony extends the concept of performance budgeting 
to two parts of the Federal budget that have not been examined 
as closely from that perspective: health insurance and tax 
expenditures.
    In both cases, the amount of Federal funds at stake is 
substantial, and important questions exist about the cost 
effectiveness of those expenditures.
    First with regard to health care, as I have said over and 
over again, rising health care costs represent the central 
fiscal challenge facing the Nation. My written testimony 
includes the chart that I have always used before this 
committee, but just for a change of pace I won't actually put 
it up.
    If health care costs continue to grow at the same rate over 
the next four decades as they did over the past four decades, 
Medicare and Medicaid would rise from under 5 percent of the 
economy today, to 20 percent by 2050; yet very little analysis 
is undertaken of whether that spending is generating 
corresponding gains in the health of enrollees, which 
presumably is the ultimate objective of the programs.
    Many treatments improve enrollees' health, and the benefits 
suggest that health spending, on average, improves health 
outcomes. But in many cases, such spending is not cost 
effective and in many cases it may not even improve health. One 
reason is that relatively little rigorous evidence is available 
about which treatments work best for which patients, or whether 
the benefits of more expensive therapies warrant their 
additional costs.
    Although estimates vary, some experts believe that less 
than half of all medical care is based on or supported by firm 
evidence about its effectiveness. Much of the research that has 
been done about Medicare spending and its impact on health has 
focused on the traditional fee-for-service portion of the 
program that serves the vast majority of its enrollees. But the 
concepts behind performance budgeting could also be applied to 
the Medicare Advantage component of the program. The data 
currently collected, however, are not sufficient to do so.
    Medicaid spending has also received relatively less 
attention. Like with Medicare, Medicaid spending at the State 
level varies substantially, even among enrollees who have 
qualified for the program, for the same reason. More research 
on the source of variation in the program's costs and its 
impact on enrollees' health is warranted.
    In sum, we are spending a substantial amount of money on 
health care. It is the central long-term fiscal challenge 
facing the Nation, and yet we are doing too little to examine 
what we are getting in return.
    The second area of the Federal budget that is often not 
examined from a performance budgeting perspective involves tax 
expenditures. Mr. Walker already laid out some of the amounts 
involved and they are obviously very significant. In each of 
the major cases, the reduction and receipts gives the 
appearance of a reduced impact on the Federal budget and on the 
economy. Indeed, most presentations of the budget omit any 
mention of tax expenditures. But tax preferences are 
effectively equivalent to collecting taxes at ordinary tax 
rates on the full potential tax base and then subsidizing the 
preferred behavior through outlays. Because selective tax 
reductions operate as expenditures for specific economic 
activities, their effectiveness can and should be evaluated in 
the same way as is done for spending programs.
    Given the size of many tax expenditures, it is striking 
that they are subjected to little analysis of their 
effectiveness in achieving their objectives. In many cases, the 
specific outcome of the tax expenditure that is desired is 
unclear, and may even conflict with objectives of other 
policies. In a number of instances where the policy goal is 
clearer and not in conflict with other policies, the tax 
incentives do not appear to yield their desired effect in a 
cost-effective manner. For example, significant empirical 
questions exist about whether tax preferences for certain kinds 
of saving vehicles such as IRAs and 401(k)s have significantly 
boosted private savings or merely subsidized saving that would 
have occurred in the absence of those incentives.
    In addition, many tax expenditures are structured in a 
relatively inefficient way. In particular, most tax 
expenditures are delivered in the form of a deduction or 
exclusion, which links the size of the tax expenditure or the 
value thereof to one's marginal tax bracket. However, unless 
one believes that there is a differential response by income 
category or broader social benefits that vary by income 
category when people do respond, from an economic efficiency 
perspective, it does not make sense to vary the subsidy rate 
per dollar of activity in any manner, either up the income 
distribution or down the income distribution. So from an 
economic efficiency perspective, unless you have evidence to 
the contrary that there is differential response rates, the 
most efficient approach to delivering a subsidy through the Tax 
Code is a uniform credit that does not vary with the 
household's income. My written testimony discusses that at more 
length.
    I would also note that perhaps the most prominent example 
of a tax expenditure that appears to be inefficient is the 
exclusion for employer-provided health insurance, and my 
written testimony discusses that issue in more detail also.
    To sum up, the growth of health care spending is the 
central long-term fiscal challenge facing the Nation, and 
existing tax expenditures entail a substantial reduction in the 
Nation's tax base. In both cases, Federal policy would be 
improved by applying principles of performance budgeting, 
evaluating whether the benefits derived warrant the resources 
provided. Thank you very much.
    Chairman Spratt. Thank you very much, Dr. Orszag.
    [The prepared statement of Peter R. Orszag follows:]

