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




   ROBBING MARY TO PAY PETER AND PAUL: THE ADMINISTRATION'S PAY-FOR-
                           PERFORMANCE SYSTEM

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

                                HEARING

                               before the

                   SUBCOMMITTEE ON FEDERAL WORKFORCE,
                    POSTAL SERVICE, AND THE DISTRICT
                              OF COLUMBIA

                                 of the

                         COMMITTEE ON OVERSIGHT
                         AND GOVERNMENT REFORM

                        HOUSE OF REPRESENTATIVES

                       ONE HUNDRED TENTH CONGRESS

                             SECOND SESSION

                               __________

                           FEBRUARY 12, 2008

                               __________

                           Serial No. 110-203

                               __________

Printed for the use of the Committee on Oversight and Government Reform


  Available via the World Wide Web: http://www.gpoaccess.gov/congress/
                               index.html
                      http://www.house.gov/reform




                  U.S. GOVERNMENT PRINTING OFFICE
53-641 PDF                WASHINGTON : 2009
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing 
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; DC 
area (202) 512-1800 Fax: (202) 512-2104  Mail: Stop IDCC, Washington, DC 
20402-0001









              COMMITTEE ON OVERSIGHT AND GOVERNMENT REFORM

                 HENRY A. WAXMAN, California, Chairman
EDOLPHUS TOWNS, New York             TOM DAVIS, Virginia
PAUL E. KANJORSKI, Pennsylvania      DAN BURTON, Indiana
CAROLYN B. MALONEY, New York         CHRISTOPHER SHAYS, Connecticut
ELIJAH E. CUMMINGS, Maryland         JOHN M. McHUGH, New York
DENNIS J. KUCINICH, Ohio             JOHN L. MICA, Florida
DANNY K. DAVIS, Illinois             MARK E. SOUDER, Indiana
JOHN F. TIERNEY, Massachusetts       TODD RUSSELL PLATTS, Pennsylvania
WM. LACY CLAY, Missouri              CHRIS CANNON, Utah
DIANE E. WATSON, California          JOHN J. DUNCAN, Jr., Tennessee
STEPHEN F. LYNCH, Massachusetts      MICHAEL R. TURNER, Ohio
BRIAN HIGGINS, New York              DARRELL E. ISSA, California
JOHN A. YARMUTH, Kentucky            KENNY MARCHANT, Texas
BRUCE L. BRALEY, Iowa                LYNN A. WESTMORELAND, Georgia
ELEANOR HOLMES NORTON, District of   PATRICK T. McHENRY, North Carolina
    Columbia                         VIRGINIA FOXX, North Carolina
BETTY McCOLLUM, Minnesota            BRIAN P. BILBRAY, California
JIM COOPER, Tennessee                BILL SALI, Idaho
CHRIS VAN HOLLEN, Maryland           JIM JORDAN, Ohio
PAUL W. HODES, New Hampshire
CHRISTOPHER S. MURPHY, Connecticut
JOHN P. SARBANES, Maryland
PETER WELCH, Vermont
------ ------

                     Phil Schiliro, Chief of Staff
                      Phil Barnett, Staff Director
                       Earley Green, Chief Clerk
                  David Marin, Minority Staff Director

Subcommittee on Federal Workforce, Postal Service, and the District of 
                                Columbia

                        DANNY K. DAVIS, Illinois
ELEANOR HOLMES NORTON, District of   KENNY MARCHANT, Texas
    Columbia                         JOHN M. McHUGH, New York
JOHN P. SARBANES, Maryland           JOHN L. MICA, Florida
ELIJAH E. CUMMINGS, Maryland         DARRELL E. ISSA, California
DENNIS J. KUCINICH, Ohio, Chairman   JIM JORDAN, Ohio
WM. LACY CLAY, Missouri
STEPHEN F. LYNCH, Massachusetts
                      Tania Shand, Staff Director













                            C O N T E N T S

                              ----------                              
                                                                   Page
Hearing held on February 12, 2008................................     1
Statement of:
    Kelley, Colleen, president, National Treasury Employees 
      Union; John Gage, president, American Federation of 
      Government Employees; Gregory Junemann, president, 
      International Federation of Professional and Technical 
      Engineers; and Carol Bonosaro, president, Senior Executives 
      Association................................................   123
        Bonosaro, Carol..........................................   168
        Gage, John...............................................   137
        Junemann, Gregory........................................   148
        Kelley, Colleen..........................................   123
    Ruiz, Diego, Executive Director, Securities and Exchange 
      Comission; Richard A. Spires, Deputy Commissioner for 
      Operational Support, Internal Revenue Service; and Ronald 
      P. Sanders, Chief Human Capital Officer, Office of the 
      Director of National Intelligence..........................    91
        Ruiz, Diego..............................................    91
        Sanders, Ronald P........................................   108
        Spires, Richard A........................................    99
    Tiefer, Charles, professor, University of Baltimore School of 
      Law; J. Russell George, Inspector General, U.S. Treasury 
      Tax Administration; Charles Fay, professor, School of 
      Management and Labor Relations, Rutgers University; Robert 
      Tobias, director, public-sector executive programs, and 
      director, ISPPI, School of Public Affairs, American 
      University; Stanley Ridley, president and CEO of Ridley & 
      Associates, LLC; and Max Stier, president and CEO, the 
      Partnership for Public Service.............................     9
        Fay, Charles.............................................    39
        George, J. Russell.......................................    29
        Ridley, Stanley..........................................    57
        Stier, Max...............................................    72
        Tiefer, Charles..........................................     9
        Tobias, Robert...........................................    40
Letters, statements, etc., submitted for the record by:
    Bonosaro, Carol, president, Senior Executives Association, 
      prepared statement of......................................   170
    Davis, Hon. Danny K., a Representative in Congress from the 
      State of Illinois, prepared statement of...................     4
    Gage, John, president, American Federation of Government 
      Employees, prepared statement of...........................   140
    George, J. Russell, Inspector General, U.S. Treasury Tax 
      Administration, prepared statement of......................    31
    Junemann, Gregory, president, International Federation of 
      Professional and Technical Engineers, prepared statement of   150
    Kelley, Colleen, president, National Treasury Employees 
      Union, prepared statement of...............................   125
    Ridley, Stanley, president and CEO of Ridley & Associates, 
      LLC, prepared statement of.................................    60
    Ruiz, Diego, Executive Director, Securities and Exchange 
      Comission, prepared statement of...........................    94
    Sanders, Ronald P., Chief Human Capital Officer, Office of 
      the Director of National Intelligence, prepared statement 
      of.........................................................   111
    Spires, Richard A., Deputy Commissioner for Operational 
      Support, Internal Revenue Service, prepared statement of...   101
    Stier, Max, president and CEO, the Partnership for Public 
      Service, prepared statement of.............................    74
    Tiefer, Charles, professor, University of Baltimore School of 
      Law, prepared statement of.................................    11
    Tobias, Robert, director, public-sector executive programs, 
      and director, ISPPI, School of Public Affairs, American 
      University, prepared statement of..........................    42

 
   ROBBING MARY TO PAY PETER AND PAUL: THE ADMINISTRATION'S PAY-FOR-
                           PERFORMANCE SYSTEM

                              ----------                              


                       TUESDAY, FEBRUARY 12, 2008

                  House of Representatives,
Subcommittee on Federal Workforce, Postal Service, 
                      and the District of Columbia,
              Committee on Oversight and Government Reform,
                                                    Washington, DC.
    The subcommittee met, pursuant to notice, at 2:05 p.m., in 
room 2154, Rayburn House Office Building, Hon. Danny K. Davis 
(chairman of the subcommittee) presiding.
    Present: Representatives Davis of Illinois, Norton, 
Sarbanes, and Kucinich.
    Staff present: Tania Shand, staff director; Lori Hayman, 
counsel; LaKeshia Myers, clerk; Mason Alinger, minority deputy 
legislative director; and Alex Cooper, minority professional 
staff member.
    Mr. Davis of Illinois. The subcommittee will come to order. 
I know that the ranking member is on his way from Texas, and, 
of course, depending on where you are in Texas, that could be 
coming a long ways. But we will go ahead and proceed with the 
hearing.
    Let me welcome members of the subcommittee, hearing 
witnesses, and all of those in attendance. I welcome you to the 
Federal Workforce, Postal Service, and the District of Columbia 
Subcommittee hearing, ``Robbing Mary to Pay Peter and Paul: The 
Administration's Pay-for-Performance System.''
    The purpose of the hearing is to examine the executive 
branch's implementation of pay-for-performance systems.
    Hearing no objection, the Chair and ranking member and 
subcommittee members will each have 5 minutes to make opening 
statements, and all Members will have 3 days to submit 
statements for the record.
    Good afternoon, and to the ranking member and subcommittee 
members and all of those present, welcome to the subcommittee's 
first hearing of the second session of the 110th Congress. 
Today's hearing continues the subcommittee's examination of the 
implementation of pay-for-performance systems at various 
Federal agencies.
    Last March, the subcommittee held a hearing on Federal 
personnel reforms, followed by a hearing in May on the 
personnel and pay reforms implemented at the Government 
Accountability Office [GAO].
    Today we turn our attention to the pay-for-performance 
systems at the Securities and Exchange Commission and the 
Internal Revenue Service. We will also take a look at the pay-
for-performance system that the Office of the Director of 
National Intelligence would like to impose on the Intelligence 
Community.
    The implementation of these systems must be evaluated with 
the same intensity that the Bush administration and other pay-
for-performance proponents advocated that these systems will be 
implemented.
    The title of the hearing, ``Robbing Mary to Pay Peter and 
Paul,'' reflects our intent to personalize the impact these 
systems have on Federal employees. Our work force is made up of 
individuals sharing a commitment to public service with 
personal goals and needs. They should not be viewed as or 
treated as a bunch of bureaucrats who can be driven to better 
performance by the prospect of monetary rewards.
    The title also helps explain how the administration's pay-
for-performance system really works. Under the system, Mary, 
who is a good performer and meeting established performance 
expectations, may not receive a cost-of-living increase [COLA], 
needed to offset inflation because her increases are needed to 
reward Peter and Paul, who were subjectively judged to be 
slightly better performers. That is the aspect of pay for 
performance that is so infrequently discussed; that in the 
absence of a significant increase in funds, performance-based 
increases are often funded by denying or reducing other 
employees' COLAs and bonuses. Also, if Mary is an African 
American, the likelihood of her being adversely impacted by the 
subjective application of the performance standards is 
increased.
    Next year, after his agency has conducted a market-based 
study, Peter will be classified as overpaid. Though Peter, with 
30 years of service, has been a good performer, he will receive 
a small bonus, but no COLA. Bonuses are not counted toward base 
pay, which will affect Peter's high three and, consequently, 
his retirement benefit.
    After a year or two, Peter, Paul and Mary are demoralized, 
and their teamwork has suffered. They are uncertain about their 
pay, have little faith in the system, and are looking for jobs 
in agencies that do not have a pay-for-performance system.
    Peter, Paul and Mary are representative of real Federal 
employees whose pay and retirement are being similarly affected 
as we speak. If these systems are not fair and equitable, 
transparent and credible, and do not have the--buy-in of 
Federal employees, I do not believe that they have a place in 
the Federal Government.
    Federal agencies cannot say that they did not know that 
credibility, employee buy-in and equity were key to the 
successful implementation of these systems. These issues were 
raised up front by this subcommittee and others, yet agencies 
are failing in all these areas. Further, these systems do not 
appear to retain employees or increase their performance as the 
administration advertised.
    I ask unanimous consent that the Treasury Inspector General 
for the Tax Administration report on the IRS's pay-for-
performance system be submitted for the record along with 
arbitrator James M. Harkless' decision that the pay-for-
performance system at the SEC resulted in discrimination 
against African-Americans and employees who are 40 and older.
    Today's witnesses are here to help us evaluate these 
systems and, where needed, to recommend corrective measures.
    [The prepared statement of Hon. Danny K. Davis follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Davis of Illinois. Now I would like to ask Delegate 
Norton if she has a statement that she would like to make.
    Ms. Norton. Yes, Mr. Chairman, beginning with expressing my 
gratitude that you have continued to do a great deal of due 
diligence here. I just regret that the subcommittee continues 
to be ignored, because it's going to make the task before us 
harder, but inevitably I want to say a few words about why I 
think the outcome is assured.
    Now, even after the GAO debacle, we now have this 
administration trying essentially to move the same pay systems 
to what amounts to a half of the work force. When I say 
``debacle,'' I don't consider getting a union a bad thing, but 
I get the GAO does. Of course, as the COLA for 2008 is 
restored, there is more work for this Congress to do.
    Let me first say a word about standards. As I read what the 
administration seeks to do, I had worse than a feeling of deja 
vu, because these issues have been litigated and settled for 
decades. To show you how far afield you are and where I come 
from, it certainly is not that there is no conceivable way for 
pay for performance to be done; nobody has even tried to 
implement what the courts have said you have to do if you want 
to create new standards.
    As it turns out, these matters came forward first in the 
discrimination laws, but the courts have said they apply across 
the board.
    Simply stated--and here is decades of court of appeals and 
Supreme Court law--simply stated, let me just move to the 
private sector first, if you want to use a pay system, a 
promotion system, a testing system, or any other term and 
condition of employment, it must be race- and sex- and age-
neutral, end of litigation.
    I don't know whether the administration enjoys a class 
action, but that's what it's going to get any day now, and 
class actions that will inevitably lose.
    Members of Congress, I have said, please save us some money 
and listen to the subcommittee, although I assure you that what 
the administration is doing now is courting not litigation, but 
legislation. That's the rule. If any of our witnesses can tell 
us that rule has been revised or changed, I would like to hear 
it, since I make it my business to keep up with Supreme Court 
litigation.
    Now, if that is the rule that was developed mostly in 
private-sector litigation, let's add on to it government, 
Federal, State or local government and ask what would the rule 
there be? I think we can agree it would at least be what the 
rule is for the private sector, but--and here is where the 
administration is brain dead, if you would forgive me--this is 
a Civil Service system. Read my lips, Civil Service system. It 
wasn't formed that way because someone woke up and said this is 
the only way to do government. It came out of hard experience 
with corruption, to be sure. But its main importance today is 
that the government is subject to due process, the due process 
clause of the fifth amendment. That's why, for example, you can 
fire a private-sector employee, but a public-sector employee 
has to have a hearing.
    Why would you have to do that? Because the fifth amendment 
kicks in, the Constitution kicks in when there is State action. 
The State action is the government employing people, and 
certain due process standards adhere.
    This administration has never understood the difference 
between a Civil Service system and a private system. I have 
just told you that even the private system now has to operate 
under race-, sex- and age-neutral standards.
    As if that weren't enough, what is truly hard to understand 
is how the administration would believe, given the actions of 
this Congress, without exception, that it could fail to provide 
money to reward people in its proposed pay system and take the 
money from COLAs or locality pay. How in the world does the 
administration think it has the statutory authority to do that?
    Let me cite for you what the Congress does each and every 
year with COLAs. The administration sends over a COLA. It 
inevitably divides military from Civil Service. Inevitably, 
overwhelmingly the Congress passes a COLA for every single 
civil servant equal to the COLA for the military. Yet the 
administration believes that it can take some people's COLAs, 
and that the Congress would just sit by and let that happen.
    If the administration wants that to happen, it seems to me 
it has to talk with the chairman, with the committee and with 
the Congress, and that would be difficult to effect, but at 
least we would be having a conversation about what you are 
after.
    So I am here, Mr. Chairman, because I think you are doing 
the right thing in bringing forward the difficulties we see, 
but also asking the administration to come in and describe in 
detail how it is able to skip over just the handful of things 
that I have pointed out that I regard as huge barriers built 
into the system, and to hear from experts who might enlighten 
us so that together we might come to some accommodation and 
would not have to rely upon, on the one hand, litigation from 
employees, or, on the other hand, legislation from the Congress 
of the United States.
    Thank you again, Mr. Chairman, for this very important 
hearing.
    Mr. Davis of Illinois. Thank you very much, Delegate 
Norton.
    Mr. Kucinich.
    Mr. Kucinich. I want to thank you for holding this hearing, 
Mr. Chairman. I think it's important that we thank those who 
work for the government for the work that they do and indicate 
that they ought to be well paid, but they ought not to be 
manipulated; that if they are doing a good job, they should be 
compensated for it, and they shouldn't be pitted against each 
other in any way.
    I want to thank all of you for the work that you do. I came 
here to make that statement, Mr. Chairman. I am glad you are 
looking at this, because this is certainly something that 
deserves to be looked at in the interest of those who serve the 
same public that we serve.
    Thank you.
    Mr. Davis of Illinois. Thank you very much.
    We will now hear from our witnesses, and I will introduce 
the first panel.
    Our first panel of witnesses is Dr. Charles Tiefer, who is 
a professor of law at the University of Baltimore School of 
Law. Prior to joining the University of Baltimore's faculty in 
1995, Dr. Tiefer served as Solicitor and Deputy and General 
Counsel of the U.S. House of Representatives for 11 years. He 
is a quoted expert on Federal, governmental and constitutional 
law.
    Dr. Tiefer, thank you so much.
    The Honorable J. Russell George was nominated by President 
Bush and confirmed by the U.S. Senate in November 2004 as the 
Treasury Inspector General for Tax Administration. Prior to 
assuming this role, Mr. George served as the inspector general 
of the Corporation for National and Community Service.
    Thank you very much, Mr. George.
    Dr. Charles Fay is a professor and chair of human resource 
management at the School of Management and Labor Relations at 
Rutgers University. He has worked over the last 5 years as a 
consultant to the Bureau of Labor Statistics on the National 
Compensation Survey. He was a Presidential appointee to the 
Federal Salary Commission and also served as chair of the 
research committee of the American Compensation Association.
    Thank you, Dr. Fay.
    Mr. Robert Tobias is currently the director of public-
sector executive programs at American University. Mr. Tobias 
was nominated by President Clinton, and the Senate confirmed 
him for a 5-year term, as a member of the Internal Revenue 
Service Oversight Board. The Board has broad and strategic 
oversight authority, responsibility for the IRS.
    Thank you, Mr. Tobias.
    Dr. Stan Ridley is president and CEO of Ridley & 
Associates, a human- and organization-development consulting 
firm in Washington, DC. Dr. Ridley has served on national 
training committees and developed training on such topics as 
managing diversity, performance appraisal and strategic 
planning.
    Dr. Ridley, we thank you.
    Mr. Max Stier is the president and CEO for the Partnership 
for Public Service. The partnership is a nonpartisan, nonprofit 
organization dedicated to revitalizing the public service 
through a campaign of educational efforts, policy, research, 
public-private partnerships and legislative advocacy.
    It is the policy of this committee that all witnesses be 
sworn. So, if you would stand and raise your right hands.
    [Witnesses sworn.]
    Mr. Davis of Illinois. The record will show that each one 
of the witnesses answered in the affirmative.
    Gentlemen, we thank you all for being here. Of course, the 
green light indicates that you have 5 minutes in which to 
summarize your statement, which is already in the record. The 
orange and yellow light indicates that your time is running 
down, and the red light indicates that it's time to stop.
    We thank all of you again for being here, and we will begin 
with Dr. Charles Tiefer.

