[House Hearing, 107 Congress]
[From the U.S. Government Publishing Office]





        THE RESULTS ACT: HAS IT MET CONGRESSIONAL EXPECTATIONS?

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

                                HEARING

                               before the

                 SUBCOMMITTEE ON GOVERNMENT EFFICIENCY,
                        FINANCIAL MANAGEMENT AND
                      INTERGOVERNMENTAL RELATIONS

                                 of the

                              COMMITTEE ON
                           GOVERNMENT REFORM

                        HOUSE OF REPRESENTATIVES

                      ONE HUNDRED SEVENTH CONGRESS

                             FIRST SESSION

                               __________

                             JUNE 19, 2001

                               __________

                           Serial No. 107-75

                               __________

       Printed for the use of the Committee on Government Reform


  Available via the World Wide Web: http://www.gpo.gov/congress/house
                      http://www.house.gov/reform

                                 ______

80-373              U.S. GOVERNMENT PRINTING OFFICE
                            WASHINGTON : 2002
____________________________________________________________________________
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpr.gov  Phone: toll free (866) 512-1800; (202) 512ï¿½091800  
Fax: (202) 512ï¿½092250 Mail: Stop SSOP, Washington, DC 20402ï¿½090001

                     COMMITTEE ON GOVERNMENT REFORM

                     DAN BURTON, Indiana, Chairman
BENJAMIN A. GILMAN, New York         HENRY A. WAXMAN, California
CONSTANCE A. MORELLA, Maryland       TOM LANTOS, California
CHRISTOPHER SHAYS, Connecticut       MAJOR R. OWENS, New York
ILEANA ROS-LEHTINEN, Florida         EDOLPHUS TOWNS, New York
JOHN M. McHUGH, New York             PAUL E. KANJORSKI, Pennsylvania
STEPHEN HORN, California             PATSY T. MINK, Hawaii
JOHN L. MICA, Florida                CAROLYN B. MALONEY, New York
THOMAS M. DAVIS, Virginia            ELEANOR HOLMES NORTON, Washington, 
MARK E. SOUDER, Indiana                  DC
JOE SCARBOROUGH, Florida             ELIJAH E. CUMMINGS, Maryland
STEVEN C. LaTOURETTE, Ohio           DENNIS J. KUCINICH, Ohio
BOB BARR, Georgia                    ROD R. BLAGOJEVICH, Illinois
DAN MILLER, Florida                  DANNY K. DAVIS, Illinois
DOUG OSE, California                 JOHN F. TIERNEY, Massachusetts
RON LEWIS, Kentucky                  JIM TURNER, Texas
JO ANN DAVIS, Virginia               THOMAS H. ALLEN, Maine
TODD RUSSELL PLATTS, Pennsylvania    JANICE D. SCHAKOWSKY, Illinois
DAVE WELDON, Florida                 WM. LACY CLAY, Missouri
CHRIS CANNON, Utah                   DIANE E. WATSON, California
ADAM H. PUTNAM, Florida              ------ ------
C.L. ``BUTCH'' OTTER, Idaho                      ------
EDWARD L. SCHROCK, Virginia          BERNARD SANDERS, Vermont 
JOHN J. DUNCAN, Jr., Tennessee           (Independent)


                      Kevin Binger, Staff Director
                 Daniel R. Moll, Deputy Staff Director
                     James C. Wilson, Chief Counsel
                     Robert A. Briggs, Chief Clerk
                 Phil Schiliro, Minority Staff Director

    Subcommittee on Government Efficiency, Financial Management and 
                      Intergovernmental Relations

                   STEPHEN HORN, California, Chairman
RON LEWIS, Kentucky                  JANICE D. SCHAKOWSKY, Illinois
DAN MILLER, Florida                  MAJOR R. OWENS, New York
DOUG OSE, California                 PAUL E. KANJORSKI, Pennsylvania
ADAM H. PUTNAM, Florida              CAROLYN B. MALONEY, New York

                               Ex Officio

DAN BURTON, Indiana                  HENRY A. WAXMAN, California
          J. Russell George, Staff Director and Chief Counsel
                 Earl Pierce, Professional Staff Member
                     Chris Barkley, Staff Assistant
           David McMillen, Minority Professional Staff Member

                            C O N T E N T S

                              ----------                              
                                                                   Page
Hearing held on June 19, 2001....................................     1
Statement of:
    O'Keefe, Sean, Deputy Director, Office of Management and 
      Budget; Christopher Mihm, Associate Director, Federal 
      Management and Workforce Issues, U.S. General Accounting 
      Office; Maurice McTigue, distinguished visiting scholar, 
      Mercatus Center, George Mason University; and John Mercer, 
      Deputy Director for Government Performance, Logicon, Inc...    11
    Thompson, Hon. Fred, a U.S. Senator from the State of 
      Tennessee..................................................    12
Letters, statements, etc., submitted for the record by:
    Horn, Hon. Stephen, a Representative in Congress from the 
      State of California:
        CRS Report to Congress...................................     2
        Prepared statement of....................................     9
    McTigue, Maurice, distinguished visiting scholar, Mercatus 
      Center, George Mason University, prepared statement of.....    53
    Mercer, John, Deputy Director for Government Performance, 
      Logicon, Inc., prepared statement of.......................    63
    Mihm, Christopher, Associate Director, Federal Management and 
      Workforce Issues, U.S. General Accounting Office, prepared 
      statement of...............................................    20
    O'Keefe, Sean, Deputy Director, Office of Management and 
      Budget, prepared statement of..............................    37

 
        THE RESULTS ACT: HAS IT MET CONGRESSIONAL EXPECTATIONS?

                              ----------                              


                         TUESDAY, JUNE 19, 2001

                  House of Representatives,
  Subcommittee on Government Efficiency, Financial 
        Management and Intergovernmental Relations,
                            Committee on Government Reform,
                                                    Washington, DC.
    The subcommittee met, pursuant to notice, at 2:35 p.m., in 
room 2154, Rayburn House Office Building, Hon. Stephen Horn 
(chairman of the subcommittee) presiding.
    Present: Representatives Horn and Putnam.
    Staff present: J. Russell George, staff director and chief 
counsel; Bonnie Heald, director of communications, Earl Pierce, 
professional staff member; Chris Barkley, staff assistant; Alex 
Hurowitz and Ryan Sullivan, interns; David McMillen, minority 
professional staff member; and Teresa Coufal, minority staff 
assistant.
    Mr. Horn. The Subcommittee on Government Efficiency, 
Financial Management and Intergovernmental Relations will come 
to order.
    Today we will examine the progress being made by the 
government's executive branch departments and agencies to 
comply with the Government Performance and Results Act of 1993, 
often called the Results Act.
    [The information referred to follows:]

    [GRAPHIC] [TIFF OMITTED] T0373.001
    
    [GRAPHIC] [TIFF OMITTED] T0373.002
    
    [GRAPHIC] [TIFF OMITTED] T0373.003
    
    [GRAPHIC] [TIFF OMITTED] T0373.004
    
    [GRAPHIC] [TIFF OMITTED] T0373.005
    
    [GRAPHIC] [TIFF OMITTED] T0373.006
    
    Mr. Horn. One of the goals of this bipartisan law was to 
improve the Federal Government's efficiency and accountability 
by shifting its focus away from a preoccupation with day-to-day 
activities. Instead, the law requires departments and agencies 
to focus on the results or outcomes of those activities. Once 
that goal is achieved, Congress will be able to make 
knowledgeable decisions on which Federal programs are worthy of 
support and which should be abandoned. The process is called 
results-oriented budgeting.
    To achieve this goal the Results Act required that 
beginning in 1997, Federal agencies were to submit long-range 
strategic plans to Congress. These plans are updated every 3 
years. The law also required agencies to submit annual 
performance plans and reports on their success in meeting those 
goals.
    Agencies have now had 2 years of experience in developing 
measurable goals and reporting on their success in achieving 
those goals. Yet, many still have difficulty linking their 
long- and short-term strategic plans to the cost of their 
activities, even though this process is supposed to form the 
basis of their budget requests.
    In 1998, House Majority Leader Representative Dick Armey of 
Texas stated that the agency plans failed to address management 
problems and lacked reliable data to verify and validate 
performance. In 1999, the General Accounting Office found that 
only 14 of the 35 Federal departments and agencies it examined 
were able to define some type of relationship between program 
activities on their proposed budgets and the performance goals 
cited in their plans.
    To paraphrase Benjamin Franklin, reforming the government, 
no matter how good the changes, requires educating and 
enlightening the people and convincing them that their 
interests will be promoted by the proposed changes. Said 
Franklin, ``This is not the work of a day,'' and so it may be 
said of the government's slowness in embracing the Results Act.
    Unquestionably, the challenge of genuine reform in the 
Federal Government is formidable. Clearly, implementation of 
the Results Act remains a work in progress. Once completed, 
however, American taxpayers will ultimately have a more 
accountable, better-managed government, which is what they want 
and most certainly deserve.
    Today, we will discuss recent agency progress in complying 
with the law. We are honored to have with us the ranking member 
of the Senate Governmental Affairs Committee, Senator Fred 
Thompson, a Republican of Tennessee, who has been an aggressive 
proponent of the Results Act. Senator Thompson has requested 
the General Accounting Office to review the most recent round 
of agency performance reports, covering fiscal year 2000. The 
General Accounting Office has just reported its findings for 
most of the agencies. And for that we are grateful.
    In addition, we will hear from a panel of witnesses who 
have closely followed the government's efforts to implement the 
Results Act. They will discuss the successes, the failures, and 
the challenges that lie ahead.
    [The prepared statement of Hon. Stephen Horn follows:]

    [GRAPHIC] [TIFF OMITTED] T0373.007
    
    [GRAPHIC] [TIFF OMITTED] T0373.008
    
    Mr. Horn. We welcome today Mr. Sean O'Keefe, the Deputy 
Director of the Office of Management and Budget; Mr. 
Christopher Mihm, Associate Director of the U.S. General 
Accounting Office; Mr. Maurice McTigue, distinguished visiting 
scholar at the Mercatus Center of George Mason University; and 
Mr. John Mercer, deputy director for government performance at 
Logicon, Inc.
    We welcome all of you and look forward to your testimony.
    We will have to wait a little while because Senator 
Thompson is in a meeting right now in the Senate in terms of 
ranking members. But he's on his way and will be here in a few 
minutes.
    And we might start with Mr. Christopher Mihm, and that way 
we'll open it up. You'll probably have a good punctual 
statement, Chris; I understand that. So I think we're going to 
swear you in.
    [Witness sworn.]
    Mr. Horn. Thank you. Please proceed.

    STATEMENTS OF SEAN O'KEEFE, DEPUTY DIRECTOR, OFFICE OF 
 MANAGEMENT AND BUDGET; CHRISTOPHER MIHM, ASSOCIATE DIRECTOR, 
     FEDERAL MANAGEMENT AND WORKFORCE ISSUES, U.S. GENERAL 
  ACCOUNTING OFFICE; MAURICE McTIGUE, DISTINGUISHED VISITING 
  SCHOLAR, MERCATUS CENTER, GEORGE MASON UNIVERSITY; AND JOHN 
  MERCER, DEPUTY DIRECTOR FOR GOVERNMENT PERFORMANCE, LOGICON 
                              INC.

    Mr. Mihm. Thank you, Mr. Chairman. And once again it is a 
great honor and pleasure to appear before you and this 
subcommittee to discuss how the Government Performance and 
Results Act can be used to assist Congress in its oversight and 
decisionmaking.
    Over the last decade, Congress and the executive branch 
have implemented a statutory framework to improve Federal 
agencies' performance and executive branch and congressional 
decisionmaking. That framework includes as its core elements 
financial management and information technology reforms which 
have been the subject of extensive support and oversight from 
this subcommittee, as well as results-oriented legislation, 
especially GPRA.
    Our work confirms the views that you expressed in your 
opening statement, sir; that is, while much work remains before 
this framework is effectively implemented across the 
government, there has been substantial progress in the last few 
years in establishing the basic infrastructure to create high-
performing Federal organizations. The task now is to move to 
the more important phase of GPRA implementation; that is, using 
results-oriented performance information on a routine basis for 
an agency's day-to-day management and congressional and 
executive branch decisionmaking.
    As a Nation, we face two overriding questions to effective 
Federal governance in the 21st century: What is the proper role 
of the Federal Government and how should government do 
business? As detailed in my written statement, GPRA can serve 
as a bridge between these two questions by linking the results 
the Federal Government seeks to achieve to the program 
approaches and resources that are necessary to achieve those 
results. In the interest of brevity, I'll hit just the 
highlights of two issues where we see GPRA making the greatest 
contribution to congressional decisionmaking. Those two issues 
are, first, instilling the results orientation and, second, 
ensuring that daily operations contribute to results.
    First, in regards to instilling the results orientation, as 
you mentioned in your opening statement, Mr. Chairman, the 
cornerstone of Federal efforts to successfully meet current and 
emerging public demands is to adopt a results orientation. That 
is to develop a clear sense of the results an agency wants to 
achieve as opposed to the products and services an agency 
produces, to move away from that preoccupation with outputs 
that you mentioned. Adopting such an orientation requires a 
cultural transformation for many agencies, one that can be 
accomplished only through the active and sustained attention of 
top leadership.
    Unfortunately, our work has shown that attention is too 
often lacking. We recently reported the findings of our 
government-wide survey of Federal managers that showed that 
many agencies faced significant challenges in instilling such 
an orientation. For example, at 11 agencies, less than half of 
the Federal managers perceived, to at least a great extent, 
that a strong top leadership commitment to achieving results 
existed. At 26 of the 28 agencies we surveyed, less than half 
of the managers perceived--and I can suspend comments there in 
deference to the Senator.
    Mr. Horn. Senator, you've got a willing member of GAO that 
will now yield some time to you. And we know you're busy today. 
So thank you very much for coming.
    We're delighted to have the former chairman of Governmental 
Affairs in the Senate. He has taken a great interest in the 
Results Act, and we want to hear what he has to say.