    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    Chairman Spratt. Let me put the question to all three on 
our panel. Is there some better systematic way that you can 
gain congressional review? Is part of the problem here 
overkill?
    For example, Director Johnson, if your information largely 
comes from PART and from GPRA, and it comes with the budget--
$2.8 trillion budget--it gets buried in a lot of budget detail. 
Do you think it would serve the interest of greater scrutiny if 
the information were prepared in some different form so it came 
to us in what was more discernible, easier to find, or 
otherwise it is scattered throughout your entire budget?
    Mr. Johnson. Yes, sir, I do. Every subcommittee is 
different. For instance, the TUB committee's budget and 
appropriations that deal with the Education Department welcome 
a lot of performance information. It is really the model for 
all of the other agencies. And yet there are other 
subcommittees that state in writing that they don't want to be 
confused by my words, all this performance information. So it 
is not a universal acceptance or rejection. It varies by 
committee, which causes to us think we have an education 
challenge before us to explain to every subcommittee how this 
information can help them do a better job of budgeting or 
appropriating.
    And every year when we talk about how to put our budget 
together, we talk about how to present this, how to focus on 
desired outcomes. The goal is not to spend $2.X billion. It is 
to accomplish, A, B, C, D. By the way, the cost to accomplish 
that is whatever it is. That ought to be the mindset, in our 
opinion. And our budgets, in our opinion, ought to be laid out 
to encourage that kind of thinking. But some committees--some 
subcommittees welcome that approach, are comfortable with that 
approach and others are not.
    Chairman Spratt. Are you familiar with the Selected 
Acquisition Report in the Department of Defense?
    Mr. Johnson. No, sir, I am not.
    Chairman Spratt. Do you have any experience with--I am sure 
you have had--accounting variation reporting systems whereby 
you establish a baseline for costs, schedule, and performance 
and then measure against that over time?
    Mr. Johnson. Well, that is the kind of rigor and approach 
that I--that is what I spend all of my time doing, is trying to 
create that sort of mindset and approach throughout the Federal 
Government.
    Chairman Spratt. Let me suggest that you take a look at the 
SAR, Selected Acquisition Report, created in about 1970. And 
somebody added the phrase--I think it was General Walker--if 
you want it to be useful, it has got to be used. It is a 
classic example of something that hasn't been intensively used, 
and therefore it has not become any more useful than it was, 
very much so, when it was first introduced.
    Mr. Johnson. Let me make one comment on that. I have every 
spring for the first 4 years that I was in this job, gone out 
and did focus groups with career managers and talked about 
goals and having performance information and outcome goals and 
so forth. And what they told me every year, no matter who I 
talked to, we have always had some description of what our goal 
is. And sometimes it has been quantifiable, sometimes it has 
been general, sometimes it has been outcome, sometimes it has 
been output. But we have never, ever been held accountable for 
achieving them. So that is this--is it useful versus used, 
their comment to me is that this information has never been 
used to drive performance.
    Chairman Spratt. One of the recommendations made a long 
time ago--in fact, I wrote an article in General Walker's 
quarterly about it--is that somehow or another for weapon 
systems in the Pentagon,--there are about 40, I think, or about 
50 that track with the SAR. The principals here and at DOD need 
to sit down early in the life of a program when it is about to 
take off and go into engineering development and establish the 
baselines that are pertinent to that particular system, things 
that you recognize the vulnerabilities that need to be watched, 
so that you don't have one template that applies to 40 
different systems that are very different and diverse.
    It would seem to me that on major programs you could have 
the same sort of attitude, that you would sit down with the 
committee and you would agree that we are going to watch these 
things, we are going to give you a periodic report on this. If 
there is going to be cost growth, it could happen here, watch 
it early. We don't have any of that kind of interactive 
activity at all.
    Mr. Johnson. There is information, not of the sort you 
want, but what you just described is a great goal to shoot at. 
There is information now--every program--the evaluation of 
every program--overall evaluation and then the recent 
performance of that program relative to its goals for all 
1,000-plus programs; $2.8 trillion is very public on 
expectmore.gov, the Web site.
    What you are talking about is regular reports; here is 
where we are running behind schedule, ahead of schedule, 
whatever, and here is what we are doing to get back on track.
    Chairman Spratt. We would call this performance budgeting, 
but part of it is just reporting anomalies that will require 
further investigation should be called to our attention, like 
Dr. Orszag is pointing to the fact that a coronary bypass has a 
fourfold variation around the country. There needs to be an 
explanation of that, or the variation of Medicare expenditure 
per capita ranges from 4,500 to 11- or $12,000 per person. That 
is maybe not really performance, but it is the sort of thing 
where the executive agency says to the Congress, you should use 
your oversight and investigative powers to look into something 
like that.
    Mr. Johnson. Right. We would welcome developing that and 
figuring out how to interact with Congress to provide exception 
reporting, or as you say, anomalies, to help you all better 
focus on how these programs are performing.
    Chairman Spratt. General Walker.
    Mr. Walker. Mr. Chairman, in addition to what you are 
talking about, I would like to raise it up a level that I think 
we need to focus on and I know you are interested in. The 
Federal Government spends $2.7 trillion a year. It issues 
thousands of pages of regulations a year. It forgoes revenues 
of $800 to $900 billion a year in tax preferences. And for the 
most part, it has no idea whether those programs, policies, and 
regulations are achieving the desired outcomes or not.
    And one of the things that has to happen, I believe, is the 
Congress, which has the constitutional responsibility for 
appropriations and which has this budget process which this 
committee is responsible for, we need to get back to basics and 
we need to say what type of outcomes are we trying to achieve 
as a Nation when legislation is authorized, reauthorized, when 
the budget process is gone through, when the appropriators 
allocate funds and when the oversight committees end up 
conducting their oversight operations.
    And that brings me back to the need for a set of key 
national outcome-based indicators: economics, safety, security, 
social, environmental. Thirty-eight percent of the budget last 
year, as you know, was discretionary. Sixty-two percent was on 
auto pilot. That has got to be on the radar screen. We have got 
to analyze that too. Furthermore, none of that counts tax 
expenditures. That has got to be on the radar screen too. We 
haven't done the basics. We haven't stepped back and said what 
are we trying to accomplish; are we doing that? And as a 
result, what the Federal budget is today in the government, it 
is an amalgamation of programs, policies, functions and 
activities from decades, many of which may be outdated, many of 
which may not be working. And given our challenges for the 21st 
century, we really need to focus at that level as well as 
making sure that we have, as you are saying, we are using the 
stuff that is being done now, we have a consistent approach and 
hopefully a streamlined and more results-oriented approach to 
the current processes that are here now.
    Chairman Spratt. Dr. Orszag?
    Mr. Orszag. Again, I think an aggregate perspective is 
important. But if there were a single thing that policymakers 
could do to improve from a performance basis what we are 
getting today and in the future, it would be examining what we 
are getting in return for our health care expenditures, 
including the variation that you noted, both in Medicare and 
Medicaid. And since so much of the growth in the future is 
going to come from increased expenditures in that area, if 
there were just one thing to do it, it would be to make sure 
that we were getting value out of those dollars. And that would 
be a substantial improvement relative to where we are today.
    Chairman Spratt. Thank you all very much.
    We have just had an ominous bell go off on a motion to 
adjourn. So that is not a good sign.
    But I am going to turn to Mr. Ryan, and let us make up as 
much as we can before we have to run over and vote and come 
back. Thank you, Mr. Ryan.
    Mr. Ryan. Thanks, Mr. Chairman. I will try to be fairly 
quick because we have 13 minutes left.
    Let me say for the record, I bristle at the notion or the 
term ``tax expenditure.'' it kind of more or less assumes that 
this is the government's money unless we benevolently expend it 
back to people.
    Be that as it may, let us focus on the health one. And if 
we are going to spend our time talking about this Nation's 
fiscal challenges, we probably should spend three-fourths of 
our time talking about health care.
    And so what I want to ask from you--and I know Dr. Orszag 
and General Walker, you spent a lot of time on this--what do 
you think we can do to address the root cause of health 
inflation, number one?
    Number two, does the health exclusion, that tax expenditure 
contribute to the problem of health inflation and is it really 
the primary reason which created the third-party payment system 
in the first place? Which I would argue is largely the biggest 
contributor towards health inflation.
    And then I have a question about TRIA. And if there is 
time, I wanted to ask you, Director Johnson, about how does the 
PART program interact with the appropriations process? Do you 
come and talk to the appropriators and bring them through these 
things work or aren't working, and is there a connection 
between how we appropriate and the analysis you are doing? So 
let me stick with that. And then a quick TRIA question if I 
have time.
    Mr. Walker. That is a lot. First let me say with regard to 
the tax preference, I think it is part of the problem. The fact 
that we spend almost--the fact that we forego revenues--and by 
the way, tax expenditure is the Treasury's term. That is a 
generally accepted term, not mine. The fact that we forego 
almost $200 billion a year in tax preferences in the health 
care area, I think is part of the problem because it 
desensitizes people to the true cost of health care. They don't 
see it on their tax return, they don't see it on their W-2. It 
is the fastest growing part of compensation expense. It is a 
huge competitiveness problem for American businesses as well as 
a fiscal problem for the Federal Government.
    So, yes, I think it is part of the problem. Secondly, I 
think part of the answer is better targeting of tax preferences 
and better targeting of government subsidies in the area of 
Medicare, et cetera. It is one thing to be eligible for a 
program or a preference. It is another to differentiate based 
upon your income and your ability.
    Thirdly, we need national evidence-based practice standards 
in health care to reduce costs, to improve consistency, to 
enhance quality and to dramatically reduce litigation risk. And 
the Federal Government ought to pave the way through the 
programs that it has direct responsibility for. We have VA 
health care. We have health care that deals with civilian and 
military employees. We have an ability to pave the way and to 
lead by example there, and that is critically important.
    Mr. Ryan. Dr. Orszag, at the end of your answer, if you 
could comment on the contributing factor to the tax expenditure 
and health care. We just passed a bill yesterday on TRIA, the 
Terrorism Risk Insurance, which I think you scored at $8.4 
billion, but it left the floor with quote-unquote no score 
circumventing the PAYGO process. If you can comment on whether 
that is going to cost the Treasury money or not, I would 
appreciate that as well.
    Mr. Orszag. Okay. Why don't I start with that? Insurance 
costs money because it provides some value to the insured 
households and businesses, even in the event that the thing you 
are insuring against doesn't occur. So when you insure your 
house against a fire, there is some value that is being 
provided and some expected cost of the insurance company, even 
if your house doesn't burn down.
    And in the context of terrorism reinsurance, we all would 
hope and pray that--that probability of an attack is very low, 
but there is unfortunately, some probability, which is why 
firms want this insurance. And that was the basis of our score. 
It was consistent with what we had done in the past. I have 
seen some reports that people were surprised by our score, but 
it was fully in line with previous analysis that CBO has done. 
The amendment that was adopted reduced the score to zero by 
removing the certainty that is essential or integral to the 
provision of insurance, because the insurance will trigger on 
only if there is a future active Congress. And under our 
scoring, we don't evaluate future legislative action. So it was 
scored at zero, as amended.
    With regard to the employer-provided tax expenditure, tax 
preference for health care, there is wide agreement among 
economists that that is a relatively inefficient approach to 
providing a subsidy through the Tax Code for health insurance. 