    STATEMENTS OF CHARLES TIEFER, PROFESSOR, UNIVERSITY OF 
BALTIMORE SCHOOL OF LAW; J. RUSSELL GEORGE, INSPECTOR GENERAL, 
   U.S. TREASURY TAX ADMINISTRATION; CHARLES FAY, PROFESSOR, 
 SCHOOL OF MANAGEMENT AND LABOR RELATIONS, RUTGERS UNIVERSITY; 
ROBERT TOBIAS, DIRECTOR, PUBLIC-SECTOR EXECUTIVE PROGRAMS, AND 
DIRECTOR, ISPPI, SCHOOL OF PUBLIC AFFAIRS, AMERICAN UNIVERSITY; 
STANLEY RIDLEY, PRESIDENT AND CEO OF RIDLEY & ASSOCIATES, LLC; 
 AND MAX STIER, PRESIDENT AND CEO, THE PARTNERSHIP FOR PUBLIC 
                            SERVICE

                  STATEMENT OF CHARLES TIEFER

    Mr. Tiefer. Thank you, Mr. Chairman, and members of the 
subcommittee. I am a professor at the University of Baltimore 
Law School, and the author of books and articles and testimony 
before House and Senate committees on Federal employment 
policies.
    In the past year, since my last testimony on these pay-for-
performance systems, the rollout of these systems has revealed 
serious costs and impacts. First and most dramatically, we have 
seen this year that they pose systemic discrimination risks. 
Unlike objective governmentwide pay increases, which are race-
neutral, these systems work by evaluative ratings, using loose 
subjective evaluation criteria; for example, whether employees 
are considered to be ``collaborating with others.''
    These systems are vulnerable to supervisors' stereotyping, 
diversity-disparaging attitudes and issues of communications. 
Just as in the past, a host of scholarly studies, of which a 
leading collection is Professor Naff's book, To Look Like 
America: Dismantling Barriers for Women and Minorities in 
Government--studies have shown that Federal promotion and 
discipline is vulnerable to such discriminatory biases when it 
works by purely subjective and loose evaluative criteria.
    Now, the particular finding that we had this year was on 
September 4th when the arbitrator's ruling agencywide about the 
SEC found systemic age and race discrimination. As Stephen Barr 
of the Washington Post summarized, the ruling ``found the SEC 
pay system led to discrimination against 324 black employees 
and 1,109 employees who were 40 or older.''
    Now, the other agencies who are still aggressively rolling 
out pay-for-performance systems, from the Defense Department 
with NSPS to the Intelligence Community, which will testify 
today, will want to contend that they have set up neutral 
systems, and they are doing it under plans, and that their 
officers have the opposite of an intent to discriminate, and 
that there's no evidence that they will intend to discriminate. 
But, as you will read, the SEC says in its statement here 
today, and as the arbitrator found, the SEC did not have intent 
to discriminate. Intent to discriminate is not necessary to 
have discrimination found in an agencywide system. What they 
had were a set of statistics that showed a prima facie case of 
rampant age and race discrimination, which followed from their 
using subjective evaluative criteria without a system to check 
what that would mean and without validation.
    Now, the SEC will also tell you today that they have 
revamped their program, but the most important lesson they 
learned, as they say toward the very end of their statement, is 
that ``the Commission has decided to temporarily separate our 
performance management system from the merit pay system.''
    Translation: They have stopped linking pay to performance. 
They have spent some years trying to come out with an 
evaluative system that can work at all, and only then will they 
consider linking it to pay. By ``some years,'' they aren't even 
going to have the performance system in place for the rest of 
the Commission under the top levels until around starting 2009. 
That's the kind of schedule that an agency soberly sets in this 
kind of transition, but it's not the schedule that other 
agencies are considering.
    Are other agencies going to produce numbers similar to the 
SEC's? If we were to check, well, absolutely, I am afraid. A 
study for this subcommittee last year by GAO Strategic Issues 
Office found that by studying the SES governmentwide, a 
shockingly low level of minorities in the SES at the Defense 
Department, half of what is found in the other departments. The 
governmentwide figure for minorities in the Senior Executive 
Service is 16 percent; the Defense Department figure a mere 8 
percent. That was the testimony given to this committee May 
2007 in this GAO report.
    The pay-for-performance system's furthermore disregard just 
like that shouts that in the Defense Department there's a big 
problem in minority manager promotion and recruiting policy. So 
the pay-for-performance system's disregard of seniority and 
experience and evaluations has sent a not-too-subtle signal to 
withhold raises from older employees, which the SEC arbitrator 
found led to a case of illegal age discrimination.
    Now, you will also hear today that the Intelligence 
Community is pushing ahead with its program, but you will hear 
very little of the skeptical and critical congressional 
oversight provision, section 308 of the Intelligence 
Authorization Conference Report, which has demanded answers 
from the DNI about the alarming threat concerning minority and 
age discrimination in the Intelligence Community, because the 
intelligence agencies have a history of much worse numbers on 
minority employees than the civilian departments. If you go 
back not that far and put it frankly, some of those 
intelligence agencies were lily white.
    Now, is DNI making a recruitment and promotion of 
minorities a top priority? No. Like the SEC, they are 
instituting a program of subjective performance evaluations on 
criteria like, ``Do they collaborate with others.''
    I might say second that the administration's no new regular 
funding basis does what the title of this hearing said: It robs 
Mary to pay Peter and Paul. I summarize in my testimony why the 
NSPS recent pay raise does not include regular funding, and I 
will refer to the balance of my statement.
    Thank you.
    Mr. Davis of Illinois. Thank you very much.
    [The prepared statement of Mr. Tiefer follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Davis of Illinois. We will go to the Honorable J. 
Russell George.

                 STATEMENT OF J. RUSSELL GEORGE

    Mr. George. Thank you, Mr. Chairman, Delegate Norton. It's 
a pleasure to be before you today to discuss the Internal 
Revenue Service's pay-for-performance system.
    The Federal Government is faced with a human capital 
crisis, and the IRS is no exception. By the end of fiscal 2010, 
66 percent of the IRS's executives, 50 percent of its senior 
managers and 36 percent of its front-line managers will be 
eligible to retire.
    To its credit, the IRS has already begun efforts to address 
its human capital challenges. It has made progress in 
transforming the agency into a more efficient, modern and 
responsive organization.
    A tool that should help recruit, retain and motivate 
managers is a pay-for-performance system, but to do so, it 
needs a good design and an even better implementation.
    TIGTA has looked closely at the IRS pay-for-performance 
system. If I were to issue a grade on the system similar to the 
report cards that former Representative Steve Horn issued when 
he was a subcommittee chairman on this committee, I would have 
to give the IRS a C.
    The IRS' efforts to implement a pay-for-performance system 
began after enactment of the IRS Restructuring and Reform Act 
of 1998. The IRS has implemented its pay-for-performance system 
in three phases. Phase 1 for senior managers was launched in 
March 2001. It placed all grades 14 or 15 mid level managers 
into one pay band. Phase 2 for department managers was launched 
in November 2001. It placed all grades 11 to 13 second-level 
managers at IRS campuses into one pay band. Phase 3, which was 
launched in 2005, placed all front-line managers from grade 5 
to 15 into 11 pay bands.
    TIGTA's July 2007 audit of the pay-for-performance system 
found three major shortcomings in its design. First, the IRS's 
pay-for-performance structure did not provide the benefits 
envisioned in the Reform Act. These benefits include the 
flexibility to assign new or different work, a greater ability 
to hire more quickly and offer more competitive salaries, and 
the ability to provide employees with better opportunities to 
enhance their knowledge and advance their careers.
    Second, the method for determining annual salary increases 
for managers on all three pay systems may result in pay 
inequities. For example, managers could receive increases that 
are less than those received by other nonmanagerial employees 
who automatically receive the annual across-the-board salary 
increases established by the President.
    Third, the Office of Personnel Management requires that the 
lowest and highest salaries for each pay band must be 
commensurate with the corresponding GS pay system's salary 
changes. As a result, managers at either end of a pay band 
would automatically receive annual across-the-board salary 
increases regardless of their performance rating. However, 
managers whose salary fall in the middle of the pay bands are 
rewarded based on their performance.
    Regarding implementation, TIGTA found that the IRS did not 
allow sufficient time to educate managers on the details of the 
front-line manager system. This resulted in increased 
opposition and decreased morale among the 6,600 front-line 
managers. Unfortunately, this lack of communication occurred 
because the IRS accelerated implementation of the system by at 
least a year in order to minimize the conversion costs in 
fiscal year 2005. This acceleration left little time to explain 
the system and no time at all to receive feedback prior to its 
implementation. Predictably this caused significant 
frustration.
    The IRS responded to TIGTA's finding by initiating a third-
party evaluation to be conducted in three phases over 5 years. 
This lengthy timeframe is of concern. We are monitoring the 
IRS' corrective actions and will conduct a followup review. In 
addition, the IRS plans to continue partnering with the 
management associations representing a number of IRS managers 
on pay-for-performance issues. The IRS has also agreed to 
communicate more effectively with employees before, during and 
after any additional changes to the system.
    However, the IRS disagreed with our recommendation to 
reinstate its policy of providing across-the-board adjustments 
to managers who receive a satisfactory or higher performance 
rating. The IRS stated that the authority for determining 
salary increases rests with the IRS Commissioner. The 
Commissioner does, indeed, have the authority to set pay 
increases, and I would not suggest that this authority be 
removed; however, the Commissioner can reinstate the policy of 
providing across-the-board adjustments and still give higher 
pay raises to highly rated managers. The Commissioner could 
allocate amounts that have previously been designated for 
within-grade step increases and quality step increases to 
recognize higher-rated managers.
    The difficulty of designing and implementing a pay-for-
performance system in an agency as large and complex as the IRS 
cannot be overstated. TIGTA is dedicated to helping the IRS 
navigate through this difficult project, as well as the many 
other human capital challenges that it faces.
    Thank you, Mr. Chairman.
    Mr. Davis of Illinois. Thank you very much.
    [The prepared statement of Mr. George follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    
    Mr. Davis of Illinois. And we will go to Dr. Charles Fay.

                    STATEMENT OF CHARLES FAY

    Mr. Fay. Thank you, Chairman Davis and Delegate Norton.
    Let me start out by saying that my whole background leads 
me to be in favor of pay for performance and performance 
management. Those two things, in conjunction together, have 
been shown in many cases in the private sector to work very 
well.
    When I look at the public-sector case, and the systems of 
the SEC, the IRS and even the DOD, I am deeply disappointed. 
Those systems don't work. I don't think they are likely to 
work. They seem to me to violate many of the requirements for a 
good pay-for-performance system that my coauthor and I set out 
in the publication that was submitted as part of my testimony 
that was sponsored by the IBM Center for the Business of 
Government.
    Let me do a couple of things; first, separate the notion of 
performance management and compensation, or pay for 
performance. Performance management is the key point. If people 
get that right, then the pay for performance becomes much 
easier, and it is clear in these systems that they have not 
gotten the performance management part right, not at all. In 
addition, as several people have noted, the SEC also messed up 
seriously on the pay-for-performance part of it. So they struck 
out in both cases.
    Let me just tell you some of the things that I see in this 
that violate the basic principles of a pay-for-performance 
system and a merit-pay system. The first of those is that 
performance criteria need to be job-specific. That's 
performance management 101. When you have criteria for 
someone's performance, those have to be specific to that job.
    Second, the general competencies and criteria that have 
been used by at least two of the systems that I have seen, DOD 
and IRS, who have published portions of them, are very general 
and very subject to bias, as a couple of people have mentioned. 
This is just asking for trouble when you ask managers to make 
decisions based on these kinds of things.
    Third, and perhaps most important from my perspective, it's 
critical that raters and the people who are rated be trained in 
how to do performance management from both sides of the aisle. 
Both the management and the director need to be able to know 
how to work this system, know how to use it correctly and to 
have some understanding for the reason for the system. In the 
documentation I see, there does not appear to be any, or at 
least not much, training of people.
    Third, the instructions should be written in plain English. 
The IRS reminds me of tax forms that I tried to fill out every 
year, and the DOD, pure ``mil spec.'' I have a Ph.D. I consider 
myself to be fairly reasonably cognizant of the English 
language, and I had a hard time figuring out what you are 
really supposed to do with some of these languages.
    Finally, I would like to stress that all of these systems 
are subjective by their nature. No performance management 
system or performance appraisal system is not subjective. They 
cannot be objective. If anyone has issues with that, think 
about the last time you had overhead allocated to a program 
that's highly objective. But it's very subjective how those 
dollars get allocated.
    So, again, I still believe people should manage 
performance. I believe the government should investigate 
performance management, should work at pay for performance, but 
I think they need to do a much better job than has been done so 
far.
    Thank you.
    Mr. Davis of Illinois. Thank you very much.
    We will go to Mr. Tobias.

                   STATEMENT OF ROBERT TOBIAS

    Mr. Tobias. Thank you very much, Mr. Chairman.
    As Inspector General George pointed out, the Internal 
Revenue Service got the authority to create pay-for-performance 
systems, and what I did was take a look at some employee 
surveys of the senior managers, the departmental managers that 
were conducted to determine whether or not the IRS achieved its 
goals.
    I think the results of these surveys reveal very clearly 
that the IRS has not achieved its goals of creating a pay-for-
performance system that links individual and organizational 
goals with compensation, or that supports better rewards for 
better performance. For example, only one in four senior 
managers agree that the pay-for-performance system is a fair 
system for evaluating or rewarding job performance; 58 percent 
of the managers stated that the pay-for-performance system had 
no impact, no impact on their motivation to increase 
performance; and 18 percent stated it had a negative impact.
    Less than 20 percent of the senior managers agreed that the 
pay-for-performance system is linked to organizational results 
or has improved business unit or IRS performance. Only 17 
percent stated that their performance was linked to their pay, 
and only 16 percent stated that their performance was linked to 
strategic business goals; 46 percent state that had they 
preferred the GS system.
    Now, I think there are several reasons why the IRS goals 
have not been achieved. First, the data does not reveal that 
employees are motivated by pay; rather, I believe Federal 
employees are motivated to increase their performance far more 
by effective leadership and an effective performance management 
system.
    Second, I think it's difficult, if not impossible, to have 
a successful pay-for-performance system if the leadership and 
the agency does not support it. In a 2006 survey of the Senior 
Executive Service, the persons who must be counted on to 
implement a pay-for-performance system, reveal that they are 
extremely unhappy with their pay-for-performance system; 83 
percent stated it had no impact on their performance, and 33 
percent indicated it had a negative impact on morale; 54 
percent indicated their pay-for-performance system had no 
impact on their motivation to increase performance, and 8 
percent indicated it had a negative impact; 44 percent felt 
their ratings were fair and accurate, but 33 percent disagree, 
and 23 percent did not know.
    Since the organizational change necessary to support the 
introduction of a pay-for-performance system is not 
enthusiastically endorsed by senior career leaders, it's no 
surprise that those they lead do not endorse it.
    Third, trust is a critical component necessary to implement 
any significant organizational change effort, and particularly 
pay for performance. Only 50 percent of the close to one-
quarter of a million Federal employee respondents to the 
Federal Human Capital Survey answered ``strongly agree'' or 
``agree'' to the statement, ``My organization's leaders 
maintain high standards of honesty and integrity;'' 20 percent 
reported either ``strongly disagree'' or ``disagree.''
    In the IRS work force, only 43 percent ``strongly agree'' 
or ``agree'' with the statement, and 23 percent ``strongly 
disagree'' or ``disagree.''
    The IRS has even more work to do if it wants to 
successfully implement a pay-for-performance system.
    Now, can the IRS achieve its stated goals in the future? I 
think--because the data shows that Federal employees are not 
necessarily motivated to increase their performance solely 
because of monetary rewards, I believe it's highly questionable 
that the IRS can successfully implement a pay-for-performance 
system.
    Because Federal employees are motivated to increase their 
performance when they are effectively led and their skills are 
matched to their mission, I believe it is possible to increase 
individual and organizational performance results with the 
design, development and implementation of a performance 
management system.
    Now, I believe Presidents, political appointees, members of 
the Senior Executive Service, midlevel managers, union leaders 
and employees all want increased performance. That's why they 
are part of public service. Harnessing that energy to actually 
improve performance in the executive branch requires the 
collaborative involvement of all parties. I believe it can be 
done, but I don't see any real evidence of a focus on improving 
performance management.
    Thank you very much, Mr. Chairman.
    Mr. Davis of Illinois. Thank you. Thank you very much.
    [The prepared statement of Mr. Tobias follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    
    Mr. Davis of Illinois. We will now proceed to Dr. Stan 
Ridley.