STATEMENT OF HON. FRED THOMPSON, A U.S. SENATOR FROM THE STATE 
                          OF TENNESSEE

    Senator Thompson. Thank you, Mr. Chairman. I'm not used to 
hearings actually starting on time, so I'm having a little 
adjustment to make here.
    But I want to congratulate you for it, and I want to thank 
you for the opportunity of coming over and spending a little 
time on something that for too long has been under the radar 
screen. I think it underlies everything the government is 
trying to do, and that is trying to develop a government that 
is more results-oriented--instead of looking at inputs, look at 
what government is actually achieving.
    As you know, the Results Act requires Federal agencies to 
develop 5-year strategic plans, annual performance plans, and 
submit annual performance reports. The act is intended to shift 
the focus of accountability from process to results. What 
matters is what these activities actually accomplish in real 
results that are important to the American people, such as 
things like workplace safety, fewer transportation accidents, 
less crime, better education and healthcare.
    All this sounds like basic common sense, and it is. We need 
to work harder to reorient the Federal Government's thinking. 
Setting results-oriented performance goals and then using them 
to track progress, make resource decisions and manage day-to-
day operations should come as second nature. Many State and 
local governments operate this way, as does much of the private 
sector. However, these concepts represent a fundamental 
cultural change in Washington.
    Moving the Federal Government in this direction has been a 
real struggle. We've now completed two rounds of performance 
reports under the Results Act covering fiscal years 1999 and 
2000. Both sets of performance reports were analyzed by the 
Mercatus Center, George Mason University, the General 
Accounting Office and the Senate Governmental Affairs Committee 
and others. All these analyses demonstrate that we have a long 
way to go before these reports will be anywhere near as 
informative and useful as they ought to be.
    Let me touch on four major problems that we see with these 
reports: While the fiscal year 2000 version shows some 
improvement in each of these four areas, these problems affect 
most agency reports for both years 1999 and 2000.
    First of all, with regard to assessing results, we can't 
tell from most of the reports whether agencies are making any 
progress toward achieving key performance results or not.
    In each of the past 2 years, I asked the GAO to determine 
from the performance reports how well agencies were actually 
achieving certain key outcomes. In all, we looked at over 90 
key outcomes across 24 major agencies. That includes such 
things as maintaining the Nation's combat readiness, 
maintaining the security of the U.S. borders, ensuring our tax 
laws are administered effectively, helping poor and 
disadvantaged families become self-sufficient, denying 
criminals access to firearms, reducing the availability and use 
of illegal drugs--things of that nature.
    We found for the year 1999, again, fiscal year 2000, that 
you can't really tell from the reports whether the agencies 
were making progress on many of their key outcomes including 
the ones that I just mentioned. With the subcommittee's 
permission, I'd like to submit for the record a table that 
summarizes our findings in this regard.
    Mr. Horn. Without objection, it will be put in the record 
at this point.
    Senator Thompson. Thank you very much.
    Second, we looked at making sense out of these overlapping 
programs that we have so many of. The performance reports don't 
tell decisionmakers or the public what's working and what's not 
within the mass of overlapping programs that exist in virtually 
every area of Federal activity. This is a huge problem when you 
consider how much overlap and duplication we have.
    I just issued a two-volume report called Government at the 
Brink. I'll leave a copy of this report for the subcommittee.
    Among other things, this report describes the extent of 
duplication and overlap that we have. Just to cite a few 
examples: 7 different agencies operate 40 different job 
training programs; 18--I'm sorry, 8 different agencies operate 
50 different programs to aid the homeless; and 17 agencies 
operate 515 research and development labs.
    These multiple programs hardly ever use consistent 
performance goals and measures that allow for comparisons among 
them. You're always comparing apples and oranges to pears. 
Nobody can seriously argue that all these programs are equally 
effective and necessary, yet we lack the performance 
information to make rational choices among the programs and 
allocate resources to where they'll do the most good.
    Another area we asked for, had to do with inadequate 
performance data. One fundamental barrier to the usefulness of 
the performance reports is the lack of reliable and timely 
performance data.
    Mr. Chairman, I think it has to do, too, with the overall 
high-risk area that the GAO tells us about, having to do with 
financial mismanagement of the Federal Government. It is a 
growing problem that erodes the basis of all else that 
government is trying to do.
    And I think of trying to come up with data, accurate data 
that we can measure, so we can see whether or not these 
agencies are doing any good, we fall flat. And it's a part of 
that bigger problem. This is the data that shows whether we're 
meeting the stated goals these agencies have for various 
problems.
    GAO reports only 3 of the 24 major agencies can produce 
credible performance data. Now, this is a law. As you know, it 
was passed back in 1993. Most inspectors general likewise 
question the credibility of their agencies' performance data.
    Congress passed a law last year intended to get agencies to 
pay attention to their data problems, and explain what they're 
doing to solve them. Unfortunately, it appears that few 
agencies took this mandate seriously. Agencies often failed to 
acknowledge specific data problems that had been highlighted by 
GAO and their IGs. I've asked the GAO to look at this in more 
depth.
    I would also like to submit for the record my request 
letter to the GAO which outlines my concerns.
    Mr. Horn. Without objection, it will be put in the record 
at this point.
    Senator Thompson. Thank you.
    Also, with regard to the area of resolving mission-critical 
management problems, I've heard the agency for years used the 
Results Act as a tool to resolve these massive management 
problems. I pointed out financial mismanagement and lack of 
ability to handle information technology projects. We spend 
billions of dollars, then throw up our hands and give up on the 
projects.
    The revolution that's happening out there in the private 
sector, as you well know, Mr. Chairman, has not been brought 
into government. We don't get the benefit of it, but we spend 
billions of dollars trying to.
    The human capital problem is that our people don't fit our 
jobs anymore. We downsize without any strategic plan at all. 
And the high-tech world we live in, when we need certain people 
in certain areas that are crucial to national security and to 
our government, we're losing those people. We're not recruiting 
new people. Our civil service system is out of date. Overlap 
and duplication, are massive, crucial, endemic management 
problems that the GAO has so eloquently told us about, and they 
threaten our ability to achieve these performance results. The 
Government in the Brink report I mentioned, catalogues these 
problems in detail.
    As you are well aware, most agencies have had critical 
financial management problems that make them highly vulnerable 
to waste, fraud and abuse. I'm happy to say that most agencies 
are now doing a better job of acknowledging these critical 
management problems in the performance report and describing in 
general terms what they're trying to do to resolve them. 
However, to generate serious action, agencies need to establish 
specific and measurable performance goals to directly address 
these problems, publicly track progress and ensure 
accountability. There is so much room for improvement here.
    Unfortunately, agencies have established direct performance 
goals for less than half of the critical management problems 
that the GAO and their IG's have identified. So here we are, 
all these years later, and finally we haven't managed to even 
identify goals for half of these problems that everybody knows 
are crucial to the operation of our government. Somebody is not 
paying attention.
    In short, there are still major deficiencies in agencies' 
performance reports and plans in the Results Act, and the act 
does not even come close to reaching its potential as a tool to 
improve government performance.
    That said, the more important point is that the act is only 
a tool. It must be used to be effective. Good plans and reports 
are not ends in and of themselves; the end result is that 
managers and policymakers use the information they provide to 
help government performance.
    So far, this hasn't happened. Results Act information has 
yet to be used significantly by either the executive branch or 
congressional decisionmakers to oversee program performance, 
allocate funding or reform non-performing programs. Likewise, 
according to a just-released GAO report, the Results Act has 
yet to take hold in the agencies as a tool for day-to-day 
management. The GAO reports leadership commitment is needed to 
sustain high levels of performance is not widely perceived 
among managers across government, and progress in fostering 
such leadership has remained stagnant.
    My staff handed me something just today, Mr. Chairman, I 
think points this out probably better than anything else. It's 
from the Federal Times of June 18th, and it says, ``Defense may 
be the next performance push.'' As you know, defense has been a 
poster child for financial mismanagement.
    Mr. Horn. Right.
    Senator Thompson. They can't balance their books, they 
can't pass an audit. They lose billions of dollars in stuff 
they can't account for. After all this Results Act talk, let's 
look and see how they look at it over there.
    It says, ``Some former and current defense insiders said 
they suspect the benefits of DOD's complying with the Results 
Act will be limited.''
    `` `Most of the blood has already been squeezed out of the 
turnip on the business side,' said Robert Sole, Director of 
Program Analysis and Evaluation for the Office of the Secretary 
of Defense.''
    ``There are not much savings that can be expected from 
changing business practices.''
    `` `The Defense Department is not focused on the Results 
Act, and the senior leadership is not interested in it,' Sole 
said. `Nor are many in Congress,' according to Robert Hale, 
former Assistant Secretary of Financial Management and 
Comptroller of the Air Force. `During my tenure as Air Force 
comptroller, I've never been asked about the Results Act by 
Defense authorizing and appropriating committees.' ''
    I think that's a pretty serious reflection on both the 
Department of Defense and Congress.
    Can we turn this around? It seems to me that the jury is 
very much out on this question. What we seem to have now is a 
chicken-and-egg situation. A huge amount of time and effort has 
gone into implementing the Results Act, but it's not yet 
produced much really useful information. For example, the 
Congressional Budget Office stated that, ``Its analysis found 
little in Fiscal Year 1999 performance report to guide the 
Congress in making choices about spending.''
    On the other hand, there is little incentive to invest lots 
more time and effort into making the information more useful, 
unless and until it's used. We need to break out of this cycle 
and start using performance information agencies and have them 
produce as best we can. Once we do this, the agencies will 
start paying attention and will start producing better 
information.
    We can't let a desire for perfection be the enemy of the 
good. The important thing now is to get the ball rolling. If we 
can't do it, we might as well hang it up and consign the 
Results Act to the scrap heap of failed management reforms. We 
might as well go on with business as usual in Washington, where 
expectations of the Federal Government are so low that we 
simply accept high levels of waste, fraud and inefficiency as 
the normal cost of operation, and where each year we basically 
throw money at programs that sound good, and simply accept with 
blind faith that they're accomplishing something.
    Obviously, I hope this doesn't happen. I'm very encouraged 
that the executive branch leadership is firmly committed to 
turning things around. OMB Director Mitch Daniels reaffirmed in 
a press conference, as he did with me on the Government at the 
Brink report, that using the Results Act to actually make 
decisions and run the government is a top priority for the 
administration. I'll certainly do everything I can to help make 
this a reality. I challenge my colleagues in the Senate and the 
House to do likewise. We can't afford to let this opportunity 
pass us by.
    Thank you, Mr. Chairman. And I would be remiss if I didn't 
acknowledge and express my appreciation for your leadership in 
this area. Thank you very much.
    Mr. Horn. Well, thank you, Senator.
    Usually we don't ask questions of Senators and our own 
colleagues, but let me go for one that I think you and I have 
talked about before; and that is, we need some of the 
leadership on both sides of the rotunda--and that would be the 
majority leaders working together with the minority leaders, 
and the people that are elected in the Senate, not just the 
staff, and the people elected in the House, not just the 
staff--and get them in a room of the people that the President 
has appointed as his Ambassadors, and get them around the table 
to ask, ``Is this what we really mean by this particular 
program?''
    And often they would find, ``Wait a minute, folks, just 
because we went into conference and tried to get this thing 
into a series of euphemisms so we can get out of here, and try 
to get on, what is the particular goal that should be sought, 
is it working, is it measurable, so forth?'' In all the papers 
you've read--and you've read a lot of them, more than most 
Members in either body--what's the best standard of measurement 
that you've seen that would make some sense?
    Senator Thompson. Government-wide or with regard to any 
particular----
    Mr. Horn. Government-wide or particular.
    Senator Thompson. Well, I haven't seen very many good ones. 
But if you were talking about specific goals that are supposed 
to be set out in a performance report, we have some that 
demonstrate the problem.
    With regard to the Commerce Department, they say that they 
have a goal to assess its success in keeping the United States 
secure from proliferation of dual-use commodities and chemical 
weapons. The Bureau of Export Administration sets goals like 
the number of strategic industry analyses completed, the number 
of enforcement outreach visits conducted, and the number of in-
use visits conducted, thus, Commerce progress related to this 
particular outcome. In other words, how many pieces of paper we 
shuffle, how many times we answer the telephone and things of 
that nature. This is endemic.
    I don't think that throughout all these--the Department of 
Energy's Office of Counterintelligence reported that they 
performed 11 inspections during fiscal year 2000, but failed to 
note what security improvements resulted from those 
inspections. In other words, total input and not output is 
endemic across government.
    I don't think that there is any government-wide performance 
standard. I think you have to look at every agency, and every 
agency has to identify what its goal is, or set-up goals are, 
and what its mission is. Why did we create it? If we were 
starting over again, would we create this agency or this 
department? And if so, why? What are we really trying to do 
here? It's not to shuffle pieces of paper, it's to save lives. 
It's not to answer telephones, it's to catch people, illegal 
people at the borders, you know. What are the results of what 
you're trying to do?
    It's extremely difficult to focus agencies in on that, on 
an agency-by-agency basis. I don't think you could ever do it 
government-wide. But I think that, with leadership, it can be 
done.
    I think we're gradually moving a little bit in the right 
direction. And that's why my comments sometimes might seem a 
little bit rough, but it's time we got some people's attention 
on this if we're serious about it. I mean, we pass a lot of 
legislation, you know, we might as well ignore, I suppose, or 
repeal; let's decide if this is one of them. If it is, let's 
get rid of it and forget it and just say that a few billion 
every year is the price we pay for democracy and don't ask any 
questions about it.
    But if we don't take that attitude, if we think we can do 
better, then let's have some leadership. The President is going 
to have to lead. Members of Congress are going to have to lead. 
It's going to have to be integrated in the authorization 
process, probably more importantly integrated into the 
appropriations process. Until people are required to give 
decent reports, No. 1, and then looking at those reports, you 
can tell that somebody is succeeding or failing and they're 
either rewarded or punished based on that--until that happens, 
this is a meaningless exercise.
    Now, as I say, I think we're glacially moving in the right 
direction. But there is going to come a time when we need to 
decide, are we serious about this or not. And thanks to people 
such as yourself over here, I still think that we've got an 
opportunity to do some good, because we've got such a low floor 
to operate off of. We ought to, by accident, be able to save a 
few billion dollars a year without really doing very much. To 
that extent, I'm optimistic.
    Mr. Horn. Well, we appreciate you coming over here and 
sharing your ideas. They're very pertinent and very useful to 
this issue.
    Senator Thompson. Thank you very much.
    Mr. Horn. Thank you.
    We will now go back to Christopher Mihm, Associate 
Director, Federal Management Work Force Issues, U.S. General 
Accounting Office, so he can go through his preparation. And 
then we'll go back to the administration represented by Sean 
O'Keefe, Deputy Director, Office of Management and Budget.
    Mr. Mihm. Thank you again, Mr. Chairman, Mr. Putnam. It's 
again an honor to be here.
    When I suspended my comments in deference to the Senator, 
we were talking about the importance of top leadership 
attention. And I would just underscore what the Senator was 
saying, both in its importance, and that attention has often 
been lacking.
    Senator Thompson hit the highlights of our recent report 
containing the survey results of Federal managers across the 24 
CFO Act agencies. And just to reiterate some of those, at 11 
agencies less than half of the managers perceived, to at least 
a great extent, that a strong top leadership commitment to 
achieving results existed. So this is less than half of the 
managers.
    At 26 of the 28 agencies, less than half of the managers 
perceived, to at least a great extent, that employees received 
positive recognition for helping the agency accomplish its 
strategic goals.
    At 22 agencies, at least half of the managers reported they 
were held accountable for the results of programs, to at least 
a great extent, but at only one agency did more than one-half 
of the managers report they had the decisionmaking authority 
that they needed to a comparable extent. In other words, at 
only one--and that's OPM--of the 28 agencies we surveyed did 
over 50 percent of the mangers say they had the decisionmaking 
authority they needed in order to achieve results.
    To build leadership commitment and to help ensure that 
managing for results become a standard way of doing business 
some agencies using performance agreements need to define 
accountability for specific goals, monitor progress, and 
evaluate results. These are basically contracts rather than 
between government and private sector, between the senior 
political leadership and the senior career leadership.
    We reported last October that performance agreements can be 
an effective mechanism to align the daily activities of 
agencies with results that take place outside those agencies. 
This then leads to the second point I was going to cover this 
afternoon.
    GPRA is showing itself to be an important tool in helping 
Congress and the executive branch assess how agencies' daily 
activities contribute to results that benefit the American 
people. As Senator Thompson underscored, GPRA provides a 
vehicle for examining agencies' internal management 
capabilities and ensuring that they are positioned to achieve 
results.
    As you know, Mr. Chairman, this past January we updated our 
high-risk and performance accountability series that outlined 
the major management challenges and program risks that Federal 
agencies face. Unfortunately, as the Senator noted, we have 
found that agencies are not consistently using GPRA to show how 
they plan to address these challenges and risks.
    For example, when we looked at agencies' fiscal year 2001 
performance plans, we found that there was a need to increase 
the breadth, depth and specificity of goals and strategies 
related to human capital and to better link them to agencies' 
program goals. They didn't have the goals they needed, and 
often when those goals were there, they weren't linked back to 
the programmatic results agencies are trying to achieve.
    In summary then, Congress and the executive branch working 
together have put in place a management infrastructure with 
GPRA as its centerpiece. However, much more needs to be done 
before this infrastructure is effectively implemented across 
the Federal Government. The planning and reporting efforts 
under GPRA nevertheless are generating new and important 
information that has not been available in the past; 
information that congressional and executive branch 
decisionmakers can use to help assess what government should do 
in the 21st century and how it should do it.
    Mr. Chairman, Mr. Putnam, this concludes my statement; and 
I would be pleased to respond to any questions you may have.
    Mr. Horn. Thank you very much.
    [The prepared statement of Mr. Mihm follows:]