It creates a bias for employer-provided health care as opposed 
to individual. It creates a bias for gold-plated health care 
relative to less generous health care plans and creates a 
variety of other distortions. I think the challenge in thinking 
about options to reform or replace it is that employers are the 
central pooling mechanism that we have in the United States for 
health care. And so if you tinker or change the tax preference 
for employer-provided health insurance, what replaces it so 
that we are not all in an individual market, which doesn't work 
well in the health sector?
    Mr. Ryan. Director Johnson, The PART in the appropriations 
process, how do we better connect the two?
    Mr. Johnson. There is currently--I forget the exact 
correlation. I think you all ran a correlation of programs that 
are PARTed on a low score and what happens to their budget; 
does it go down? And there is a positive correlation. There is 
attention being paid whether programs work or not. More 
attention is paid in some committees than others, but there is 
a positive correlation. Every year we send up----
    Mr. Ryan. In the administration's budget?
    Mr. Johnson. Pardon me.
    Mr. Ryan. In the administration's budget.
    Mr. Johnson. It is not only in the budget, but money that 
is actually appropriated. It is not only what goes on in our 
budget, but it is not true that every ineffective or results-
not-demonstrated program is recommended for a big reduction. 
Sometimes it is such a priority, we need to work to make it 
fix, and we need to fund it in the meantime. In some cases, we 
need more money to allow them to measure the quality in our 
parks and so forth and so on. So nothing automatic happens 
because of the current performance of a program. Nor should it.
    So there is information that--we will send up like--several 
years ago it was 50, and last year it was 140 programs that we 
recommend for zeroing out or significant reductions. Usually a 
third, 40 percent, 25 percent of those are because of the way 
they PARTed. There are other reasons for the majority of those 
programs, why they were recommended for a big reduction. They 
are duplicative; they work, but they are duplicative, or it is 
not a priority or we can't afford it anymore or whatever. There 
is a variety of reasons. Performance is not the only reason, 
honestly, to drive your all's budgeting thinking or 
appropriations thinking.
    But right now, performance is being looked at. And it can 
be looked at more and it can be looked at more consistently 
across all subcommittees. But most importantly, we can do a 
better job using this information to drive performance within 
the executive branch, which David talks about, within the 
discretionary programs; and then in the future, in the 
nondiscretionary programs.
    Mr. Ryan. Thank you. I see we probably have time for 
somebody else to ask something.
    Chairman Spratt. We are down to 6 minutes. Okay, good. If 
you will bear with us, we will run and vote. We have got 6 
minutes for this vote and then the two 5-minute votes following 
it. So we should be back in about 15 minutes, barring some 
antics on the floor. The committee stands in recess.
    [Recess.]
    Chairman Spratt. Call the committee back to order and now 
recognize Mr. Edwards of Texas for 5 minutes.
    Mr. Edwards. Thank you, Mr. Chairman. Mr. Chairman, thank 
you for holding this hearing. I think while there are sometimes 
partisan differences on this Budget Committee regarding what 
the priorities and responsibilities of the Federal Government 
ought to be, one issue in principle that ought to bring us all 
together is the idea that when we are spending taxpayers' 
dollars for programs, we should work together to see that those 
dollars are being spent effectively, efficiently, with clear 
goals in mind. And I think I can say as a Democrat to my 
Democratic colleagues, I think in years past sometimes we have 
tended to pass a bill based on--or funding based on a good 
intention, but good intentions aren't good enough when it comes 
to spending taxpayers' dollars. And we who believe in 
government ought to be the first in line to say we are going to 
be very tough on ensuring that programs are accomplishing their 
goals, because if they aren't, ultimately taxpayers will lose 
their credibility in the programs themselves.
    So I salute you, Mr. Chairman, for holding this important 
hearing. I thank all of the witnesses for your input and your 
leadership in this area as well.
    In that spirit, Mr. Johnson, let me say one reason I took a 
few more minutes to come back from our votes is I went 
downstairs from the House floor to talk to the subcommittee 
staff that I work with from my vantage point as Chairman this 
year of the Military Construction and Appropriations 
Subcommittee. And I asked if they had ever, in this year or in 
previous years, had ever met with OMB staff to review the 
analysis that your agency has carried out regarding the 
effectiveness of military construction and VA programs. And 
their answer was, no, not this year, and not in their memory.
    So whether that is my subcommittee's fault or OMB's doesn't 
really matter. But let me just say, in the spirit of my opening 
comments, I would welcome your staff meeting with me or my 
staff as we start to go to conference on the MILCON-VA bill and 
look at what we can glean from that analysis.
    And an example, the kind of question that is so basic to me 
that still can't get answered very well, is the VA health care 
system generally a good bang for the buck? My general belief is 
yes, it is. It is an outstanding system. And the conventional 
wisdom is that we provide quality care for fewer dollars than 
the private and nonprofit and for-profit health care system. 
But I am not sure I could quantify that judgment.
    So by the same token, I was very involved in the new 
public-private partnership program we have had in military 
housing. And I think generally it has been accepted by the 
departments of the agencies within the Department of Defense as 
an outstanding program. But I haven't seen any kind of economic 
analysis to compare that to the old system. So I, for one, who 
at least chairs one appropriations subcommittee, would welcome 
more interaction between us on this.
    Mr. Johnson. I will make sure that happens, sir.
    Mr. Edwards. Thank you.
    I would--Mr. Walker had brought up, Mr. Johnson, the issue 
of analyzing tax preferences. Whether we call them tax 
expenditures legally or tax preferences doesn't bother me one 
way or the other, but let us call them tax preferences out of 
respect for others. The fact is that GAO has put out a report 
saying that they do represent a huge loss of revenues based on 
preferences that some companies, some groups of individuals, 
some industries get compared to others. It also says that OMB 
has only rarely used the performance-based analysis on tax 
preferences. Could you tell me what the administration's 
official position is on that issue? And should we expand 
performance-based analysis to the effectiveness and efficiency 
and the clarity of goals of tax preferences?
    Mr. Johnson. I know it was the administration's official 
position to develop the PART to analyze programs, both 
discretionary and nondiscretionary programs. And so the feeling 
about using something--using it or something like it to 
evaluate tax programs, it was not no, but not yet. That is 
still our position. And unofficially it sounds like it is 
something we ought to do, and we ought to review the mechanism, 
or decide that that mechanism--the PART mechanism or some 
version of the PART mechanism, and figure out what we want to 
do.
    But our position has never been that we didn't want to do 
it; it was just not yet. We wanted to focus on how programs 
worked, because we wanted to use that then as the basis for 
driving performance management, people management within 
programs, and cause the Federal Government to become more 
purposeful about the way they spend the money.
    Mr. Edwards. Right. Okay. Thank you. I see my time is up, 
so I will withhold my additional question. Thank you, Mr. 
Johnson.
    Thank you, Mr. Chairman.
    Chairman Spratt. Thank you, Mr. Edwards.
    Mr. Doggett is not here.
    Mr. Scott of Virginia.
    Mr. Scott. Thank you, Mr. Chairman.
    Mr. Walker, you indicated that--I think implied that tax 
expenditures were hidden in the budget process. Doesn't PAYGO 
expose these expenditures, at least the prospectively, if we 
try to pass them in the past, in the future?
    Mr. Walker. My understanding is there is some prospective 
application, but not ones that already exist, if you will. They 
are in the base. And that is what I am talking about.
    Mr. Scott. Okay. And when we were talking about performance 
measures, is that a Budget Committee function, or is it a 
suggestion that authorizing committees on reauthorization 
should consider, should be proactive in ascertaining whether or 
not the programs have been performing, and then kind of report 
to the Budget Committee or to the Appropriations Committee?
    Mr. Walker. I believe that the country would benefit and 
the Congress would benefit for having a set of key outcome-
based indicators for budgeting, authorization, reauthorization, 
appropriations, and oversight purposes. When you are 
considering legislation to begin with, one of the basic 
questions ought to be what are you trying to accomplish, and 
how are you going to end up measuring whether or not it is 
making a difference? And so I think it crosses all those 
different constitutional responsibilities.
    Mr. Scott. I serve as Chair of the Crime Subcommittee. What 
kind of performance budgeting policies would you suggest for 
budgeting in the area of crime policy?
    Mr. Walker. Crime rates, the nature of violent crime. I 
mean, there are issues that relate to actual outcomes. What are 
we doing, you know? Those would be a couple of examples in that 
area.
    Let me give you a couple other examples that are outside 
that area: math proficiency, science proficiency, life 
expectancy at birth, infant mortality, household savings rates. 
There are a number of things that right now, quite frankly, we 
are measuring, and many other countries are measuring, but we 
are not using it for purposes of informing our decisions on 
spending programs, tax policies, and regulatory actions.
    Mr. Scott. And if the expenditure on a crime policy issue 
did not lead to a reduction in crime, reduction in violence, 
then we ought to reconsider making that expenditure?
    Mr. Walker. Right. You need to have more evidence in order 
to be able to make informed decisions about which programs and 
policies are working and which aren't, how to redirect limited 
resources and authorities to achieve the maximum impact.
    Mr. Scott. This is hard to understand, because most crime 
policy is slogan generated; that is, you come up with a good 
slogan and you codify it, and it helps politicians get elected. 
What you are suggesting is we ought to use our common sense and 
actually reduce crime. How does that help us get reelected?
    Mr. Walker. My objective, Mr. Scott, is to help the 
Congress help the country, help the American people.
    Mr. Scott. Could you say a word about--a lot of the 
testimony is on health care, and it seems to me if you don't 
have employer-based health care, there would be no incentive--
with the tax benefit--there would be no incentive for the 
employers to actually provide the health care.
    Mr. Walker. I will let Mr. Orszag----
    Mr. Scott. We are trying to get more people insured. What 
would happen if we did not have this tax expenditure?
    Mr. Walker. Let me answer that and then go to Mr. Orszag.
    First, I don't know that Mr. Orszag is talking about 
eliminating any tax preference. From my standpoint, I think you 
have to recognize that there are a variety of forces that cause 
employers to offer health care, not just the tax preferences.
    Mr. Scott. If you don't have the tax preference, that is 
the rationale.
    Mr. Walker. No, no. Well, the number one benefit of 
preference by employees--and I used to practice as the 
worldwide head of a human capital consulting firm before I had 
this job--the number one employee benefit of preference for 
employees is health care. Number two is health care. Number 
three is health care. So the market says employees want health 
care as part of their basic compensation package. The tax 
incentives are icing on the cake, but they are by no means 
going to reduce the demand or desire of employees for health 
care.
    And I will let Mr. Orszag----
    Mr. Orszag. Yeah, Mr. Scott, if I could just add, clearly 
the current tax preference provides one incentive for employers 
to offer health care. But in analyzing proposals like the 
administration's proposal to transform the current tax 
preference into something else, we had to struggle with what 
other things, what other aspects of the employer-provided 
package would still persist? And our estimate suggests there 
would be a reduction in employer-provided coverage, but it 
would still be the bulk of coverage. And the reason is 
administrative cost savings from large employers being able to 
pool people together, and the pooling mechanism through 
employers. So unless you specifically----
    Mr. Scott. That you get a group rate.
    Mr. Orszag. You get a group rate, and you get 
administrative cost savings because it is not just a group rate 
by pooling people together, but you are doing it in an 
employment setting, where it is easier to sign people up, and 
the back office kind of steps are also easier.
    Mr. Scott. Thank you, Mr. Chairman.
    Chairman Spratt. Thank you, Mr. Scott.
    Mr. Boyd of Florida?
    Mr. Boyd. Thank you, Mr. Chairman.
    Mr. Chairman, I am very pleased that you have decided to 
take on this issue in this hearing. And also it is very obvious 