                  STATEMENT OF STANLEY RIDLEY

    Mr. Ridley. Thank you, Mr. Chairman.
    Most of us remember the political quip, ``It's the economy, 
hmm.'' To cut to the basics here, it's the basics, hmm, and the 
people, hmm.
    What Delegate Norton pointed out was that to design and 
implement an effective performance appraisal, performance 
management, and even a pay-for-performance system, the 
knowledge and skills have been here for decades. People get in 
the way.
    I should mention that my roles in the government over a 
decade have included being a coach, an organizational change 
management expert, a performance appraisal and management 
developer, teaching courses on supervising and human resources, 
etc.
    The bottom line is that these things do not work as they 
should because they are either designed inappropriately, or, 
more often, they are implemented in a way that are not 
consistent with the basics of how they're supposed to be done. 
That is the fundamental problem.
    When you look at the SEC, an article is attached my 
testimony about pass/fail. Over a decade ago I used a group of 
employees in the government, all levels, and we went through an 
exercise the appropriate way, and they discerned that pass/fail 
had no chance of being successful. About 6 months later, pass/
fail was chosen. Did anybody mention pass/fail today? In the 
SEC case, they used pass/fail. The bottom line is that the 
basics and cycle metrics would tell you that pass/fail could 
never work, and a group of employees, when given the right 
information, were able to make that decision more than a decade 
ago.
    The bottom line is that the knowledge and skills are there 
to make any system that is designed to motivate employees to 
perform at their maximum and then to compensate them 
appropriately, but if we do not employ those the right way, it 
would never work.
    The key thing I learned about 10 years ago was that the 
biggest problem in terms of the government--they have some 
excellent employees, and I am not saying that to be PC--
accountability. That was such an issue that I had to define it, 
and by that I meant consequences, positive or negative, based 
on how well you do what you are supposed to do.
    How is it that we have the knowledge and skills to design 
these programs and implement them the correct way, but they 
have failed? It is not a knowledge or a skill issue. It is the 
failure to use the knowledge and skill issue and to adhere to 
whatever it is that you develop.
    One of the things I should say about the Federal 
Government, when I read these documents, boy, they sure look 
good. You even see in my testimony, I made a faux pas when I 
made a comment, although I did use the word ``appear,'' about 
what I call the pay side of pay for performance. I say, well, 
it appears like that is not too much of a problem, OK? But if 
you delineated an example of how that particular approach has 
been misused--and that is what I have seen over time, either 
underuse, nonuse or misuse of the basics that will allow you to 
design and implement these systems correctly, and the way in 
which you make sure that people actually do that is that you 
consistently and appropriately hold them accountable.
    Any time you see a system that has failed, I will show you, 
I can show you, that there was a failure to appropriately and 
consistently hold people accountable for what it is that they 
do.
    That's the short version of my presentation. I will make a 
couple of recommendations, because that's what this really 
should be about, as to the performance appraisal side of this.
    We have--the OPM developed what is called a PAAT, a 
performance appraisal assessment tool. It's a good tool. It 
shows that they have made progress over the last 10 years. But 
that tool is not sufficiently aligned with what we want pay for 
performance to be. A concrete example, the tool says that if 
you meet the criteria at the 80 percent level, you are 
certified. The problem is that tool includes some very good 
items. Ask yourself this question: If the item is important, 
why is it that you can fail on 20 percent of them, and then you 
are certified? These are basics.
    The next thing in that tool, they have items that use the 
word ``adequate.'' Why would you have as a standard--Delegate 
Norton, you mentioned standards--adequate in a tool that's for 
the whole government that is supposed to let us know what we 
are supposed to be able to do.
    Other inconsistencies, these are basics, is the pay for 
performance is supposed to be results-focused. Great concept, 
no problem. Yet in the actual performance appraisal plans, 
individuals are required to have at least one performance 
element that is results-focused. That's what we call internal 
inconsistency. If results focus in what it's about, and I have 
five elements, why is it that only one is required to be 
results-focused? These are all basics.
    To cut to the chase further, in the PAAT, a number of 
excellent things they have are either suggested or recommended. 
If they are so good, why is it not required?
    When you start talking about pay for performance and the 
valuable performance management, how about this as a 
recommendation? A sizable bonus should be awarded to each 
agency executive whose performance appraisal system is shown to 
be clearly valid and reliable in design, and here is the key 
thing: design and practice. Where we come up short most times 
is in the practice, based on a qualified objective review and a 
100 percent pass score.
    My point is really simple. The knowledge and skills 
necessary to have an appropriate performance appraisal, 
performance management, pay for performance, whatever you call 
it, those requisite knowledge and skills have been here for 
decades. We misuse, underuse or fail to use the appropriate 
criteria.
    For example, I was surprised when you mentioned this misuse 
about COLAs. That shows a fundamental lack of understanding 
about what we call systems. See, a performance management 
approach should involve the PAAT system so you shouldn't be 
robbing Peter to pay Paul if it's unfair.
    So when you sit down to design these things, what you tend 
to do, we design processes, not systems. Concrete example I use 
is a car. Why would you suggest that you could have an 
effective performance appraisal system, and you don't have to 
have a real performance plan, a valid and reliable one?
    I had mentioned in my testimony, it's kind of analogous to 
having a car and saying, we recommend that you have a steering 
wheel and a speedometer, but you don't have to have one. The 
performance plan is supposed to guide that person throughout 
the year, but you are supposed to get clear expectations up 
front and then reinforce then.
    The last one I will mention is the acronym that I used. It 
is so simple, but if it is used consistently in the design and 
implementation and evaluation of any system, that system will 
work better. It's called CARE. You must have Clear, Aligned and 
Reinforced Expectations. Any system that you all point out that 
has failed, I will show that you there is a failure in one or 
more of those four pieces.
    Thank you.
    Mr. Davis of Illinois. Thank you very much.
    [The prepared statement of Mr. Ridley follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    
    Mr. Davis of Illinois. We will go to Mr. Max Stier.