    [GRAPHIC] [TIFF OMITTED] T0373.009
    
    [GRAPHIC] [TIFF OMITTED] T0373.010
    
    [GRAPHIC] [TIFF OMITTED] T0373.011
    
    [GRAPHIC] [TIFF OMITTED] T0373.012
    
    [GRAPHIC] [TIFF OMITTED] T0373.013
    
    [GRAPHIC] [TIFF OMITTED] T0373.014
    
    [GRAPHIC] [TIFF OMITTED] T0373.015
    
    [GRAPHIC] [TIFF OMITTED] T0373.016
    
    [GRAPHIC] [TIFF OMITTED] T0373.017
    
    [GRAPHIC] [TIFF OMITTED] T0373.018
    
    [GRAPHIC] [TIFF OMITTED] T0373.019
    
    [GRAPHIC] [TIFF OMITTED] T0373.020
    
    Mr. Horn. We'll now have Mr. O'Keefe come to the table. And 
we might as well swear in everybody at once here. So if the 
gentlemen from George Mason University and Logicon will raise 
your right hand.
    [Witnesses sworn.]
    Mr. Horn. The clerk will note all witnesses have taken the 
oath.
    And we will now get to the representative of the 
administration, Mr. O'Keefe, Deputy Director, Office of 
Management and Budget. We know you're a very busy person, but 
we'd certainly like to get your testimony in the record and 
some questions.
    Mr. O'Keefe. Thank you, Mr. Chairman. It's a pleasure to be 
with you and Mr. Putnam this afternoon. I'd like to submit for 
the record the prepared statement and summarize that if I 
could, for just a couple of moments.
    Mr. Horn. All of these statements go in the record 
automatically once we call you in.
    Just remember on that pleasure bit that you're under oath.
    Mr. O'Keefe. In that case, I'm just delighted to be here to 
see you, sir.
    For all that I've learned about the Government Performance 
and Results Act over the last several years, I owe a tremendous 
debt of gratitude to a great faculty colleague friend of mine 
at Syracuse University by the name of Patricia Ingraham, who is 
the director of the Alan K. Campbell Institute for Public 
Affairs there, and has been conducting a government performance 
project for several years now.
    Since enactment of the act, Ms. Ingrahm, Government 
Executive magazine and George Washington University have been 
engaged in an effort I'm sure your subcommittee is well aware 
of to measure State, local and Federal performance criteria and 
to publish those results on an annual basis. As a consequence, 
it has had a very compelling effect in changing the approach of 
many different levels of government--different government 
institutions--and resulting in a routine and regular assessment 
of how the performance is conducted.
    I'm a native New Orleanian from, New Orleans, LA, where the 
city itself, under this particular criteria that the project 
was conducted, I think overall got a D minus. As a result, the 
citizenry of that city found it necessary and appropriate to 
comment on the city management. Their comments were quite loud 
and well understood in terms of the level of satisfaction 
relative to those performance criteria which had not been 
conducted for a long time.
    It made the city management far more defensive about 
exactly what they were doing, but in the process of doing so, 
drew attention to the conduct of activity there, which had not 
been felt for a very long time. So, at a minimum, this 
particular effort on the part of so many different 
institutions, be it Federal, State or local, for the purpose of 
trying to identify where the shortcomings or advantages and 
successes of various public institutions have resided, is 
nonetheless, I think, an opportunity to focus the attention on 
what citizens expect relative to what is appropriate. It also 
measures what that performance requirement is. Sometimes it can 
have some very dramatic results as a result of that attention.
    Having said that, I'm sure, Mr. Chairman, it comes as no 
surprise to you, that I concur wholeheartedly to your 
assessment and that of Senator Thompson. We find ourselves in 
the Federal Government, across the board in every department 
and agency, implementing the Government Performance and Results 
Act. I think it's been discouraging at best.
    Again, having seen this from a distance and looking at the 
effort--as Senator Thompson alluded to, in State and local 
governments--there are so many that are so far ahead of where 
we are that as a consequence, I think this has been not a 
particularly impressive implementation effort. Although we've 
done a great job, from what I can gather, in trying to 
determine how to comply with the law at its minimums; and that, 
as a result, may in turn be the nature of the problem we're 
confronting.
    This is a way of expressing a little more, I think, to what 
Senator Thompson referred to as the ``cultural phenomenon'' 
that we're dealing with here. GPRA, with all due respect, has 
been treated by Federal agencies and departments, by and large 
as another reporting requirement; something else that needs to 
be complied with. And as a consequence of that, for it to be 
useful, for it to be really useful for any management purpose, 
it has to be introduced into the regular day-in-and-day-out 
management processes that are conducted throughout every 
Federal department and agency. It's nowhere near there. There 
are a very few interesting examples of how it's beginning to 
take hold, but those are noticeable by their distinguished 
nature of being so few, not because it's pervasive.
    I am influenced very heavily by a mind-set that is captured 
as follows: government should be results-oriented; guided not 
by process, but guided by performance. There comes a time when 
every program must be judged either a success or a failure. 
When we find success, we should repeat it, share it and make it 
the standard. When we find failure, we must call it by its 
name.
    A government action that fails in its purpose must be 
reformed or ended. That was President Bush's comment during the 
campaign, and he has lived by that since Inauguration Day of 
this year.
    And, as a function of the management agenda that he has 
just completed, there are five primary issues on that agenda. 
No. 1 on that agenda, is the integration of the the performance 
criteria within the budget itself. This is, in my estimation, 
the only way that we're ever going to see a tangible kind of 
improvement, not in terms of reporting requirements, not in 
terms of producing lots of strategic plans that make mighty 
fine doorstops, but instead become a management objective that 
is laid-out and measured. Therefore, success or failure is 
determined each year in the budget process by whether or not 
those resources are provided to recognize the appropriate means 
by which to accomplish that task.
    In a career engaged in the resource management business, by 
and large the dominant part of my professional career in public 
service, I have found it frustrating that the process culture 
in the administration and, with all due respect, in Congress as 
well, is largely input-oriented. We look at individual parts, 
different items, whatever else. We measure success or failure 
by the percentage differential, the delta between last year and 
this year.
    Headlines of newspapers celebrate increases by, you know, 
multiple percentage points, and lament decreases by the same 
amount as if somehow that was a measure of how well or poorly 
various programs are performing. As a consequence of that, we 
are doomed to the proposition, under this kind of approach and 
this kind of process, that every single year looks at that 
delta as if everything that was performed the prior year was 
done absolutely to perfection. As a result all we're arguing 
about is dollars at the margin, the differential between last 
year and this year.
    That's the approach that is taken within every department 
and agency. It's taken by the various committees of Congress, 
as well. In looking at those relative measures of failure or 
success, we spend more time analyzing that difference than 
anything else and presume again that the composition or conduct 
of how programs are conducted is just fine, when all we're 
doing is just measuring increments.
    Well, beginning in 2003 as a consequence of the President's 
initiative, the No. 1 item on his five-item management agenda, 
the budget will incorporate specific linkages of performance 
criteria and indicators with the budget requests for very 
specific programs. Let me quickly give you the criteria for 
those programs, or those projects, in those agencies and 
departments that will be reflected in the 2003 budget that will 
carry these criteria and what they're going to call for.
    In order to apply this, it's got to be clearly stated. No. 
1, there has to be a specified, desired outcome that is 
articulated; and it can't be, with all due respect to my good 
friends and colleagues at the State Department, stating that 
for the purposes of accomplishment of agenda for the State 
Department, achievement of world peace is the outcome. It needs 
to be a lot more specific than that.
    It needs to be narrowed down to a specific program 
objective that can be seen, and has a result that is measurable 
for which we can see the distinction. That's going to take time 
and a lot of negotiation. An awful lot of platitudes can be 
passed off as outcomes. As a result, there needs to be an 
effort for sorting through it all. In order to really follow 
through to make this a meaningful effort, it's going to require 
that we be diligent and purposeful about how we define those 
outcomes. If we fail in our definitions, we will have to wonder 
about why the program is in business in the first place and 
resort back to President Bush's quote on this point.
    Second, there is going to be an examination of multiple 
means to accomplish that outcome, not just the one stock way 
it's going or the way it's been done. It has to be an 
examination of what the alternatives are, how you achieve that 
same result that is defined.
    Third, there has to be a third identification of the 
outputs. What are we using as a means to determine whether or 
not that result, that outcome, has been achieved? What would we 
use for the purpose of defining that as a performance measure?
    The fourth, that there be a complete--and I mean in its 
truest sense, a complete inventory of all of the inputs 
required. By virtue of the fact that we look at the budget as 
an input-oriented kind of process from agencies and departments 
all the way through Congress--we are fixated and more focused 
on itemizing individual inputs in ways and arrays that suit 
budget officers, not program managers. As a consequence, it is 
easier for them to array those kinds of expenses and identify 
them in order to comply with the requirements that you can 
spend no more than what's been appropriated for those purposes, 
and less so on the program's success.
    So, as a consequence, all the inputs required have to be 
tallied up, rather than buried in lots of different locations; 
and that's going to take work.
    So, when you think through these criteria, it basically 
means that there's going to be a selected number of programs 
and projects that can pass through this kind of test that 
require this many answers to that many circumstances.
    And last, but maybe most important, from my view with a 
bias as a resource manager, is the identification of cost-per-
output, which includes all the costs to accomplish and achieve 
the task. Right now, what we do more often than not is satisfy 
ourselves as long as we captured most of the costs; maybe it's 
a good enough reflection. Now, until we get to the stage where 
we are accurately measuring all the expenses it takes to carry 
out the task and the manner in which it's being done--we'll 
never be able to appropriately judge what the cost or the 
relative performance comparison would be of any other way to 
achieve the same result. As a result, this becomes the fool's 
errand that we have been trapped by for so many years of 
permitting or being permissive about how we capture costs as we 
go through this.
    So, in that regard--and I'm very hopeful that this 
committee will be helpful in this endeavor in the weeks ahead--
we hope to submit or advance to you a legislative proposal to 
begin this long effort, which it's going to take, in order to 
capture all these costs to truly measure what the cost-per-
output will be. And it begins within the weeks ahead.
    Again, a couple of initiatives we're going to advance for 
the purpose of trying to calculate just the cost of all 
expenses to support individual full-time-equivalent civilian 
personnel throughout the Federal Government. As it stands now 
there are lots of different ways to measure that, and lots of 
different ways in which the sources and costs of individuals to 
be supported are, in some cases, budgeted directly within a 
department or agency, and in other cases, they're budgeted 
centrally through other parts of the Federal Government and 
allocated back.
    As a means to at least try to corral all the parts that go 
with that, this is the first step in a long series of efforts 
we hope to enlist your support and endorsement of, to at least 
begin that process of capturing all expenses necessary and then 
working through the criteria described here a moment ago.
    In conclusion, I would say that as the 2003 budget is 
submitted, there will be a selected number of programs that are 
going to meet these criteria. These are very rigid, extremely 
specific kinds of criteria that have to be complied with, and 
errors or compromises along the way of how you would conduct 
this particular approach are, in turn, the very things that 
would bring it to its downfall.
    So, as a result, we have to be very diligent in making sure 
only those programs we incorporate within the 2003 budget fully 
meet these criteria. Absent doing that, all we've done is 
simply created yet another case in which we're starting 
criteria that cannot be accurately measured. But in that 
regard, we hope to bring life to GPRA, use the tools that were 
enacted several years ago for the purposes for which they were 
intended, and carry them out in a way that we can assure a 
little more reasonably, that indeed, they are an accurate 
reflection of how well the performance of various programs will 
exist.
    Again, thank you for the opportunity to testify, Mr. 
Chairman. It's a pleasure to see you.
    [The prepared statement of Mr. O'Keefe follows:]