that these three panelists have a good grasp of the principles 
that would be required to take on this monumental task.
    When I was in the Florida Legislature in 1994, I chaired a 
committee that wrote a piece of legislation called performance-
based budgeting. That ultimately came into law, was signed by 
then Governor Lawton Chiles, and now we have 13 or so years of 
history of performance-based budgeting in Florida. And that has 
been hailed by some, including the National Conference of State 
Legislatures, as one of the premiere performance-based 
reporting and budgeting laws and States in the country.
    I would ask those of you on the panel, are you familiar 
with this, and have you studied it, and is that or some other 
similar system one that we can use as a model? Because this is 
not going to be easy.
    Mr. Johnson. There was a hearing 2 years ago, I think, and 
Mike Conaway and Congressman Cuellar were on the panel with me, 
and Congressman Cuellar was talking about the performance-based 
budgeting in Texas. I am sure it wasn't as good as it was in 
Florida, but maybe it was close. And a lot of the things that 
he pointed out were being done in Texas he was recommending be 
done here, that they have this kind of information and that 
kind of information. And the thing that came out of that 
hearing was actually the kind of information he is calling for 
exists, it is this performance information, but it is not used, 
which is--we haven't done a good job of communicating it, and 
maybe the interest level here, or piquing the interest, 
tweaking the interest level up here.
    As you make this comment, I was writing down we need to 
look at, in our world, all the examples of performance-based 
budgeting in the States, and what are the lessons learned? How 
do different States use the information to make smarter budget 
decisions, appropriations decisions? And are there any 
suggestions about what we might be doing differently at OMB or 
in the executive branch, or what this committee or the 
appropriators might be doing differently? I bet that we haven't 
studied that to see if there are any lessons learned as much as 
we should have.
    Mr. Boyd. Before we go to the other panelists to answer 
that question, I might add we went all over the country 
studying other governmental entities--Texas was one of them--
but we found lots of our good work or good examples of good 
systems in cities, municipalities, and counties.
    So, Mr. Walker?
    Mr. Walker. Mr. Boyd, I am somewhat familiar with that. As 
you know, I am a Florida CPA and lived in Florida a number of 
years. I know that Florida was an early adopter of performance-
based budgeting. I know you had something to do with that. I 
know GAO did some work in the mid-1990s which included looking 
at a number of performance-based uses at State level, and 
Florida was a leader at that time. We haven't done work on it 
in recent years, and, quite frankly, we would entertain a 
request from this committee to take a look at what is the state 
of performance-based budgeting should you desire us to do that, 
because I think it clearly is meritorious.
    Mr. Orszag. The only thing I would add is my understanding, 
for whatever it is worth, is that at the State level 
performance-based budgeting has not been widely applied to two 
of the areas that we were discussing here, in particular, for 
example, State Medicaid programs and exactly what they are 
accomplishing or not, and similarly SCHIP, and also tax 
expenditures. So they are very useful for the areas that they 
are applied to, but, at least from a Federal budget 
perspective, there are big pieces that are not as analyzed.
    Mr. Boyd. It is obvious--well, I can tell you from having 
been through this exercise at a lower level of government that 
you have to--it is not about just having information, it is 
about creating a framework or a structure in which you use that 
information with definable and measurable principles or 
principles of measurement, performance, duplication, waste, 
fraud, and--all of those things come into play. But you have to 
have a structure which tells the bureaucracy of a government 
that this is the way we are going to do it, and that structure 
has to be in law. Obviously, that bureaucracy is a very, very 
difficult one to break through here in Washington, D.C. Are you 
folks optimistic that we could do that if Congress and the 
administration put our heads to it?
    Mr. Johnson. I am confident, primarily because we are 
getting agencies to do this, to focus on results by holding 
them accountable for it. What we have found is--and it would be 
even stronger, it would be even more compelling if there were 
some congressional reasons for doing it, if some of this was in 
statute. We have found that when you sit around with agency 
leadership and say, here is what we want to do, here is how we 
want to run the railroad, don't you agree; you have goals and 
you have performance, you report on it quarterly or every 6 
months or whatever, and you have corrective action plans and 
performance action plans, we want to make this very public, 
don't you agree this is a good idea? And they will say, well, 
yes, but let us do it this way. Then we say, here is the way we 
are going to run the railroad, and then every quarter let us 
assess who is doing what they said they were going to do, and 
when are we getting to the point where it is a desired level of 
performance management?
    We do that. There is no law that says this will happen, but 
it is the President, starting with the President on down, that 
says, this is the way we are going to do it. And so you can 
say, well, then what if another President came in and it wasn't 
a priority, would you lose it? Maybe. But my sense is that the 
people in the agencies understand this is good for the 
agencies. This is something that by and large it will make HUD 
work better. It will make Interior work better. And so it will 
make it a better place to work, a better place to be a public 
servant. And so the employees get it. This is not something 
that is bad for them. But they want to be trained how to do it. 
They want to have the money to be able to do it. They want to 
have the time to be able to do it.
    And so I think we have--what we need to do is figure out 
how to help agencies do it even more successfully going 
forward. It exists in the executive branch now. It is not 
locked in. It is not--doesn't permeate their entire being, but 
it is something they get, it is something they are doing 
because we are holding them accountable for doing it.
    Mr. Boyd. Mr. Chairman, I know my time has run out. Are we 
going to have a second round so that we may continue with----
    Chairman Spratt. Go ahead, Mr. Boyd, if you would like.
    Mr. Boyd. Okay. I would ask the other panelists to respond.
    Mr. Walker. First, Mr. Boyd, you already have one law that 
is called the Government Performance and Results Act. And as I 
noted in my testimony, we need full implementation of that law. 
We don't believe the executive branch has fully implemented the 
governmentwide performance plan. We think they need to do more 
on the horizontal dimension, and they also need to consider tax 
preferences, as I have noted.
    Secondly, we have also recommended in my testimony that the 
Congress ought to consider a congressional performance 
resolution. And this committee obviously could take the lead.
    I think this committee could also take the lead on looking 
at the concept of key national indicators. And I think that is 
something that, quite frankly, is meritorious of moving forward 
with a public-private partnership to take that concept and make 
it a reality, as other countries have. We have done some work 
on that. We are happy to do more work on that.
    I also think that this committee could also think about 
doing something similar to what the Senate Budget Committee did 
back early in my tenure, and that is to form some task forces 
on health care, key national indicators, selected issues that 
you may have an interest in to see what might be a proper way 
forward not just for this committee, but for the Congress. And 
we would be happy to work with you on that.
    So first we have got to fully implement the existing laws, 
then we need to figure out what else can be done 
administratively. And I come back to what I said earlier: Even 
if the information exists, it has got to be used. There is 
information that exists right now that is not being used. And 
that is a cultural issue; it is not a legislative issue. It is 
the fact that people aren't used to it, they are not being held 
accountable for it, and therefore why bother?
    Mr. Orszag. If I could just tunnel in on one area, in 
addition to the comments that have already been made. With 
regard, for example, to Medicare, there would be statutory 
changes that would be required in order to significantly expand 
the information base that we have, which is currently lacking. 
We simply do not have enough information on what works and what 
doesn't, what is causing the variation in Medicare spending, 
what type of spending is worth it and what is not, what works 
better, what works not so well. And for Medicare to then 
incorporate that information into its payment system so that we 
would move from a fee-for-service to a fee-for-value, one would 
require statutory changes also. So in that particular area, as 
just one example, there are statutory changes that would be 
necessary to build the information base and then to apply the 
information.
    Mr. Johnson. If I may have one more comment, a good bit of 
GPRA has been implemented. Agencies do a lot of reporting about 
performance. In my opinion, our opinion, the PART information 
gives us more consistent, higher quality, outcome-oriented 
performance information with which to comply with GPRA 
requirements.
    I think most people would agree that little change in 
performance has come about as a result of GPRA. It is not 
because--for the programs that we are managing. It is not so 
much for the mandatory, and we have got nothing going on in the 
tax whatever you want to call it program. But for the 
discretionary programs that we pay a lot of attention to and 
are trying to hold, increasingly so, managers responsible for 
performing well, little attention is paid to what is reported 
to Congress about performance. It is not as specific in the GPA 
format as it can be, and, I think, as it should be, and when it 
is reported, very little attention is given to it. Very 
little--there is very little, so what? This program doesn't 
work. What are you going to do to cause this program to work at 
an exceptional level versus a marginal level? And this isn't a 
budget issue so much as at a congressional level, and I think 
we can even do a better job at the executive branch.
    One of the things that has really helped sort of set the 
groundwork for change in the executive branch is 3 years ago or 
so, they changed the law that allowed the Senior Executive 
Service, 8,000 of them, 90 percent of which are career, to be 
held accountable for the performance--the performance 
measures--program performance measures in their performance 
goals. So they are not held accountable for having proper 
competencies; they are held accountable for the performance of 
their programs. And their pay increase can be tied to their 
performance rating.
    We could have had this law 3 years ago and not been able to 
abide by it because we didn't have the performance measures 
used to evaluate a manager's performance. Now we have 
performance measures. They are not perfect, never will be, but 
they are very, very good. We have the data now to increasingly 
every year hold managers accountable for the performance of 
their programs. That is where real rubber meets the road is 
when individual executives in programs and all of their staff 
are held accountable for how their programs work. That can take 
place in the executive branch with the help of some 
legislation, and if we work together to prevent some 
legislation that would keep that from happening. There are some 
interests in this town that would like that not to happen, as 
you can imagine.
    Mr. Boyd. Thank you, Mr. Chairman, for your indulgence.
    Chairman Spratt. We will go next to Mr. Cooper and then Mr. 
Porter of Nevada.
    Mr. Cooper.
    Mr. Cooper. Thank you, Mr. Chairman. I can't think of a 
more important topic than helping taxpayers get more value for 
their money.
    We have excellent witnesses here today. I am disappointed 
at Member response. This testimony should be required reading 
for all of our colleagues so that we can get more on top of 
these issues.
    First let me focus on the Medicare discussion that Mr. 
Orszag was having with Mr. Boyd. I want to commend Peter for 
beefing up the CBO staff in the health care area, and I think 
there are some very explosive parts of the testimony for 
Members who really understand. First the map on page 5, the 
geographical differences in Medicare reimbursement are 
stunning. Completely different pay levels for doing exactly the 
same medical procedure, completely unjustifiable, but yet it 
has been law for 40-plus years. And we seem to live with those 
discrepancies.
    But my real point is the chart on page 8, which shows the 
relationship between quality of care and expenditure of money 
on care. There are a lot of dots on the map. I wish that CBO in 
the future would put each State's identity on those dots, and 
also do what I think the Commonwealth Foundation has done, 
which is to draw a trend line over those dots so that you can 
see the inverse relationship between money and quality. So, in 
other words, the more you spend, you don't get more quality; 
the more you spend, the less quality you get, which is the 
opposite of what it should be. So I hope CBO wasn't pulling any 
punches in this case, but an overlay trend line would be 
helpful for Members so they could see clearly the relationship.
