Financial Regulators: Agencies Have Implemented Key Performance  
Management Practices, but Opportunities for Improvement Exist	 
(18-JUN-07, GAO-07-678).					 
                                                                 
Congress granted financial regulators flexibility to establish	 
their own compensation systems and required certain agencies to  
seek to maintain comparability with each other in pay and	 
benefits to help the agencies overcome impediments to recruiting 
and retaining employees and avoid competing for the same	 
employees. In response to a request, this report reviews (1) how 
the performance-based pay systems of 10 financial regulators are 
aligned with six key practices for effective performance	 
management systems, (2) the actions these agencies have taken to 
assess and implement comparability in pay and benefits, and (3)  
the extent to which employees in selected occupations have moved 
between or left any of the agencies. GAO analyzed agency guidance
and policies, agency data on performance ratings and pay	 
increases, agency pay and benefits surveys, data from the Central
Personnel Data File, and interviewed agency officials.		 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-07-678 					        
    ACCNO:   A70906						        
  TITLE:     Financial Regulators: Agencies Have Implemented Key      
Performance Management Practices, but Opportunities for 	 
Improvement Exist						 
     DATE:   06/18/2007 
  SUBJECT:   Federal agencies					 
	     Federal employees					 
	     Merit compensation 				 
	     Pay						 
	     Pay rates						 
	     Performance appraisal				 
	     Performance management				 
	     Performance measures				 
	     Policy evaluation					 
	     Regulatory agencies				 
	     Salary increases					 
	     Performance-based pay				 
	     Transparency					 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-07-678

Report to Congressional Requesters

June 2007

FINANCIAL REGULATORS

Agencies Have Implemented Key Performance Management Practices, but
Opportunities for Improvement Exist

Contents

Letter 1
Results in Brief 4
Background 7
Financial Regulators Generally Have Linked Pay to Performance
and Made Distinctions in Performance, but Opportunities Exist
for Improvements 9
Agencies Have Taken Various Actions to Seek to Maintain Pay and
Benefits Comparability 26
Few Employees Have Moved among the Financial Regulators, Most
Resigned from Federal Employment 31
Conclusions 35
Recommendations for Executive Action 36
Agency Comments and Our Evaluation 37

Appendices

Appendix I: Objectives, Scope, and Methodology 40
Appendix II: Information on Agencies 47
Appendix III: Financial Regulators Have Implemented Key Practices in
Varying Ways 50
Appendix IV: Actions Taken by Financial Regulators to Seek to Maintain
Pay and Benefits Comparability and Pay and Benefits Data 87
Appendix V: Analysis of Movement Data of Financial Regulator
Employees from Fiscal Years 1990 through 2006 112
Appendix VI: Comments from the U.S. Commodity Futures Trading
Commission 136
Appendix VII: Comments from the Board of Governors of the Federal
Reserve System 138
Appendix VIII: Comments from the Federal Housing Finance Board 139
Appendix IX: Comments from the National Credit Union
Administration 140
Appendix X: Comments from the Office of Federal Housing Enterprise
Oversight 144
Appendix XI: Comments from the Securities and Exchange Commission 145
Appendix XII: GAO Contacts and Staff Acknowledgments 146

Tables

Table 1: Federal Financial Regulator Pay Comparability Legislative
Provisions 8
Table 2: Selected OPM Locality Pay Percentages Compared to
Financial Regulatorsï¿½ Locality Pay Percentages, Fiscal Year
2006 30
Table 3: Performance Appraisal Cycle by Agency 41
Table 4: Agency Information 47
Table 5: Pay and Benefits Surveys That Federal Financial
Regulators Conducted through External Compensation
Consultants, 1991-2006 88
Table 6: Selected Examples of Benchmarks Agencies Have Used to
Assess Pay and Benefits Comparability 92
Table 7: Selected Examples of Recent Pay and Benefits
Adjustments Resulting from Agenciesï¿½ Comparability
Assessments 93
Table 8: Agenciesï¿½ Current Methods for Determining Locality Pay
Percentages and Adjustments 95
Table 9: List of Benefits Offered by the 10 Financial Regulators 98
Table 10: Number of Financial Regulator Employees in
Mission-Critical and other Occupations Who Moved to
another Financial Regulator, Fiscal Years 1990ï¿½2006 114
Table 11: Commodity Futures Trading Commission Employment
and Movement Data, Fiscal Years 1990ï¿½2006 116
Table 12: Farm Credit Administration Employment and Movement
Data, Fiscal Years 1990-2006 118
Table 13: Federal Deposit Insurance Corporation Employment and
Movement Data, Fiscal Years 1990ï¿½2006 120
Table 14: Federal Housing Finance Board Employment and
Movement Data, Fiscal Years 1990ï¿½2006 122
Table 15: National Credit Union Administration Employment and
Employee Movement Data, Fiscal Years 1990ï¿½2006 124
Table 16: Office of the Comptroller of the Currency Employment and
Employee Movement Data, Fiscal Years 1990ï¿½2006 126
Table 17: Office of Federal Housing Enterprise Oversight
Employment and Employee Movement Data, Fiscal Years
1990ï¿½2006 128
Table 18: Office of Thrift Supervision Employment and Employee
Movement Data, Fiscal Years 1990ï¿½2006 130
Table 19: Securities and Exchange Commission Employment and
Employee Movement Data, Fiscal Years 1990ï¿½2006 132
Contents
Page iii GAO-07-678 Financial Regulators
Table 20: Other Federal Agencies Employment and Employee
Movement Data, Fiscal Years 1990ï¿½2006

Figures

Figure 1: Percentage of Regulatory Employees by Rating Level for
Systems with Multiple Rating Levels, for Completed
Appraisal Cycles Specified 22
Figure 2: Percentage of Regulatory Employees by Rating Level and
Merit Increase Category for Agencies with Pass/Fail
Performance Rating Systems, for Completed Appraisal
Cycles Specified 24
Figure 3: Nonexecutive Minimum and Maximum Pay Ranges and
Average Actual Pay for Mission-Critical Occupations by
Regulator, 2006 28
Figure 4: Attrition among Financial Regulators, Fiscal Years 1990ï¿½
2006 33
Figure 5: Excerpt from an OCC Commissioned Examinerï¿½s
Individual Performance Plan 56
Figure 6: Excerpt from an FCA Individual Performance Plan 62
Figure 7: Example of OFHEOï¿½s Worksheet for Weighting
Performance Elements 65
Figure 8: Average Number of Employees in Mission-Critical and
other Occupations Moving among the 9 Financial
Regulators, Fiscal Years 1990ï¿½2006 113