    [GRAPHIC] [TIFF OMITTED] T0373.021
    
    [GRAPHIC] [TIFF OMITTED] T0373.022
    
    [GRAPHIC] [TIFF OMITTED] T0373.023
    
    Mr. Horn. Thank you.
    Let me ask a few questions before you leave here. Are you 
putting together a team within OMB to help you on the 
measurements and the bringing together and getting rid of the 
stagnation, but getting performance-based management and budget 
all on the same thing by the year 2003? Is that on track now? 
What are your plans as to how you'll get some help?
    Mr. O'Keefe. Again, I am of the mind that the talent for 
this resides throughout this entire Federal Government. This is 
a very professional workforce. There is no question, if 
properly challenged and focused in the direction of this 
endeavor, there is no absence of talent to accomplish this 
task.
    What I defined are real basic fundamentals of how you go 
about devising performance relative to budget criteria. This is 
not something that in my professional experience, in lots of 
different places, both here on Capitol Hill, as well as in an 
agency and department and on the White House staff, there has 
never been any absence, in my mind, of talent sufficient to 
carry this out. It means we must be more specific about the 
guidance we want to see implemented.
    In that regard, I think we are very well equipped from a 
staff-talent standpoint within the Office of Management and 
Budget to devise that guidance properly and to do it in a way 
that is more specific. And again we have to be. The biggest 
challenge is not necessarily whether there is sufficient depth 
of talent within the organizations, it's being specific about 
the guidelines that meet these criteria that will pass muster 
for inclusion in the 2003 budget and thereafter. I expect there 
will be a limited number that we'll see in 2003, but the ones 
we'll see will truly be a reflection of the concept I've talked 
about today.
    Mr. Horn. Well, what will you do with the budget examiners 
in this role? They're going to be going through the same thing 
they've done for 40 years, and it's going to occur in certain 
months. So where's the programmatic analysis, as opposed to 
simply the budget analysis? And do you have those people in 
mind to help you with that?
    Mr. O'Keefe. Well, sir, I must confess, in 1978, I started 
off at GS-0 step, minus 23. I am therefore a Luddite and 
testimony to the fact that we're trainable. There is, I think, 
lots of opportunity to make sure that the talent pool we have--
the very professional financial management talent we have 
throughout the Federal Government in departments and agencies--
can begin to look at these issues as program analysts and more 
in terms of analyzing this particular approach to it rather 
than concentrating on the increments.
    Again, with all due respect, sir, it is the entirety of the 
process that motivates our financial managers, our resource 
managers throughout the Federal Government to focus on the 
incremental differences between last year and this year. 
Because that's what every department and agency seeks, that's 
what the headline of every newspaper demands, and that's what 
the Congress looks at in making appropriations year after year. 
Until we break that circumstance, we should not be surprised 
that resource managers respond exactly as we've trained them 
to.
    This is an effort to begin going down another road, which 
is again based in its logic on basic simplicity, asking the 
fundamental question, what is it you seek as a result and an 
outcome? Why is it in business in the first place? What are the 
elements and expenses required to carry it out? What are the 
different ways you go about doing it? How do you take process 
steps out and then measure it, based on its success or failure? 
It's not that rough.
    As a result, it just means we've got to be disciplined 
about it and start to provide the motivations, as well as the 
incentives, for people to begin to look at it that way. And 
that's why we're carefully selecting those programs and 
projects that will meet these criteria rather than trying to do 
it across the board.
    Mr. Horn. Have you had a chace to look at the Mercatus 
studies that----
    Mr. O'Keefe. Yes, sir.
    Mr. Horn. What's your reaction?
    Mr. O'Keefe. I am well familiar with the results and the 
accomplishments of Mr. McTigue in that regard. I guess, suffice 
it to say, that if the results and achievements that he and the 
Mercatus study reveal in this circumstance can be achieved in 
the circumstances they had to deal with, by goodness, we should 
have a leg up on the challenges we have to sword with. Because 
at least it's focused in the same direction, it's focused on 
the same results that we're looking for in terms of 
improvement.
    I don't think there's a Federal manager out there that's 
driven by malice or different political, philosophical 
objectives that is diametrically opposed to a democratic 
system. As a consequence, we're all on the same page in that 
regard, and it should be a much easier hurdle to deal with.
    Mr. Horn. Let me yield 10 minutes to the gentleman from 
Florida for questioning the witnesses.
    Mr. Putnam. Thank you, Mr. Chairman.
    Beginning with Mr. O'Keefe, what institutional changes 
would you recommend to the Congress in terms of the bifurcated 
authorization and appropriations process, the fact that 
agencies have barely had an opportunity to begin spending their 
budget when they're making preparations for the next budget 
cycle? What institutional changes would--should Congress make 
to have a more performance-driven executive branch?
    Mr. O'Keefe. Well, again I would suggest that it is 
incumbent upon the administration, as it has been for as long 
as, you know, the Budget Act of 1921, that the President 
propose the programs and various resource levels that he 
considers to be appropriate and necessary to conduct and carry 
out the conduct of the Federal Government.
    As a result, it is incumbent upon you to set the debate in 
that regard, and the Congress, in turn, to determine what is 
the best way for the power of the purse to be exerted and to 
determine what you think is the most efficient way to dispose 
of that particular set of proposals.
    And as a result, in my mind, I don't think organizationally 
or institutionally there is any dramatic change that needs to 
occur other than simply to be on the same proposition we've 
discussed here, which is that the outcome, the result, that 
we're seeking is what will become the criteria. And, therefore, 
flexibility of how the appropriations structure may be made, 
how the various programs may be justified, is more driven by 
that particular objective rather than what is the input and how 
many fulltime equivalents are there at Depot X or Y or whatever 
else.
    That will all flow from the information. It will all be 
readily apparent and visible. If, instead, the result or the 
concentration is looking at what the performance result is, I 
don't think there needs to be a dramatic change in the way the 
authorizing and appropriations process works throughout the 
Congress, other than an acceptance of a change in recognition 
of those committees in examining how you measure what success 
is.
    Mr. Putnam. You don't think there should be an increase in 
resources and time spent by the Congress on oversight, as the 
Senator pointed out in panel one?
    Mr. O'Keefe. Well, sir, again I've spent half of my public 
service time as part of the congressional institution and the 
other half in the executive branch in various positions. In the 
course of that time, I've learned that I've never been in a 
position where I've asked for more congressional oversight, 
sir.
    Mr. Putnam. Mr. Mercer, do you think that there are any 
congressional institutional changes that could result in a more 
performance-driven Federal Government?
    Mr. Mercer. Certainly. And would you like me to address 
those now or wait until after I've made my statement? What 
would be your preference?
    Mr. Horn. Why don't we do both? Let's answer the questions 
now and then you're going to have a chance to give your 
statement.
    Mr. Mercer. If I were king and could impose some changes on 
Congress that I thought would make it more results-oriented, 
two thoughts come to mind.
    One is a thought that was originally reflected in the 
earliest drafts of the Results Act, which would have required 
that Congress include measurable performance goals in any 
authorization or appropriation legislation; otherwise, the 
legislation would be subject to a point-of-order. This 
provision would have applied to both House and Senate 
legislation. That's one change.
    I understand the House rules have been amended to sort of 
head in that direction beginning this year by requiring some 
notion of addressing performance--what results programs are 
expected to achieve--in the report that accompanies the 
legislation to the Rules Committee. But I'm going to be very 
interested to see how high the Rules Committee sets that bar, 
because the reason they didn't put many teeth into the 
requirement was that the committees objected to the notion of 
when they send legislation to the Rules Committee, they would 
actually have to say what the program is supposed to 
accomplish, which is exactly the same standard we expect the 
agencies to meet.
    Another change I might make is with respect to oversight. 
If I had my preference, I would require each committee to issue 
a report at the beginning of the year stating all the issues 
over which they have jurisdiction. And you can be sure that 
would be an expansive list; committees like to think their 
jurisdiction would be fairly large. Then I would require them, 
next to that, to cite when they're going to have an oversight 
hearing on that, when was the last one and when is the next 
one. That might indicate that they haven't had and don't plan 
to have any oversight hearings, and hopefully that would 
embarrass them into scheduling one.
    I would also have a definition of what oversight is. In my 
experience--I had 8 years in the Senate with the oversight 
committee there, the Governmental Affairs Committee, in 
watching congressional oversight--and generally in other 
committees it was picking at particular issues, but they rarely 
would ask questions about the financial management of a 
program: Why does this agency not get a clean opinion? They 
rarely, if ever--I say rarely, probably never--had a copy of 
the agency's strategic plan, or annual performance plan in 
front of them and went through that as a basis for oversight. 
But I would define ``oversight'' as including those general 
issues. I would call them out and say that this is what an 
oversight hearing has to look at in order to be called, 
dignified, with the term ``oversight.''
    Then I would require them to issue a report on each hearing 
that they've had as to what their findings were with respect to 
financial management, what their findings were with respect to 
the quality of the strategic plan, what their findings were 
with respect to the quality of the recent annual performance 
plan--the most recent annual performance report--and maybe some 
Clinger-Cohen issues. I would define, in other words, 
``oversight'' as including those issues that you have to 
address, and to issue a report on it.
    Now, Congress doesn't usually want to put requirements on 
itself, certainly nothing that specific. But like I say, if I 
were king, that's what I would do.
    Mr. Putnam. Thank you.
    Mr. O'Keefe, what consequences will OMB contemplate for 
agencies who fail to comply with the Results Act?
    Mr. O'Keefe. I'm parsing through the very words you've used 
in that question, because I think that's the nature of the 
issue we're consorting with right now. The agencies and 
departments, by and large, comply.
    Mr. Putnam. They haven't produced a clean audit report on a 
consistent basis in the history of the government, and three 
have managed to produce a snapshot over the course of the last 
several years. So clearly there's a lack of compliance.
    Mr. O'Keefe. I guess we're parsing through words here, 
because with all due respect, sir, the act itself doesn't say 
you have to have a clean opinion. It says you have to have an 
opinion, you have to have an audit. The CFO Act of 1990 says 
that.
    So, as a result, in terms of filling out the paperwork and 
complying, the agencies and departments are filling out papers. 
They're doing strategic plans or what passes for them. But 
again, in many respects I don't think they make much more than 
mighty fine doorstops. We're focused more on compliance and 
less on how you use the tools of the act to actually fulfill 
its objectives.
    So, as a result, if I said, here are the penalties for not 
complying, what I would get is slavish adherence, slavish 
compliance; they would make sure they got every single paper 
that is defined by the act submitted on time right there, and 
it would be meaningless--right now they don't use it for 
management purposes.
    What I think is victory or success is getting Cabinet 
officers, as President Bush did, to sign up to the proposition 
that the No. 1 item in the criteria for his management agenda, 
that is going to be implemented in every department, is to 
begin the long process of infusing the performance criteria to 
the budget. And that means identifying what an outcome is, what 
the result is you want, why it's in business in the first 
place, going through all those rigid criteria and ultimately 
then coming up with a determination about how successful or not 
a program is and, therefore, what resources should it have to 
carry out those objectives.
    That's a different way of looking at it than being traffic 
cops or enforcers at OMB, which we're going to do. We can 
always ask that they comply with this. Frankly, I find that to 
be not nearly so significant a mission as the one the President 
has identified as No. 1 on his management agenda.
    Mr. Putnam. I agree it's not nearly as significant, but 
since they can't even accomplish the baby step, I'm pretty 
pessimistic on their accomplishing the long haul. If they don't 
know how much money they spent, then I'm pretty cynical about 
their knowing where it went.
    So I'm hopeful, as we embark on this, there are a number of 
States, including Florida, that are now several years into a 
performance-based budgeting program. There are a number of 
States whose mistakes we can learn from, whose successes we can 
learn from. And we look forward to the continuation of this 
discussion.
    My time has expired, Mr. Chairman.
    Mr. Horn. You can have 5 minutes more if that will help.
    Mr. Putnam. No. Thank you.
    Mr. Horn. OK. I still am not quite clear on what you're 
thinking is in that 2003 section where you relate performance 
and budget. Are there going to be various pilot programs, or is 
it going to be across the board in the government? What's the 
thinking on this?
    Mr. O'Keefe. It will not be across the board. It has to 
meet the criteria that I outlined in the commentary, which is, 
first and foremost, you have to identify what the outcome is 
and have everyone agree with what that's supposed to be. Again, 
it can't be as lofty or as ethereal or as platitudinous as 
achievement of world peace. It has been to be definable as, for 
example, the proposal that has been submitted to the Congress, 
as a matter of fact, in an amendment submitted just last week, 
the establishment of an AIDS trust fund.
    As a result, that would, in my mind, lend itself nicely, 
very successfully to the kind of criteria we're talking about 
here, to determine how much we should put into that particular 
task; because the identification of the outcome, the objective 
in that circumstance, has been very specifically identified by 
Secretary Colin Powell and Secretary Tommy Thompson. They have 
signed-up to what they want the result of that to be. They 
said, ``Here is the consequence we want to see happen as a 
result of creating this financial mechanism.''
    From there, there are a lot of different ways to do it. 
Spirited debates over whether or not you should, in the 
achievement of the outcome, try to reduce and influence the 
proliferation, expansion of that horrendous disease in a series 
of very, very spirited approaches or, excuse me, spirited 
debate about what approaches are more effective, whether it's 
treatment, whether it's something as fundamental as sanitation 
programs--a range of different approaches--that in developing 
circumstances, there's an advocacy for lots of different 
schools of thought.
    So during the course of the 2003 review, my bet is--and, 
again, just picking this one at random--there is likely to be a 
very conscientious effort applied toward saying, what are the 
various methods that we could go about achieving the result 
that both the Secretary of State and the Secretary of Health 
and Human Services have identified as the outcome. They can 
say, ``That's what I want to see happen,'' and say, ``Here are 
the various approaches we could use to accomplish that task. 
Here are the inputs that are necessary to go do it.'' How many 
people, what kind of assets, things, etc., do you need to do 
that?
    In the case of treatment, it's one set of professional 
skills. In the case of improving, again, living conditions, 
sanitation circumstances, whatever, in whatever location 
they're trying to achieve this, it could be more of an economic 
development-related kind of approach. Two very different ways 
to go do it, requiring very different inputs, very different 
kinds of assets and capabilities to bring to bear aren't 
mixable parts; you have to be able to figure out what it is you 
want to do.
    Last, to determine what is the output desired, what is it 
we're seeking to go do, how would we measure success or failure 
or movement in the direction of achieving that success or 
failure? Is it the number of folks who have contracted the 
disease or reduction thereof? Is it the number of people 
treated? Whatever the criterion is going to be, that's the--
there are varying ways to go about doing it.
    And last, to look at what is the cost-per-output, once 
you've decided what that output is, as a measure of determining 
what the result is. What is the cost to accomplish that, and 
how do we corral up those expenses and make them relative to 
all the other alternatives?
    That's the kind of program I think is going to lend itself 
most specifically, again just at random, to achievement of the 
result we're looking for; as opposed to this year you put X 
number of bucks into it, and last year you put Y number of 
dollars into it, and are we better off relative to this year 
versus last because that percentage is high or low?
    Mr. Horn. Let's apply that scheme that you've just spelled 
out, which certainly is one way to go at it.
    Secretary Thompson spent a few weeks up in Baltimore 
looking into the so-called HCFA, the Health Care Financing 
Administration. And as he said, everywhere he goes, people are 
griping about HCFA; and one of his ideas was everybody likes 
Medicare and Medicaid and maybe we can have the Medicare 
organization and Medicaid--and that's M-O-M, mom, and nobody 
gets mad at mom--so maybe that would get HCFA off the books.
    But at HCFA let's apply what they do there. They have fees 
that are allowed or not allowed by professional doctors. They 
have intermediaries that are very difficult to really have much 
to say by the planning group of the Health Care Financing 
Administration; and a lot of it is handled by the so-called 
intermediaries, and I think that's one thing that ought to be 
looked at. If it takes legislation, we ought to get it.