    Mr. Orszag. I would hope that I am getting known for not 
pulling punches. So, no, that was not on purpose. We can look 
into the suggestions that you made for future publications.
    Mr. Cooper. Thank you.
    I thought the other most explosive element of the testimony 
was regarding tax expenditures, and here Comptroller Walker 
excels. Most Members of this body do not realize that the Ways 
and Means Committee's pocket jurisdiction is larger than the 
total discretionary spending of every other committee in 
Congress combined. Now, I am not against the Ways and Means 
Committee, but that is an amazing side feature of their 
jurisdiction, to be able to effectively spend $800, $900 
billion a year, and that is not even their primary portfolio.
    So I would like to ask Mr. Walker what we can do to try to 
rein in these expenditures in drag, basically, because they are 
masquerading as something else, and I think to a curious 
effect. What should we do about this expenditure problem?
    Mr. Walker. First, we need more transparency with regard to 
these tax preferences and the cost of these tax preferences. We 
need to have more transparency in the financial statements. We 
should have more transparency as part of the annual process the 
budget goes through, you know, maybe the congressional budget 
resolution that we recommended.
    Secondly, we need to also subject them to an outcome-based 
analysis, and hopefully that the Congress can move forward with 
this concept of key national indicators through a public-
private partnership. Then we should subject not just 
discretionary spending, but mandatory spending as well as tax 
preferences to an analysis of whether or not they are making an 
outcome-based difference.
    I think whenever you are passing new legislation and 
reauthorizing existing policies and programs, you need to 
determine what you are trying to accomplish on an outcome-based 
measure. Put it right in the legislation and make sure, 
therefore, that there is a periodic analysis of that both on 
the tax side and the spending side that Congress automatically 
will consider as part of its normal constitutional 
responsibilities. Tax preferences are largely off the radar 
screen. You can't manage what you don't measure. You can't hold 
accountable what there is not adequate transparency for.
    Mr. Cooper. I thank the gentleman for his response. This 
effectively--it seems to me you are saying that as 
uncontrollable as some entitlement spending is, tax expenditure 
spending could be even more uncontrollable, because oftentimes 
we don't even know what it is.
    There is a Bush Treasury official, I think her name was Pam 
Olson, who testified before this committee a couple years ago 
saying that tax expenditure spending is basically unmeasured 
and unmeasurable, unverified and unverifiable. That should 
alarm anyone who is interested in good government. And yet year 
in, year out we seem to do very little about it.
    I see that my time has expired, Mr. Chairman. Thank you for 
this important hearing.
    Chairman Spratt. Thank you, Mr. Cooper.
    Mr. Porter?
    Mr. Porter. Thank you, Mr. Chairman. I appreciate this 
opportunity today. And we have three of the most talented 
individuals in Washington. And I say that I do respect, I 
appreciate your opinion, and have worked closely with you for a 
number of years, and I appreciate you being here. And we could 
choose who has talent or not, if you would like. I know, Mr. 
Walker, you were just in Las Vegas. Did you see O.J. While you 
were there? Anything we need to know about----
    Mr. Walker. It is amazing, you know. I was in there doing 
the Fiscal Wakeup Tour the day that O.J. got arrested, and I 
didn't come across him.
    Mr. Porter. You weren't at the Palace Station, obviously.
    Mr. Walker. Don't know him, haven't seen him, and don't 
really worry that I haven't.
    Mr. Porter. In all seriousness, we appreciate you being in 
Nevada. And I understand it was very well received. And thank 
you for your insights.
    And again, thank you, Mr. Chairman.
    And I guess I sometimes try to oversimplify things in 
trying to compile information and put it in words that maybe 
our constituents can understand, but it seems to me--and this 
may be more of a comment than a question, but it seems to me we 
need a day of reckoning in Congress, that there is 1 day of the 
year when we look at expenditures, we look at revenues, how 
much we have spent, or how much we have promised to spend, and 
on that particular day everything comes to a point where we see 
how much we don't have. And I know we go with continuing 
resolutions sometimes, and there is omnibus bills, but it 
doesn't seem to me, unless I am missing something, that there 
is 1 day of reckoning, other than tax day, of course, which I 
think should be election day, April 15th, where this Congress, 
this body--and I believe that we could do it together in a 
bipartisan way--look at this is where we are today, this is 
what we have promised, and this is what we don't have. Is there 
such a mechanism in place? Can we do that?
    Mr. Walker. I can tell you that I lead strategic planning 
for auditor generals around the world, and also I am on the 
board of the international organization, and some countries 
have such a thing. And I think one of the things that this 
Congress ought to consider is one time a year have the Director 
of OMB, the Comptroller General of the United States, and the 
Director of CBO report to the Congress as a whole--and I am 
talking about not just sending a paper, but I am talking about 
physically be there with the Congress--to present an assessment 
of where we are, and where we are headed, and the different 
perspectives. I think that would be invaluable. And I know that 
happens in some countries. Unfortunately it doesn't happen in 
ours.
    Mr. Porter. I think for families we look at our budget, and 
at the end of the year you look, and you look at your credit 
cards, and you look at your bills and how much income you have. 
There is a day with our families when we know we are short or 
we are doing okay. And the same with business. We have that. At 
the end of the year we reconcile our books. And I would like to 
work with you, if we could make that presentation to this body 
at some point so that we can see where--the American people can 
see where we are. I would appreciate that.
    Mr. Orszag. I was just going to add that one of the 
purposes of CBO's budget outlook that is released in January 
and then in August is to give a snapshot of both where we have 
been and where we are going. And one of the purposes of the 
budget resolution, to the extent it is adopted on an annual 
basis, is to try to take that information into account. So I 
agree there is much more that can be done.
    Mr. Porter. Be more transparent. I think you are providing 
the information, but maybe we can consolidate it more concise.
    Mr. Orszag. Yes.
    Mr. Porter. I would appreciate that.
    Mr. Walker. It is not just that, Mr. Porter. There is no 
question that CBO provides a lot of valuable information, and 
so does GAO, and so does OMB. But there is a difference between 
sending up electronic or print documents up here and appearing 
before the body of the whole and making an oral presentation, 
using charts and graphs that convey a lot of information very 
concisely. That is what is missing, and that is what is needed.
    Mr. Porter. Thank you.
    Thank you, Mr. Chairman.
    Chairman Spratt. Mr. Etheridge is next.
    Mr. Etheridge, would you yield just for a second and let me 
say a word?
    Mr. Etheridge. Yes.
    Chairman Spratt. This is the point I was going to make 
about the SAR. I recall when it was created in the Pentagon, I 
was working for the Comptroller when it was created. Twelve 
years later, when I came back here, came into Congress, I went 
over and started pulling out SARs just to see how they looked. 
They looked the same way they had 12 years ago. They weren't 
more useful because they hadn't been used.
    One of the things that happens when you engage and use a 
document is it becomes better and better adapted to its 
intended purposes. When it sits on the shelf, nobody uses the 
end product, it just gives you the same old output every year.
    Mr. Etheridge.
    Mr. Etheridge. Thank you, Mr. Chairman, and thank you for 
the hearing.
    Let me thank each of you for being here. And I won't 
elaborate on the accolades, because we are honored to have you 
here. Let me just make a couple of points and ask a question.
    I served as a State legislator before I came here, and 
chaired the Appropriations Committee, and then served 8 years 
as State superintendent and ran one of the largest budgets in 
our State. And what you say is true. But the real challenge 
gets below really that, as you well know. And let me just give 
an example, and then I am going to ask my question, because I 
do think that performance-based budgeting really holds some 
potential, really maximum potential, for public agencies if it 
is used and used properly.
    For example, let me just give you one example, that we are 
underfunding education right now, in my view, if we are really 
going to continue to be a strong country. And if we are pairing 
it off against other things, I think that is dangerous. The 
administration sent a budget up cutting education by 1.5 
billion, eliminating 44 various agencies. Sounds like a great 
idea if you don't know the difference, but when you start 
looking at it, it was a tool that was called Program Assessment 
Rating Tool. Sounds good. Makes sense. However, when you find 
out that, number one, none of the public participants were in 
it, none of the stakeholders were consulted, they just reached 
in and parceled them out, threw them up on the table, and the 
problem you have with that is the folks in the field are 
dealing with training the next generation of workers, and they 
don't have input.
    That is not a very good performance-based process, in my 
view, having worked with it. People need to be engaged and 
believe it is fair. And we are getting close to a reckoning 
day. So I think you are absolutely right.
    So let me ask you a question. And I do want to applaud the 
work that each of you are doing and thank you for it. It is 
very important to our work. To develop options, though, for 
policymakers to address the looming train wreck that you have 
been talking about, and I think it is one we have got to deal 
with on the budgets, on the horizons, in a variety of ways, and 
I want to thank Chairman Spratt for calling this meeting, let 
me just ask each one of you for one suggestion, just one, and 
you have given us a whole bunch, but just one that we could use 
to accomplish that purpose. Just one. Whoever wants to start 
first. Mr. Johnson?
    Mr. Johnson. I think that this committee or Congress or the 
administration should publicly declare that they want to be 
held accountable for the government working well and better 
every year in a demonstrable, quantifiable degree.
    Mr. Etheridge. That is more like a policy statement. Okay.
    Mr. Johnson. Just that commitment will drive behavior, will 
drive a level of transparency that doesn't exist now. It will 
drive an attention to performance or nonperformance that 
doesn't exist now. And it will drive real-life things happening 
unlike that are happening now.
    Mr. Etheridge. Very good. Thank you.
    Mr. Walker?
    Mr. Walker. I testified before that with regard to our 
fiscal challenge, the best thing you can do is to reimpose 
tough statutory budget controls on both sides of the ledger, 
dealing not just with discretionary spending, but also 
mandatory reconsideration triggers for mandatory spending, and 
also to put tax preferences as part of that radar screen. You 
have to do that.
    Now, as far as performance budgeting, I would say moving 
towards an annual congressional performance-based budget 
resolution--now, it is going to take you--you are going to have 
to do some work to get there, because you are right, there is a 
lot of debate about, well, which indicators do you want to use 
in order to assess performance? But agreeing that there is a 
desire to move to some type of annual congressional 
performance-based budget resolution and then putting in place 
mechanisms to make that a reality over time.
    Mr. Etheridge. Starts that dialogue.
    Mr. Orszag. I would say if there is one step that, looking 
out over the future, you wanted--and you wanted to adopt a 
perspective to get more for your money from the Federal budget, 
it is in health care. And perhaps the most single auspicious 
thing that could be done if you wanted to move in that 
direction is to vastly expand the amount of comparative 
effectiveness research that is done; that is, examining what 
works and what doesn't through registries, through trials, et 
cetera, and then tying financial incentives and other 
incentives to the results of that research.
    Mr. Etheridge. Thank you. Thank all three of you.
    Thank you, Mr. Chairman. I yield back.
    Chairman Spratt. Thank you, Mr. Etheridge.
    Mr. Becerra?
    Mr. Becerra. Thank you, Mr. Chairman, and thank you to the 
witnesses for your testimony.
    I am not sure if you explored this earlier or not, but I am 
interested in hearing your thoughts about the expenditure side. 
And, Comptroller General Walker, let me start with you and ask 
have you all done any analysis of the charitable deduction 
expenditure and whether or not we as a Nation are getting the 
value of that lost revenue in returned programming and services 
for Americans through that charitable deduction?
    Mr. Walker. We have done some work with regard to tax 
preferences in general. I do not recall any particular work 
that has focused specifically with regard to the charitable 
deduction and the value for money implication, but I will check 
and provide for the record a specific answer.
    [The information follows:]