                     STATEMENT OF MAX STIER

    Mr. Stier. Thank you.
    Thank you as well, Congresswoman Norton.
    It's a pleasure to be here. It's very important that you 
hold this hearing.
    I would ask that this committee take a step backward for a 
moment and ask about not just pay for performance, but pay for 
performance more generally, because what I would argue is that 
the government is making distinctions around personnel issues 
in a variety of areas that don't receive sufficient attention. 
So, for example, whether it is hiring or promotion or decisions 
around pay, government managers are making decisions that are 
based upon systems that sometimes are working and sometimes 
not. It's only because we are looking at efforts to change the 
system right now that there's real attention being paid here, 
and I think it would be worthwhile to focus on performance 
issues more generally, understanding how it is we are making 
distinctions, and how to use those distinctions in a way that 
truly and effectively and appropriately promote performance.
    Rather than looking backward, looking at issues and the 
stakes or problems, I would propose six different 
recommendation that we would make; first and foremost and most 
important, the clear need to collect data and publish that data 
on a regular basis. We see examples here of that, when Mr. 
Tobias talks about the Human Capital Survey. In fact, his data 
points to the fact that it is not pay-for-performance systems 
that are generating problems in the work force, it's actually 
the whole system itself.
    General Schedule employees are raising issues about the pay 
that they are receiving, about the leadership that they are 
receiving. These are issues that are systemic and that we need 
to address governmentwide. We need specific data efforts, 
collection efforts. When you are talking about system change, 
we propose seven different buckets of data that ought to be 
collected around recruitment, retention, skills gap, 
performance distinctions, performance culture and, ultimately, 
as the last witness has testified, around implementation. That 
data is quite important because it will allow you and other 
decisionmakers to understand what is happening in real time 
when these changes are being implemented.
    Second, fundamentally, I believe, and you see this again 
and again, you need true engagement and buy-in from the 
employees and employee organizations. Unless that happens, 
there is simply not going to be any success. Ultimately, when 
you think about performance management systems, you are 
ultimately trying to encourage employees to provide this, as 
Bob would say, their discretionary energy. How can they do that 
if they don't buy into the system? Therefore, that has to be a 
foundational element of success.
    Third, before moving to pay systems, again, and this is 
something you have heard from other witnesses, you really need 
to make sure that your performance evaluation and your larger 
performance management system is working.
    Fourth, managers have to receive the training, the support 
in order to be able to successfully use these systems. When you 
are asking them to make distinctions, you need to make sure 
that they are getting the tools that will enable them to make 
them appropriately. Ultimately they need to be held 
accountable. They need to understand that organizationally this 
is an important part of their work. By and large these are 
challenging, difficult issues, and, given a choice, most 
managers would rather do other things. So it's very important 
to make sure that they understand that they themselves are 
being judged on their performance with respect to the way that 
they are judging the employees that they are supervising.
    Fifth, we need to make sure that we give these systems time 
to actually change and time to actually work. It's important to 
collect the data. It's important also to realize that change is 
difficult of this sort all the time, and you are not going see 
a turnaround immediately.
    If you look at some of the data points from some of the 
demo projects, you actually see increased buy-in from the 
employee base over time, and that is something I think again is 
vital; that we make sure we understand that these things, A, 
are not likely to be gotten right the very first time, that 
they are going to take time to actually work through the issue 
and to get the buy-in that's necessary.
    Finally, No. 6, I would propose that this committee and 
Congress more generally has a critical role in providing the 
resources to make sure that the systems are resourced 
appropriately, and that the work force is resourced 
appropriately, and that you give this kind of attention to the 
changes that are necessary.
    So thank you very much.
    [The prepared statement of Mr. Stier follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    
    Mr. Davis of Illinois. Thank you very much, and I want to 
thank each of you.
    I want to ask each of you one question. I will begin with 
you, sir.
    In your testimony, you recommend that GAO conduct a study 
and audit of statistics that involve pay for performance. Would 
you explain why?
    Mr. Stier. Certainly, Mr. Chairman. In the past year, GAO's 
Office of Strategic Studies has done two extensive surveys, one 
governmentwide about the SES and the other governmentwide about 
Hispanics, of employment statistics. They have experience, and 
they have the data base. There is the CPPF data base. It exists 
already. It is not just that it has to be brought into 
existence.
    Using this data base and perhaps a limited supplemental 
information, which this same office has used before, they did a 
report in 2004 about the demonstration pay-for-performance 
projects in existence at the time. Using their existing skill 
set and their existing data base, they could produce the 
statistics that would show who the raises have been going to in 
SES and NSPS and the IRS managers system and so forth. The 
statistics that they could produce would determine which units 
in the government have the kind of statistics that the 
arbitrator found at the SEC, that the arbitrator ruled that the 
SEC made out a prima facie case of discrimination.
    If they simply run the numbers up--I say simply, it's a 
considerable effort to do that--but if they were to take the 
systems that they are operating now, we could find out which 
units have what is in effect a prima facie case of 
discrimination going on using the pay-for-performance system.
    Mr. Davis of Illinois. Let me ask. Why do you think that 
the Federal pay-for-performance systems are failing? And what 
if any legislative changes would you recommend?
    Mr. Tiefer. Well, I think they are failing to take them in 
a different order than I originally stated it. The No. 1 reason 
that they're failing is that there is no new regular funding, 
basically no new funding being put in for the performance 
bonuses. And if you look at this--if you survey the literature 
on pay-for-performance, what makes for success is if new money 
is put in so that the employees accept--employee acceptance 
goes with the system. Even if some employees don't get bonuses, 
they say, well, we didn't lose anything from it, so we are not 
going to fight the system; it's not embittering us.
    The fact that no new money is put in is exactly why the 
title of this hearing, robbing Mary to--I'm sorry--I get it 
mixed up. I get it mixed up. I get it mixed up between Mary, 
Peter and Paul. It's the same as the pay-for-performance 
system. It is hard to figure out who it is coming from and who 
it is going to. But if you don't put new money into it, you 
know, you don't have any net winners; just as many losers as 
winners.
    And that is unfortunate because--let me add--a second--the 
second reason is that we're using not--we're using a value of 
ratings which were rushed into operation without giving them 
time to reach the validated stage. Ms. Norton talked about--and 
rightly so--the criteria that are used to determine whether you 
have employment discrimination in the private sector. If you're 
going to have an impact, if you are going to have an employment 
system that may have a statistical impact that is not age 
neutral and is not race neutral, then it has to be the private 
sector title 7 litigation as shown validated. The systems we're 
talking about here have not been validated. They couldn't be 
validated. Those are the two basic reasons.
    Mr. Davis of Illinois. Thank you very much. Dr. Fay, last 
year you testified before the subcommittee regarding a market 
based compensation study conducted by Watson Wyatt for GAO. 
Given your findings, would you recommend that GAO continue to 
use the Watts Wyatt study as a basis for determining pay at the 
GAO?
    Mr. Fay. I wouldn't use that study for the basis of 
anything.
    Mr. Davis of Illinois. Could you elaborate?
    Mr. Fay. Well, as I testified before, the method of data 
collection for that survey, the definition of jobs, the job 
comparability, that is, the market matches that they made were 
inadequate. The surveys they used did not have, in my judgment, 
the appropriate sample comparators, and the way they utilized 
the data to come up with the numbers they did--well, the second 
or third time they did it--were I think misleading and 
inaccurate or gave inaccurate results.
    Mr. Davis of Illinois. Well, let me ask you. If validated 
performance management systems are a prerequisite to the 
implementation of successful pay-for-performance systems, why 
do you think agencies have focused more on implementing 
credible performance management systems?
    Mr. Fay. Well, I would disagree that--it is very difficult 
to validate a performance appraisal. Performance measures are 
usually what is called a criterion variable. In selection, for 
example, you predict in the selection process, you predict 
toward performance. You have to use content validation methods 
with performance appraisal systems. And my dissertation was on 
errors and performance rating. And I recall all too well how 
many rating errors can dance on the head of a pin. It is not 
easy to do. And for that reason, there is a number of people 
who have said it is really critical that you get buy in from 
employees and employee organizations. This has to be a 
collaborative process, whereby people recognize that what 
they're doing is aimed toward improving performance of the 
organization as a whole and rewarding those who most help that; 
not a system where you have kind of a gotcha after the fact.
    This is a--one person mentioned that planning is critical. 
If performance planning is done correctly by a manager and his 
or her direct reports, much of the rest of it falls into place. 
The problem is when you have these generic criteria--which by 
the way in my opinion are against the OCCP rules on dealing 
with selection and so forth and validation of decisions, 
primarily selection decisions. They state specifically in those 
rulings that they have to be job-specific. And these are not 
job-specific. And you see people twisting, trying to get an 
employee's actual duties and responsibilities and behavior and 
so forth, outcomes to match those. If you're going to be rating 
performance, you should decide what is important in performance 
on an individual and discuss it with them, coach and counsel 
them over the year to get them there. And the summary rating at 
that point becomes a--almost a nonevent. In these systems, the 
summary rating is to me the total event.
    Mr. Davis of Illinois. Thank you very much.
    Delegate Norton.
    Ms. Norton. Mr. Chairman, let me try to get my arms around 
some of this.
    Mr. Stier, in describing failure or success on page 4 of 
your testimony, problems with unsuccessful systems often 
traced--and then you list--communication with employees, overly 
ambitious timelines, ambitious goals, inadequate guidance and 
poor assessment. Do you believe that you can implement--because 
you have been a defender of these systems. And as I've 
indicated, there are circumstances in which these systems are 
more than defensible. Do you believe you can--you can have a 
successful pay-for-performance system if there has been no 
validation of the performance criteria?
    Mr. Stier. I would say--I mean, I think the position of the 
partnership is very much akin to--I'm sorry--the position of 
the partnership for public service is very, very much akin to 
what you describe, which is that these are very difficult 
systems to get right. It is possible to get right, but they 
require quite a bit of effort and effort that is really very 
much focused on the individual manager in that relationship 
between the manager and the----
    Ms. Norton. What do you think the systems--the criteria--
the criteria for judging the employee would need to be 
validated?
    Mr. Stier. I think that, again, as some of the earlier 
witnesses stated, that you're not going to be able to validate 
each individual performance plan that the employee has. You can 
and ought to be collecting information on whether there is an 
adverse impact with any system that is being put in place.
    Ms. Norton. You just described what happens if a plan has 
not been val--in other words, you're all saying that you do not 
believe that a system would be valid if it discriminated on the 
basis of race, sex or age?
    Mr. Stier. Correct, correct. I think you have to look----
    Ms. Norton. The only way in which we found to keep that 
from happening is to require employees to suspend what has now 
been if I can be clear, since 1960, billions of dollars, an 
incalculable amount of dollars, in order to validate systems so 
that indeed they can use them. And here I've not gotten to the 
Federal employ--the Federal Government has paid--and I don't 
know how much--but it also has paid a boatload of dollars. But 
in the private sector, which led the way on validation, people 
went to the trouble of the kind Dr. Fay has described of 
validating each job. And you have just told me what you can 
expect to do that.
    Mr. Stier. To be clear, my point is that I think you're 
absolutely right that we've under-invested in ensuring that 
these are the right systems. I think----
    Ms. Norton. I just want to know--you can't take--the 
position that you're taking is a position to which I'm 
sympathetic. You are not the best defender of that position if 
you seem to be, as you--as part of your answer seems to say, 
that even in the Federal Government, those systems can be used 
if they show discrimination. This is very late in the game for 
either the private sector--and I want to stress this--or the 
public sector, either sector to put into place performance 
criteria that discriminate, except the private sector doesn't 
have the overlay of the Civil Service system so the arbitrator 
throws the whole kit and caboodle out. Do you think that would, 
if it were appealed to court, survive?
    Mr. Stier. Congresswoman Norton, just to be clear, if I 
have left you with the impression----
    Ms. Norton. You certainly have left me with the impression 
that you are generalizing across the board. This testimony does 
not even mention validation.
    Mr. Stier. Oh, I think it does. It certainly talks about 
adverse impact, and it refers to the finding that you're 
describing there.
    Ms. Norton. You can't avoid--you can't avoid what we're 
investigating here. What does the Government have to do in 
order to avoid adverse impact? And that's what I'm asking you. 
Isn't the case law--isn't the state of art of the profession 
that you have to take the trouble and you have to spend the 
money to validate whether you're a GM, Toyota or the Government 
of the United States?
    Mr. Stier. I will--there are other experts on the 
validation process here on this panel. What my understanding 
is, you certainly need to look to see if there is any adverse 
impact. If you will look at our testimony, we are very 
supportive of the idea----
    Ms. Norton. I take it you're for validation. That's all I 
wanted to make sure.
    Mr. Stier. Certainly.
    Ms. Norton. But you have something in your testimony that 
concerned me. You haven't paid--something else in your 
testimony. You see, the over--what you are trying to do overall 
is important. But you don't do a service by mitigating how 
difficult it is going to be. And you particularly don't do us a 
service--if I may say so, I've been on the other end of this 
where I have been part of an enforcement agency that forced the 
private sector to spend all of this money. Here I now come to 
the Congress of the United States having remembered all that, 
billions of dollars that the private sector had to spend. 
Which, yeah, they'll say the same thing, could it have gone for 
something else. And we want to apply a lesser criteria to a 
system that has also due process built into it. On page 4, you 
tell us--boy, I think this would be a headline, and I'm going 
to let them speak for themselves. But you tell us that the NTEU 
has reached agreement with the National Credit Union 
Administration on a compensation system. See? It can be done. 
One of the--one of the great unions in--who is a tough union, 
has already reached agreement that will reward the employees 
commensurate with their performance. Did you think we weren't 
going to check that out? Especially since--as a matter of fact, 
we know they can bargain for pay-for-performance. And then you 
say previously NTEO, EEU--and again I just think this ought to 
be a banner headline, particularly in the Washington Post also 
where so many Federal employees are--also reached a collective 
bargaining agreement with the Security and Exchange Commission 
and the Federal Deposit Insurance Corporation on their 
performance pay system. Why didn't you say the arbitrator had 
thrown it out?
    Mr. Stier. Congresswoman, I'm very confused here as to what 
your concern is. These are all factually correct statements. We 
have not misrepresented anything at all. And I'm happy to 
explain any----
    Ms. Norton. I have just--you can explain--you can correct 
what I just asked you. I have put facts before you. I said they 
can bargain for pay-for-performance.
    Mr. Stier. I'm sorry. As I understood what you said, is you 
said we had made a statement that there had been an agreement 
between the National Credit Union Administration [NTEU], and in 
fact, that is correct. There has been an agreement. I'm not 
understanding exactly----
    Ms. Norton. According to the union, there has been no such 
agreement. I'd be very interested in hearing that statement 
because there has in fact been a----
    Ms. Norton. And according to the union--and according to 
the union, with the SEC, there was impasse. And you know what 
impasse means?
    Mr. Stier. I do.
    Ms. Norton. It means that you can't reach agreement with 
your union, so you then go to have--have the system impose----
    Mr. Stier. Congresswoman Norton, the testimony is intended 
to be helpful to you. I apologize if it is not. I don't believe 
there are any factual inaccuracies, but we'll of course correct 
them if there are any.
    Ms. Norton. I'm going to let them speak for themselves.
    Mr. Stier. I'd like to hear if there are any----
    Ms. Norton. This is a fine art, and this is a very 
technical matter. And really what we have to get down--down to 
because I find most of your points, points I'm in agreement 
with, and I think everybody else on the panel would be in 
agreement with your points about, you know, getting employees 
to buy in and the rest. You know, that doesn't get to what is 
the nub of the problem, extremely technical, reams of 
litigation, lots of money spent both on litigation and on 
validating system, shouldn't have to go back and start all over 
again.
    I want to ask--ask you--you all this question, all of you. 
The notion that most of you at least have indicated that a 
system has to be validated in order to be valid if you're going 
to judge people. And you can't judge people on the basis of 
subjective criteria. Now, employees all get evaluated now. Are 
the systems by which employees get evaluated now valid? How do 
you explain that? Do they use any subjective criteria?
    Mr. Ridley. I would say, much too often, they clearly are 
not. And that is based on years of experience doing this.
    One quick thing about--in terms of my colleague who noticed 
that they are subjective, this is one of the things that I call 
manager/supervisor learning disabilities. That if it is really 
all that subjective, why do we go through the time trying to do 
what Delegate Norton is talking about? If it is all subjective 
as we just said, then we just said, we don't need all these 
systems. The fact of the matter is, is that subjectivity and 
objectivity are matters of degree. And what you want to do is 
curtail the subjectivity; increase the objectivity. And how do 
you do that? You dial up to get as much clarity as you can so 
at the very least you have a shared understanding of what is 
expected concretely from the supervisor. And I say that your 
report must be written well, and you say, what does that mean? 
I will be specific with you. I would mean that the right 
content must be in there. If it is not, that is a problem. I 
would suggest it needs to be well organized. If it is all over 
the place, that is a problem. If there is a problem with 
format, spelling--you need to be specific. And my point here--
and I'll let this go--is that we can do these things. We can do 
them well. But any time you engage in a false dichotomy in 
terms of something being either subjective or objective, you 
get no place. It's a matter of degrees.
    Ms. Norton. I want to get to whether or not we can do them 
so easily if we just kind of do them well.
    Mr. Tobias. I think you're using----
    Ms. Norton. I'm really looking for the answer to my 
question about what is it that keeps the present system from 
being attacked? Now, in his testimony--in his testimony, Mr. 
Stier correctly says Title V recognizes that incentives and 
recognition for excellence are appropriate. Then he says but 
the current 1949 era general service system does not reward 
high performance. It rewards above all else longevity. Well, he 
is right. Why does it do that, gentlemen? Why didn't it start 
out doing exactly what Mr. Stier wanted to do? That is the 
point I'm trying to get somebody to speak to since it is 
obvious that if you can do it other than by time and service or 
the like, that you'd expect the Federal Government to go to 
that system.
    Mr. Tobias.
    Mr. Tobias. I think that you're using the term validated as 
a term of art. And as a term of art, there is no performance 
evaluation system in the Federal Government today that has been 
validated as you're using that term. With respect to the issue 
of longevity, in many agencies, within grade increases have 
been equated with longevity. That is not the intent of the 
statute. And if the statute were incremented properly, 
performance would be an integral part of whether or not people 
receive within-grade increases or not.
    However, to get back to your original point, there is no 
performance evaluation system in the Federal Government that 
has been validated as you're using that term.
    Ms. Norton. Well, I believe that is the honest answer, and 
I believe there is a reason for it. And if you want to be 
helpful to us, you can help us get to the point where you could 
set up a performance-based pay or any other system--obviously 
you have a system which, confronting the difficulty of finding 
and identifying objective criteria, default on that and go to 
an overall system that is not inherently unfair but is not the 
best system. It says at least if an employee lasts long enough 
and is satisfactory, then that employee ought to receive an 
increment in pay. And essentially what you are confronting and 
what you especially, Mr. Stier, are avoiding, is that it is the 
difficulty, some would say, in a system of 3 million employees, 
perhaps impossibility, of coming up with a system that would 
meet validation standards that sends the Federal Government to 
broader standards, to avoid just the kind of litigation that we 
have already seen at the SEC, the same kind of results we've 
seen at the GAO.
    So that if there are short cuts to what the Supreme Court 
has made the public and private sector do in order to implement 
systems of pay or any other term and condition of employment 
since the late 1960's, if there are short cuts, it is--that is 
what we're looking for. If there are not short cuts, then we 
need to tell the Federal Government the truth. We need to send 
the Federal Government to AT&T, who got sued in a nationwide 
class action suit, and say how did you, in fact, finally get 
out of the consent decree? And then AT&T will tell the OPM we 
did so first by saying a lot of back pay and second by 
validating everything in this place, and it cost us billions of 
dollars. I don't think you do us any service by saying that the 
Federal Government, which paid the private sector to do that 
and has made other Federal agencies do that, can now implement 
the same kind of system on its own without going through the 
same rigorous process and adding in the additional step that 
will be required because it is a Civil Service system. If you 
mitigate the difficulty, then you invite SEC-type overturns and 
a lot of waste of the taxpayers' money.
    And, Mr. Chairman, I must say that I think, if not before 
the end of this year, but quickly, as soon as we can, we have 
to find a way to roll back the pay-for-performance so that we 
do not subject ourselves to inevitable litigation, lose Federal 
employees at the height of when we're trying to keep them, at 
the height of when we can't compete with the private sector on 
many grounds. This has been a real lose-lose for us. And if you 
want to sit down with any of us who have been in touch with 
this work for a long time, I am open to seeing if there are 
ways--some of the broadbanding ways have some--some suggestions 
within them.
    What I'm not open to is saying to the Federal Government on 
the part of this subcommittee, go ahead, try it out; if you get 
sued, you've got a lot of lawyers, let them take care of it.
    Mr. Stier. Congresswoman Norton, I'd love to take you up on 
the offer.
    Ms. Norton. I'd be glad to meet with you.
    Mr. Stier. Thank you.
    Mr. Davis of Illinois. Thank you very much, Delegate 
Norton.
    Mr. Tobias, you noted that you collaborated with the 
Partnership for Public Service on a study that found that 
Federal employees are not motivated to increase their 
performance solely because of monetary rewards. Federal 
employees are motivated to increase their performance when they 
are effectively led and their skills are matched to the 
agency's mission. Despite having access to the same 
information, you and Mr. Stier have reached two different 
conclusions. You suggest that it is highly unlikely that the 
IRS can successfully implement a pay-for-performance system. 
Mr. Stier suggests moving forward with these systems.
    If Federal agencies have already shown not to seek employee 
buy-in and have difficulty linking agency mission with 
measurable standards, is it fair to continue to allow Federal 
employees to labor under these systems?
    Mr. Tobias. Well, I think the risk, Mr. Chairman, of a 
failed pay-for-performance system is what is shown in the data 
from the IRS, and that is, rather than motivating, morale drops 
and people perform at a lower level than they would with a pay-
for-performance system. I think that is the real risk of a 
malimplemented, maldesigned, maladministered pay-for-
performance system. I think it is a significant risk, and the 
IRS hasn't gotten over the hump based on their own survey data.
    Mr. Davis of Illinois. Both you and Mr. Stier recommend 
closer involvement and oversight of these systems by Congress. 
You note that the Controller General deserves congressional 
committee jurisdiction to hold oversight hearings on each of 
the major agencies once a year. The subcommittee has not taken 
no for an answer, but agencies have been reluctant to testify 
before this subcommittee on these issues and have ignored our 
recommendations and findings. This includes the GAO and one of 
today's witnesses. This leaves the subcommittee with no choice 
but to legislate to effect change which can be long and 
difficult. Given these facts, is it reasonable to expect 
Congress to assist with and examine every agency's pay-for-
performance to ensure that they are fair and equitable, 
credible and transparent and have the support of agency 
employees?
    Mr. Tobias. I think if the--if the data were developed by 
the agency, I think it would be rather easy for Congress to 
determine whether the goals and objectives are being achieved. 
IRS had stated goals. The data showed that the goals weren't 
being achieved. It--it is not a complex matter to measure the 
effectiveness based on what employee surveys say and whether or 
not the organizational goals and objectives are being achieved. 
I think it is more that Congress is interested in performance 
and performance results. And that is really the thrust of my 
testimony. I believe that it takes the interest of Congress, 
the dedicated time of a President and the support of political 
appointees to really have a performance management--an 
effective performance management system. And as I said in my 
testimony, I believe that the implementation of an effective 
performance management system with the corollary of a better 
led work force would have a significant, positive impact on 
performance in the Federal Government.
    Mr. Davis of Illinois. Thank you.
    Dr. Fay, a senior management analyst with DNI wrote in a 
November 12, 2007, commentary for the Federal Times that pay-
for-performance suffers from two false assumptions: First is 
that what is the best for business must be best for government; 
and the second is that pay-for-performance will be effective 
for the entirety of a work force as diverse the Civil Service. 
What are your views on this analyst's assessment?
    Mr. Fay. I would agree that what is good in the private 
sector is not necessarily appropriate for the public sector. I 
think both sectors have an interest in improving performance of 
the work force, and I think they might need to go about it in 
different ways. With respect to your second question, could you 
repeat that, please? The second question.
    Mr. Davis of Illinois. I think it was only one, and that is 
your assessment of the analyst's----
    Mr. Fay. Well, his second point.
    Mr. Davis of Illinois. Let's see. His second point, I 
believe--oh, his second is that pay-for-performance will be 
effective for the entirety of a work force as diverse----
    Mr. Fay. Generally, I think that is probably true. That is, 
the same system would be--the kind of system you use with one 
job family is likely to be different from that used with 
another. And as I recall, the rationale that was developed by 
the agencies and accepted by OPM for each unit having its own 
performance management system, performance appraisal system was 
that there were two diverse set of needs across agencies for 
one system to cover everybody. And my reading of the DOD system 
was that is probably true of DOD as well, having a single 
system with some subsets to cover all the civilian employees of 
DOD strikes me that it is unlikely that it is going to be 
highly workable.
    Again, in the private sector what works best is a system 
that is customized to each employee's job. As I said, the EEO 
rulings of the past suggest that any criterion measure, any 
performance measure that is not based on job-specific 
information is of necessity biased and illegal.
    Mr. Davis of Illinois. Thank you very much.
    And Mr. Philips, let me thank you for joining us. Let me 
ask you, what can the IRS do to improve the effectiveness of 
its pay-for-performance system?
    Mr. Phillips. Thank you, Chairman Davis.
    My name is Mike Phillips, and I'm the deputy inspector 
general for audit for TIGDA. And--well, we've heard some of the 
same themes here today. The IRS took its system of 11 grades 
for frontline managers in 2005 and converted it to an 11 pay 
band system for its frontline managers. The problem was they 
could not reach consensus--senior management could not reach 
consensus in terms of--because of the variety of--of 
positions--of which positions--and how large the band should 
be.
    So what we would recommend is to look to consolidate their 
bands from 11 to a smaller number to move more toward the 
flexibility and the opportunities for managers in those 
different pay bands to diversify their knowledge as well as 
their experience and also present them greater opportunities to 
advance their careers within those individual pay bands.
    We also felt like--feel that those managers who are 
receiving at least a fully satisfactory or a net or higher 
assessment--performance assessment should receive at least the 
salary raise that is equivalent to the across-the-board 
adjustment that is given to all nonmanagerial IRS employees 
under the GS based system.
    And then, finally, for those truly exceptional and 
outstanding managers, we feel like that the IRS needs to work 
with the Office of Personnel Management to look for additional 
flexibilities to provide appropriate salary increases that 
would maybe even take those managers above the pay band that 
they are in to recognize their performance.
    Mr. Davis of Illinois. Well, let me ask you, do you have 
any concerns about the IRS's decision to hire a contractor to 
evaluate the IRS's pay-for-performance system?
    Mr. Phillips. Yes, Chairman Davis. The IRS has worked with 
a third-party contractor to--to do a three-phased assessment of 
its pay-for-performance system over a 5-year period. And we 
really feel that is too long a period for making any 
significant changes to the system. Particularly at this point 
in time, as Mr. George mentioned, the IRS is faced, like all 
other Federal agencies, with a tremendous wave of retirements 
at its executive and managerial ranks over the next few years. 
So we really feel that it needs to shorten the amount of time 
that it is going to take to assess the system.
    Mr. Davis of Illinois. Did your audit examine the impact of 
the IRS's pay-for-performance system on minorities and older 
workers?
    Mr. Phillips. No, sir. We do not look at those aspects.
    Mr. Davis of Illinois. Would you recommend that such a 
study be conducted?
    Mr. Phillips. We are planning, as the IRS continues to 
implement its system, to do further study in that area, and we 
would take that certainly under consideration.
    Mr. Davis of Illinois. Thank you very much.
    And let me thank all of the witnesses for the hearing this 
afternoon. We thank you for your insight as well as for your 
patience and thank you indeed. You're excused.
    And we'll seat our second panel. While our witnesses are 
being seated, I'll go ahead and introduce them. We have Mr. 
Diego Ruiz, who is the Executive Director of the Securities and 
Exchange Commission. For 10 years, Mr. Ruiz served as a 
business executive with Univision Communications, Inc., a 
leading Spanish language media company. In January 2006, Mr. 
Ruiz left Univision to serve at the Federal Communications 
Commission in Washington, DC, as Deputy Chief of the Office of 
Strategic Planning and Policy Analysis.
    Mr. Richard Spires is the Deputy Commissioner for 
Operational Support at the Internal Revenue Service. He is 
responsible for overseeing the development of policy for IRS 
Personnel Services, Technology and Security. Mr. Spires 
previously served as the IRS's Chief Information Officer.
    Dr. Ronald Sanders is Associate Director of the National 
Intelligence for Human Capital.
    And I believe that constitutes our panel. Gentlemen, it is 
our custom that all witnesses be sworn in. So if you will would 
stand and raise your right hands.
    [Witnesses sworn.]
    Mr. Davis of Illinois. The record will show that the 
witnesses answered in the affirmative. And, gentlemen, we're 
delighted that you're here. Of course, the drill is that the 
green light indicates that you have 5 minutes in which to 
summarize your statement, which is already included in the 
record. The yellow light indicates that the time is running 
out. Of course, the red light indicates that it is time to 
stop.
    So we thank you very much, and we'll begin with Mr. Ruiz.