    And so there would be a little more flexibility. But we 
have doctors across the country that say, I can't practice 
medicine at those rates and something ought to be done. And I 
think something ought to be done. And I believe, Chris--again, 
remind me of the figure for Medicare. It's--as I remember, it 
was something like $13.5 billion misuse.
    Mr. Mihm. Improper payments. That's my recollection yes, 
sir.
    Mr. Horn. That's not hay. And the Columbus Army Processing 
Center on checks know that's not hay. Those people are spewing 
out and have been over 5 years--maybe it's solved now--but $1 
billion worth of checks, and they just went wondering where the 
paper was to back it up. And doesn't that lead to malfeasance, 
to fraud, to abuse and so forth? And are we serious about that? 
What can we do?
    Mr. O'Keefe. Well, I think, you know, you've raised--a 
priority should be placed toward those programs which are 
identified, for example on the GAO high-risk list, that are 
perennial favorites.
    Having said that, I think you've got to be able to identify 
them as specifically as you've just done. If, instead, I were 
to select one that Congressman Putnam, for example, referred to 
of saying, ``Gee, they can't pass a financial statement, a 
clean audit opinion.'' Well, in and of itself a clean audit 
opinion doesn't make me rest any better. It just doesn't. I 
don't think I'm going to sit back and feel like, boy, we've 
really licked the problem today.
    If the Defense Department next year suddenly gets a clean 
audit opinion. I'm not sure I believe that any more than the 
one they've got right now. If, instead, the answer is, as 
Secretary Rumsfeld has said, the achievement of improvements to 
the financial systems, which he is dedicated to doing, in turn 
is going to provide greater visibility over management of 
programs, visibility over cost--so I can answer the kind of 
questions that you posed, very rightly posed--then, all right. 
This is the truest form of oversight in that context.
    If he can answer those kinds of situations, if the result, 
oh, by the way, happens to be a clean opinion all the better. 
But achievement of that outcome in and of itself is not a 
result. It is a consequence, a happy one, but it isn't an 
objective all by itself; and it is the end condition of what is 
exposed there. It should be a management device. It should be 
something that should inform management decisionmaking, not an 
end in and of itself.
    Whereas you rightly say or, I think, correctly point out 
that the kind of conditions we look at of Medicare, Medicaid, 
etc., kinds of payments in which the erroneous payment--that's 
one that is a ripe example of a set of program objectives that 
should lend itself quite nicely to the approach that I've 
defined here. Those will be the kinds of programs we're going 
to look at first.
    But, again, I don't want to prejudge the outcome of what 
ultimately is going to pass muster in this case. I think, again 
not to sound too rigid on this point, but I'm really quite 
convinced that it is a criterion that has to be adhered to. 
It's got to meet all the gates I've talked about. You've got to 
be able to identify clearly enough what the outcome is, and 
have everybody agree to what that is.
    You have to be able to identify all the inputs, be able to 
develop a cost-per-output, have all the costs necessary and 
included in that particular equation to be able to measure that 
properly. If you don't meet all those criteria, you're kidding 
yourself. You've basically just gone ahead and developed 
something that, in turn, may yield a different answer or a 
wrong answer relative to other alternatives and other 
approaches on how to accomplish the same objective.
    I'd rather make sure that the programs we select, the areas 
we select to apply this, lend themselves best toward this 
solution and then start to work through the successes, as 
opposed to try to slap something on that will pass for 
compliance in this area.
    In a previous incarnation, I made a mistake of having done 
that before. I don't want to do it again.
    Mr. Horn. You've had a lot of experience across the 
government looking at it from Congress, looking at it within 
the major complex organizations that are part of the executive 
branch. And I wonder if you would agree that there's a little, 
simple thing that ought to be before executives every Monday 
morning if they're working with their fellow administrators in 
a particular area; and that is simply to get the accounting 
processes, so the Secretary of the Navy or Army or Air Force or 
Defense can see percentage-wise what was spent in the last 
month. And if it's--let's say 12 percent of the year has gone 
by, and they've got 24 percent of the money going, then the 
question is, what is that buying?
    Is it buying equipment? It might be prudent to do that. 
Let's get all the equipment out before the prices go up.
    Let's--if you've got personnel intensive, just as I found 
in the university, that you expect those to be on a little more 
prudence of, say, 15 percent year expanded, you should be about 
15 percent for personnel. Because that's just clear that 
there--now, is that helpful to a management group or isn't it? 
And if so, what else could be done to put on their plate, to 
say to the Secretary of Defense, ``Here's where it is, Chief, 
this month, in case you have to move money around.''
    Mr. O'Keefe. No. Positively that is a very helpful 
accounting tool, and there's no question that kind of 
visibility is desirable, no doubt about it whatsoever.
    I'll tell you that based on, I guess, my impression in the 
course of the professional experiences I've had the privilege 
of working through in public service, by and large--there are 
some very notable exceptions, but by and large--the general 
proposition, the financial management community throughout the 
Federal Government is reasonably good at spending money only on 
those programs for which Congress has provided the money, and 
spending only those amounts that Congress has provided.
    As a general rule, pretty high marks. Big time exceptions 
to that, and when they happen, they're front page news; but as 
a general proposition, pretty good at those two principles.
    If you chase them back, those are the same two principles 
that were the centerpiece of the Budget and Accounting Act of 
1921. As a result, we haven't really progressed a whole lot in 
terms of maturation of the systems to do much more than simply 
identify as a matter of accountancy where it's going to and 
restricting the amount that can be available for those 
purposes.
    It's taken a long time. Again, in the last 10 years, an 
amazing array of tools that Congress has enacted that are far 
more modern in forcing the administration to look at this 
differently--the CFO Act, GPRA, a range of different 
initiatives that have gone on in order to focus the attention 
more in the direction of thinking precisely in the manner 
you've described. How do you make it real-time for management 
information purposes as opposed to either compliance or 
demonstrating that the thieves didn't run away with the 
Treasury?
    Matter of fact, the two things that were the fundamentals 
of what motivated the 1921 act to be enacted in the first 
place. It was a success. Great. Now, let's declare it a success 
and move on to the next phase. I think that's what GPRA and the 
CFO Act and others have helped do.
    Mr. Horn. Let me ask you one more question, and then you're 
free, and we'll hear from our friends at George Mason.
    The President's budget proposes a new account in the 
General Services Administration that provides what appears to 
be a new, or at least expanded, role for the Office of 
Management and Budget involving the direct control of a 
program, the Electronic Government Fund.
    Does this added responsibility require new specific 
authorization? If not, why not? And who controls it? And is 
this a supplemental fund?
    Mr. O'Keefe. Let me provide you a far more explicit answer 
for the record, if you would permit me, sir. Because in terms 
of the first part of your question of what legal authorities 
are required, if memory serves me right, we have included 
language, or at least a provision, for an account within the 
General Services Administration for the purposes of 
administering their E-Government Fund for that purpose.
    But in terms of exactly the legislative language required, 
let me defer and give you a better answer for the record 
itself.
    In terms of what its purpose is, I can speak to that; and 
it is----
    Mr. Horn. Without objection, it will be put in the record 
at this point.
    Mr. O'Keefe. Thank you, Mr. Chairman.
    The purpose and the objective of why it was set up this way 
is an attempt to leverage the $45-plus billion we spend every 
single year for information technology throughout the Federal 
Government. That's a lot of money to go out for the purposes of 
developing, maintaining, continuing or introducing new 
information technology.
    And as a consequence, it is typically the case that it's 
not a point that we haven't availed ourselves throughout the 
Federal process of the most modern or the most useful or the 
most up-to-date or the most contemporary information technology 
advances. Typically, the argument from the Defense Department 
for example has been, ``Oh, yes, we're busy introducing, you 
know, last generation's technology into today's systems that 
are coming out.''
    In this particular case, the technology is moving so 
rapidly, and we are so far ahead of it, that in many respects 
it is far more capacity than we can actually utilize as 
efficiently as we could. So, as a consequence, the E-Government 
Fund is intended to try to leverage those cases specifically in 
which there is commonality across departments and agencies, and 
which will have an opportunity to try to have a user base that 
is adaptable for lots of different applications, those 
circumstances which lend themselves not necessarily to an 
individual discipline but a multitude of disciplines.
    So it's more integrated management systems, those are the 
kinds of things we're looking for, again as a way to stay ahead 
of and encourage the kinds of opportunities for utilizing the 
advances in information technology on as wide an application as 
we possibly can government-wide, and to motivate different 
departments and agencies to participate in that program.
    It is purposely not designed as a means to substitute for 
the modernization efforts of any individual agency or 
department. That clearly is the effort we're about throughout 
the entire Federal process and will again be more evidenced in 
the 2003 budget submission to you, to try to leverage that 
across the entire Federal Government; but also to require 
agencies and departments to engage in the information 
technology modernization necessary. At $45 billion a year, that 
is not an absence of resources; it's a question of where it's 
being applied and how efficiently.
    Mr. Horn. I think that makes a lot of sense in many ways. 
In the Debt Collection Act of 1996 we provided an incentive for 
departments that would bring in the debts, and they would get 
certain percentages just for that purpose of improving their 
computing capacity.
    And I think some of those things would help in terms of not 
just waiting for the annual budget, but dealing with the 
problem of the new software, new hardware. And it would make 
some sense, I would think, because as you say, it's going 
awfully fast.
    Mr. O'Keefe. Yes, sir.
    Mr. Horn. Well, you needed to be back downtown. We're glad 
you came. Thank you very much.
    Mr. O'Keefe. Thank you, Mr. Chairman. I appreciate your 
patience.
    Mr. Horn. We will now go back to our friends, Mr. Maurice 
McTigue, distinguished visiting scholar, Mercatus Center, 
George Mason University; and Mr. John Mercer, deputy director 
for Government Performance at Logicon.
    So, Mr. McTigue, it's all yours.
    Mr. McTigue. Thank you, Mr. Chairman.
    I wanted to start by saying that at this stage, I think 
we've been through the boring part of the Government 
Performance and Results Act and now we're getting to the 
exciting part. The first two stages were really planning and 
implementation. Now we have some information. And the exciting 
part is about, what do we do with the information that's being 
produced?
    The Government Performance and Results Act doesn't do 
anything on its own. It's a tool. Compliance with the 
Government Performance and Results Act doesn't do anything 
either unless the information produced is used in some way to 
actually change the quality of the performance of the 
organizations that are covered by the Performance and Results 
Act.
    I want to pick out five issues and spend a moment or two on 
them. And the first one is examining outcomes, not agencies. 
And I think that we're going to be getting a change in the 
behavior and the performance of government only if we start to 
look at outcomes as the issue, not an agency as an entity in 
its own right and it's worth preserving just because it's been 
there for 140 years or something like that.
    The agency is actually the deliverer of services, and we 
can judge its performance in delivering the service.
    But the politicians and the public should be interested in, 
what is the result that you're aiming to achieve by the use of 
that organization. If it's in defense, it's readiness and 
superiority. So how much has our readiness and superiority 
improved during the year is the issue, not the Department of 
Defense?
    Most outcomes have multiple agencies and programs that are 
addressing those outcomes. And it's our view that it would be 
much better to look at the outcome and all of the programs and 
agencies that are trying to impact that outcome and make 
comparisons between the level of impact or success that they're 
having on bringing about improvement in that issue.
    If you were to do that, you introduce competition between 
government activity, and in the introduction of that 
competition, all of the benefits that normally come with 
competition come as well. You will have new avenues of 
innovation, discovery; and all of those processes will look for 
better ways of being able to achieve the result, because 
effectively people will see that they're competing for a common 
pool of money rather than having a guaranteed appropriation.
    That brings me to the second part, which I think was 
addressed in part by your colleague, Mr. Putnam, in one of his 
questions; and that is linking performance and appropriations. 
Unless there is a consequence for either complying or not 
complying, and either using or not using the information, then 
of course you're not going to change behavior.
    In Congress, I think there are two cultures that need to be 
addressed in Congress itself. The first of those cultures is 
the culture of the committees of Congress. And I think that 
without having a structural change, if committees would start 
to view their activities that the experts on issues would be 
the oversight and authorizing committees that know this 
particular issue in great depth, and what they become then is 
the research arm of the appropriators and the appropriators; 
pick that expert opinion up and make decisions based upon that. 
Essentially what you're doing is linking performance and 
appropriation.
    Until such time there is a linkage between performance and 
the allocation of money, we're not going to see a major change 
in the way in which government organizations work. That will 
bring about a very rapid change, of course, if such a situation 
were to occur. It's very encouraging to see that the 
administration is looking at exactly that, as Mr. O'Keefe has 
just set out for us.
    The third issue is requiring agencies to measure outcomes 
of importance. And I think that in a number of areas the 
government isn't well served in actually addressing a primary 
issue in terms of what is the state of knowledge on that issue.
    For many years, you've had programs that have worked on the 
issue of homelessness. But who knows anything about the state 
of homelessness in the United States at this particular point 
in time? Why is it that on one of the wealthiest countries in 
the world we have one of the highest rates of homelessness? 
What is it about the programs that are addressing that issue 
that is currently not successful?
    Well, one of the issues that might view, that might 
actually be there, that hasn't been identified in the past, is 
that many of the activities of governments have been directed 
at the consequences of problems rather than the causes, the 
political processes, that there are homeless people, so it 
starts to put in programs to house the homeless people. But 
who's out there looking at the cause?
    And unless you address the cause, what you do by addressing 
the consequence is that you start to build in dependency. So 
the programs you're putting in place are actually creating 
dependency rather than solving the dependency that was your 
original intent. And it is by focusing on outcomes that you 
start to identify that there are factors here that we're not 
addressing, that are making this problem worse rather than 
better.
    I think a classic case is the feeding programs that are run 
by USDA, instituted in the first instance, as I understand it, 
to use up agricultural surpluses. But you now have some of the 
biggest feeding programs in the world, and are quite probably 
creating a lot of dependency as a result of that. So the 
problem is having negative effects in many instances rather 
than positive effects. Until such time as somebody starts 
looking at the causes of the hunger, I think that the program 
is going to have some negative effects while it actually feeds 
hungry people.
    The last comment I want to make is this: changing the 
political value equation. What I'm talking about here is that 
political commitment to issues are too frequently measured in 
terms of how many additional dollars you spend on that issue. 
An election is coming up: Congress spends more dollars on the 
drug issues because drugs are topical at that time. But there's 
no indication whatsoever whether those additional dollars are 
going to have positive or negative effects. And I think in that 
case, in some instances, the expenditure of dollars has had 
negative rather than positive effects.
    So if you were actually to follow through, in an ideal 
world, the concepts that come out of the Government Performance 
and Results Act, what you would see is that the value equation 
would gradually change so that things weren't measured in terms 
of how many additional dollars you spent, and political 
commitment wouldn't be measured in terms of how many additional 
dollars you spent; it would be measured in terms of what was 
the public benefit that arose from the expenditure of those 
dollars, how successful have we been in eliminating that 
problem or diminishing that problem, how successful have we 
been at enhancing that good?
    In my view, that is the real value of the Government 
Performance and Results Act. Until such time as you go through 
all of those other stages, you won't start to impact that value 
equation. But when you succeed in impacting that value 
equation, then I think the political scenario and the success 
ratio of government will change dramatically.
    Thank you, Mr. Chairman. I'm very happy to answer 
questions.
    Mr. Horn. Well, we thank you.
    [Note.--The publication entitled, ``2nd Annual Performance 
Report Scorecard: Which Federal Agencies Inform the Public,'' 
may be found in subcommittee files.]
    [The prepared statement of Mr. McTigue follows:]