    ******** COMMITTEE INSERT ********

    Mr. Becerra. Let me offer any of the witnesses if they have 
any comment with regard to any work that may have been done to 
analyze that particular expenditure, the charitable deduction, 
and its value to America.
    Mr. Orszag. There has been some academic work done. I will 
have to check on whether CBO has published work. The academic 
work has raised some interesting possibilities. So, for 
example, there is a recent study that shows a match for a 
charitable contribution may be more effective in eliciting 
charitable contributions than money back in your pocket, which 
is basically what a tax--the current tax preference does. So 
you give a dollar to a charity and there is a matching 
contribution, that, according to one study, generated more 
activity than you give a dollar to the charity and you get 
something back in your pocket in exchange.
    Mr. Becerra. Dr. Orszag, I am actually interested in the 
output side of that transaction, where we then determine how 
that charity uses the money. I would like to find out if anyone 
has analyzed how those organizations that receive that 
charitable deduction employ those dollars that they get as a 
result of taxpayers having less money to have to pay into the 
Treasury. Money forgone, my understanding, it is about $32 
billion that the Treasury forgoes because people make these 
charitable deductions to these nonprofit organizations. And I 
am interested in finding out, for example, if we are getting 
our dollars' worth. I know there are a lot of poor people in 
this country. I am wondering how many of those charitable 
contributions end up helping people who are poor, end up 
helping people who have no health care.
    Mr. Walker. You are raising a very interesting point. 
First, there is audit authority and evaluation authority that 
we have, for example, when you are dealing with Federal grants. 
All right. So they are on the radar screen, because they go 
through the appropriations process and the budget process. But 
tax preferences don't. That is not Federal money. It is forgone 
Federal revenues. And therefore, you would even have an issue 
of whether or not we would have the audit authority, you know, 
with regard to a tax preference. That is not Federal money. It 
is forgone revenues.
    Mr. Becerra. Let me ask and see if any of you wish to 
volunteer an opinion on whether or not we are making the best 
investment of tax dollars or forgone tax dollars by providing 
this particular tax expenditure, the charitable deduction.
    Mr. Orszag. Let me just say there are a variety of tax 
preferences associated with the nonprofit and charitable 
sector, not just the contribution one that you noted, but also 
the fact that they do not pay tax on their net revenue.
    Mr. Becerra. Correct.
    Mr. Orszag. And there have been questions that have been 
raised about the degree to which some of those entities are 
performing services and doing things that are quasi-commercial 
in nature. And the dividing line between a purely charitable 
activity and a quasi-commercial activity has become blurred. 
CBO has done some work on that in the past and will be doing 
more in the future. And I guess I would leave it at that, that 
there have been questions raised about that line.
    And then you are raising another issue, which is even 
within the charitable--you know, even within that more clearly 
charitable set of activities, how effective are they in 
actually achieving their objectives?
    Mr. Becerra. Right.
    Mr. Orszag. Again, I think that is something that we don't 
have good answers on.
    Mr. Becerra. General?
    Mr. Walker. A few comments real quick. First, there is 
something called UBTI, unrelated business taxable income. And, 
for example, AARP pays a lot of money on--a lot of tax on that, 
because they make a lot of money with regard to their insurance 
programs. Furthermore, we have done work at GAO for the Senate 
on the issue of not-for-profit entities, for example, 
hospitals, and whether or not there is a meaningful difference 
at the type of services that they are providing versus for-
profit hospitals, because obviously they are gaining a benefit 
through tax-exempt status, and presumably there should be a 
public welfare gain from that.
    Mr. Becerra. And, Mr. Chairman, I see my time has expired. 
If I could just explore one final question.
    Is there something, General Walker, that we could do to 
perhaps expand that examination? Because I think it is a 
legitimate concern to find out if there is a difference between 
what a not-for-profit hospital is providing to the public 
versus a for-profit hospital, and whether or not there is a 
reason to give a not-for-profit hospital a tax advantage over a 
for-profit hospital in the work that they do in a community.
    I also think there is a reason to explore whether or not 
many foundations are doing the work that we expect of them, 
especially family foundations are doing the work we expect of 
them to provide the general welfare of the public for the 
forgone taxes that they would otherwise pay. Is there something 
there that you think would be worth exploring by the GAO?
    Mr. Walker. First, let me do two things. One, I will 
provide back to your office information on what we have done 
and what we are doing with regard to not-for-profit hospitals. 
Secondly, I also want to think about whether or not there may 
be additional authorities that we might need in order to be 
able to do evaluations in situations where the Federal 
Government is forgoing billions of dollars of revenue and has a 
legitimate interest in trying to understand what public good is 
being obtained from that. There may be some need to look at our 
access authority and our audit authority here.
    Mr. Becerra. I look forward to hearing back from you on 
that.
    Mr. Chairman, I thank you very much. Yield back.
    Chairman Spratt. Thank you, Mr. Becerra.
    Mr. Ryan, do you have any further questions?
    Mr. Ryan. No.
    Chairman Spratt. Let me say to our witnesses today, 
Director Johnson, General Walker, Director Orszag, thank you 
very much for your time, and more than that for your insightful 
advice on a topic that we clearly need to pay more attention 
to.
    I also--earlier on I recognized the contribution of Dr. 
Posner, and Barry Anderson as well. Dr. Posner is here.
    [The prepared statement of Mr. Posner follows:]

 Prepared Statement of Paul L. Posner, Director, Public Administration 
                    Program, George Mason University

    Mr. Chairman and Members of the Committee, I appreciate this 
opportunity to share my perspectives on the prospects for performance 
budgeting in general and the Program Assessment Rating Tool (PART) 
process in particular. In brief, performance informed budgeting has 
become institutionalized at the federal level through two successive 
Administrations. The use of performance data to inform budget decisions 
can potentially improve the decision making process and the outcomes of 
government by ensuring that performance goals and results are taken 
seriously at all levels. The benefits of strengthening the link between 
resources and results suggest that some form of performance budgeting 
is here to stay--it would not be surprising if the next Administration 
continued this agenda with initiatives of its own. While we should 
never expect to ``depoliticize'' the inherently political process of 
budgeting, performance information provides valuable new perspectives 
that can help transform the kind of debate--budget options can be 
compared not just based on their potential costs but on their relative 
contribution to policy outcomes we care about as a nation.
    In rethinking how to design the next round of performance 
budgeting, the PART process of this Administration will form an 
important foundation. Now that nearly all programs in the budget have 
been covered by at least one PART review, policymakers have an 
opportunity to create a process to better inform the difficult 
budgetary tradeoffs the nation will be facing in the years to come. 
Rather than engaging in yet another round of reviews revisiting the 
same programs, it is time to build on the base of measures and 
assessments developed under both GPRA and PART to do broader- based and 
more selective assessments of broad program areas. Such assessments 
should review the relative contribution of multiple programs to 
overarching program goals, e.g. improving food safety, or providing for 
low income housing, and should include all major tools of government 
used by the federal government to reach those broad outcomes, including 
tax expenditures. The assessments should also be performed in an open 
and inclusive manner that ensures the engagement of significant 
stakeholders and the Congress itself. While PART served the President's 
interests, a revised assessment process could achieve greater 
credibility and support if it were more open and inclusive. Enhancing 
the prospects for Congressional attention to performance issues is 
vital if we are to promote this agenda for the future.