 STATEMENTS OF DIEGO RUIZ, EXECUTIVE DIRECTOR, SECURITIES AND 
EXCHANGE COMISSION; RICHARD A. SPIRES, DEPUTY COMMISSIONER FOR 
 OPERATIONAL SUPPORT, INTERNAL REVENUE SERVICE; AND RONALD P. 
SANDERS, CHIEF HUMAN CAPITAL OFFICER, OFFICE OF THE DIRECTOR OF 
                     NATIONAL INTELLIGENCE

                    STATEMENT OF DIEGO RUIZ

    Mr. Ruiz. Thank you, Mr. Chairman. I appreciate the 
opportunity to appear before you today to discuss the SEC's 
pay-for-performance system. The SEC is phasing in a new pay-
for-performance system that was developed in cooperation with 
the National Treasury Employees Union, which we believe will 
provide meaningful rewards to employees for strong performance 
and will facilitate the accomplishment of the SEC's statutory 
mission to protect investors, promote capital formation and 
foster market efficiency.
    In response to chronic difficulties by the SEC in 
recruiting and retaining mission critical staff, in December 
2001, Congress passed the Pay Parity Act which authorized the 
SEC to increase its pay and benefits to levels comparable to 
those of other financial regulatory agencies while adhering to 
merit system principles. The SEC, with the support of OMB, 
obtained additional funding to implement a new compensation 
system that would improve base pay while increasing 
accountability by linking pay increases to individual 
performance.
    In early 2002, after the development of the new 
compensation and pay-for-performance systems, the SEC and the 
NTEU entered into compensation negotiations but were unable to 
reach agreement over several issues. Subsequently, the matter 
was submitted to Federal Services Impasse Panel, and in 
November 2002, the FSIP issued a decision and order that 
supported the SEC's arguments and the substance of the SEC's 
proposal, including that the SEC's compensation proposal was 
the product of extensive research carefully tailored to meet 
the specific needs of the SEC and the comparability and the 
overall fairness were established. The FSIP order allowed the 
agency to provide an annual merit increase based on individual 
performance. In the same year that the SEC obtained pay parody, 
the SEC's mission was expanded by the enactment of the 
Sarbanes-Oxley Act. The additional responsibilities and the 
requirements of the act demanded a corresponding increase in 
the SEC's mission critical staff.
    The SEC's then relatively new compensation system proved 
its effectiveness by allowing the SEC to increase staffing by a 
third in less than 12 months. In addition, in the time since 
the compensation system was established, attrition rates have 
fallen dramatically to as low as 6 percent, which at that time 
was a 10-year low for the SEC and well below governmentwide 
averages.
    As part of the SEC's continuous re-assessment and 
refinement of the pay-for-performance system, in May 2003, the 
SEC adopted our current two-tiered management system to assess 
individual performance, which can be rated either acceptable or 
unacceptable based on several agency-wide success factors. To 
link performance with compensation, only employees who receive 
an acceptable assessment are eligible for, although not 
guaranteed, a merit increase.
    After the first merit pay cycle in 2003, the NTEU filed a 
grievance against the merit pay process alleging that it 
discriminated against employees in several protected classes. 
Although the arbitrator ruled that the statistical evidence 
concerning the commission's compensation system in 2003 
supported a finding of impact discrimination for two classes of 
employees, the arbitrator's ruling made clear there was no 
evidence of any intentional discrimination on part of the 
commission or any of its employees.
    The administrator did not make a ruling on the appropriate 
remedy and has asked that the parties present briefs on how to 
resolve the issue. The briefs have been submitted to the 
arbitrator, and briefing on all issues should be complete this 
month. We await the arbitrator's decision.
    Even before the arbitrator's ruling, the SEC had identified 
a number of areas where the two-tiered system could be 
improved. Thus, in September 2006, the commission established 
the Performance and Accountability Branch within the Office of 
Human Resources, which is leading the implementation of a new 
five-tiered performance management program for the commission 
that will help establish unambiguous criteria from which fair, 
credible and transparent rating and merit increase decisions 
can be made.
    The new five-tier program is designed specifically for the 
commission's unique business needs. A joint labor management 
team comprised of senior human resources staff and NTEU 
officials is working collaboratively on our transition to this 
program. This transition began in September 2006 with a pilot 
program involving all SEC Office of Human Resources personnel. 
Following significant adjustments based on lessons learned from 
the pilot program and feedback from SEC managers and 
supervisors, the commission's management is now being 
transitioned into the new system. We anticipate transitioning 
the rest of the commission beginning in 2009. The commission is 
purposely taking its time in starting with leadership to allow 
them to become comfortable with a new process before being 
required to manage subordinates on it. Also, the commission has 
decided to temporarily separate our performance management 
system from the merit-based system until the new performance 
management system is completely implemented. This will allow 
the agency to focus all its efforts on effectively implementing 
all aspects of the new system before relying on it to provide 
performance information to support paid decisions. During the 
transition to the five-tier system, all employees receiving an 
acceptable performance rating will receive an equivalent share 
of the funds the agency has available for merit-pay increases. 
The new performance management program was developed based on 
best practices both from other Federal agencies and the private 
sector. Additionally, it responds to several performance-
related recommendations from the GAO and OPM.
    Finally, the new program addresses issues raised by the 
SEC's own Office of the Inspector General. To underscore the 
SEC's resolve to improve continuously in this area, the 
Inspector General has agreed to perform another full audit of 
the performance management program in 2 years. Thank you for 
providing me the opportunity to update you on the SEC's pay-
for-performance system, and I would be happy to address any 
questions you may have.
    [The prepared statement of Mr. Ruiz follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    
    Mr. Davis of Illinois. Thank you very much. Mr. Spires.

                 STATEMENT OF RICHARD A. SPIRES

    Mr. Spires. Thank you, Chairman Davis.
    I'm pleased to be here today to discuss the Internal 
Revenue Service's efforts to implement pay-for-performance and 
respond to questions from the subcommittee. This is an 
important issue as the Federal Government continues to look at 
ways to recruit and retain talented managers. While I've worked 
at the IRS in various capacities since 2004, I've spent more 
than 20 years in private industry where pay-for-performance is 
commonplace and, from the perspective of the companies with 
which I was associated, has had great success.
    I recognize that there is not a perfect correlation between 
government and private enterprise and what works in one may not 
in the other. And in my 4-year tenure at the IRS, I've seen 
some of the reasons why. However, the development of a strong 
pool of talented employees is such a critical issue for any 
enterprise; it is important that innovative programs be 
attempted.
    In many respects, the IRS has been at the forefront of the 
pay-for-performance program in the Federal Government. We've 
been dealing with it for over 7 years as we've implemented such 
a system for our more than 7,000 managers. Though there have 
been some bumps along the way, the creation of pay bands and 
compensating employees for the quality of their work rather 
than their tenure with the agency has helped the IRS respond to 
the challenges presented in turning the agency into a modern 
and more efficient organization.
    My written statement lays out much of the background of how 
we got into pay-for-performance and describes in some detail 
how we implemented the program and discusses some of the 
obstacles we faced.
    I want to focus my remarks this afternoon on two things. 
First, I want to outline the areas in which pay-for-performance 
has benefited our agency. Second, I want to offer some of the 
lessons we've learned so that the agencies that follow us can 
benefit from our experiences and have an easier transition.
    Perhaps the greatest benefit of pay-for-performance for the 
IRS has been the opportunities afforded to us in implementing 
the dramatic overhaul of the agency mandated by the IRS 
Restructuring and Reform Act of 1998. Specifically, the 
implementation of a new performance management system allowed 
us to link manager performance to the functional goals of the 
organization. Managers and their supervisors jointly developed 
specific performance commitments as part of an annual 
performance plan that are designed to further the goals of the 
functional unit and the IRS. The pay flexibilities have enabled 
the IRS to strengthen the linkage between manager performance 
and the overall IRS goals.
    Despite these benefits, the road has not always been smooth 
and without controversy. Let me offer several lessons we've 
learned and, frankly, are still learning that may benefit other 
agencies in the Federal Government. First, agencies should move 
deliberately and cautiously to implement the program that is 
right for their organization, recognizing that any change in 
the way employees are paid will raise concerns on their part. 
Second, communication is critical. Employees must understand 
how the program will work and how they will be affected. There 
also must be forums that have their questions answered. Third, 
an effective performance evaluation system must be in place. 
Employees must understand the basis for their evaluation, and 
there should be a review system in place to make sure 
evaluations are being made on a consistent and fair basis. 
Fourth, supervisors and employees must be trained properly on 
how to use the system and make sound evaluations. Fifth, 
ongoing program evaluation is essential to ensure that the pay-
for-performance system is operating as intended, and agencies 
must be willing to modify and revise to meet the changing needs 
of their organization. And finally, evaluations must be made 
free of any discrimination based on race, gender, age or 
national origin.
    I'm proud to say that an overall evaluation of our program 
by a third party contractor found that, since fiscal year 2004, 
there has been no disparate impact on any group of our 
managers. The contractor analyzed the trends of the ratings 
data grouped by race, gender, age and national origin. In each 
group, ratings trended in a similar path to the average ratings 
across all groups.
    Thank you, again, Mr. Chairman, for the opportunity to be 
here. And I'll be happy to respond to any questions.
    [The prepared statement of Mr. Spires follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    
    Mr. Davis of Illinois. Thank you very much, Dr. Sanders.