    [GRAPHIC] [TIFF OMITTED] T0373.024
    
    [GRAPHIC] [TIFF OMITTED] T0373.025
    
    [GRAPHIC] [TIFF OMITTED] T0373.026
    
    [GRAPHIC] [TIFF OMITTED] T0373.027
    
    [GRAPHIC] [TIFF OMITTED] T0373.028
    
    [GRAPHIC] [TIFF OMITTED] T0373.029
    
    Mr. Horn. And let's go back now to Mr. Mercer. Do you have 
some comments on either Mr. O'Keefe's remarks, Mr. Mihm's and 
Mr. McTigue's. Plus any other things you want to get on the 
record.
    Mr. Mercer. I understand my full written statement will be 
put in the record.
    Mr. Horn. Automatic.
    Mr. Mercer. Of course. And I'll just hit some of the key 
points and like to lay some items out.
    First of all, I'm John Mercer and I'm here testifying 
before you today solely in my capacity as the former counsel to 
the Senate Governmental Affairs Committee who led the 
development of the Government Performance and Results Act of 
1993. I'm not testifying as a representative of any other 
interest.
    I first want to say how much I appreciate the continued 
interest this subcommittee has shown in GPRA and particularly 
your leadership, Mr. Chairman.
    First, a little bit of background on the legislation. I 
served from 1989 to 1997 as Republican counsel to the Senate 
Governmental Affairs Committee. GPRA, as a specific piece of 
legislation, began with a conversation I had with Senator Bill 
Roth, for whom I worked, in January 1990, about a real-life 
example of a performance-based management and budget system 
that had proven to be very effective and which I thought 
applicable to the Federal Government. So he asked me to develop 
the legislation on his behalf.
    And I would be remiss if I didn't acknowledge the very 
important input I got in developing that from Walter Groszyk of 
OMB.
    The law was intended to point the Federal Government toward 
a vision of improved government performance and then to begin 
moving it down the road toward fulfilling this vision. GPRA, in 
other words, was intended to lay a firm foundation upon which 
to build a more complete structure of performance management, 
but wasn't intended to achieve that full vision by itself. 
Clearly, other reforms would be needed.
    GPRA intended to encourage this primarily through two 
means, increased transparency and accountability.
    There are some who sincerely believe that this effort 
wouldn't work because government is inherently inefficient, in 
this view, because it has little or no competition. You hear 
that often. But this is exactly what GPRA intended to provide, 
competition for every Federal agency and program by making them 
compete against their own past performance. When performance is 
tracked and reported over time, there is inevitably pressure to 
show steady improvement.
    The actual inspiration for GPRA was the performance-based 
management and budget system of the city of Sunnyvale, CA. I 
had served there as mayor and city councilmember. I think it's 
relevant to look at that city.
    Mr. Horn. I was fascinated by your sort of case study 
there. Was John Deever city manager before or after or during 
it?
    Mr. Mercer. John Deever was city manager in 1973 when the 
city first started developing performance audits for programs; 
and he put some of the elements in place.
    As you know, he left for Long Beach, I think in 1977, and 
Tom Lewcock came in as city manager and really turned it into 
performance-based budgeting. Lewcock came in 1978-79 and 
developed the performance-based budgeting system.
    I'll say that when I was developing GPRA, a team from GAO 
and OMB went out and visited the city. This was in 1991. The 
following year in congressional testimony--I think this is 
interesting because it gives an indication of what we'd like to 
see happen in the Federal Government--OMB said in testimony 
before the Governmental Affairs Committee, ``As indicated, the 
city of Sunnyvale, California, stands out as the single best 
example of a comprehensive approach to performance measurement 
that we have found in the United States.''
    And I should say that New Zealand is not in the United 
States. Now this is the key sentence.
    ``One underlying reason for the success achieved in 
Sunnyvale is the fact that every program manager uses the 
system to plan, manage and assess progress on a day-to-day 
basis.'' We hope to get to that in the Federal Government, but 
as we've heard, we're a long way from that. Sunnyvale's 
performance budget is actually a fully integrated program 
performance plan and annual budget, using performance-based 
budgeting with full cost accounting.
    Now, GPRA itself contains no specific requirement for real 
performance-based budgeting nor do the requirements for 
performance planning reach every activity and employee of a 
department. However, these elements were important aspects of 
the vision underlying GPRA.
    I'll talk a little bit more about performance budgeting in 
my statement, but I'd like to make a couple of key points about 
it.
    I think it's important to understand that a true 
performance budget is not simply an object class budget with 
some program goals attached. Real performance-based budgeting 
gives a meaningful indication of how the dollars are expected 
to turn into results, not necessarily with scientific precision 
but at least through a general chain of cause and effect.
    The most effective governmental performance-based budgeting 
does this by showing how dollars fund day-to-day activities, 
how those activities generate outputs, and then what outcomes 
should result. The basic building block of a sophisticated 
performance budgeting and management system, in my opinion, is 
the cost-per-unit of activity which rolls up into cost-per-unit 
of output. This is a powerful format because it directly 
measures what most managers actually manage on a day-to-day 
basis, dollar expenditures and staff activity, in order to 
achieve certain outputs. These elements serve as the 
underpinnings for achieving higher level objectives including 
program outcomes.
    The earliest drafts of the statute, in fact, contain 
provisions requiring agency plans and reports to include, 
``trends in costs-per-unit of result, unit of service or other 
unit of output.'' Unfortunately, I had to remove those 
provisions when I discovered that agencies did not have in 
place and weren't required to have in place the requisite cost 
accounting systems.
    I am pleased to say that in 1998 the Federal Accounting 
Standards Advisory Board issued a requirement that agencies 
develop this type of information. And quoting just briefly from 
it, what they said is that, ``The managerial cost accounting 
statement and standards contained in the statement are aimed at 
providing reliable and timely information on the full cost of 
Federal programs, their activities and outputs.'' Those 
standards will provide a method for identifying the unit cost 
of all government activities.
    Now, I recognize agencies have scarcely begun to do this, 
but at least it's a requirement. And that's a fundamental 
building-block of a really powerful performance-based budgeting 
system, amongst other uses.
    There's more potentially good news, and we heard it 
reflected here today in Mr. O'Keefe's statement, that in the 
President's budget he announced that agencies will be asked to 
submit performance-based budgets this September for a selected 
set of programs. I'm hopeful they will set the bar high enough 
that we will see in those performance budgets that they pilot 
how activities are funded and those activities turn into 
outputs which, in turn, become outcomes.
    Now, what's the big deal about performance-based budgeting? 
I have outlined in my written statement a little bit more some 
of the uses of it, but just in general, it not only shows that 
there is a relationship between dollars and results, but done 
right, it gives some indication of what that relationship is, 
how those dollars become results. When you do that, you can 
then begin to, at least in a general sense, see what the impact 
of increased dollars or budget cuts could have.
    It allows for more informed contracting-out decisions, 
because you see the full costs, if you do this right, of the 
results you get when you think about contracting. It's 
certainly useful in promoting performance management. And 
perhaps my favorite use of it, as we use it throughout 
Sunnyvale, is that program goals became two-part goals.
    We hear about program goals under GPRA right now as what 
level of result. When you put a cost factor, particularly unit 
cost, then you can have goals that are two-part, achieve a 
certain result at a certain cost-per-unit. When you do that and 
you link that to pay-for-performance systems or some other 
methodology for tracking this, you can begin to create 
incentives for actually reducing the cost of government.
    Right now, the incentive in every Federal agency is to 
spend every nickel of your budget. Now, under GPRA, it's spend 
every nickel and get a certain level of result. In Sunnyvale, 
it was to get a certain level of result at a certain unit cost, 
and by the way, if you drive down the unit cost you're eligible 
for a bonus. So at the end of the fiscal year there was a 
disincentive to sweep money out the door, because that just 
raised your unit costs and blew your chances of getting a 
bonus.
    And now GPRA's relevance to day-to-day management; I have 
to say that I had expected that within a couple of years of the 
performance plans, these plans would have cascaded down to the 
lowest levels of the organization. This has not happened and it 
is a major reason that the government-wide movement toward 
managing for results has been impeded.
    Too many Federal managers still believe that GPRA does not 
apply to them; and in a literal, legal sense I suppose it 
doesn't. But for GPRA to work, every Federal employee must be 
involved.
    At the lowest organizational level there must be an annual 
plan that shows how the activities of the manager and the staff 
directly support achievement of the next-higher level plan. 
Each higher level plan should, in turn, do the same until all 
such plans can be traced clearly up through the organization to 
support the agency-wide plan. Otherwise, departmental and 
agency plans will be little more than wish lists, if you can't 
literally trace that plan back down in the organization to day-
to-day activities. There are methodologies for formally linking 
that, and I have appended an example at the end of my 
testimony.
    At this point, I'll say that the final area--and I won't go 
into it because I touched on it earlier, and I'd be glad to 
address it in the questions--I've been disappointed in the lack 
of use of this by Congress. There is still, as we all know, 
much room for improvement there, in things that should be done.
    Mr. Chairman, I thank you for this opportunity to talk 
about my favorite subject.
    [The prepared statement of Mr. Mercer follows:]