Abbreviations

CFTC Commodity Futures Trading Commission
CPDF Central Personnel Data File
EEO Equal Employment Opportunity
FCA Farm Credit Administration
FDIC Federal Deposit Insurance Corporation
FHFB Federal Housing Finance Board
FIRREA Financial Institutions Reform, Recovery and Enforcement Act
IT information technology
NCUA National Credit Union Administration
OCC Office of the Comptroller of the Currency
OFHEO Office of Federal Housing Enterprise Oversight
OPM Office of Personnel Management
OTS Office of Thrift Supervision
SEC Securities and Exchange Commission
SSP Senior Staff Position

Letter

June 18, 2007

Congressional Requesters:

The federal government must adapt to a range of major trends and
challenges in the nation and the world, and to respond, it must have the
institutional capacity to plan more strategically, identify and react more
expeditiously, and focus on achieving results. Critical to the success of
this transformation are the federal government's people--its human
capital. Yet the government has not transformed, in many cases, how it
classifies, compensates, develops, and motivates its employees to achieve
maximum results within available resources and existing authorities. One
of the questions being addressed as the federal government transforms is
how to update its compensation system to be more market based and
performance oriented.^1 In this type of system, organizations consider the
skills, knowledge, and performance of employees as well as the labor
market when making pay decisions.

Congress has recognized the need for flexibility in how selected agencies
compensate employees. Congress granted the federal financial regulatory
agencies the flexibility to establish their own compensation systems
recognizing that the existing approach to compensating employees could
impede these agencies' ability to recruit and retain employees critical to
meeting their organizational missions. In addition to the flexibility
provided to the agencies over the years, Congress also directed most of
the agencies to seek to maintain pay comparability and to consult with
each other in doing so to ensure the agencies do not compete with each
other for employees.^2 The 10 federal financial regulatory agencies
reviewed in this report are the Commodity Futures Trading Commission
(CFTC), the Farm Credit Administration (FCA), the Federal Deposit
Insurance Corporation (FDIC), the Federal Housing Finance Board (FHFB),
the Board of Governors of the Federal Reserve System (Federal Reserve
Board), the National Credit Union Administration (NCUA), the Office of the
Comptroller of the Currency (OCC), the Office of Federal Housing
Enterprise Oversight (OFHEO), the Office of Thrift Supervision (OTS), and
the Securities and Exchange Commission (SEC).

^1GAO, 21st Century Challenges: Reexamining the Base of the Federal Government, GAO-
05-325SP (Washington, D.C.: February 2005).

^2See, e.g., section 1206 of the Financial Institutions Reform, Recovery, and Enforcement Act
of 1989 (FIRREA), Pub. L. No. 101-73. See also, H. Conf. Rpt. No. 101-222, 457-458 (1989).
While FIRREA uses ï¿½compensationï¿½ to mean ï¿½pay,ï¿½ for purposes of this report,
compensation is defined as employee pay and benefits.

In our prior work, we identified key practices for effective performance
management systems that collectively create a "line of sight" showing how
team, unit, and individual performance can contribute to overall
organizational goals and help individuals understand the connection
between their daily activities and the organization's success.^3

In your February 2006 letter, you noted that the financial regulatory
agencies are at different stages of implementing their performance-based
pay systems and compensation authorities and that an examination of how
they are implementing their systems could be valuable to other agencies
pursuing performance-based pay systems. Our past work looking at agencies'
implementation of performance management and performance-based pay systems
has shown that better linking pay to performance is very much a work in
progress at the federal level.^4

In response to your request, this report examines (1) how the
performance-based pay systems of 10 federal financial regulatory agencies
are aligned with six key practices for effective performance management
systems, (2) the actions 10 federal regulatory agencies have taken to
assess and implement comparability in pay and benefits with each other,
and (3) the extent to which employees in selected occupations have moved
between or left any of the agencies.

For purposes of this review, we focused on six key practices for effective
performance management, which are more closely related to planning,
rating, and rewarding individual performance:^5

^3GAO, Human Capital: Preliminary Observations on the Administrationï¿½s Draft Proposed
ï¿½Working for America Act,ï¿½ GAO-06-142T (Washington, D.C.: Oct. 5, 2005); Human Capital:
Senior Executive Performance Management Can Be Significantly Strengthened to
Achieve Results, GAO-04-614 (Washington, D.C.: May 26, 2004); and Results-Oriented
Cultures, Creating a Clear Linkage between Individual and Organizational Success,
GAO-03-488 (Washington, D.C.: Mar. 14, 2003).

^4GAO, Human Capital: Implementing Pay for Performance at Selected Personnel
Demonstration Projects, GAO-04-83 (Washington, D.C.: Jan. 23, 2004).

^5The other three key practices are: provide and routinely use performance information to
track organizational priorities, require follow-up actions to address organizational priorities,
and maintain continuity during transitions.

1.Align individual performance expectations with organizational goals.

2.Connect performance expectations to crosscutting goals.

3.Use competencies to provide a fuller assessment of performance.

4.Link pay to individual and organizational performance.

5.Make meaningful distinctions in performance.

6.Involve employees and stakeholders to gain ownership of performance
management systems.

In this report, we present important aspects of the agencies'
implementation of the practices of linking pay to individual and
organizational performance, which includes providing adequate safeguards
to help ensure fairness and guard against abuse, and making meaningful
distinctions in performance. We discuss the agencies' implementation of
the four other practices included in the list in appendix III.

To address our first objective, we analyzed the selected agencies'
performance management and pay systems' guidance, policies, and procedures
and other related documents; interviewed key agency officials and
representatives of unions or other employee groups; examined a small,
select set of employees' individual performance plans to illustrate annual
performance expectations for employees; and analyzed agencies' performance
management data, such as the distribution of performance ratings and
performance-based pay increases. The individual performance plans,
selected in conjunction with agency officials to reflect a mix of key
locations, occupations, and grade levels at each agency, as well as the
performance management data from each agency that we examined, pertained
to each agency's last completed performance appraisal cycle when we began
this review.^6

To address our second objective, we analyzed relevant statutes and
legislative histories for the selected agencies; reviewed the most recent
pay and benefits surveys from these agencies; obtained agency pay and
benefits data; and discussed agencies' informal interactions to assess pay
comparability with agency officials. For the third objective, we analyzed
data from the Central Personnel Data File (CPDF), which includes
information on pay, benefits, personnel actions, and other data to support
statistical analyses of executive branch personnel management programs, to
determine employee movement.