                 STATEMENT OF RONALD P. SANDERS

    Mr. Sanders. Thank you, Mr. Chairman.
    It is good to be back. Thank you for the invitation to 
testify at today's hearing. Specifically I understand that the 
subcommittee is interested in learning more about our plans for 
designing, developing and implementing a more modern 
compensation system for civilian employees in the U.S. 
Intelligence Community--I'm sorry. I've got it now.
    Of interest to these proceedings, we do have a few IC 
elements that are covered by Title V of the U.S. Code to 
include analytic functions in the Departments of Energy, State 
and Treasury. Unfortunately, I'm unable to share any details of 
that proposed system with the subcommittee at this time. We're 
still very much in a predecisional phase. In this regard, the 
most salient features and likely those that are of most 
interest to the subcommittee remain under deliberation and 
discussion amongst the six Cabinet Departments and seven 
agencies and elements that comprise the IC, with the various IC 
directives that will enable and establish the system's 
framework still in formal interagency coordination.
    As you may know, Section 308 of H.R. 2082, the Intelligence 
Authorization Act for Fiscal 2008, require the Office of the 
Director of National Intelligence to submit a report of the 
IC's pay modernization initiative. And although that bill is 
not yet law, we fully intend to comply with the congressional 
wishes. We'd be glad to provide your subcommittee a copy of 
that report as soon as it has been submitted to the 
Intelligence Oversight Committees on the Hill.
    What I can discuss with the subcommittee today is a related 
initiative now proved to establish common performance 
management requirements for the IC civilian work force. Those 
requirements are set by an IC directive issued by Director 
McConnell last November, and I'd be pleased to do so. Also I 
can discuss one of our legislative proposals from last year 
that did not make it into H.R. 2082 but may be of some interest 
to the subcommittee given your focus today.
    First, our performance and management directive. The DNI 
believes strongly that a common set of core performance 
management policies are an essential requirement for the IC as 
a way of strengthening the community. In the past, performance 
appraisal systems varied widely across the IC, and they did not 
consistently reinforce the common behaviors and values which 
are critical to the modern Intelligence profession, such as 
collaboration and critical thinking.
    . To remedy this, the DNI's 100-day plan called for the 
completion of a directive of establishing performance 
management requirements for the IC civilian work force. That 
directive does not establish a common system. Instead it 
establishes common core requirements and processes for managing 
the performance of IC employees that are to be incorporated 
into the performance systems established and administered by 
the 16 separate IC elements and their 6 parent Cabinet 
Departments.
    Employees will be evaluated on what they achieve, their 
results, performance objectives developed jointly between 
manager and employee, and in the manner in which those results 
were achieved. How, their manner of performance, with the 
latter focusing on six common performance elements, which by 
the way have been validated. They include such transformational 
competencies such as critical thinking, collaboration, personal 
leadership and integrity and technical expertise. These go to 
the heart of Intelligence reform.
    This performance management directive does not cover our 
senior executive or equivalent senior level positions. They'll 
soon be covered by a similar policy. All departments and 
agencies in the IC are to implement the directive not later 
than October 1, 2008 for application in the 2009 appraisal 
cycle. I would note here that the implementation of the ICD 
does not require any special or unique statutory authority. It 
can be implemented as is by all IC elements, including those 
covered by Title V. I'd also note that the directive is a 
necessary antecedent of a modern performance-based compensation 
system; although it is a separate and critical strategic human 
capital initiative in its own right. For example, the IC deems 
common performance management requirements are essential to the 
success of our Civilian Joint Duty Program. They also provide a 
mechanism to reward and reinforce our core IC values, as well 
as some of the transformational behaviors I mentioned before.
    Let me turn now quickly to a related legislative proposal. 
As part of the administration's fiscal 2008 Intelligence 
Authorization request, we asked Congress to give the DNI the 
ability to extend already authorized personnel flexibilities 
from one IC agency to another; this in order to maintain a 
level playing field across the community. In short, we wanted 
to be able to share the myriad special personnel flexibilities, 
for example, deployment incentives, foreign language incentive 
pay, various scholarship authorities, etc., that have been 
authorized for one or more but not all of our agencies by 
Congress over the years. And in the longer term, we wanted to 
allow our smaller IC elements, those covered by Title V, to be 
able to take advantage of our new pay-for-performance system.
    As you know, there is no direct legal authority for those 
elements for Title V employees to be covered by that new 
system, and we're concerned that, as the rest of the community 
moves to a system over the next few years that is more 
performance-based and market-sensitive, our smaller elements 
may be placed at a competitive disadvantage. To remedy this, we 
propose that the DNI, with the concurrence of the head of the 
department or agency, could authorize adoption of a flexibility 
already granted to another IC element. But those IC elements 
with employees covered by Title V, the director of OPM would 
also have a say. Although this was included in the Senate 
Intelligence Authorization, it was not included in the 
conference report, nor has that conference report become law. 
We're now in the process of developing our legislative 
proposals for the 2009 Intelligence Authorization. Thus I must 
add that, while we again requesting this provision, it has not 
yet cleared the Office of Management and Budget.
    Thank you very much, Mr. Chairman. I'll look forward to 
answering your questions.
    [The prepared statement of Mr. Sanders follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Davis of Illinois. Thank you very much, Dr. Sanders.
    Mr. Ruiz, let me begin with you. In retrospect, do you--do 
you think it was prudent for the SEC to implement a pay-for-
performance system without buy-in from the National Treasury 
Employees Union?
    Mr. Ruiz. Thank you, Mr. Chairman. As I stated in my 
testimony, the circumstances around the time of our setting up 
the pay-for-performance system that we are currently under were 
unique in the life of the agency. We were undergoing a number 
of years of pretty significant attrition, having a hard time 
remaining competitive with other Federal regulators and 
retaining staff. And we received from the Congress additional 
funding to be able to increase pay for our staff to be more 
competitive with other areas of the government. So we were 
operating under the significant constraints and operating under 
a fairly accelerated schedule at the time. We take very 
seriously any concerns that--the concerns that have been raise 
about the 2003 merit pay cycle. And as I noted, we're in the 
process of discussing the remedy phase of that arbitration 
decision with the arbitrator and with NTEU.
    Mr. Davis of Illinois. Let me see if I'm understanding. Are 
you saying that, given a different set of circumstances, where 
some of the pressures that you were experiencing in the agency 
at that time, if those were not present, then you would find 
employee buy-in to be desirable, a desirable part?
    Mr. Ruiz. Absolutely. I think it is. At the end of the day, 
any performance management system is meant to improve the 
quality of the work force and the caliber of the work being 
performed by the agency, and I don't think that is doable in 
any meaningful way unless the entire team, both management and 
employees, has clear objectives; there is a commonality in 
terms of the goals for the agency. And so I think it is a very 
important component of any successful pay-for-performance 
system that--any performance management system period, that 
there be significant commonality of objectives among those 
staff.
    Mr. Davis of Illinois. While the arbitrator found no 
evidence of intentional discrimination on the part of the SEC, 
hundreds of minorities and older workers were adversely 
impacted by the system. How does the SEC intend to make the 
affected employees whole or to rectify that situation?
    Mr. Ruiz. Let me just say, Mr. Chairman, that the case is 
still going through the arbitration process. And so I have to 
be careful of exactly what I say about it so as not to prejudge 
the manner. But as you've stated, the arbitrator did not find 
any evidence of intentional discrimination, but there--
statistically--a statistical--regarding the disparate impact. 
We've submitted briefs to the arbitrator. He has asked the SEC 
to do so as well, specifically regarding the remedy--and we're 
awaiting for the arbitrator's decision on that.
    Mr. Davis of Illinois. Now, the arbitrator's decision 
applies only to the first merit based cycle in 2003. The 
National Treasury's Employee Union has filed grievances against 
the merit pay system for 2004, 2005, 2006 and etc. Has the SEC 
made any attempt to settle with the union regarding the 
subsequent years this problematic pay-for-performance system 
has been in place?
    Mr. Ruiz. Mr. Chairman, we've been in discussions with the 
union concerning the 2003 case, including discussions and 
that--seeing if there are grounds for settlement on that. In 
the subsequent years, 2004 and beyond, I do not believe we have 
engaged in those discussions. And again, this is simply because 
we're still going through the process of the 2003 merit cycle 
arbitration decision. And I believe it is not yet--the issue is 
not yet right to gather the lessons we can draw from the 2003 
cycle and apply it to those other years. But we certainly, I 
think, have a good working relationship with our union and have 
been engaged in discussions throughout this process and are 
open to further discussions in the future, certainly.
    Mr. Davis of Illinois. I note that instead of denying the 
SEC employees a COLA, that the SEC can deny its employees an 
automatic within-grade increase. Could you explain for the 
committee what an automatic within-grade increase is and 
whether or not an employee who meets expectations can be denied 
such an increase?
    Mr. Ruiz. I am sorry, Mr. Chairman, I didn't hear the 
beginning of your question. I didn't quite make it out. Could 
you state it again?
    Mr. Davis of Illinois. Instead of denying a COLA, that the 
SEC can deny employees an automatic within-grade increase. And 
I am trying to find out what is an automatic within-grade 
increase, and whether or not an employee who meets expectations 
can be denied such an increase.
    Mr. Ruiz. Under our current system, Mr. Chairman, we have 
basically a two-tier performance process. An employee can be 
classified either as acceptable or unacceptable, which, I 
think, was referred to in the earlier panel as essentially 
pass/fail. So employees that are rated at the unacceptable 
level are not eligible for a merit pay increase. They do 
receive the full COLA, as does the rest of the Federal 
Government.
    If an employee is classified at the acceptable level, 
again, under the current system that we are under, they can be 
given a range of steps in merit pay increases each year. And, 
of course, we are an appropriated agency, and so this is 
subject to what our level of funding is for the year.
    So the decision on whether or not an employee receives a 
merit pay increase, again, I am talking about merit pay above 
and beyond the COLA, is dependent on that unacceptable rating.
    I should say, just by way of context, in this past 
performance cycle, I don't believe that we have had any 
unacceptable ratings.
    Mr. Davis of Illinois. But it is considered merit pay, and 
if a determination is made that the employee does not merit 
this, for all practical purposes, bonus, that it can be denied 
them?
    Mr. Ruiz. The merit pay component, yes, that is correct.
    Mr. Davis of Illinois. Thank you very much.
    Mr. Spires, let me ask you, can you explain how budgetary 
constraints prohibit the IRS from establishing a policy 
providing mandatory minimum increase in salary, which is 
equivalent to an across-the-board adjustment?
    Mr. Spires. The statute RRA 98 delegated decisions 
regarding those increases of COLA to--I believe, the Secretary 
of the Treasury would then delegate it to the IRS Commissioner. 
We have made a decision that we would not mandate that, that we 
would live by that statute.
    However, the practical application is that since we have 
implemented a pay-for-performance system in the IRS for 
managers 7 years ago, we have always granted that COLA increase 
for all managers that are fully successfully rated or above.
    Mr. Davis of Illinois. Do you believe that the IRS pay-for-
performance system is actually providing an incentive to 
managers, given that employees who are rated fully satisfactory 
receive both step increases and the across-the-board-
adjustment, while managers who are rated fully satisfactory are 
assured of neither?
    Mr. Spires. I believe, Mr. Chairman, that our pay-for-
performance system, while certainly not perfect, is helping the 
organization meet its overall objectives. We did have some 
bumps in the road, and I allude back to my testimony where I 
talk about the fact, and TIGTA alluded to this, we didn't 
communicate and train the management staff as well as we needed 
to on the system. And I think that's led to some of the kinds 
of employee satisfaction ratings that Dr. Tobias stated on the 
previous panel.
    That being said, time-specific objectives to each manager 
in their performance plan that roll up to that organization's 
overall goals and objectives, that then roll up to the overall 
IRS mission, I think, has really helped us in overall 
performance. If you look at the overall performance of the IRS 
over the last 5 years, in almost every category we measure, we 
have performed markedly better, in taxpayer service, in our 
compliance function.
    In our modernization, compliance, for instance, from 2002 
to 2007, we increased our enforcement revenues more than 70 
percent in an era of essentially flat budgets, and we have 
10,000 employees less than we had then.
    I think that speaks to good management, and I believe that 
a pay-for-performance system--and echoing what Dr. Tobias said, 
a good performance management system are keys to have made that 
happen.
    Mr. Davis of Illinois. Let me ask, has the IRS made a 
decision on whether to implement a pay-for-performance system 
for the rest of the approximately 900,000 employees?
    Mr. Spires. Sir, we have not. We--as I said earlier, it's 
not perfect. We do believe we need more time to get it right 
for the management ranks. We are still assessing whether it 
makes sense to engage with the NTEU and discussions regarding 
doing it for nonmanagerial employers. We have not made any 
decision yet as to whether we are going to move forward.
    Mr. Davis of Illinois. Recognizing that there is some 
decrease in morale among some managers within the agencies, and 
employees that might be considering managerial positions, what 
is the IRS doing to try and beef that up or to have cut it off?
    Mr. Spires. Well, we recognize that there may be a 
perception of a disincentive, if you will, for nonmanagerial 
employees to move into the managerial ranks. We are studying 
that right now, sir. This goes well beyond pay-for-performance 
issues, but given the pay compression that you see in the 
Federal Government pay scale, what can we do to incentivize 
those that really have the capability to manage, to move into 
the management ranks?
    I personally believe, and I think it's bearing out in the 
IRS, that the pay-for-performance system does not need to be a 
disincentive. I think with the kinds of studying we are doing 
regarding ours, the training we are doing, the education, I 
believe we can move past that. But there are other 
disincentives. Particularly in the IRS, we have issues around 
mobility. Many times we ask those going into management ranks 
to actually relocate, given the disparate nature of the IRS and 
our function. There are just issues like that also loom large 
in trying to get employees interested in becoming managers.
    Mr. Davis of Illinois. Can you tell me what the critical 
pay authority is?
    Mr. Spires. Sure. As part of the enactment of RRA 98, the 
IRS was given the authority to hire up to 40 individuals meant 
to come out of the private sector into the executive ranks in 
order to help the IRS bring best practices, appropriate best 
practices, from the private sector into the government, and it 
was more pay flexibility, although none of the critical pay 
executives can earn more than the Vice President of the United 
States.
    Mr. Davis of Illinois. Have you been able to evaluate that 
to determine what it is that the IRS might have been able to 
accomplish with this authority and with these individuals that 
it was not able to accomplish without them?
    Mr. Spires. Well, you are right in my wheelhouse, sir, 
because I am one of those critical pay executives.
    Most of them, about 75 percent, and I believe we have about 
25 on board right now, although I am not sure of the exact 
number, most of them are in our information technology 
organization, our IT organization.
    Most of them were brought in because they brought in 
specific skills where we felt we were lacking in the IRS, in 
information technology, management in particular, and I believe 
that it has been a major benefit to the organization. We have 
great individuals in our IT organization that are career 
people, but there's a new set of technologies, a new set of 
disciplines that they weren't familiar with. By being able to 
bring in people from the private sector with those skills for a 
4-year period, I think, has been a very big help in moving our 
whole modernization program along on the IT side.
    Mr. Davis of Illinois. Thank you very much.
    Dr. Sanders, let me ask you, you testified that you cannot 
provide any details on the DNI's proposed pay-for-performance 
system because it is in a predecisional phase.
    Could you explain what the predecisional phase means?
    Mr. Sanders. Certainly, Mr. Chairman.
    When you began the hearing, you talked about the pay-for-
performance system that the Director of National Intelligence 
wants to impose on the Intelligence Community.
    Mr. Chairman, the DNI does not have the statutory authority 
to do that. Instead, we must achieve agreement amongst the 16 
intelligence agencies and elements and their six parent Cabinet 
departments, who have six different personnel systems in 
different titles of the U.S. Code.
    Again, only a very small minority of our employees are 
covered by Title V. The rest are Title X and Title XV and other 
parts of the law. So given that the DNI cannot impose this 
system on them, we must achieve agreement amongst them.
    Literally 2 years from last week, then principal Director 
of National Intelligence, General Mike Hayden, commissioned a 
feasibility study to look at whether we could and should 
develop a common compensation framework for the Intelligence 
Community. This was one of the critical recommendations made by 
the President's Commission on the Intelligence Capabilities of 
the United States with regard to weapons of mass destruction--
let me take a deep breath--other wise known as the Silverman-
Robb Commission, and they viewed it and we view it as an 
essential way of trying to glue this community together, 6 
departments, 16 agencies, all with different chains of command.
    And one of the things I lose sleep over, Mr. Chairman, is 
different pay and other personnel practices across that 
community. If we have those kinds of disparities, there will be 
no community, and we won't be able to achieve the purposes of 
the Intelligence Reform Act.
    Mr. Davis of Illinois. Now, let me make sure, because I 
understand that this Thursday the Director of the DNI, Mike 
McConnell, is scheduled to testify before an open Senate Select 
Intelligence hearing on DNI's authorities and personnel issues. 
Will Mr. McConnell provide the Select Intelligence Committee 
with details of its pay-for-performance proposal, or do you 
know?
    Mr. Sanders. No, again, Mr. Chairman, he is in no different 
situation than I am. The directives that we are developing over 
those 16 agencies and 6 departments are, in effect, treaties. 
We need everyone's agreement to move forward in a common way. 
We don't have that agreement yet. We have been at this for 2 
years. This is very much an event-driven process, and the first 
and most critical event is to try to reach some consensus 
amongst the intelligence agencies and elements to proceed 
forward in a common way.
    Mr. Davis of Illinois. The administration--the fiscal year 
2008 budget contained funds to increase diversity in the 
Intelligence Community. Each year the DNI reports to the House 
Permanent Select Committee on Intelligence on diversity in six 
of the largest intelligence components.
    In fiscal year 2006, the DNI reported that the IC was 21 
percent minority compared with 37 percent of the overall 
population, and 39 percent women compared with 51 percent of 
the overall population. DNI has acknowledged that the 
Intelligence Community needs to make progress on diversity.
    Given the adverse impact pay-for-performance systems seem 
to have on minorities and older workers, how can you assure the 
subcommittee that the DNI system will be any different?
    Mr. Sanders. Let me respond in a couple of ways. First, we 
are in the process of completing our fiscal 2007 report on EEO 
and diversity in the Intelligence Community, and I am not sure 
it has been submitted to our oversight committees, but, as it 
is, as we do so, you will see that we have made modest 
progress.
    We have made modest progress each year for the last 3 years 
in almost every category; for example, recruiting and promotion 
rates. The percentages exceed our representation rates. So 
while admittedly our representation rate is below, for example, 
the civilian labor force, the fact is that we are hiring at a 
greater rate than our current representation rate for women and 
minorities. So we are making modest progress. That's progress 
we don't want to retrench in that regard. We need a work force 
that looks like America and, frankly, to be able to deal with 
all the diverse peoples of the world.
    It's a very high priority of Director McConnell's. We would 
not be moving forward with pay for performance if we thought 
for one instant that it would have adverse impact. Pay-for-
performance systems are not inherently discriminatory.
    Frankly, I agree completely with Congresswoman Norton on 
the notion of trying to set up a validated system so that 
behaviors you are rewarding people on are, in fact, resident in 
the work that they do. We have spent a lot of money and a lot 
of time engaging hundreds of our employees, our best subject 
matter experts, to develop the performance elements that I 
talked about earlier, collaboration, critical thinking and the 
like, as well as performance standards, specific behavioral 
definitions of what those look like.
    I would just like to add, my colleague on the previous 
panel who seemed to scoff at the notion of evaluating employees 
on collaboration--Mr. Chairman, you know this--the 9/11 
Commission said one of the causes of that tragedy was that we 
had not collaborated and shared information across the 
Intelligence Community.
    We have 16 agencies and stovepipes and 6 Cabinet 
departments. If we don't put it in our performance evaluation 
system, and we don't validate it, and we don't reinforce those 
kinds of behaviors, we won't improve, and our improvement is a 
matter of national security.
    So we are not going to--we are not going to shortcut this. 
We have spent the money to validate our elements and our 
standards. We have oversight built into our proposed rules. We 
have training built into our proposed rules, not just general 
training for managers and supervisors and how to administer the 
system, but we are going to include specifically training on 
how to guard against implicit bias, subconscious bias.
    So we have standards that are validated. We have oversight 
mechanisms, literally five levels of oversight from the first 
level of supervisor all the way up to a body called the IC 
Human Capital Board that will look for and guard against 
unlawful discrimination. In fact, the charter board 
specifically says, among other things, it will guard against 
unlawful discrimination.
    So we have accountability, we have transparency, we have 
oversight, we have the standards, we have the training to make 
sure that we don't have any adverse impact on women and 
minorities. Our goals there have been too hard won to risk 
them.
    Mr. Davis of Illinois. Thank you.
    The last question I think I have is given the statutory 
authority you are seeking for civilian personnel changes fall 
within Title V of this subcommittee's jurisdiction, when do you 
intend to brief the members of this subcommittee on the pay 
provisions of your proposal, and do you have a timeline for 
each phase of your pay-for-performance proposal?
    Mr. Sanders. I am happy to brief you, your colleagues and 
your staff as soon as we get past the predecisional stage. I 
can tell you, Mr. Chairman, no one in this room more than I do 
wants to get past that predecisional stage. This has been 2 
years in the making. It is event-driven and time-phased. This 
would literally be implemented through fiscal 2012.
    So we want to do this right. It will start with a single 
agency and proceed in gradually expanding phases until the 
entire Intelligence Community is covered.
    Mr. Chairman, we want to make sure you are on board, 
because one of the things I worry about--and we do need your 
committee's indulgence on this--at some point we have to be 
able to extend the system to our Title V agencies. In relative 
terms they are very small, they are vulnerable to interagency 
competition, and we don't want to leave them behind. So we are 
going to have to brief you; we are going to have to get you on 
board; we are going to have to convince you that this pay 
system won't work, that we have learned from lots of other 
mistakes, including many I have made myself, so that we can get 
it right the first time and then, with your permission, be able 
to extend it to our smaller elements so they are literally not 
cherry-picked as they move forward.
    Mr. Davis of Illinois. Well, I have no more questions.
    I want to thank you gentlemen very much for your testimony 
and for your patience. You are excused.
    Thank you very much.
    Mr. Davis of Illinois. While we are seating our third and 
last panel, I will go ahead with the introductions. We are 
pleased to have Ms. Colleen Kelley. She is the president of the 
National Treasury Employees Union [NTEU], the Nation's largest 
independent Federal-sector union, representing employees in 31 
different government agencies. As the union's top elected 
official, she leads NTEU's efforts to achieve the dignity and 
respect Federal employees deserve.
    Ms. Kelley represents NTEU before Federal agencies and the 
media and testifies before Congress on issues of importance to 
NTEU members and Federal employees.
    Mr. John Gage is the national president of the American 
Federation of Government Employees [AFGE], of the AFL-CIO. Mr. 
Gage watches over the rights of some 600,000 Federal and D.C. 
Government employees. Mr. Gage was elected national president 
at AFGE's 2003 national convention in Las Vegas, Nevada.
    Mr. Gregory Junemann was unanimously selected to serve as 
president of the International Federation of Professional and 
Technical Engineers [IFPTE], AFL-CIO, at the union's 54th 
convention in March 2001.
    On Tuesday, May 8th, the IFPTE filed a petition to hold an 
election at the Government Accountability Office. On September 
19, 2007, GAO analysts voted overwhelmingly to join the 
International Federation of Professional and Technical 
Engineers [IFPTE]. The vote was 897 to 485, a 2 to 1 margin in 
favor of IFPTE representation for 1,800-plus analysts at the 
Government Accountability Office.
    Ms. Carol Bonosaro is president of the Senior Executives 
Association [SEA], the professional association representing 
the top career executives in the Federal Government. Ms. 
Bonosaro was, herself, a senior executive until her retirement 
from Federal service in 1986 to become SEA's full-time 
president.
    Let me thank all of you for being here, for your patience, 
your indulgence, and, of course, as is tradition with this 
committee, all witnesses are asked to be sworn in. So if you 
would stand and raise your right hands.
    [Witnesses sworn.]
    Mr. Davis of Illinois. The record will show that the 
witnesses answered in the affirmative.
    Again, thank you all very much for being here with us, and 
we will begin with you, Ms. Kelley.

  STATEMENTS OF COLLEEN KELLEY, PRESIDENT, NATIONAL TREASURY 
 EMPLOYEES UNION; JOHN GAGE, PRESIDENT, AMERICAN FEDERATION OF 
      GOVERNMENT EMPLOYEES; GREGORY JUNEMANN, PRESIDENT, 
    INTERNATIONAL FEDERATION OF PROFESSIONAL AND TECHNICAL 
  ENGINEERS; AND CAROL BONOSARO, PRESIDENT, SENIOR EXECUTIVES 
                          ASSOCIATION

                  STATEMENT OF COLLEEN KELLEY

    Ms. Kelley. Thank you very much, Chairman Davis. I 
appreciate the invitation and the opportunity to speak with you 
today. This distinguished subcommittee has heard testimony 
today from Federal agency heads and renowned experts from 
academia. I am here today to present the viewpoint of the tens 
of thousands of dedicated public servants who are currently on 
the ground working in the government. These Federal employees 
are the ones for whom alternative pay systems are a looming 
reality, not just an abstract concept.
    The President's fiscal year 2009 budget submission 
reaffirmed a commitment to replace the current GS system with a 
``modern classification, pay, and performance management system 
that is both results-driven and market-based.''
    OPM just released its December 2007 report on performance-
based pay systems in Federal agencies touting their success in 
leading to a better government. I am here to refute that.
    No. 1, the notion that the GS system needs to be replaced 
is not true; and, two, that current pay-for-performance systems 
have shown widespread success. To the contrary, NTEU's 
experience shows that the alternative pay systems at many of 
the agencies in the OPM report are characterized by a slew of 
grievances, arbitrations, litigation, high attrition rates and 
rock-bottom employee morale.
    Nothing in this OPM report or any other government study I 
have found presents data documenting the need to eliminate the 
GS system. The GS system is market-based. It has the goal of 
achieving comparability with the private sector through 32 
different locality pay areas, and employees receive raises 
based on merit, which is synonymous with performance and 
achieving results.
    The GS system is transparent. It has rules, standards and 
evaluations which must be written. If managers currently have 
trouble with the GS system it does not make sense to go to a 
more subjective pay system.
    The Transportation Security Administration's past pay 
system is a prime example of failure. Employees are constantly 
tested, but if they fail, they are not told what they did 
wrong. The training is minimal, and a majority of airport 
screeners don't know what is expected to get a pay raise. The 
uncertainty impasse has resulted in the highest attrition rate 
in the government. The TSA past system should be eliminated, 
and legislation should be enacted to put TSA screeners onto the 
GS pay system.
    NTEU strongly believes that in the absence of a statutorily 
defined pay system like the GS system, pay should be subjective 
to collective bargaining as it is in the private sector. At the 
SEC and FDIC, NTEU bargains for pay on behalf of its employees, 
yet problems still exist.
    These alternative pay systems must be seen as fair, as 
credible and as transparent. Employees must know what their 
work expectations are and what they need to do to improve. 
Unfortunately, the SEC has failed on all accounts.
    In September 2007, as you know, NTEU won an important legal 
battle when an arbitrator ruled for the union and for employees 
that SEC's implementation of its 2003 pay-for-performance 
system was illegal. This faulty system was found to be 
discriminatory against African American employees above grade 8 
and employees aged 40 and older.
    The SEC system used a set of vague and subjective what they 
call agency success factors to determine whether and how much 
of a merit increase an employee would receive. The generic 
factors were not linked to employees' job duties, and they 
applied to every position within the SEC. They were based on 
amorphous criteria.
    NTEU warned the SEC that employees would not know how to 
satisfy the vague standards, and that arbitrary treatment would 
occur. This was compounded by a lack of training and guidance 
for managers.
    As you noted, four additional grievances are pending, yet 
the SEC continues to determine pay increases based on the 
flawed system currently in place.
    FDIC's pay system, too, was problematic. The system had 
established a separate set of what they call corporate 
contributions. These factors were used to determine an employee 
pay raise increase, which generated a great deal of resentment 
and did little to motivate employees to foster teamwork. So to 
its credit, however, working with NTEU, the FDIC has agreed to 
suspend its pay-for-performance system, hiring employees for 
the 2007 performance cycle. However, IRS managers have not 
embraced their pay-banding system. In their public comments on 
OPM's regs, the Federal Managers Association spoke against 
these forced pay quotas.
    In conclusion, I would like to emphasize a few things. 
There is no hard evidence that the current pay system for 
Federal employees needs to be changed. The current experiments 
with alternative pay systems are failing, and, most 
importantly, pay systems must be fair, credible and transparent 
to be successful.
    Thank you for the opportunity again, Mr. Chairman, and I 
would be happy to answer any questions.
    Mr. Davis of Illinois. Thank you very much.
    [The prepared statement of Ms. Kelley follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Davis of Illinois. Mr. Gage.