    [GRAPHIC] [TIFF OMITTED] T0373.030
    
    [GRAPHIC] [TIFF OMITTED] T0373.031
    
    [GRAPHIC] [TIFF OMITTED] T0373.032
    
    [GRAPHIC] [TIFF OMITTED] T0373.033
    
    [GRAPHIC] [TIFF OMITTED] T0373.034
    
    [GRAPHIC] [TIFF OMITTED] T0373.035
    
    [GRAPHIC] [TIFF OMITTED] T0373.036
    
    [GRAPHIC] [TIFF OMITTED] T0373.037
    
    [GRAPHIC] [TIFF OMITTED] T0373.038
    
    [GRAPHIC] [TIFF OMITTED] T0373.039
    
    [GRAPHIC] [TIFF OMITTED] T0373.040
    
    [GRAPHIC] [TIFF OMITTED] T0373.041
    
    [GRAPHIC] [TIFF OMITTED] T0373.042
    
    [GRAPHIC] [TIFF OMITTED] T0373.043
    
    [GRAPHIC] [TIFF OMITTED] T0373.044
    
    [GRAPHIC] [TIFF OMITTED] T0373.045
    
    [GRAPHIC] [TIFF OMITTED] T0373.046
    
    [GRAPHIC] [TIFF OMITTED] T0373.047
    
    [GRAPHIC] [TIFF OMITTED] T0373.048
    
    [GRAPHIC] [TIFF OMITTED] T0373.049
    
    [GRAPHIC] [TIFF OMITTED] T0373.050
    
    [GRAPHIC] [TIFF OMITTED] T0373.051
    
    [GRAPHIC] [TIFF OMITTED] T0373.052
    
    [GRAPHIC] [TIFF OMITTED] T0373.053
    
    [GRAPHIC] [TIFF OMITTED] T0373.054
    
    [GRAPHIC] [TIFF OMITTED] T0373.055
    
    [GRAPHIC] [TIFF OMITTED] T0373.056
    
    [GRAPHIC] [TIFF OMITTED] T0373.057
    
    [GRAPHIC] [TIFF OMITTED] T0373.058
    
    [GRAPHIC] [TIFF OMITTED] T0373.059
    
    [GRAPHIC] [TIFF OMITTED] T0373.060
    
    [GRAPHIC] [TIFF OMITTED] T0373.061
    
    Mr. Horn. I take it Sunnyvale still uses those processes?
    Mr. Mercer. Absolutely. Their most recent budget, an 
example from it--the budget they adopted this month--is the 
appendix. I use an example from police services. If you look at 
that, the appendix in my testimony, you will see what is 
frankly the most sophisticated performance-based budgeting 
system in the United States. Countries come from all over the 
world to look at it.
    They're not going to stop that. It's very powerful stuff. 
The citizens love the fact that the level of result keeps going 
up and the cost of government drops steadily, and they get lots 
of attention.
    Mr. Horn. Where would be the second or third or fourth city 
that would have this type of approach?
    Mr. Mercer. Well, I don't know. Which is not to say that 
nobody does it now, but I know if they did, it's fairly recent. 
I don't know of any city that links it to day-to-day 
activities, so that in these performance budgets you see all of 
the work hours for all the employees covered, linked to 
activities. Somebody may have done that.
    But just short of that, actually having goals and driving 
it down and that sort of thing, Phoenix, AZ, ought to be 
recognized. It's obviously a much larger city than Sunnyvale 
and was one of the co-winners of the Bertlesman Award for being 
the best-run city in the world. Christchurch, New Zealand, was 
the other one, and Phoenix, AZ.
    Cities in the West with council-manager forms of government 
tend to be where you would want to look for examples. If you're 
looking at big cities, then San Diego and Dallas and cities 
like that. But if you're looking for the best of the best, 
well, it's Sunnyvale.
    Mr. Horn. How about the State of Oregon, are you familiar 
with what they're doing?
    Mr. Mercer. The Oregon benchmarks?
    Mr. Horn. Right.
    Mr. Mercer. Yes. I'm not familiar with what they're doing 
now. When I was counsel to the committee, I was familiar with 
what they were doing. It got a lot of publicity.
    I was probably less impressed than most people that looked 
at it only because--not because their heart wasn't in the right 
place and they weren't a leader in a lot of this--they had lots 
and lots of measures for things; arguably, too many in some 
instances. But in any event--and they had a lot of input from 
the public in developing these--all that was very good.
    But they didn't drive it down to the actual operations of 
the government. That is, you set--these are the benchmarks for 
where we want--certain indicators of health, education, 
whatever, to be 5 years, 10 years, 20 years from now for the 
State, that was a wish list. When I looked behind the curtain, 
I didn't see anything that said, OK, to get there 5 years from 
now here's what we have to do for each of the next 4, 5 years. 
And that means, you know, here's how many work hours we're 
going to apply to this task which will create this outcome, 
which should generate the outcome--at least we hope. That, they 
haven't done, so here's a great wish list; now let's go off 
about our business and hope we hit the targets.
    Unless you drive it down--and in my opinion, you have to 
formally link it down to day-to-day activities--like I say, 
you've got to cascade it down. In Sunnyvale, there are task 
codes. There's a five-digit task code for everything everybody 
does. And some people would look at that and say, that's 
overkill. I would say the proof is in the pudding. It works.
    They do it right. And they drive down the unit cost of 
government. And they drive up performance because it's linked 
to day-to-day activities in a formal, meaningful way.
    Mr. Horn. Mr. McTigue, the New Zealand plan has a lot of 
aspects of that, I gather. Could you give us a little idea of 
what would happen in Christchurch, let's say? Or the government 
generally? What did they do with their Ambassadors who have to 
account for everything in the embassy, or is it out of their 
pocket?
    Mr. McTigue. Can I just link it into something that Mr. 
O'Keefe said a few moments ago in a response to a question from 
you, Mr. Chairman? He was talking about the approach that the 
administration is taking at the moment in identifying very 
clearly the outcome, what is the issue that you're addressing; 
and then looking around at the multitude of different ways in 
which you might address that in making choices among those that 
might be biased, impacting that, and they might be quite 
different activities. That was at the heart of most of the 
reform of the central government of New Zealand. By looking 
just for exactly what is the outcome if it's dependency, then 
let's look at the variety of programs that impact dependency.
    Then taking something that Mr. Mercer said, what we looked 
for was not really the unit cost, but the cost-per-unit of 
success which is slightly different, because some of these 
programs might have higher costs but better success rates or 
they might be dealing with a different cohort of people and the 
cost might be different. So we looked at costs-per-unit of 
success. And then what you left on the table was a decision for 
the elected politicians to make in terms of, is the value to 
the community in this program justified given the known level 
of success and the cost-per-unit of success. And in some cases 
the answer was yes.
    I can give you a clear example from my own experience where 
there were two programs that were seeking additional funding, 
both designed to try and get people back into the work force. 
One of them was a very basic program that had a high success 
rate, 70 percent of their participants got back into work at a 
relatively low cost of $256 per week per person.
    There was another program designed for at-risk youth that 
had a similar success rate, about 70 percent of their people 
went to work, but at a much higher cost, $932 per week per 
participant. The difference, though, was for those people who 
got jobs out of that youth program, their criminality rate 
dropped by 60 percent. That far outweighed the additional cost.
    But what you had was all of the elements on the table, so 
that the political process was able to make a good decision and 
say, ``yes, this program at four times the cost is good value 
because keeping people out of the criminal justice system more 
than compensates for that additional cost.''
    In the case of a place like Christchurch, it manages itself 
in a very similar way to that which has been outlined for 
Sunnyvale or for Phoenix. But it's a concentration on what are 
the essential services that the public needs and delivering 
those services in the best possible way, using competition in 
many instances as a meaning of finding the best way of 
delivering the services; and in some instances, deciding that 
these are services that the government should no longer 
deliver, that there are other people who would be better or 
more appropriate to deliver these programs. And to a degree, 
that parallels some of the President's ideas with the concept 
of using faith-based organizations; not because they were 
faith-based, but just because they happened to have a very high 
success level in dealing with people in dependency, dealing 
with families that were at-risk and things like that.
    It was their success level rather than the fact that they 
were faith-based. But that's what we're really looking for, 
who's the best provider, and the government should actually buy 
the services from the best provider of those services.
    Mr. Horn. Mr. Mihm, what's your thinking now? Your 
colleagues in the General Accounting Office have spent a lot of 
time looking at these proposals. Can we learn anything from 
Sunnyvale, Christchurch, Oregon and the Mercatus Center of 
George Mason University?
    Mr. Mihm. I think, Mr. Chairman, we can learn an awful lot 
from those examples. I have been fortunate to be involved in 
some of those studies. I certainly was very fortunate and 
honored to assist the subcommittee in its examinations of 
Australia and New Zealand management reforms a year--I guess 2 
years ago.
    One of the interesting points of the whole trend toward 
results-oriented management is that all the Western democracies 
and State and local governments are basically moving in the 
same direction, and that is despite differences in cultures and 
histories and forms of government, whether it be Western or 
Republican forms, as we have here. There is a worldwide trend 
basically toward a greater focus on results, a greater focus on 
trying to link performance and results as part of the budget of 
the appropriations process.
    I would underscore a couple of points that you've heard, I 
think, running throughout the entire theme. The first is how 
important it is to bring good cost information, program cost 
information, performance information and budget information 
together. And the power of the management reforms that Congress 
has put in place were really begining to pay off, once those 
three actually come together.
    The second point, though, is how difficult that is going to 
be. I was intrigued by the list that Mr. O'Keefe listed of the 
very high bar that they're going to have in order to accept 
programs into this new budget approach. I think it's great that 
bar is high. What I would remind us all, though, is the very 
difficult time that we all had with the performance budgeting 
pilots under GPRA.
    I know this subcommittee worked--expended quite a bit of 
energy with OMB over the last 3 years to, first, get them to 
identify pilots and then, second, to find out what was going on 
for those. The bar for those pilots was much lower than what 
Mr. O'Keefe is suggesting. This is in no way saying you ought 
to adjust the bar.
    Mr. Horn. My conclusion on that was that it was laughable.
    Mr. Mihm. And one of the challenges that they had, sir, as 
you know, was just finding enough places around government that 
had program cost information. And as a result, they did come up 
with some pilots that were perhaps not as robust or prominent 
as the authors envisioned when they wrote the provision for the 
performance budget pilot.
    So I would just underscore that this really is the long 
haul. John and Morris have underscored that what's going on in 
Sunnyvale, and New Zealand began well over a decade ago. What's 
been impressive with the United States is that we've put the 
infrastructure in place just over the last half-dozen or so 
years, but we're still lagging behind the rest of the world in 
terms of being able to use that infrastructure to be making 
decisions.
    Mr. Horn. So where do we go from here in terms of the 
General Accounting Office? You're doing another study now for 
Senator Thompson, I believe.
    Mr. Mihm. Yes, sir. There's a couple of things that we're 
doing.
    We're looking at each of the annual performance plans and 
reports, as the Senator indicated, for usually up to four or 
five key outcome areas for each agency, and will be reporting 
out in separate reports on each of the 24 largest agencies. 
Over the next basically 2 weeks or so those reports will come 
out. This is part of our commitment to move beyond--as we've 
all been discussing--questions of just compliance and quality 
with GPRA to more substantive discussions as to the value of 
GPRA to informed decisionmaking.
    The second thing that we're continuing to do is look at the 
attention that agencies are paying to the major management 
issues. This subcommittee and certainly my colleagues back at 
the GAO, the Comptroller General, have invested an awful lot of 
attention and energy into making sure that agencies pay due 
attention to their major management problems; those high-risk 
areas to make sure there are people accountable for getting off 
those lists. This list has been around, as you know, since 1990 
and many of these things have been on there right from the very 
beginning. We need to have people accountable for getting this 
thing off of there, so we don't have to come up each year and 
give the Congress a new and somewhat augmented list.
    The third thing that we're doing is paying particular 
attention to human capital in the annual performance plans. Too 
often the human capital, the personnel aspect, has been seen as 
just yet another functional area to be stove-piped and, OK, 
we'll have the personnel office kind of write us some nice 
goals. And that's not to deprecate those; those can be 
important goals, but they're not integrated into the 
programmatic thinking that is taking place in the agencies.
    Then, I guess the final area we are spending a lot of 
attention on are the performance management aspects, as to how 
do we specifically link programmatic outcomes to day-to-day 
difficulties within agencies? We've had a lot of discussion 
here about the budget aspect of that. We're focused on a lot of 
the incentives that are given to managers and how individual 
managers are held accountable.
    Mr. Horn. The General Accounting Office has found that the 
agency use of performance information and decisionmaking has 
declined since 1997. What do you think might have caused that 
decline and how might Congress help resolve the problem?
    Mr. Mihm. Well, the finding that you're mentioning there is 
from our survey of Federal managers, and it is one of the most 
disturbing aspects of the responses that we get.
    We surveyed managers in 1997 and again in 2000. As you 
indicate, we have got lower levels of reported use of 
performance information in a variety of very important 
categories, that is, managers using it to allocate resources, 
to set program priorities, to coordinate with cost-cutting 
programs, which has been a major topic of discussion here 
today.
    And what was most disturbing about that, Mr. Chairman, is, 