^6The smallest number of performance plans we examined from an agency was 1, in a case
where the performance plans for all employees are completely standardized, and the largest
number of plans we reviewed from an agency was 32. See table 3 in appendix I for more
information on the dates of the performance appraisal cycles we reviewed at each agency.
We conducted various inspections and electronic testing of agency data
obtained for the first objective for reasonableness and the presence of
any obvious or potential errors in accuracy and completeness. We also
reviewed related agency documentation, interviewed agency officials
knowledgeable about the data, and brought to the attention of these
officials any concerns or discrepancies we found with the data for
correction or updating. On the basis of these procedures, we believe the
data are sufficiently reliable for use in the analyses presented in this
report. Our data collection strategies were not designed to guarantee that
we would identify all potential examples of how agencies may have
implemented the various practices. An agency may have implemented a
particular practice even if it is not specifically mentioned in the
report. We did not independently verify the data in the pay and benefits
comparability surveys the consultants conducted for the agencies or the
pay and benefits data we received from the agencies. On the basis of our
data reliability testing of CPDF data, we believe the CPDF data are
sufficiently reliable for this review. Appendix I provides additional
information on our scope and methodology. We conducted our work from
February 2006 through June 2007 in accordance with generally accepted
government auditing standards.

Results in Brief

Overall, the federal financial regulators have implemented key practices
for effective performance management systems in ways that consider the
unique needs of their organizational cultures and structures, but some
have opportunities to improve implementation of certain practices as they
continue to refine their systems. All of the regulators awarded some pay
increases during the appraisal cycles we reviewed that were linked to
employees' performance ratings. However, CFTC and SEC also provided
across-the-board increases to employees, even to the few employees who had
not received acceptable performance ratings, weakening the linkage of pay
to performance. Officials from both agencies stated that in the future,
annual pay adjustments will not be awarded to unsuccessful performers.
While all of the regulators built safeguards into their performance
management systems to enhance credibility and fairness, SEC did not
establish and communicate performance standards to its nonexecutives,
which could compromise the credibility of the system. Several regulators
did not fully implement the safeguard of providing overall ratings and pay
increase results to all employees, affecting employees' ability to
understand where they stood in their organizations. As a result, some
regulators have opportunities to strengthen an important safeguard for
providing transparency within their performance-based pay systems.

The financial regulators have hired external compensation consultants to
conduct individual, formal comparability surveys, exchanged pay and
benefits information, explored the feasibility of conducting a common
survey, or adjusted pay and benefits to seek to maintain comparability
requirements. Although officials said that they have adjusted pay and
benefits partly based on the results of their comparability efforts, there
is some variation in pay ranges and benefit packages among the agencies.
For example, the regulators have varying minimum and maximum pay ranges
for similar job series and offer different benefits such as tuition
reimbursement and supplemental retirement plans. When discussing the
reasons that may contribute to this variation, agency officials said that
the year the agencies first became subject to comparability provisions,
budget constraints, and the needs and preferences of the workforce play a
role in compensation decisions and contribute to this variation.
Furthermore, officials emphasized that it was not their goal to have
identical pay and benefits packages; rather, they considered pay and
benefits as a total package when seeking to maintain pay and benefits
comparability and when setting pay policies aimed at recruiting and
retaining employees.

From fiscal years 1990 through 2006, the movement of employees among the
financial regulators was very low and presented no discernible trend over
the period, but 86 percent (13,433) of the 15,627 employees leaving the
regulators voluntarily (i.e., moving or resigning), resigned from the
federal government.^7 The number of employees who moved to another
financial regulator ranged from a low of 16 of 1,362 who moved or resigned
in fiscal year 1997 to a high of 97 of 1,229 who moved or resigned in
fiscal year 1991. The total number of financial regulator employees was
15,400 (1997) and 19,796 (1991) during those 2 years. (Federal Reserve
Board data are excluded from the federal government's personnel database
and are not included in these analyses.) Some agency officials told us
that they believe that the Financial Institutions Reform, Recovery and
Enforcement Act (FIRREA) comparability provision and similar provisions in
subsequent laws applicable to financial regulators have been effective in
ensuring that regulators' pay and benefits are generally comparable among
the 10 agencies, which probably helps minimize employee movement among
financial regulatory agencies. The movement of mission-critical employees,
including accountants, attorneys, auditors, examiners, economists,
financial analysts, investigators, information technology specialists, and
business specialists, to a different financial regulator also produced no
discernible trends.

^7Resignations do not include employees who left an agency due to retirement.

This report includes recommendations to CFTC, FCA FHFB, NCUA, OFHEO, and
SEC. FCA, FHFB, and OFHEO should communicate the overall results of the
performance appraisal and pay increase decisions to all employees
agencywide while protecting individual confidentiality. NCUA, CFTC, and
SEC should work with unions to communicate the overall results of the
performance appraisal and pay increase decisions to all employees
agencywide while protecting individual confidentiality. SEC should clearly
communicate the criteria for making performance rating and pay increase
decisions to nonexecutive employees and should assess senior executives'
performance at the end of the performance appraisal cycle regardless of
the amount of funding available for performance-based pay increases.

We provided a copy of the draft report to the Chairman, Commodity Futures
Trading Commission; Chairman of the Board and Chief Executive Officer,
Farm Credit Administration; Chairman, Federal Deposit Insurance
Corporation; Chairman, Federal Housing Finance Board; Chairman, Board of
Governors of the Federal Reserve System; Chairman, National Credit Union
Administration; Comptroller of the Currency, Office of the Comptroller of
the Currency; Director, Office of Federal Housing Enterprise Oversight;
Director, Office of Thrift Supervision; and Chairman, Securities and
Exchange Commission, for review and comment. We received written comments
from six of the agencies. They generally agreed with the findings and
recommendations. See appendixes VI, VII, VIII, IX, X, and XI for letters
received from CFTC, the Federal Reserve Board, FHFB, NCUA, OFHEO, and SEC.
These six agencies, along with the other four, also provided clarifying
and technical comments, which we have incorporated as appropriate. Several
agencies described actions they plan to take to address the recommendation
to communicate the overall results of the performance appraisal and pay
increase decisions to all employees on an agency-wide basis while
protecting individual confidentiality.

Background

The 10 federal financial regulatory agencies in our review vary in size,
mission, funding structure, whether they bargain with a union, and how
long they have been implementing aspects of performance-based pay systems.
For example, FHFB is the smallest agency with just over 120 employees,
while FDIC, the largest agency, had more than 4,300 employees as of
September 2006 and has been implementing pay for performance since 1998.
Likewise, these agencies regulate a range of activities including banking
and securities and futures. Appendix II includes the financial regulators'
missions, funding structures, and whether they are unionized and bargain
with a union over pay and benefits.