         THE SURPRISING STAYING POWER OF PERFORMANCE BUDGETING

    Performance based reforms have had a long history in the United 
States at all levels of government. Often led by state and local 
initiatives, public administrators at all levels have become gripped by 
waves of performance reforms intended to improve performance and 
enhance public confidence in government. Ushered in with great 
expectations, reforms such as Planning-Programing-Budgeting, Zero Based 
Budgeting, and Total Quality Management, achieved significant 
improvements but are widely acknowledged to have fallen well short of 
their mark in institutionalizing a sustainable focus on performance 
within government over the longer term. This checkered history of 
reforms encouraged a certain amount of cynicism about the efficacy of 
performance management to achieve lasting success in government.
    One lesson learned from these past initiatives is that the ultimate 
success of performance reforms will be predicated on their integration 
with the most important process engaged in by public managers every 
year--the budget process.1 The introduction of performance goals and 
metrics into the budget process gained the moniker of ``performance 
budgeting'' and this has become a fundamental feature of current 
performance reforms at federal, state and local levels of government, 
as well as in most OECD nations.2
    Notwithstanding the failures of prior incarnations of performance 
management, since 1993, performance management reforms in general, and 
performance budgeting in particular, has achieved a level of continuity 
and sustainability that is at once both surprising and tenuous. The 
Government Performance and Results Act (GPRA), passed in 1993, has 
ushered in a period where performance information and justifications 
have become widely accepted in federal agencies, in OMB reviews and 
even, sporadically, in certain committees of the Congress itself. 
Notably, the reform has not only survived through two Administrations 
of different parties, and has become a vital part of the management 
reform strategies of both the Clinton and Bush Administrations.
    Under the GPRA, agencies were required to prepare strategic and 
annual performance plans and reports that covered the program 
activities in their budgets. While agencies had primary responsibility 
for preparing these plans, OMB reviewed the plans as part of the budget 
process and eventually worked with agencies to integrate performance 
plans into agency budget justifications. GAO found in its ten-year 
retrospective that federal managers reported having significantly more 
types of performance measures and that GPRA had begun to facilitate 
better linkage between performance information and planning and budget 
decisionmaking, although more remained to be done to promote the use of 
this information for resource allocation. 3 The surprising staying 
power of GPRA is partly due to its statutory origins, as reformers 
realized that past initiatives undertaken without the support of the 
Congress failed to transcend the terms of their executive political 
champions

            ADMINISTRATION PERFORMANCE BUDGETING INITIATIVES

    The Bush Administration built on the GPRA infrastructure to carry 
performance budgeting initiatives further along the continuum. While 
the agencies had made progress in developing the ``supply side'' of 
performance plans and measures under GPRA, the demand for this 
information by budget decisionmakers proved to be episodic at best, 
particularly in the Congress. Although performance reformers hoped that 
building credible performance plans would in themselves prove 
compelling enough to foster their own demand, these high hopes proved 
to be unfounded. Some fear that agencies may eventually lose interest 
in generating the supply of information if the supply is unrequited by 
the demand and actual use of the information.
    The Administration sought to strengthen the linkage of performance 
with budgeting. First, they included budget and performance integration 
(BPI) as a key initiative within the President's Management Agenda and 
evaluated agencies' progress periodically by using a scorecard 
published in the budget. As part of this initiative, OMB encouraged 
agencies to restructure their budget presentations and underlying 
account structures to better align budget resources with performance 
goals. Beginning with the fiscal year 2005 budget, OMB required 
agencies to submit a ``performance budget'' that would integrate the 
annual performance plan and the congressional budget justification into 
one document. While framed as a strategy to promote the GPRA agenda of 
linking plans with budgets, this initiative was viewed by appropriators 
and some agency managers as laying down the proverbial gauntlet by 
eliminating traditional information presentations that long served as 
the foundation for congressional appropriations and oversight. 4
    The Administration's Program Assessment Rating Tool (PART) added a 
program review and assessment component to performance budgeting. The 
PART initiative was justified as a way to promote greater attention to 
the performance perspective that was originally fostered by GPRA. 
Moreover, the initiative also was premised on the need to transcend the 
traditional focus of budget process from one that focused 
disproportionate attention on incremental changes to existing programs 
to a process where the base itself is periodically reexamined. Given 
the fiscal challenges facing the nation both now and over the longer 
term, such a periodic reexamination can be healthy for any political 
system.5
    The PART process consists of a series of questions rating the 
purpose and design, planning, management and results of each program in 
the budget. Programs receive ratings based on the answers to these 
questions, with ratings ranging from effective to ineffective and a 
separate category, results not demonstrated, when programs fail to have 
sufficient information or documentation. To date, OMB has assessed the 
performance of nearly 1000 programs, covering 96 percent of all 
programs in the budget.
    While PART can be viewed as building on GPRA, there are differences 
to be sure. One observer contrasted PART and GPRA by noting that GPRA 
was a ``passive'' strategy relying on the power of well developed plans 
and metrics to create their own demand. In contrast, PART constituted a 
more active strategy, where performance information would form the 
basis for explicit judgments of a program's effectiveness that would 
become part of the executive budget formulation process. Moreover, 
while GPRA plans and metrics were largely developed by federal agencies 
themselves, PART marked a new departure by placing control of 
performance assessments in the hands of OMB. It was the President's 
budget agency that designed the process and made the final judgments on 
program assessments, with the active participation of the agencies. At 
least in the initial years of implementation, agencies were encouraged 
to replace GPRA goals and measures with those developed under PART, 
substituting OMB's judgment for that of the agencies.
    The PART process has itself been assessed by GAO and others in the 
federal community. The process has institutionalized more formal and 
systematic program reviews in the executive budget process and has jump 
started a more regular conversation between OMB and the agencies on the 
performance of their programs. This has arguably led to a greater focus 
on performance as part of the budget review process. Agencies have a 
more powerful incentive to improve their performance data and measures 
to protect themselves in this review process in the future. Some agency 
officials suggest that PART has stimulated initiatives to focus on 
management issues that the agency may not have addressed on its own. 
While funding levels have been changed in some cases, changes to 
program design and management were a more prevalent outcome of the PART 
process, according to the GAO.6
    However, the PART process raises some vexing questions about the 
ability of any process to develop the appearance of a ``bottom line'' 
rating for complex government programs. Unlike the private sector, 
there is no single bottom-line for most federal programs. Given the 
multiple purposes and goals reflected in most programs, OMB staff 
invariably had to make judgments on how to answer yes/no questions on 
such issues as whether the program was effective in reaching its goals 
and whether the costs of the program were reasonable. Notwithstanding 
claims that PART rests on ``objective'' ratings, this is often not 
possible in the high stakes world of federal budgeting where multiple 
stakeholders have their own interpretations of seemingly 
straightforward numbers. A single performance indicator, whether it is 
trends in drug abuse, serious crimes or welfare caseloads, is subject 
to multiple interpretations about what the performance data mean and 
how it should be used in making decisions. The GAO assessment of PART 
concluded that the experiences with the rating tool illustrate ``the 
inherent limitations of any tool to provide a single performance answer 
or judgment on complex federal programs with multiple goals.''7
    Moreover, the PART process used discrete budget program activities 
as its unit of analysis for assessments. The assessment of detailed 
budgetary activities have the advantage of more closely tying 
performance assessments to the budget, but at the price of accepting 
the current budget structure as a given. However, there is a tradeoff--
while performance data may gain greater influence in budgeting, this 
may come at the expense of the breadth and openness that characterize 
strategic planning processes. The PART focus on relatively narrow 
budget accounts and activities contrasts with the broader planning and 
goal orientation of the GRPA planning process. The key question 
agencies are charged with answering in their GPRA plans and metrics is 
what difference does the agency and its programs make for outcomes that 
matter to the various publics in a democratic society? In many cases, 
whether it be low income housing, job training, food safety or child 
nutrition, such outcomes are achieved from the results of a number of 
programs and governmental tools, often crossing agency lines.
    Finally, any assessment will invariably reflect the interests of 
the entity that controls the process. As part of the President's budget 
preparation, PART clearly must serve the President's interests. 
However, it is unlikely that the broad range of actors whose input is 
critical to decisions will use performance information unless they 
believe it is credible and reliable and reflects a consensus about 
performance goals among a community of interested parties. This 
consensus building process is critical--if the goals in the plans gain 
broad support, their use in the budget will become more accepted and 
credible. The measures used to demonstrate progress toward a goal, no 
matter how worthwhile, cannot appear to serve a single set of interests 
without potentially discouraging use of this information by others. In 
the case of PART, the control of the process by OMB has arguably served 
to erode the support for the assessment process by such key players as 
Congressional appropriators and authorizers. The absence of broad based 
support will limit the impact that such a process can have longer term 
on the sustained development and improvement of linkages between 
performance and budgeting. I might add that most efforts to integrate 
performance information into budgeting suffer to some degree from the 
absence of open collaboration with stakeholders and the Congress, 
reflecting the closed nature of the budget process where deliberations 
are considered to be ``predecisional'' and not subject to public 
disclosure our debate. This is true not only for the budgetary figures 
and decisions themselves, but also for performance plans and PART 
scores that comprise the President's budget under both Clinton and Bush 
Administrations.

                        WERE DO WE GO FROM HERE?