we were asking, if you will, a methodologically impure 
question; that is, we were asking managers to report on 
themselves which--my colleagues have designed questionnaires 
saying, you never want to do that because you'll always get an 
inflated and more positive answer because you know people will 
try and give you what they think you want to hear--we still got 
bad responses so that kind of even heightened my concern on 
that. We're getting behind the data to try and find out what's 
going on.
    I think one of the problems--there's a couple of things 
that can't be going on there. One is that we're--whereas in the 
early years of GPRA--that is, in 1993-94 up through 1997--we 
had an awful lot of attention, a lot of congressional 
attention, certainly a lot of attention, at least a lot of 
rhetorical attention, from OMB, over the last couple of years 
there hasn't been an equivalent level of attention.
    It runs the very real risk of becoming, as Mr. O'Keefe was 
mentioning, a paperwork-driven exercise in agencies that 
becomes overly focused on measurement and not focused on what 
it really ought to be about, which is agencies in Congress 
engaging in conversations about, what are we trying to achieve, 
how are we doing and how are we improving performance? Which 
gets directly to the second part of your question, what can 
Congress do on this?
    The single most important thing that Congress can do is 
continue to have oversight hearings like this, continue to 
bring up performance and performance-related questions and 
programmatic oversight when you meet with the people from GSA 
and the other committees under--or other agencies under this 
subcommittee's jurisdiction, and certainly in your 
appropriation capacities to start or to continue to ask 
performance-related questions during the budget process. 
People's hearts and minds will follow when they see 
congressional leadership on that and OMB leadership.
    Mr. Horn. Couldn't you think of a sort of matrix-type 
questionnaire where you get the managers of the Federal 
Government to say not what they do, but what they think the 
other managers do? And I think we could get a lot of data from 
that, where they wouldn't be saying, I did it. Don't put bars 
in the State and all that. But it just seems to me that's the 
way to get at it and not say, ``this is what I do or don't do, 
but my, you know, fellow manager does that.''
    Mr. Mihm. We did ask a series of questions related to that 
about the extent to which top leadership and their agency used 
performance information for a variety of tasks. We even asked 
questions, and I've conveniently forgotten the responses, about 
whether or not they felt GAO was using performance information 
in assessing their progress. As I said, it can be newly 
forgotten, the responses to that particular question. But we do 
try and tease that out from them.
    Mr. Horn. And have you identified any of the positive 
trends, as opposed to the negative trends between this year's 
performance reports and last year's? Is there anything that's 
making that difference?
    Mr. Mihm. I think we're seeing a couple of things that are 
positive, in our view. We see--while the weaknesses in agency 
data certainly continues to be a very real problem--greater 
attention by agencies within their annual performance plans and 
reports to explaining those data limitations, to being able to 
tell Congress and other stakeholders and decisionmakers, here 
are the decisionmaking consequences of this poor data. So 
that's one positive achievement.
    Second, we're seeing greater attention to cost-cutting 
programs and the need to effectively coordinate those cost-
cutting programs--still very much a work in progress. We don't 
have the common performance measures for similar functions that 
Mr. McTigue and Mr. Mercer mentioned, but at least we're 
getting recognition that other agencies are involved.
    Third, a lot of the effort that's been expended over the 
last few years to try and get agencies to articulate outcome-
oriented sets of goals is beginning to pay off. We do have 
some--as Mr. O'Keefe mentioned, there are still the world peace 
or what typically is also known as the ``end world hunger'' 
goals where, you know, just give us more money and we promise 
hunger goes away.
    I think those are becoming fewer and farther between. We 
are seeing greater understanding of agencies that they need to 
focus on results.
    The real problem that we're still seeing within agencies is 
this linking daily activities, linking budgets, linking the use 
of resources to results that take place outside those agencies. 
That is still the fundamental struggle that they're facing.
    Mr. Horn. The ability to link performance information to 
cost seems to be the common failure. What solution would you 
suggest?
    Mr. Mihm. Well, I think one of the single most important 
things we need to do is to continue to put pressure on agencies 
to effectively implement the statutory framework that Congress 
has done.
    In getting back a little bit to Mr. Putnam's question 
about, are there structural changes, my view is that Congress 
has done its part. It's now up to the executive branch 
agencies; and Congress can still do more and OMB can do more. 
It's up to them to step up to the plate.
    Congress has made known, through FFMIA, that the cost 
accounting standards are to be implemented. As Mr. Mercer 
mentioned, we're still quite a ways away from there. We're 
still, as Mr. O'Keefe mentioned, quite a ways away from 
agencies even being able to produce clean audit opinion.
    As you well know, we chair--the Comptroller General sits 
basically in this chair and explains why we offer a disclaimer 
on the government-wide financial statement.
    So we need to still, as a Federal Government, continue to 
make progress in implementing the statutory framework, the 
tools that Congress has already given to agencies. We can 
explore whether other tools are needed, but that's the first 
step. Let's implement what Congress has already told us to 
implement.
    Mr. Horn. Some of us believe in moving to a 2-year budget 
appropriations cycle. How does the Comptroller General of the 
United States feel about that? Do we know?
    Mr. Mihm. I don't know sir. I know we have testified on it. 
And rather than give you something that I would have to correct 
for the record, let me go back, find out and then get back to 
you immediately with the proper answer.
    Mr. Horn. We'll put it in the record at this point, without 
objection.
    This will be my last question to you.
    Some programs produce their benefits on an annual basis, 
while other government programs, such as those that conduct 
research and development, find it more difficult to estimate 
the timing of the scientific discovery. That's obvious--or the 
measure of a result from a research program.
    And do you have any advice on how the government agencies 
that conduct these types of programs can provide meaningful 
performance metrics? And that would be Agriculture, NIH, HHS, 
some of the others, or social science and so forth; what's your 
feeling on that?
    Mr. Mihm. Actually I'll start and say, there are provisions 
within the Government Performance and Results Act that allow 
for agencies to use qualitative goals, or even with the 
authorization of OMB, where it's characterized as the 
alternative form of measurement. The National Science 
Foundation is one agency.
    At GAO we use a qualitative goal as the alternative form of 
measurement. That is exactly designed to get at situations 
where the performance, or the outcome, is exceedingly hard to 
measure and where the outcome takes place or can be achieved 
over years.
    Just to really drive home the point you were making in the 
question, Mr. Chairman, I was once talking with someone from 
the National Science Foundation, and they put it to me in a 
very direct way; they said, how do we set annual performance 
goals this year for people who are going through post-doctoral 
fellowships in part funded or supported for National Science 
Foundation when their outcome will be the discovery of cures 
for diseases 20 and 25 years from now that we don't even know 
the name of yet, because the people that are now doing cutting-
edge AIDS research were in post-doc programs before we even 
know AIDS existed? So how do we set goals like that?
    Well, there has been work on that. And the National Science 
Foundation has done quite a bit of work on this alternative 
form of measurement.
    The key thing that we advise agencies is, don't turn this 
into a technical debate. Sit down with Congress, sit down with 
the authorizers, sit down with the appropriators and your other 
stakeholders and develop a common understanding about how 
progress will be measured, what we're trying to achieve, how 
we're going to know we're on the right track. If agencies do 
that, then they've done GPRA the way GPRA should be done and 
have not turned it merely into a measurement exercise.
    Mr. Horn. You're very right on that. And that's exactly 
what we want to do.
    Yes, Mr. Mercer.
    Mr. Mercer. I'd like to respond to that same question. Two 
thoughts: First, to echo what Mr. Mihm just pointed out, that 
the statute actually does envision alternative forms of 
measurement other than objective, quantifiable, measurable, 
which is what it prefers first and foremost. In wrestling with 
the notion of, what kind of measurable goal would I expect from 
the State Department for some of its diplomacy programs, I knew 
I had to have an escape hatch in the legislation, something 
other than just throwing up your hands and saying, well, we 
can't measure anything.
    So I took a whack at it, and it stayed in the law. It 
allows you to come up with something on your own, but getting 
approval from OMB first of all. Then it suggests one, and this 
is the one that's used.
    It says, first define a successful program, you know, a 
written narrative, a subjective statement. And then, being a 
little queasy about leaving that in there just like that, 
thinking, well, you could write something--``we want to improve 
the quality of life of people living below the poverty level'' 
or some mushy thing. It says, you have to have a second 
statement, and that is of a minimally effective program. In 
other words, presumably something that shows some positive 
value, but not enough that you could call it a success.
    In other words, success had to be measured in context; and 
the context was something that was less by definition, less 
than successful, but not a total failure. Presumably, if it 
didn't meet either standard, then it would be a failure. It 
says, you have to write the definitions in terms that would 
allow an outside, objective analysis of looking at your 
program's results for the year to be able to determine which of 
those categories it fit into.
    As Mr. Mihm points out, the National Science Foundation has 
made pretty good use of that. And, in fact, it's probably one 
of the most underutilized aspects of GPRA; and I hate to point 
it out, because I love to see measurable goals, but recognizing 
that's not always possible, especially for outcomes.
    Mr. Horn. Could you get us that section of the law that you 
had something to say with? And we'll put it in the record at 
this point.
    Mr. Mercer. Sure. I'll cite that.
    But I'd like to answer more broadly because it's an answer 
I give, and it's almost tongue in cheek, but I'm just sort of 
overstating it, but it literally--there are some elements of 
truth to this test--when confronted with a program that says, 
there aren't ways, there are not any good ways to measure the 
effectiveness of what I do, when a manager says that for their 
program, then what we did in Sunnyvale--though we did it more 
delicately than I will state it here--you say to the manager, 
you're saying that the nature of this program that you manage 
is such that it's not easy to come up with measures of success?
    He says, that's right.
    Well, we pay you a pretty good salary; but it occurs to me 
that we could probably save ourselves a lot of money if we 
replaced you with somebody that we pay half as much, get 
somebody in here pretty mediocre, because as you indicate, 
nobody could tell the difference.
    Of course, at that point, the manager will start to give 
you ideas on how to measure effectiveness. Because if you let 
them say that this program can't have goals for effectiveness, 
then what you're really saying is, we can't tell the difference 
between excellence and mediocrity.
    If they can't write down, using even the alternative format 
that GPRA allows, a clear distinction that an objective 
observer looking at the program could tell which category it 
fits into, if they can't tell the difference between excellence 
and mediocrity, then why don't we just slash our personnel 
budgets, hire a lot of mediocritites and nobody is going to 
know the difference. We will save ourselves some money.
    That is usually a motivation for people to tell you what 
they're doing is really good, and you can measure it. Often it 
would be somebody else complaining about it.
    Mr. Horn. Mr. McTigue.
    Mr. McTigue. Mr. Chairman, just very briefly, the first 
thing is that you should not absolve anybody from providing you 
with a measure just because what they do is difficult to 
measure. There are varieties of ways in which you can find 
surrogates which will show, maybe not the progress on a 
particular program, but the progress on a body of work, what's 
happening to the volume of knowledge that's being acquired in 
this area. And I think that's very appropriate for places like 
the National Cancer Institute, the science foundations, etc.
    Getting into the detail that Mr. Mihm was referring to, in 
my view, is a way of trying to escape being measured. In the 
private sector, there's a huge range of measures that are used 
to look at research and development and technology development 
and taking into account risk and the fact that not all of the 
investments are going to produce a positive result.
    So there is a whole body of knowledge there where the 
private sector also uses surrogate measures to be able to show 
whether or not there is value in these activities. So I think 
that it can be done. The thing that's important is that it 
should not be excused because it's difficult to do.
    Mr. Horn. Thank you.
    Mr. Mihm, anything to add?
    Mr. Mihm. No, sir, unless you have additional questions.
    Mr. Horn. If there are no additional comments, why we're 
now adjourned. The staff will be put together for Russell 
George, staff director/chief counsel; Bonnie Heald, director of 
communications; Earl Pierce, professional staff member; Scott 
Fagan, assistant to the committee; Chris Barkley, staff 
assistant; Alex Hurowitz, Ryan Sullivan, and Fariha Khaliq, 
interns.
    Minority staff: Michelle Ash, minority counsel; Earley 
Green, assistant minority clerk; and Theresa Coufal, minority 
staff assistant.
    And court reporter: Julie Thomas. Thank you, Julie, very 
much. We appreciate it.
    [Whereupon, at 4:48 p.m., the subcommittee was adjourned.]