Under Title 5 of the U.S. Code, the financial regulatory agencies have the
flexibility to establish their own compensation programs without regard to
various statutory provisions on classification and pay for executive
branch agencies. At the same time these financial regulators received
increased flexibility regarding compensation, Congress also generally
required that they seek compensation comparability with each other. A
provision in FIRREA requires six agencies--FDIC, OCC, NCUA, FHFB, FCA, and
OTS--in establishing and adjusting compensation and benefits, to inform
each other and Congress of such compensation and benefits, and to seek to
maintain comparability regarding compensation and benefits.^8 Additional
FIRREA provisions require FCA, FHFB, NCUA, OCC, and OTS to seek to
maintain compensation and benefit comparability with the Federal Reserve
Board.^9 Although the Federal Reserve Board is under no obligation to seek
to maintain compensation or benefit comparability with these or any of the
other financial regulators, it has agreed to share compensation
information with the other financial regulators.

The other three agencies are subject to their own compensation
comparability provisions. As required by its 1992 enabling legislation,
OFHEO must maintain comparability with the compensation of employees from
the Federal Reserve Board, OCC, FDIC, and OTS and consult with those
agencies in that regard.^10 In 2002 legislation, SEC and CFTC were placed
under comparability requirements. SEC must consult with and seek to
maintain compensation comparability with FDIC, OCC, NCUA, FHFB, FCA, OTS
and CFTC.^11 However, as shown in table 1, this legislation did not
require these agencies to seek to maintain compensation comparability with
SEC. Similarly, CFTC must consult and seek to maintain compensation
comparability with the six FIRREA agencies, but those agencies are not
required to seek to maintain compensation comparability with CFTC.^12

^8Pub. L. No. 101-73, section 1206, 103 Stat. 183, 523 (Aug. 9, 1989).

^9Sections 301, 702, 1202-3, and 1210 of Pub. L. No. 101-73.

^10Section 1315 of the Federal Housing Enterprises Financial Safety and Soundness Act of
1992, Pub. L. No. 102-550, 106 Stat. 3941 (1992).

^11Section 8(a) of the Investor and Capital Markets Fee Relief Act, Pub. L. No. 107-123, 115
Stat. 2390 (2002).

^12Section 10702(a) of the Farm Security and Rural Investment Act of 2002, Pub. L. No. 107-
171, 116 Stat. 516 (2002).

Table 1: Federal Financial Regulator Pay Comparability Legislative
Provisions

                                        

                               Comparability                        
                                 agencies                           
Agency Comparability  Year of     FIRREA     CFTC Federal OFHEO SEC 
        provisions   provision   agencies         Reserve           
                                                                    
                                    FCA            Board            FDIC FHFB NCUA OCC OTS 
CFTC   Post-FIRREA   2002            x        x      x      x    x   x                     
FCA    FIRREA        1989                     x      x      x    x   x         x           
FDIC   FIRREA        1989            x               x      x    x   x                     
FHFB   FIRREA        1989            x        x             x    x   x         x           
NCUA   FIRREA        1989            x        x      x           x   x         x           
OCC    FIRREA        1989            x        x      x      x        x         x           
OFHEO  Post-FIRREA   1992                     x                  x   x         x           
OTS    FIRREA        1989            x        x      x      x    x             x           
SEC    Post-FIRREA   2002            x        x      x      x    x   x    x                
FRB    Not required  
       but shares    
       information   
       with a number 
       of agencies   
       regarding     
       compensation  
       and benefits  

Source: GAO analysis of comparability legislative provisions.

We previously identified key practices for effective performance
management based on public sector organizations' experiences both here and
abroad.^13 High-performing organizations seek to create pay, incentive,
and reward systems that clearly link employee knowledge, skills, and
contributions to organizational results. Performance-based systems reward
employees according to their performance by using performance ratings as
the basis for pay increases. Linking pay to performance can help to create
a performance-oriented culture by providing monetary incentives to become
a top- performing employee. At the same time, as a precondition to linking
pay to performance, performance management systems need to provide
adequate safeguards to ensure fairness and guard against abuse. Providing
adequate safeguards that help to ensure transparency can improve the
credibility of the performance-based pay system by promoting fairness and
trust. Safeguards can include establishing clear criteria for making
rating decisions and determining merit increases, and providing overall
results of performance rating and pay increase decisions to all employees,
while protecting confidentiality. Effective performance management systems
also make meaningful distinctions between acceptable and outstanding
performance of individuals and appropriately reward those who perform at
the highest level. As we have previously reported, effective performance
management systems can provide management with the objective and
fact-based information it needs to reward top performance and provide the
necessary information and documentation to deal with poor performers.^14

^13GAO-03-488.

^14GAO-04-614.

Financial Regulators Generally Have Linked Pay to Performance and Made
Distinctions in Performance, but Opportunities Exist for Improvements

Overall, the federal financial regulators have implemented key practices
for effective performance management systems in ways that consider the
unique needs of their organizational cultures and structures, but some
have not fully implemented certain practices. For purposes of this
section, we focus on the regulators' implementation of the two key
practices of (1) linking pay to performance (which includes building in
safeguards), and (2) making meaningful distinctions in performance. First,
we found that while the regulators generally linked pay to performance,
two regulators awarded across-the-board increases to employees regardless
of their performance. Second, while most regulators generally used
safeguards in varying ways to increase transparency, one did not establish
and communicate performance standards to its nonexecutives, which resulted
in questions about how decisions were made and could compromise the
credibility of the performance system. Third, many regulators did not
fully implement the safeguard of providing overall ratings and pay results
to all employees, which reduced the transparency of their
performance-based pay systems. Fourth, we found that while most regulators
used multiple rating levels to make meaningful distinctions in
performance, employees were usually concentrated in one or two rating
categories and all had very few poor performers. Finally, one agency did
not complete performance ratings for senior officers due to lack of
funding for pay increases, thereby missing an opportunity to provide
valuable feedback. For information about the other four key practices as
well as additional material pertaining to the linking pay to performance
practice, see appendix III.^15

Financial Regulators Generally Have Linked Pay to Performance, but Two
Regulators Still Provided Increases to Performers at All Levels

All of the regulators awarded some performance-based increases during the
appraisal cycles we reviewed that were linked to employees' performance
ratings, although two financial regulators also provided annual pay
adjustments to employees, regardless of performance, during the appraisal
cycles we reviewed. Specifically, CFTC provided an across-the-board pay
increase to all employees to be equivalent to the cost of living
adjustment received by General Schedule employees of the federal
government in January 2006. During the 2005 appraisal cycle, SEC also
provided all employees an across-the-board pay adjustment of 2.1 percent,
regardless of their performance.^16 SEC officials noted that this
across-the-board pay adjustment was in accordance with the negotiated
compensation agreement with the union.^17 While the percentages of
employees rated as unsuccessful or unacceptable at CFTC and SEC during
those cycles were extremely small (less than 1 percent), these agencies
lost opportunities to reinforce the linkage of pay to performance in their
performance management systems. CFTC officials told us that the
performance-based pay portion of the new performance management system
that will begin on July 1, 2007, will require a minimum threshold
performance rating for an
employee to be eligible for a pay increase.^18 SEC and its union are
currently negotiating implementation of a new Compensation and Benefits
Agreement, which provides that employees rated as unacceptable will not
receive annual pay adjustments. SEC officials acknowledged that a negative
perception occurs when employees who are not performing satisfactorily
receive a pay increase.