                     STATEMENT OF JOHN GAGE

    Mr. Gage. Thank you, Mr. Chairman. It's a pleasure to 
appear before you today.
    I will focus my remarks on DOD's national security 
personnel system and TSA's pay system for transportation 
security officers. Recent press reports of the 2008 NSPS payout 
for employees have been highly misleading, and we urge this 
subcommittee to demand data from DOD to explain how the system 
has been applied, because what they have provided so far makes 
any evaluation impossible.
    But what we do know is subjectivity and bias plague the 
entire system. NSPS is supposed to set base pay and future 
salary increases on individual performance, judged by a 
hierarchy of supervisors, but senior managers at DOD have told 
us that different pay pools have different rules for 
distribution, and that supervisors have an order to tell their 
subordinates only their narrative ratings, not their numeric 
ratings, so that the latter can be changed after the fact. 
Surely it violates the principle of transparency if an employee 
cannot even see their supervisor's rating.
    Because NSPS is not supposed to exceed the cost of the GS 
system, it must fit performance ratings into a normal 
distribution or bell curve. In practice, this means that 
numerical ratings can be changed not because of failure of 
employees to reach performance objectives, but to align with 
preset ceilings on the number of 5, 4, 3s necessary to match 
preordained funding for pay pool distributions. Far too many 
managers have carefully rated their subordinates as objectively 
as possible, only to be forced by the pay pool manager to 
reduce their ratings in order to get to the bell curve.
    But it becomes more complicated than that. Employees in the 
same pool who were rated 3, for example, might not get the same 
number of shares. Further, the money put into the shares varies 
enormously. In some places, a share was worth 1 percent of 
salary; in others, it was worth 1.5; in others, it was worth 2. 
In some workplaces an employee who got a 3 could get more than 
someone else who got a 4.
    Just to make matters even more complex, pay pool managers 
can decide how much of a share can be put into an employee's 
salary increase versus cash bonus, and this is enormously 
important. Obviously, this can affect the employee's standard 
of living not only while he is still working, but profoundly 
into retirement.
    In addition, there appears to be a bias in favor of 
employees who work higher up the chain of command or closer to 
the Pentagon. I would think Members of Congress would be 
particularly interested in what is so obviously an inside-the-
Beltway bias.
    We are also concern about the elimination of merit 
promotion. Under the GS system, an employee's ability to get 
promoted is clear from the position description. If the job is 
based on the career ladder, say, and starts as a GS-5, goes to 
a 7 and then a 9, employees know what is expected of them, and 
they can look forward to those promotions if their performance 
so warrants.
    Under NSPS, promotions are likely to be far rarer. 
Employees might be given additional duties by their supervisor 
in order to advance inside the pay band, but there will be no 
pathway, no clear pathway, to that advancement. Nor is there a 
requirement that a job at the higher level be open to 
competition. Bias and favoritism are inevitable.
    Mr. Chairman, I have dwelled on the details of NSPS for 
good reason. It has been so easy for the administration to spin 
pay for performance as a great idea. Who can argue with 
rewarding excellence and punishing the lazy and incompetent? 
AFGE argued against the establishment of the authorities that 
gave DOD, DHS and TSA the ability to create new pay systems. We 
have testified numerous times before, and we reaffirmed today 
that no matter how successful people may think pay for 
performance is in private-sector settings, it is inappropriate 
for the government where merit system principles must be 
upheld, where teamwork is paramount, where politics always 
threatens to corrupt the workplace, and where profit has no 
place.
    But the administration argued that they can be trusted to 
create fair systems, that best practices would be followed, and 
that we should just wait and see how happy and motivated 
everyone will be once they get rewarded for their 
contributions.
    But the reality of pay for performance is different. In any 
pay system like NSPS, with this much flexibility, the results 
must be exposed to a great deal of sunshine. The fairness, 
transparency and accountability promised by NSPS can only be 
evaluated looking at the numbers.
    We urge the committee to acquire the data listed in the 
appendix that we are provided and to make it public. If the 
system is any good, it will withstand the scrutiny. If it is 
not, AFGE and other unions will negotiate for its improvement.
    Now, transportation security officers. Just briefly, 
despite the public's call for a Federalized, well-trained and 
well-compensated screener work force following 9/11, TSOs 
continue to be drastically underpaid. The average salary is 
$30,000, approximately equal to that of a GS-5, while other law 
enforcement officers at DHS are classified at much higher 
grades.
    TSOs are also subject to the unaccountable and highly 
subjective performance-based pay system known as PASS. While it 
is virtually impossible for us to obtain data or even basic 
information about how this system is supposed to work, to make 
matters worse, TSA continually changes the system. While 
understandably confused about the details, employees tell us 
that PASS is based on favoritism, not performance.
    Last December, TSA disclosed that TSO officers--that TSO 
officers would receive a smaller pay raise in 2008 than in 
2007, even if they receive the same performance rating as the 
previous year. TSA consistently ranks at the bottom of any 
survey of employee morale.
    We all know it from the tragedy suffered on 9/11 that this 
work force is too important to be treated so callously. It is 
time to provide a rational pay system for these workers before 
the attrition rate climbs any higher. AFGE urges the 
subcommittee to end the PASS system and instead place TSOs 
under the pay system that it applies to other Federal workers, 
including their colleagues throughout the DHS.
    That concludes my statement, and I will be happy to address 
any questions.
    Mr. Davis of Illinois. Thank you very much, Mr. Gage.
    [The prepared statement of Mr. Gage follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Davis of Illinois. Mr. Junemann.

                 STATEMENT OF GREGORY JUNEMANN

    Mr. Junemann. Thank you, Mr. Chairman, and thank you and 
Congressman Sarbanes for allowing me here to testify today.
    IFPTE is a labor organization representing over 80,000 
highly skilled professional and technical workers in the 
private, Federal and public sectors throughout the United 
States and Canada. IFPTE represents up to 20,000 workers of the 
Department of Defense, civil servants and 1,800-plus analysts 
within the Government Accountability Office.
    IFPTE also represents Federal employees at the Department 
of Interior, Department of Energy, NOAA, NASA, EPA, CRS, the 
Executive Office of Immigration Review, and the administrative 
law judges within the Social Security Administration.
    Having said that, IFPTE has significant experience with pay 
for performance, particularly at DOD and GAO, where our members 
are either currently experiencing this firsthand or have been 
threatened by its implementation over the past years.
    IFPTE is somewhat unique when it comes to this 
controversial issue. For one, in the private sector we have 
represented members who for years have been on a pay-for-
performance system. While most of our Federal locals have 
clearly said no to pay for performance, some of our members do 
welcome the concept and believe that if properly and fairly 
implemented, it may be able to work. This is the case among our 
GAO membership.
    One thing is clear, however. The proper application of pay 
for performance in the Federal Government has yet to be 
realized, and one could argue that its success is many years 
away, if it is possible at all.
    It's been the IFPTE's experience that pay-for-performance 
schemes, whether they are in the short demo projects or largest 
agencywide efforts, such as DHS or DOD, for the most part have 
not worked in the Federal Government. Despite what some 
supporters would contend, these pay systems have provided 
little evidence that productivity of the workers is enhanced.
    In my written testimony, I also argue that the recent 
report of a 7.6 percent pay increase in the first year of 
spiral 1 of DOD is inflated and unsubstantiated. Quite to the 
contrary, in fact, IFPTE has seen negative impacts on the 
concept of teamwork, morale, potential problems related to 
safety in such places as shipyards, and discrimination toward 
women and people of color in many instances.
    One of the more troubling trends with pay for performance 
recently has been the misguided belief by management that it is 
somehow acceptable to pick and choose who receives 
congressionally mandated yearly pay increases intended to all 
Federal workers who meet satisfactory ratings. Since when was 
it the prerogative of management to unilaterally decide to 
circumvent Congress' intent on annual pay increases through 
pay-for-performance systems?
    IFPTE believes that regardless of what pay system an 
employee is under, when it comes to the yearly pay increases 
approved by Congress for Federal employees, there should be no 
losers and winners.
    We at IFPTE are well aware of this ``race to the bottom'' 
practice, both at GAO and at DOD through NSPS. In order to even 
start thinking about an effective pay-for-performance system, I 
think we can all agree that it would have to have a strong 
employee buy-in; otherwise, morale, recruitment and retention 
will suffer, and these are things the Federal Government can 
ill afford.
    As a foundation for any pay-for-performance system, IFPTE 
would argue that, at the very least, pay parity among all 
Federal employees who have satisfactorily or meet the 
expectation ratings should be achieved. This can be done by 
mandating the annual congressionally approved pay raises 
guaranteed to all Federal workers, as is the intent of 
Congress. Given management's prerogative to ignore Congress' 
intent on the pay raise, IFPTE believes that guaranteed pay 
parity between agencies and various pay systems can only be 
achieved through legislation. Obviously, this is legislation 
that the union would ask your subcommittee to champion through 
Congress.
    In my written testimony, I elaborate specifically on some 
of the problems within DOD and GAO relative to the pay-for-
performance system, and I welcome the opportunity to discuss 
these with you.
    Again, I thank the subcommittee for taking on this very 
important project and issue, and I look forward to any 
questions you might have.
    Mr. Davis of Illinois. Thank you very much.
    [The prepared statement of Mr. Junemann follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Davis of Illinois. Ms. Bonosaro.