    PART and GPRA established the valuable expectation that performance 
will inform budget decisions. Continuity is important--the attention 
devoted to performance information by both Clinton and Bush 
Administrations have prompted agencies to take the generation of goals 
and measures and data more seriously. The momentum will hopefully 
continue. The long term budget models of GAO, CBO and OMB have reached 
the consensus that the federal deficits of today are but a prelude to 
the serious and unsustainable deficits of tomorrow. As the baby boom 
retires, the budget challenges will become even more daunting--and the 
first baby boomers retire next year. A performance based assessment 
process will help the nation update its priorities as all programs 
increasingly compete for increasingly limited resources. The question 
is--how do we build on the recent reforms to establish a performance 
assessment process that will be sustainable and relevant over the 
longer term?
    PART has served to continue and heighten the focus of agencies on 
performance and has prompted some to redouble their efforts to more 
systematically evaluate their own programs and improve the base of 
information upon which such evaluations rest. However, now that nearly 
all programs have been reviewed, it is an ideal time to rethink the 
assessment process. Simply continuing the current process through yet 
another round of assessments would not constitute the best use of 
scarce analytic resources in the agencies and the budget community. The 
completion of the cycle of PART reviews, instead, should provide the 
occasion to refocus the attention of the budget community on a broader, 
yet more targeted assessment process that would be more inclusive and 
open than PART has proved to be. The process I have in mind would 
refocus assessments on the broader outcomes that individual federal 
programs are attempting to influence together, building on the results 
of both GPRA planning and individual PART reviews.
    The experiences of other nations can be examined to help us rethink 
our own process. Other nations, too, have undertaken comprehensive 
reexamination efforts. New Zealand, Australia, Canada, and the 
Netherlands, for example, have undergone performance-based budgeting 
and performance management reforms aimed at reprioritizing the base of 
their respective budgets that spanned a number of years. In Canada, an 
OECD study concluded that a program review exercise delivered $18.8 
billion in savings over three years, contributing to the achievement of 
the nation's deficit reduction targets in the mid 1990's.
    In the Netherlands, reconsideration reviews are conducted on both 
particular programs as well as broader crosscutting areas selected for 
each budget cycle, with participation by working groups of central 
budget and departmental staff as well as external experts, resulting in 
a public report with recommendations to be considered. According to 
OECD, the process has been in place since 1981 and has lead to 
significant savings as well as many reforms of major policy areas. 8 In 
their broader crosscutting reviews, which they call Interdepartmental 
Reviews, their system is different than PART in several respects. First 
it is selective, with about 10 reviews each year. Second, it has a 
broader focus as reviews address an entire policy area or 
governmentwide management concern. Third the reviews are more 
collaborative involving a partnership the budget office, agencies and 
outside researchers and academics.
    Our own recent experience as well as that of other nations suggests 
it is time to consider a revised assessment process within the context 
of continuing and enhancing the connection between performance planning 
and budgeting.. Such a process should include the following elements:
     Continuing and enhancing the GPRA planning process as the 
foundation for the assessment process. This process has succeeded in 
generating the ``supply side'' of performance budgeting, producing 
plans and reports with progressively greater coverage and credibility. 
Such a process can be enhanced by the implementation of a 
goverrnmentwide performance plan to capture those important outcomes 
that are the produce of the programs of several different agencies.
     Continuing efforts to tie performance plans and metrics 
into the budget formulation process in both executive and legislative 
stages. This should encompass changes to budget presentations and 
structures to highlight the performance implications of budget 
decisions--an initiative begun under the Clinton Administration and 
carried forward during the Bush Administration under the Budget and 
Performance Integration. The goal of these initiatives is to change the 
debate to focus more on how given levels of resources will achieve 
particular performance goals and targets. Accountability should 
increasingly come to focus on the setting and accomplishment of 
discrete performance targets. These targets can be articulated both in 
budget presentations within the executive process as well as in 
appropriations debates and reports.
     Targeting future assessments based on such factors as the 
relative priorities, costs, and risks associated with related clusters 
of programs and activities addressing common strategic and performance 
goals. More selective reviews would help ration scarce analytic 
resources as well as focus decision makers' attention on the most 
pressing policy and program issues. The Netherlands Interdepartmental 
Reviews provide an illustration of a long standing assessment process 
that selects several broad program areas for review each fiscal year, 
in a process that has been established for over 25 years.
     Adopting a broader unit of analysis keyed to program 
outcomes, not budget accounts. PART's focus is overly narrow and fails 
to focus attention on the most important areas for governmental 
policymaking--the relative contribution of different programs and tools 
to policy outcomes. Thus, rather than producing separate reviews of the 
many different job training programs in differing budget years, a more 
comprehensive review process would cover all significant programs 
together in one assessment. Such a review should cover all significant 
governmental tools address broader outcomes, including the more 
indirect tools such as tax expenditures and regulatory programs.
     Establishing a more open process that is more independent 
of OMB and the agencies. The Netherlands process entails reviews by 
committees often chaired by independent experts, with participation by 
budget and agency officials. There are many options to open up the 
assessment process here. One option that could be considered would be 
to anchor the process in a nonprofit organization while guided by OMB 
and the Congress. While including a broader base of stakeholders, such 
assessments must be truly independent and neutral to attain the 
credibility necessary to influence budgetary decisions.
    Mr. Chairman, it is vital that such a revised assessment process 
engage the Congress. Given the strong role Congress plays in both 
budgeting and management oversight, Congressional involvement will 
prove to be vital in reinforcing the importance of performance 
assessment and ensuring that such a process reflects a broader base of 
values and interests. PART has served the President, but as a result 
was not perceived to serve the interests of other actors in the system, 
most notably the Congress. While Congress has largely ignored PART 
assessments and performance budgets, such a posture risks giving the 
President unimpeded access to the performance high ground. As we have 
learned from PART and even GPRA, performance reviews are anything but 
objective, and turn on highly contestable formulations of program 
goals, selection of particular measures and highlighting of particular 
results. Performance information would be more credible if Congress 
were engaged in selecting reviews, highlighting key differences with 
executive interpretations and using their own information resources to 
challenge executive data.
    Several alternatives might be considered to enhance Congressional 
involvement in program assessment:
     A more collaborative approach could be adopted where the 
new President and the Congress could reach agreement about those areas 
to be assessed in each budget year. Congress could help ensure that 
areas that are ripe for reexamination, such as those up for 
reauthorization, would get attention in the executive review process. 
The resulting assessment process could have greater credibility in the 
process. Such a process would require the Congress to articulate its 
oversight and reexamination priorities centrally and OMB to invite 
Congress to help determine executive assessment priorities.
     Congress could request the GAO to undertake a series of 
assessments of broad program areas deemed in need of reexamination by 
the leadership. The agency would engage in an evaluation synthesis, 
systematically examining the evidence from other studies, including its 
own, of the effectiveness of federal program design for the area in 
question. Such studies could, in effect, constitute Congress' own 
assessment process oriented to the oversight and performance priorities 
expressed by its own leadership in contrast to those articulated by the 
President.
     Congress could adopt its own performance assessment and 
review process linked to the budget process. The Budget Committees 
could be the vehicle to accomplish this by reporting out a 
``congressional performance resolution'' as part of the budget 
resolution. Such a resolution could be the vehicle to engage the full 
Congress in debate over those areas most ripe for review and assessment 
each year. The resolution could be viewed as a requirement that 
committees undertake the assessments through hearings, GAO studies and 
other vehicles that they deem appropriate to reexamine the program 
areas identified in the resolution. This would more forcefully place 
Congress in the forefront of the performance budget process. Just as 
the budget resolution helps identify priorities for scarce resources, 
the performance resolution could jump start a healthy process for 
prioritizing those areas that most warrant a reexamination of goals, 
program design or management processes.
     Budget committees could not only spearhead such a 
resolution but also become a focal point for undertaking more 
systematic oversight of important crosscutting issues, perhaps in 
concert with other committees. Similar to a process undertaken by the 
Senate Budget committee in the l990's, a budget committee driven 
oversight process could add value to the traditional oversight 
conducted by other committees by conducting assessments that both cut 
across current committee jurisdictions and have significance for 
current and future budgets.
    Mr. Chairman, this concludes my statement. I would be glad to 
respond to any further inquiries by the Committee.

                                ENDNOTES

    \1\ U.S. General Accounting Office, Performance Budgeting: Past 
Initiatives Offer Insights for GPRA Implementation (Washington, D.C.: 
GAO, 1997)
    \2\ Organization of Economic Cooperation and Development, 
Performance Information in the Budget Process: Results of the OECD 2005 
Questionnaire (Paris: OECD, 2005)
    \3\ U.S. Government Accountability Office, Results-Oriented 
Government: GPRA Has Established a Solid Foundation for Achieving 
Greater Results (Washington, D.C.: GAO, 2004, GAO -04-38)
    \4\ U.S. Government Accountability Office, Performance Budgeting: 
Efforts to Restructure Budgets to Better Align Resources With 
Performance (Washington, D.C.: GAO, 2005) GAO-05-117SP
    \5\ U.S. Government Accountability Office, 21st Century Challenges: 
Reexamining the Base of the Federal Government (Washington, D.C.: GAO, 
2005) GAO-05-325SP
    \6\ U.S. Government Accountability Office, Performance Budgeting: 
PART Focuses Attention on Program Performance, but More Can Be Done to 
Engage Congress (Washington, D.C.: GAO, 2006, GAO-06-26)
    \7\ U.S. Government Accountability Office, Performance Budgeting: 
Observations on the Use of OMB's Program Assessment Rating Tool for the 
FY 2004 Budget (Washington, D.C.: GAO, 2004, GAO-04-174)
    \8\ OECD, Reallocation: The Role of Budget Institutions (Paris, 
OECD: 2005)

    Chairman Spratt. We appreciate you coming, and we 
appreciate your very methodical and clear set of proposals that 
you sent in your testimony. Thank you very much indeed.
    The hearing is adjourned.
    [Whereupon, at 12:21 p.m., the committee was adjourned.]

                                  