^15The other four practices are (1) align individual performance expectations with
organizational goals, (2) connect performance expectations to crosscutting goals, (3) use
competencies to provide a fuller assessment of performance, and (4) involve employees and
stakeholders to gain ownership of performance management systems; the financial
regulators have implemented these important practices in various ways.

^16SEC had two distinct performance appraisal cycles, one for senior officers and one for all
other employees. The completed cycles we examined were as follows: for senior officers
from October 2004 to September 2005; for all other employees from May 2005 to April 2006.

^17According to SEC officials, all employees received the annual across-the-board pay
increase that all GS employees received.

^18During the completed performance appraisal cycle we reviewed for CFTC (July 1, 2005, to
June 30, 2006), the agency operated under a performance management system in which the
only increases directly linked to performance ratings were some performance bonuses. In
October 2006, CFTC introduced a new performance management system directive that
affects the performance planning and appraisal processes. The agency will introduce an
accompanying new pay policy in July 2007 that will complete the transition to a
performance-based pay system under which merit increases will be linked to performance
ratings. The new system was developed by an agency committee with employee and union
input and the final approved system took effect on October 1, 2006. (The American
Federation of Government Employees union has chapters at CFTCï¿½s Chicago and New York
offices.) CFTC officials informed us that they do not plan to continue the annual pay
adjustments in the new system, but will continue to use locality pay percentages equivalent
to General Schedule executive order locality percentage increases.

Most of the financial regulators used their rating systems to
differentiate individual performance to award performance-based increases
and reward top performers during the appraisal cycles we reviewed.
Furthermore, all of the agencies also provided increases that, while not
directly linked to performance ratings, considered employee performance in
some way. These increases included special bonuses or awards given to
individuals or teams for special accomplishments or contributions, as well
as promotions and within-pay-band increases. For example, FCA provided
Achievement or Special Act Awards to employees for significant
achievements or innovations towards a special program, project, or
assignment that contributed to the agency's or organizational unit's
mission, goals, and objectives. To receive these awards, employees had to
have performed their regular duties at least at a fully successful level
of performance. FCA also provided some pay increases for competitive and
noncompetitive promotions during the completed appraisal cycle we
reviewed.

Pay increases linked to performance ratings accounted for only part of the
total increases awarded to individual employees during the appraisal
cycles we reviewed. See appendix III for more information on the different
ways in which the regulators translated performance ratings into pay
increases and budgeted for performance-based increases, as well as more
information on other pay increases that involved considerations of
performance.

Financial Regulators Generally Used Safeguards to Increase Transparency,
but SEC Did Not Establish or Communicate Performance Standards for its
Nonexecutive Employees

All of the financial regulatory agencies have built safeguards into their
performance management systems to enhance the credibility and fairness of
their systems, although they varied in how safeguards have been
implemented. For example, with the exception of SEC, the agencies have
used the safeguard of establishing and communicating (1) standards for
differentiating among performance rating categories and (2) criteria for
performance-based pay decisions, thus enhancing transparency, which can
improve employee confidence in the performance management system. (See
app. III for information on the financial regulators' implementation of
additional safeguards.)

CFTC's four-level rating system (i.e., unsuccessful, successful, highly
successful, and exemplary) defined the successful level of performance for
areas that CFTC had identified as critical to employees' job performance,
and included some information on how to distinguish variations from the
successful level of performance. However, an employee representative at
CFTC maintained that the rating level descriptions did not sufficiently
communicate to employees the skills and behaviors employees needed to
demonstrate in order to move, for example, from the "successful" to the
"highly successful" level. Employee representatives stated that even
though there was helpful guidance on distinguishing between levels of
performance in a CFTC manual, these descriptions were hard to understand
and most employees did not refer to the CFTC manual for guidance. An
agency official told us that the revised performance management system
that went into effect in October 2006 is a five-level system, and includes
descriptions of all five performance levels rather than only the
successful rating level described in the system it replaced.

Similarly, OFHEO defined how employees would be rated on its five-level
rating scale for each of the performance elements included in their
performance plans. These performance standards defined the middle level of
performance (fully successful), and included what the rater should look at
to determine if an employee is performing better or worse than this
benchmark. An employee's performance for each element was assessed and a
total score was determined. OFHEO further distinguished between "high" and
"low" levels within rating categories. For example, a rating of
"outstanding" would be classified as being in either the high or low level
of the outstanding rating category based on the performance score the
employee received. Merit increases at OFHEO have been determined directly
by employees' performance ratings, so employees could ascertain the merit
increases they would receive for given performance ratings. For example,
an employee rated "high" commendable receives a higher merit pay increase
than one who is rated "low" commendable. OFHEO employee working group
members noted that both supervisors and employees understand how the
performance elements and standards have been applied through rating
decisions, and they stated that employees generally understood what was
expected of them to attain higher levels of performance and associated
merit increases. However, an employee working group member also commented
that when distinguishing between performance rating levels, some managers
seemed to apply the performance standards more effectively than others,
which could result in differences in how rating decisions were made.

At FDIC, for nonexecutive/nonmanager employees to be eligible for
performance-based pay increases, employees had to first earn a "meets
expectations" rating. Then, in a second process called the "Pay for
Performance" system, FDIC nonexecutive/nonmanagers were placed into one of
four pay groups, based on an assessment of total performance and corporate
contributions as compared with other employees in the same pay pool.^19
The pay for performance program was essentially comparative, meaning that
the contributions and performance of each employee were evaluated and
rewarded on a relative basis within his or her pay pool, as compared to
peers. According to union representatives, employees were not informed
about how management made the distinctions in pay increase groupings.
According to FDIC officials, there are no definitive descriptions or
definitions of the performance levels for each of the three pay groups
because employees are assessed compared to each other, not against fixed
standards. Officials also said that information on the system for
determining pay groups was provided to all employees in early 2006 after
the compensation and benefits agreement became effective, when the system
was first rolled out, and is explained to new employees at orientation. We
did not determine how widespread the concern about how management made
distinctions in pay increase groupings was among FDIC employees.