                  STATEMENT OF CAROL BONOSARO

    Ms. Bonosaro. The Senior Executives Association very much 
appreciates the opportunity to share our experiences and views 
related to the current SES pay-for-performance system. With the 
creation of the SES in 1978, its members were subject to what 
we thought was, indeed, a pay-for-performance system requiring 
performance standards, detailed appraisals, and providing 
performance awards when warranted.
    The SES pay system was changed in 2003, and for the past 
three annual performance cycles, senior executives have been 
under a new system, the principal features of which are the 
absence of locality pay, absolute and unreviewable agency 
discretion in determining annual salary adjustments, and the 
availability of higher pay caps to pay higher salaries to 
senior executives if OPM certifies an agency's SES pay and 
performance management system is making meaningful distinctions 
based on relative performance.
    Our opinions about the SES pay and performance system are 
formed with frequent interaction with our members and by a 
comprehensive survey of the SES that SEA undertook just over a 
year a half ago. Both the survey results and the continued 
feedback from our membership show that changes to the pay-for-
performance system need to be made. Our goal in recommending 
changes to that system is to develop one which is viewed fair 
and reasonable by both those subjected to it and those who 
might aspire to the SES.
    One of the more disturbing findings of SEA's survey was the 
opinion of 47 percent of the respondents that GS-14s and 15s 
are losing interest in applying to the SES. Anecdotal evidence 
we have continued to receive indicates that the narrowing gap 
between SES pay and the General Schedule, coupled with the 
uncertainty and inconsistency of the SES system, the loss of 
locality pay, increased responsibilities and fewer rights, 
results in a less attractive Senior Executive Service.
    A significant finding of our survey, which persists to this 
day, is the perception that agency quotas, not actual 
performance levels, sometimes drive decisions about performance 
ratings and salary adjustments. Quotas are, of course, illegal 
according to the regulation, but de facto or notional quotas 
seem to flourish. Most agencies perceive that a key factor in 
receiving OPM certification is to reduce the number of 
outstanding ratings, so downward pressures on rating levels 
exist within many agencies.
    The certification process itself is a problem. It's an 
arcane, time-consuming and cumbersome process which seems to 
change each year. It must also be done every 1 or 2 years, and 
often the decision whether to certify does not come until well 
into the performance cycle.
    Another concern is inconsistency among agencies and 
sometimes from year to year within an agency on how the 
performance systems are implemented. This inconsistency makes 
it difficult to assess which agencies are doing the best job of 
rewarding good performance.
    In the 3 years of experience under the new SES performance 
system, one of the most striking results is the very low salary 
adjustments that have occurred. In the most recent year for 
which data is available, 2006, those senior executives rated 
fully successful received an average salary increase of 2 
percent, far below the increase received by the General 
Schedule. Higher-rated executives fared only slightly better 
with those whose performance exceeded the successful--fully 
successful level receiving a 3 percent salary adjustment, and 
those whose performance was outstanding receiving 3.7 percent.
    For the rating cycle that just ended, we have learned of a 
fully successful senior executive who received an annual salary 
increase of $323, less than one-third of 1 percent of the 
minimum SES salary. We have also learned of an outstanding 
senior executive who was denied any salary increase or any 
performance award.
    Attached to my written testimony is a copy of SEA's 
recommendations for legislative changes to fix the SES system. 
Our two principal provisions are to assure a minimum annual 
increase for those SES'ers at the fully successful or higher 
level, and to include performance rewards and retention 
allowances in the high three retirement calculation.
    We also have recommendations concerning a longer 
certification period, minimum funding of SES salary 
adjustments, a minimum increase in pay for new senior 
executives, rules for pay tiers which a number of agencies have 
developed and are developing, feedback to senior executives, 
and greater transparency in the administration of these 
systems.
    It is our hope that with the adoption of SEA's 
recommendations, the SES pay system will be one that adequately 
and fairly compensates those who perform the most challenging 
and important jobs in the career Civil Service, and a very 
important one which will attract quality candidates for the 
future.
    Thank you.
    Mr. Davis of Illinois. Let me thank you very much.
    [The prepared statement of Ms. Bonosaro follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Davis of Illinois. Let me thank all of you for your 
testimony.
    Let me ask one boilerplate kind of question, and perhaps I 
will begin with you, Ms. Kelley. What do you think is driving 
the movement toward pay for performance in the public sector?
    Ms. Kelley. I think that agencies are looking for ways to 
be able to hand managers the unilateral authority to determine 
pay raises without having a lot of accountability required or 
attached to that.
    You know, the complaints that we hear about the GS system 
really are not based on fact at all. If a--you know, I hear 
things from agencies such as pay raises are automatic, and they 
reward longevity, and that is not how the GS system is built. 
If that is how managers are implementing it, the problem is 
with the implementation by the managers, the problem is not 
with the system. So rather than dealing with training managers 
and holding them accountable for giving appropriate ratings 
within the GS system, they try to find fault with the system.
    And also, we are looking for ways, I believe, to move the 
limited funds--because agencies have been faced with, in many 
cases, flat-line budgets--and they are looking for ways to take 
the money from the front-line employees and use it to recognize 
or reward the few that they see fit, or to use it with other 
things within the agency.
    I just think they are not using the GS system 
appropriately, and I think that they are really not interested 
in recognizing and rewarding the front-line Federal employees. 
Every Federal employee I talk to, they tell me they would like 
to know what it is they need to do not just to succeed, but to 
excel at what they do. They want to be told what that is so 
that they can strive to achieve it. At the end of that, when 
they do succeed, they want the appropriate recommendation and 
reward attached to it, and the interested agencies are not 
willing to do that.
    Mr. Davis of Illinois. Mr. Gage.
    Mr. Gage. I am probably of--well, to state it frankly, I 
think we just have to look at where this stuff came from. It 
came from people who have almost a hatred for government 
service, who want to shrink government service; in fact, who 
want to drown it in a bathtub, as one said.
    I think, very clearly, that this whole pay for performance 
is designed for one reason, and that is to lower overall 
Federal pay. I think, too, that when you see some of the 
schemes, if we could just get the numbers on what DOD or any of 
these agencies are doing, I think it will really show exactly 
what this is all about.
    But the one thing I would like to bring up to you, Mr. 
Chairman, about it, it's--it's so wrong where you give people a 
bonus that would probably have added up to the pay raise they 
got under the GS, but that money does not count for their 
pensions. That's really disingenuous, and I think that's--that 
in itself, these agencies these agencies won't be able to stop 
themselves from saving money by not being able to put it into 
the Federal pension system.
    Mr. Davis of Illinois. All right. Mr. Junemann.
    Mr. Junemann. Yes, I guess I could certainly echo what's 
been said here by my two colleagues on this. When I look at 
what happened with the NSPS, really shortly after the Bush 
administration came on, some people came over from Heritage 
Foundation and brought this idea of scuttling the GS system. 
And I had no idea it was since 1949 until I started reading 
these papers from the Heritage Foundation that said, OK, now, 
it's old, it's Truman-era legislation, and now it has to be 
replaced.
    A lot of what President Kelley talked about was really 
true; that it is--in essence, there is the ability to measure 
and to rate employees within the GS system.
    When we were in what we used to call ``no-gotiations'' when 
we were developing the NSPS meet-and-confer process, that was 
told to us: Well, it's an automatic increase. Well, it's really 
not, and if it's implemented that way because the supervisors 
weren't doing it properly or improperly trained, then that's 
what should have been fixed.
    But I think it was really--it was sort of an idea that 
people had been sitting on for years and years, and then they 
finally saw they had the power to implement it, and they did. I 
think it was nothing more than wanting to, you know, lower the 
pay of Federal workers by people who have some sort of animus 
against Federal workers, and then they put it in NSPS, they 
wrapped it into that, and then said, in addition to that, let's 
put national security on the table with it and see if we can 
get rid of collective bargaining at the same time.
    So I don't think there was a heck of a lot good about this. 
Nobody can tell me that everybody is a winner on these systems, 
because that's not what we have seen. There's been just 
thousands and thousands of losers. That has been our 
experience.
    Mr. Davis of Illinois. Ms. Bonosaro.
    Ms. Bonosaro. Well, I am going to speak from my parochial 
perspective of how this came to be in the Senior Executive 
Service. I think, initially, I thought this is the management 
theory du jour. We like to look at what the private sector is 
doing, say to them, this would be a good idea.
    Actually, I think virtually every administration comes in 
and very often looks for a way to manage the Senior Executive 
Service, because people who come in from the outside are used 
to, perhaps, very often having the ability to pick their own 
team, and they find a lot of career executives in place that 
are pretty much going to be there, at least during their, 
``get-acquainted period.''
    So we went through something called recertification a 
number of years ago that was proposed in the first Bush 
administration. It has since been thrown overboard. But that 
was looked at a way--it was looked at a way to get a handle on 
the Senior Executive Service.
    I think to some degree this system is looked at that way, 
that you can--yes, you are assessing performance, but you are 
going to be able to send some messages. And with pay adjustment 
decisions, it is a system which can be manipulated. But I look 
back, too, to initially the reason that this system was able to 
get put in place. We had 70 percent of all senior executives 
capped and earning exactly the same salary.
    The day that I met with then-OMB Director Mitch Daniels and 
OPM Director Kay Coles James, and they said, we are going to 
solve this problem of the pay compression, but not everyone who 
is capped is going to get a pay raise. And I walked out 
scratching my head for a few days, because I couldn't think out 
of the box.
    And, obviously, what they had come up with was a way to 
eliminate the S ranks and come up with a pay-for-performance 
system, because there was also a sense that a lot of these 
executives somehow or other were vastly overrated. There could 
not possibly be this many people who were performing that well. 
I think that was a very unfortunate premise, by the way, and a 
wrong one.
    Mr. Davis of Illinois. Thank you very much.
    I see that we've been joined by Mr. Sarbanes. Is the 
election over? I thought there was--well, some people don't 
have to worry.
    Mr. Sarbanes. We don't take anything for granted, but I 
voted this morning. I did vote for myself. So hopefully that--
thank you, Mr. Chairman. And thank you to the panel.
    Just a couple of comments, and then one or two questions. 
Listening to the testimony today, I was able to hear some of it 
on the radio as well before I got here. You would imagine that 
to put a system like pay for performance into place, you would 
need to do it under the best of circumstances to make it work, 
and it seems that this has been done really under the worst of 
circumstances. I'll leave aside for a moment whether I think it 
can work anyhow. But everyone--if you assumed under the best of 
circumstances you could implement it, you had the worst 
circumstances here. You had inadequate funding, so you get the 
result that Mr. Gage referred to where you have certain limited 
amount of resources for each performance pool, and then you get 
into this sort of allocation system which ends up resulting in 
very arbitrary systems, highly subjective decisions that are 
designed to just meet an overall agenda that is in place.
    You have the fact that this is being done at a time when 
we're still far away from comparability between what Federal 
employees are making and what their counterparts are making in 
the private sector, and obviously that is something you need to 
move to, it seems to me, first before you start tinkering 
around with these other things.
    And then you have the other fact that those implementing 
the system betrayed at every turn a fundamental lack of 
respect, and a number of you have alluded to it, for the 
Federal worker. And what they're trying to do--and the fact 
that they're--they're not just bureaucrats as they'd have you 
believe, they are people that are really committed to what 
they're doing. They have a sense of mission, and they want to 
perform well. They want to perform well, If you give them the 
tools and the resources and the support that they deserve and 
the leadership that they deserve.
    And so this effort has been undertaken with a sort of 
thinly disguised agenda, it seems to me, to attack the 
integrity and the function of the Federal worker, and that is, 
I think, why there is so much resentment to it. It also comes 
at a time sadly when we're trying to recruit into the ranks of 
our agencies and the Federal Government the best and the 
brightest to serve there and to address some of the attrition 
that has occurred, some just because of the passage of time, 
but some as a result of this conduct on the part of the 
administration which has undermined morale.
    There is hundreds of thousands of jobs that need to be 
filled in the coming years, and we need to make sure that the 
Federal agencies are modeling absolutely the best behavior and 
the best kind of leadership when it comes to rewarding people 
for what they do and showing them basic respect.
    The chairman did ask a question that I was going to ask, 
which was sort of to describe your views of what--what the 
agenda was here. I think that the administration--well, let me 
ask you this: Do you think, for example, with respect to 
collective bargaining rights, that the NSPS was launched and 
pursued with the purpose of attacking collective bargaining 
rights, among other goals that it had, or was it launched with 
maybe more responsible objectives and then got hijacked by a 
very strong ideology within this administration along the way? 
I mean, I would be interested in your comments on that.
    Mr. Gage. That is interesting.
    By the way, I voted not far from where your brother lives 
up in Baltimore today, too.
    But I think that--you know, I almost don't want to look 
back over the last 4 years, the fight we had about our 
collective bargaining rights which we saw were--it was just so 
wrong to take away our rights that used any type of a national 
security reason for it, and--but I always thought the pay for 
performance was really what they wanted. And to put in pay for 
performance, at least the pay for performance that I think they 
really want, it would be a lot easier to do without a union 
there and without any voice of employees to be able to bargain 
any fairness or transparency into this system. So maybe that is 
just a conspiracy part of me, but through this whole exercise, 
I always felt that the pay for performance was what they really 
wanted, and to get rid of the unions was a step there.
    Mr. Sarbanes. Yeah. Let me--can I just ask a quick 
question, Mr. Chairman? I know my time has run out. But to look 
positively toward the future in terms of fixing--I mean, you've 
all commented on the fact that the General Schedule can work 
well, that the systems that were in place can work well if they 
got the right kind of resources and leadership behind them. 
Maybe speak for 30 seconds if you--anyone who wants to jump 
in--on your hope for how that can happen moving forward?
    Ms. Kelley. Well, I would hope that there is a recognition 
of the front-line employees that are there, and that they 
should be supported in the work that they're doing. You know, 
the overall question of pay for performance, at first blush it 
is a pretty good sound bite, and many Federal employees at 
first blush say, I'd love to be under a pay-for-performance 
system; maybe then I would be paid appropriately, equitable to 
the private sector. Then when they recognize pay for 
performance under this administration is about no more funding 
for the agency, so it goes to the name of the hearing today, 
robbing Mary to pay Peter and Paul, and also to the fact that 
managers will be--have this authority to decide who gets how 
much money without any accountability, then they recognize that 
it is a good sound bite, and that is all that is.
    Mr. Gage. I don't think this system will work. I don't 
think it is salvageable. I do think that many of the ideas that 
our union has put forth, that Colleen has put forth about the 
existing GS, we can do a lot on that classification system. And 
it was really ironic to hear OMB and OPM say it is an outdated 
system, and it has to be scrapped, where we've been arguing for 
years that it needed to be adjusted, and it needed to be fixed. 
And to see them say, no, it so out--and they were the ones that 
were blocking trying to fix it in many areas.
    But I'm very optimistic that I think that we can--now that 
this issue is so much on the table, that the ideas that we have 
for improving on the Civil Service can really maybe come 
forward again, and that we can put out some really positive 
ideas on how we can improve the Civil Service, protect the 
Civil Service, and really make our country terrific.
    And one thing you said about it, Congressman, why would 
they do this or start this in DOD and DHS, you know, two of the 
most--well, we've got two wars going on, and our DOD employees 
are working mandatory overtime right on down the line. And DHS, 
40 agencies coming together, the whole thing is a mishmash. And 
to try to overlay that with a new personnel system, I think, 
was really short-sighted and wrong.
    Mr. Sarbanes. Thank you.
    Mr. Junemann. I think one of the things that you brought up 
before that kind of addresses this really helps. Now, under 
DOD--because thanks to your legislation that the Congress 
pushed through, we actually sort of have a two-tier thing going 
on here. Where there is a union in place, they are still under 
the GS system. Where there is no union, they fall under the 
spiral thing.
    Now, our union is not alone in this. We're involved heavily 
with recruiting, you know, college graduates. And we can really 
use and have used--when my members go out and meet with college 
grads to try to get them to come into the naval shipyards as a 
for instance, they can look at the GS system and say, OK, if 
you have a 3 point or higher, you start with a GS-7. You're 
going to move to a GS-9 over a certain period of time, and a 
year later you'll go to a GS-11. If you're under a 3 point, you 
start at a GS-5, and they can show them this is how you're 
going to move through the ranks. And if you're able to get into 
a more accelerated program, you can get into a GS-12, and here 
is how long that is going to take. And it works. I mean, it 
really works with people where they can see, OK, you know, I'm 
going to actually have to perform to do this, but there is 
guaranteed raises in there. There is a guarantee that if I do 
something, I can make it to the next level and the next level.
    Under the spiral system, they don't have that. And I think 
what this is going to show, sort of the good news of all of 
this, is, you know, the agencies that are unionized, that are 
under the GS system, are going to find that they're able to 
actually recruit the best and the brightest because they're 
going to be motivated by this.
    And I really need to speak to something I think that Dr. 
Tobias brought up earlier where he talked about the fact that 
money is not a motivator, yet--and we agree with that and--at 
least I do. I agree with that; however, I still say, well, this 
hurts morale. Well, because I think that the employees that I 
represent are motivated by the mission of the agency they work 
for, whether it is GAO, NASA, CRS, the Department of Defense, 
they're motivated by that, that they see that their efforts 
make a difference and that they really--and that they have to 
make a difference, and that they really perform a service for 
the taxpayer. However, they also understand that--when they get 
cheated.
    So it is not like that they are working for the money, but 
when it is not there, you know, when they're not adequately 
compensated, when suddenly they find that, you know--that the 
director of their agency gets COLA and they don't, you know, 
they feel cheated by that, and that is where the morale sets 
in. And that's when maybe they start looking around to other 
agencies or in the private government to maybe find something 
better.
    So you can't have both with this. I mean, they're not 
motivated by the money under a merit system; however, you know, 
the lack of proper recognition in their paycheck also--you 
know, it hurts morale.
    Ms. Bonosaro. Let's see, we need, I think, three things, 
and first is a behavior change, which I would be very surprised 
to see, but be very welcome, which is for those who are running 
these systems in the agencies at least with regard to Senior 
Executive Service to totally divest themselves of the notion 
that there is or should be a normal distribution curve of 
ratings, and to rate people honestly and these executives 
honestly and fairly. I think that would first go a long way to 
starting to fix this system.
    But I do think that we need some structural changes. I 
mean, if you were a fully successful senior executive, and the 
OPM Director says that should be and is a very good rating, 
there is no reason you should be getting an annual pay 
adjustment that is far less than the people that you supervise. 
I mean, a 2 percent adjustment for a fully successful executive 
makes no sense.
    I'm afraid that we're going to need legislation to fix 
those kinds of issues, including the--counting the performance 
of words in the high three and some other issues.
    And there is a third issue. I think it is a long-range one, 
but it is getting shorter- and shorter-range every day, and 
that is how the General Schedule is--is creeping far into the 
SES pay scale, and that is turning off a lot of potential, 
really superb SES candidates. And we're going to have to take a 
look at that SES pay cap and see what we can do with it, 
because right now I think most rational GS-15s would say, why 
do I want the additional responsibility, fewer rights, no 
annual pay adjustment, no locality pay, etc? Yes, I love my 
job, I love my country, I love public service, but--so I think 
that is the third one we're going to have to fix, and it is a 
tough fix, but we're going to have to address it at some point.
    Mr. Sarbanes. Thank you.
    Mr. Davis of Illinois. Thank you very much, and I have a 
couple of last questions.
    Ms. Kelley, you refer to the term ``amorphous criteria'' 
used in the SEC pay system. Could you explain what you mean by 
that term?
    Ms. Kelley. It is a criteria that is unknown to employees, 
it cannot be defined. In the SEC they call it an agency 
contribution factor, and in the FDIC they call it a corporate 
success factor, but no one can tell you what that is. They kind 
of describe it as ``I'll know it when I see it.'' And it has 
zero correlation to your performance appraisal or your 
evaluation based on the requirements and criteria for your job. 
They intentionally delinked the rating to the--what they call a 
merit increase.
    Mr. Davis of Illinois. Given--and perhaps this might be our 
last question. Given all of the problems and the controversy 
that we hear, one side I suppose we get certain kinds of 
arguments and discussions. From other sides, we get a different 
set of discussions and arguments and criteria. And, of course, 
we sit kind of in the seat where ultimately decisions have to 
get made. And if you were in the business of recommending to 
Congress that we do something about this controversy, what 
would your recommendation be? Perhaps we'll begin with you, Ms. 
Bonosaro.
    Ms. Bonosaro. Well, our recommendation is attached to my 
testimony. We have a legislative package that we think will 
help fix this, because we really are concerned an awful lot of 
senior executives are eligible to retire and probably will go 
out. I mean, as we were saying earlier, pay is not the 
motivator, but it sure can be a demotivator when it is handled 
incorrectly. So I'm afraid that for us nothing short of getting 
that bill moving is going to do it, because just some of this 
is not going to happen without that, and they are the 
recommendations I mentioned earlier. So I won't take your time 
up with them again.
    Mr. Davis of Illinois. Well, you know, the blues singers 
say the best things in life are free, but you can give it to 
the birds and bees because I need money.
    Mr. Junemann. And that is pretty much my answer. I think 
what it is going to take to fix this is funding; not only just 
funding for training--and I would say not to fix the pay-for-
performance system, but to fix the GS system. You know, you're 
going to need additional funding for training. If there is 
going to be anything resembling pay-for-performance systems, 
whether they are at GAO or wherever, there has to be--it has to 
be properly funded.
    If we're going to use a given pool--and let me use an 
example. And we tried discussing this with the DOD folks when 
we were going through the negotiations. What we do in the 
private sector is we negotiate a cap of money, and it is 
funded, you know. So we say, OK, everybody is going to get, you 
know, a 2 percent across the board, and then there will be 
merit above that, say, up to 6 percent. That money is funded 
because what happens otherwise is you've got the four of us 
going for three $20 bills, and somebody is going to come up 
short. And that is what happens when you have this small capped 
pool where everybody can't win no matter what their performance 
is. And then what happens is--and we've alluded to it--is team 
work suffers, because if I see that she is not doing her job 
properly, and yet if I help her, if I train her, if I help make 
her look better, it might cost me my pay, you know, that can't 
work.
    And this also even leads to, I think, the issue of 
training--I mean, the issue of safety, you know, within 
nuclear--within nuclear refueling, that if somebody again saw 
that they had made a mistake and they thought, geez, maybe I 
better tell somebody--and I'm not talking anything dramatic or 
drastic, but even if it is something small, you know, that 
they'd go back and tell their supervisor, I think I goofed 
something up here, we better go back and take a hard look at 
it, and they know it's going to cost them money if they do 
that. You know, that is a position we really can't be in.
    But I think the No. 1 thing that has happened--you know, a 
system like this, two things really. It has to be developed 
with the employees in mind, whether they're unionized or not. 
It works better when there is a union, but it has to be 
developed from the ground up with employee buy-in, and that 
means just have an idea, not throw a program at them and expect 
them to sign onto it. And then it has to be funded properly so 
that, again, the money is there to reward people, and money is 
there to train not only employees affected, but the supervisors 
who are doing the evaluations.
    Mr. Davis of Illinois. Thank you.
    Mr. Gage.
    Mr. Gage. Yeah. I'll be brief.
    I think just two things. I think first the classification 
system, it really could be an opportunity to look at that and 
remove barriers that really stop people from moving up within 
the GS system because of really antiquated barriers which don't 
take into account how well the person works and how well they 
perform. It is just some qualification that has been in there 
for 30 years. Now, I'd like to see the classification system 
really tweaked and fixed. Now, I don't think you can remove it 
as a basis, though, of the Civil Service.
    And the second thing, a real practical thing--and we've 
been trying to push this, you know, for years--is more use of 
this career ladder, a system that we already have in the 
government. Many agencies use them. It is very similar to pay 
banding, although it has set criteria for what you have to do 
to be promoted. And these things have really worked good. I 
don't know why we're moving away from things that have worked 
well into this unknown.
    But, again, Mr. Chairman, I just think this pay for 
performance is built on false premises, and I don't think it 
will ever work. I think that putting as much time into the 
classification system and some techniques in the Federal 
Government would pay off much more than to continue on with 
this expensive disaster.
    Ms. Kelley. I would ask that if the GS system is not going 
to be the pay system for Federal employees, that the right to 
collectively bargain pay be given to the unions and the Federal 
sector, because if there is not going to be the GS system, a 
known and established system, then that's what should happen. 
We should have the right to collectively bargain.
    The GS system is not perfect. There are changes that could 
be made to it to allow the agencies more opportunity to reward 
employees at the high end, and we have offered more than once 
to work with the agencies to adjust that system to those 
realities. They have never taken us up on that offer. We have 
given them very specific proposals on how the GS system could 
be used to be better able to attract, retain and reward. And 
without collective bargaining, there is nothing there that 
requires them to do that, to talk to us.
    Funding is an issue because of the forced distributions 
that you see in all of these alternative pay-for-performance 
systems. To say that 25 percent of the employees will be given 
the top raise, well, if only 25 percent of the employees are 
performing at the top level, then that's appropriate. But what 
if you have an agency with 60, 70, 80 percent top performers? 
Why penalize them because of a funding issue or having to take 
from one to give to somebody else? So I would ask for your help 
there.
    I would also ask for any opportunity you have to hold 
agencies accountable for what they do with these alternate 
systems. You heard from the SEC today, a 2003 arbitration win, 
that they still have not made these employees whole who were 
harmed. You know, the idea that they have not stepped up to 
this issue, not made those employees whole and stepped up to 
the fact that they made the same mistake in 4, 5, 6 and 7, and 
here we are in 2008, and yet they think it is OK to be--you 
know, writing briefs to have the arbitrator tell them what the 
remedy will be. I think that is unconscionable. They should be 
held accountable for that and they are not.
    And the idea that anyone points to SEC, FDIC or the IRS 
manager's pay banding as a success should be an opportunity for 
anyone who can influence this to say that these are all 
failures. And, you know, this sound bite needs to stop, and 
they really need to get down to the business of managing the 
work force and properly compensating and rewarding them, which 
they are not doing today.
    Mr. Davis of Illinois. Well, thank you all so very much. 
Mr. Sarbanes, if you have no further----
    Mr. Sarbanes. I just wanted to say that I'm very excited 
about the prospects for doing the right thing going forward, 
because, you know, there is this phrase, if it is not broke, 
don't fix it. But a better way of saying that is if it is not 
broken, make it better. And there is no shortage of good ideas 
that you have identified in a very responsible way, which 
signals that if we get the kind of leadership in place at the 
levels where it can make a difference in collaboration with you 
in taking up the ideas that you've offered, we can really start 
to set a new standard and model as the Federal Government 
should, the kind of behavior that every business in society can 
emulate. So I'm very excited about what is to come based on 
what I've heard here today.
    Thank you for the hearing, Mr. Chairman.
    Mr. Davis of Illinois. Well, thank you all so much. We 
appreciate your being here, your testimony, your patience. And 
perhaps good, better and best; never let it rest until your 
good becomes better and your better becomes best. Maybe we'll 
end up with the very best system that decisions can make.
    Thank you so much. This hearing is adjourned.
    [Whereupon, at 5:25 p.m., the subcommittee was adjourned.]