In contrast, SEC officials did not establish standards upon which to base
rating decisions for nonexecutive employees, nor did they communicate
criteria used to make performance-based pay decisions to these employees.
For its nonexecutive employees, SEC used a two-level rating system in
which individuals' performance was rated as acceptable or unacceptable.
According to agency management, SEC followed the definitions under Title 5
that are used by the rest of the government for differentiating between
acceptable and unacceptable performance.^20 However, SEC did not establish
written performance standards for appraising employees' performance as
acceptable or unacceptable.

^19In the collective bargaining agreement for years 2006-2009, pay group 1 will receive a 5
percent pay increase and 1 percent lump sum payment, pay group 2 will receive a 3.2
percent pay increase and 1 percent lump sum, and pay group 3 will receive a 2.4 percent pay
increase. Pay group 4, containing individuals who did not receive a ï¿½meets expectationsï¿½
rating in the first appraisal process, will receive no increase.

To determine performance-based pay increase amounts for nonexecutive
employees, SEC developed a second phase process that involved making
distinctions in contributions for those individuals who received a summary
performance rating of acceptable. As part of the second phase, employees
and their supervisors submitted contribution statements summarizing the
employees' accomplishments during the appraisal cycle. Using the summary
statements and the supervisors' own assessments, supervisors placed
employees into one of four categories: (1) made contributions of the
highest quality, (2) made contributions of high quality, (3) made
contributions of quality, and (4) made no significant contribution beyond
an acceptable level of performance. Next, a compensation committee within
each office or division evaluated the contribution statements and the
supervisors' placements. For each employee, the committee recommended a
merit pay increase ranging from zero to 4.41 percent (corresponding to
"steps" 0 to 3) to an official from each office or division, who made the
final determination of the employee's merit increase.^21

However, SEC did not develop criteria to differentiate between the four
contribution categories that the compensation committees considered when
recommending merit pay step increase amounts. In addition, SEC employee
representatives told us that it was not clear to employees how the
contribution statements and the subsequent supervisory recommendations
were translated into the decisions about the four contribution categories
into which employees would be placed. SEC officials noted that employees
received copies of narratives written by their supervisors to describe the
employees' contributions; however, they acknowledged that the system could
be more transparent. According to SEC officials, in an effort to increase
transparency in the future, they plan to share with employees information
on supervisors' preliminary recommendations on ratings that are provided
to the compensation committee, so that employees can see into which of the
four contribution categories they were recommended for placement and the
supporting documentation. If the committee changes an initial
recommendation from a supervisor, SEC will provide the employee with the
rationale for the change. An agency official indicated they are developing
broad statements, such as "the committee had a broader perspective of
employee contributions," that address a range of possible reasons for
changes.

^20Section 4301 of Title 5 of the U.S. Code defines unacceptable performance as failure to
meet established performance standards in one or more critical performance elements.

^21SEC used ï¿½stepsï¿½ to indicate performance-based pay increases. Zero steps meant no
increase, one step corresponded to an increase of 1.47 percent, two steps to an increase of
2.94 percent, and three steps to an increase of 4.41 percent.

Some Financial Regulators Did Not Fully Implement the Safeguard of
Providing Overall Ratings and Pay Increase Results to All Employees, Which
Would Increase Transparency in Their Performance-Based Pay Systems

The extent to which the financial regulators shared the overall results of
performance ratings and pay increase decisions with all employees varied,
and some agencies did not make this information widely available to
employees. We have previously reported that the safeguard of communicating
the overall results of performance appraisal and pay increase decisions
while protecting individual confidentiality can improve transparency by
letting employees know where they stand in the organization.^22 An
employee's summary performance rating conveys information about how well
an employee has performed against established performance standards, which
is important, but not sufficient to provide a clear picture of how the
employee's performance compares with that of other employees within the
organization. When the organization communicates where an employee stands,
management can gain credibility by having honestly disclosed to the
employee the basis for making pay, promotion, or developmental opportunity
decisions that may have been based upon relative performance.

The Federal Reserve Board communicated the overall results of the
performance appraisal decisions to all employees by sharing annual
performance rating distributions with all employees, disaggregated by
division. Since this system for determining the amounts of
performance-based increases for individuals based on their performance
ratings is essentially driven by formula, employees know what their merit
increases will be relative to others after receiving their performance
ratings.

^22GAO-06-142T and GAO-04-614.

At FDIC, the distribution of pay group assignments for all nonexecutive/
nonmanager employees who passed the first assessment process is fixed by
the negotiated agreement with the union, so those employees know how
performance-based pay increases will be distributed and the amounts of
increases received by the various pay groups. Further, FDIC officials told
us that, in accordance with the collective bargaining agreement, after
completion of each annual pay for performance cycle they share data on the
results of the pay grouping decisions for employees covered by the
bargaining unit contract with union representatives. These include summary
pay group data analyzed according to the agreement with the union, such as
certain demographic data and individual rating information.

According to an agency official, OCC began to post some limited
information on the average size of some performance-based pay increases on
the agency intranet in November 2006. The information included the
average, agencywide percentage increases for merit increases, merit
bonuses, and special act and spot awards, as well as the percentage of
employees receiving the increases.

During the performance appraisal cycle we reviewed, OTS shared with union
representatives some data on average pay increases. The agency did not
share ratings distribution data with the union, and did not make either
performance-based pay increase or rating results information available to
all employees. However, in November 2006, OTS distributed to all employees
information for the recently completed appraisal cycle on the percentage
of employees who received each performance rating level and the average
pay increase percentages to be received by people at each level. The
information was disaggregated by regions and Washington, D.C.

While SEC did not make the results of performance rating decisions
available to all employees, officials said that they reported information
on performance awards (bonuses) to the union and that, under
implementation of the compensation and benefits agreement currently being
negotiated, they plan to publish aggregated information on performance
ratings under the planned new performance management system for
nonexecutive employees. SEC officials also told us that they plan to
provide information at the lowest possible organizational level while
still protecting individual confidentiality.

The remaining five financial regulators did not share overall data on
ratings or performance increases widely with all employees, although in
some cases some information was shared with managers. The following
outlines how information was shared:

oCFTC shared information on the results of ratings and award decisions
with managers on a Pay Parity Governance Committee, but not with all
employees, for the appraisal cycle we reviewed. CFTC officials told us
that there is no prohibition against sharing this type of information
under the new performance management system directive, and they are aware
that there is some interest among employees in receiving it. They said
that the pay parity committee will determine whether there is value in
releasing this information to all employees in the future.

oAt FHFB, an official told us that office directors see all the ratings
within their offices and make the decisions about the performance-based
pay increases for employees, but this information is not shared across
offices or with all employees. However, the director of the Office of
Supervision, FHFB's largest office, has shared information with all staff
in the office on the ranges of pay increases corresponding to different
performance rating levels and base salary levels that were received by
staff within the office for a given year, as well as the standards used to
assign the merit increase amounts.

oOfficials at OFHEO told us that just last year they started sharing
information on the results of ratings and pay increase decisions with
management, but that they have not yet shared this type of information
with all employees.

oFCA officials told us that they do not share aggregate results of the
performance rating and pay increase decisions with all employees. They
explained that, under a previous administration, in early 2000, an
executive summary was prepared and posted that all employees could
potentially access, which contained information on the results of ratings
and pay increases. However, this information was not broadly disseminated
directly to employees.

oNCUA shares information on the results of the merit pay decisions with
directors, but not with all employees. An NCUA official told us that it is
up to the directors to decide whether or not to share this information
with their staff. In comments on the draft report, NCUA explained that
this is one of the issues involved in its current negotiations over pay
and benefits with the National Treasury Employees Union, and that the
agency's proposal to the union does provide for this type of transparency.

Agencies provided a variety of reasons for not sharing overall ratings and
pay increase information more widely. Officials from FHFB and FCA told us
that the relatively small size of their agencies, 122 and 248 employees,
respectively, makes it harder to share this type of information while
protecting individual confidentiality and that an FHFB official was not
aware of employee demand for this type of information. FCA officials also
mentioned that the emphasis in their performance management system is on
rating individual employees against the standards, not against other
employees and they wanted employees to focus on their individual ratings
and performance. According to union representatives at OCC, the union has
made multiple requests for data on the results of the performance rating
and pay increase decisions but management has declined to share
information that would enable the union to, in their words, perform a
meaningful independent analysis of the ratings and pay increase decisions.
OCC officials told us that they prefer not to share with employees
disaggregated information on ratings and pay increase distributions
because organizational units administer the process differently. For
example, the percentages of individuals rated at the highest level (4) and
next highest level (3) vary from unit to unit. Because units receive fixed
pools of funds for performance-based increases, the average size of a
merit increase that an employee receiving a level 4 may receive can vary
from unit to unit, depending on how many individuals receive the highest
rating.^23 OCC officials told us that sharing information on average merit
increases by unit with employees, without sufficient context of the
factors considered when making these decisions, including more detailed
rating information (which is privacy protected), could lead to
misinterpretation of the data.

However, not sharing information on the results of the performance rating
and pay increase decisions processes can detract from the goal of a
transparent and fair performance management system. This information needs
to be presented in ways that protect individual confidentiality, such as
by aggregating it. Without access to this type of information, individual
employees can lose a valuable opportunity to understand how their
performance stands relative to others in their organization. In cases
where agencies negotiate agreements with unions, an important
consideration is to reach agreement to share aggregate results of the
rating and pay increase decisions with employees, while protecting
individual confidentiality.

^23See appendix III for a discussion of different ways in which agencies budgeted for
performance-based increases, including use of funding pools.

While Financial Regulators Generally Used Multiple Rating Levels to Make
Meaningful Distinctions in Performance, Employees at Most Agencies Were
Concentrated in One or Two Rating Categories, with Very Few Poor
Performers

While most of the financial regulatory agencies used multiple rating
levels to assess employee performance and make distinctions in
performance, at most agencies employees were concentrated in one or two
rating categories and very few received poor performance ratings. By using
multiple-level rating systems, agencies have the capability to make
meaningful distinctions in performance. Effective performance management
systems make meaningful distinctions between acceptable and outstanding
performance of individuals and appropriately reward those who perform at
the highest level. As we have previously reported, performance management
systems can provide management with the objective and fact-based
information it needs to reward top performers and provide the necessary
information and documentation to deal with poor performers.^24 More
specifically, using multiple rating levels provides a useful framework for
making distinctions in performance by allowing an agency to differentiate,
at a minimum, between poor, acceptable, and outstanding performance. We
have reported that two-level rating systems by definition will generally
not provide meaningful distinctions in performance ratings, with possible
exceptions for employees in entry-level or developmental bands.^25

Eight agencies used four or more rating levels. For example, as described
earlier, OFHEO used a five-level rating category system to appraise
employee performance and contributions toward achieving agency goals, and
further distinguished between high and low performance scores within
rating categories. As shown in figure 1, at the eight agencies with four-
or five-level rating systems, the largest percentage of employees fell
into the second highest rating category, except at OFHEO and the Federal
Reserve Board. At OFHEO, more than half of the employees were placed into
the high or low levels of the top rating category. Conversely, at the
Federal Reserve Board (excluding economists), almost half of the employees
fell into the third highest or middle (commendable) rating category.
Across the eight agencies shown in figure 1, the percentage of employees
who fell into the highest rating category varied from 10.6 percent for
economists at the Federal Reserve Board, to 55 percent of employees at
OFHEO.

^24GAO-03-488.

^25GAO-06-142T.

Figure 1: Percentage of Regulatory Employees by Rating Level for Systems
with Multiple Rating Levels, for Completed Appraisal Cycles Specified

Note: The dates of the performance appraisal cycles varied across
agencies, as indicated in this figure and in table 3, appendix I. SEC had
two distinct cycles affecting two different groups of employees.

^aOFHEO distinguished between "high" and "low" level performance scores
within rating categories. The percentage values shown in the graphic
include employees with scores at both the high and low levels within
categories. For example, the percentage value shown in the graphic for the
"Outstanding" rating category includes those employees rated as high-level
outstanding and low-level outstanding combined. According to OFHEO
officials, about 32 percent of employees fell into the low-level
outstanding category.

^bThe new CFTC performance management system that went into effect in
October 2006 has five rating levels.

SEC and FDIC used two-level rating systems (essentially pass/fail systems)
to appraise the performance of certain groups of employees. Although
two-level rating systems by definition will not provide meaningful
distinctions in performance ratings, both SEC and FDIC used a second
process to determine performance-based pay increases and effectively make
more meaningful performance distinctions. As figure 2 shows, the highest
percentage of employees at FDIC fell into the second highest of four
categories, in keeping with the fixed percentages included in the
negotiated agreement with the union. At SEC, the largest percentage of
employees fell into the third highest of four rating categories.

Figure 2: Percentage of Regulatory Employees by Rating Level and Merit
Increase Category for Agencies with Pass/Fail Performance Rating Systems,
for Completed Appraisal Cycles Specified

Note: At SEC, senior officers were not rated during this rating cycle
because, according to officials, the agency did not have the budget to
fund any merit increases. SK employees at SEC include all employees other
than senior officers.

As shown in figures 1 and 2, the percentage of employees rated as poor
performers at each agency was very small during the completed performance
appraisal cycles we reviewed. Employees rated at below the successful and
meets expectations rating levels accounted for less than 3 percent of
employees across the agencies.^26 OTS had zero employees in the bottom two
rating categories combined--all OTS employees received fully successful or
higher ratings. Similarly at NCUA, no executives and 2.1 percent of
nonexecutives were rated below minimally successful.

While the financial regulators rated very few employees as poor
performers, all of the agencies have established procedures to deal with
poor performers. When an employee does not perform up to a threshold
standard for satisfactory performance, most agencies place the employee on
a performance improvement plan or provide counseling for the employee, and
the employee does not receive a performance-based increase at the end of
the performance cycle. For example, OTS has addressed poor performance by
working with the employee to improve his or her area of deficiency. An
employee who receives a rating at the unacceptable level is placed on a
performance improvement plan for a minimum of 90 days. Specifically, OTS
policy advises supervisors to develop a performance improvement plan by
identifying the performance areas in which the employee is deficient and
the types of improvements, including specific work products and steps to
be followed which the employee must complete to attain the fully
successful performance level. In addition, according to OTS policy, the
agency may provide the employee with closer supervision, or on-the-job or
formal training.

However, governmentwide, 29.7 percent of employees indicated in the 2006
Office of Personnel Management (OPM) Federal Human Capital Survey that
they agreed or strongly agreed that differences in performance within the
work unit were recognized in a meaningful way. Positive responses to this
question for the eight financial regulators who participated in the survey
ranged from 24.9 percent for CFTC to 41.6 percent for OCC. None of these
agencies had a majority of their employees provide positive responses to
this question, and only three of the eight agencies had more than one
third of their employees provide positive responses to this question.

^26Rating levels at the agencies that accounted for less than 3 percent of employees included
the following categories: ï¿½unacceptable,ï¿½ ï¿½unsatisfactory,ï¿½ ï¿½unsuccessful,ï¿½ ï¿½minimally
acceptable,ï¿½ ï¿½minimally successful,ï¿½ ï¿½marginal,ï¿½ and ï¿½does not meet expectations.ï¿½

SEC Did Not Complete Performance Ratings for Senior Officers, Missing an
Opportunity to Provide Valuable Feedback

While it may have been an isolated incident, for senior officers, SEC
effectively did not make distinctions in rating their performance during
the appraisal cycle we reviewed because the agency did not complete
performance ratings for them in 2005. According to SEC officials, no funds
were available for performance-based bonuses (which are normally dependent
on performance ratings) during that assessment cycle. As a result,
divisions performed assessments of senior officers, but the assessment
process was not completed and their ratings were not signed by the
Chairman for the October 1, 2004, to September 30, 2005, performance
appraisal cycle. A recent SEC Inspector General report confirmed that
senior officers in SEC's Enforcement Division did not prepare performance
review documents for the performance cycle that ended on September 30,
2005, and recommended that required steps of the senior officer
performance appraisal process be conducted in accordance with Commission
policy, even when merit increases are not awarded.^27 All senior officers
received annual across-the-board salary increases during that cycle.
Conducting performance appraisals and making distinctions in performance
are important not only for determining performance-based pay increases,
but for providing feedback to help employees improve their performance and
assess how their work contributed to achieving organizational goals. By
not appraising their performance, SEC missed an opportunity to provide
valuable feedback to senior officers.

Agencies Have Taken Various Actions to Seek to Maintain Pay and Benefits
Comparability

Financial regulators have hired external compensation consultants to
conduct individual, formal comparability surveys, exchanged pay and
benefits information, explored the feasibility of conducting a common
survey, and adjusted pay and benefits to seek to maintain pay and benefits
comparability. The majority of the financial regulators conducted pay
comparability surveys that have included other financial regulators and in
some instances, private-sector entities. To compare pay across agencies,
consultants send questionnaires on behalf of the sponsoring agency and ask
participating agencies to match the jobs based on the job descriptions
provided. To compare benefits, consultants use various methods, such as
side-by-side comparisons of benefits and calculation of total cost of
benefits per employee. In addition to these surveys, human capital
officials at the 10 financial regulators have formed an interagency group
to exchange information and consult on topics such as updates on merit pay
ranges and bonuses. However, agency officials told us that because many of
the financial regulators conduct separate comparability surveys, their
staffs have to respond to numerous and often overlapping inquiries, which
can be inefficient. To begin addressing the inefficiencies of this
process, the agencies formed a subcommittee in December 2006 to study the
feasibility of conducting a common survey on pay and benefits. According
to agency officials, the subcommittee also has discussed the feasibility
of establishing a Web-based data system to make the most current pay and
benefits information available to participating agencies.

^27Securities and Exchange Commission, Office of the Inspector General, Enforcement
Performance Management, Audit Report No. 423 (Feb. 8, 2007).

In the absence of a legislative definition of what constitutes
comparability, agency officials told us that they use various methods to
assess pay and benefits comparability after they have obtained relevant
data from the other agencies. For example, FDIC has sought to set its
total pay ranges (base pay plus locality pay) for specific occupations and
grade levels within 10 percent of the average of FIRREA agencies. FCA used
the average market rate paid by other financial regulators as a benchmark.
Finally, partly on the basis of the results of the comparability surveys
and discussions among the agencies, the financial regulators have adjusted
their pay and benefits policies in their efforts to seek to maintain
comparability. For example, as a result of gaining pay flexibilities, CFTC
implemented new pay ranges for its 2003 pay schedule, and increased base
pay by 20 percent for all eligible employees to partially close the 25
percent gap between CFTC and FIRREA agencies. Appendix IV provides
additional information on our analysis of individual agency actions.

While the regulators have taken actions to seek to maintain comparability
in their pay and benefits, there are some variations in base pay ranges
and benefit packages among the agencies. Figure 3 shows the base pay
ranges (minimum and maximum) for the mission-critical occupations,
excluding executives at the 10 agencies.^28 As shown in the same figure,
the actual average base pay among the 10 agencies also varies somewhat in
relation to the agencies' respective base pay ranges, which according to
agency officials, could be affected by the average length of service of
employees, and the fact that some agencies tend to hire employees at
certain grade levels.

^28See appendix I for our criteria for selecting mission-critical occupations in our study. We
excluded executives from the analysis of average actual pay and pay scale. In addition, with
the exception of the Federal Reserve Board and OFHEO, which do not have separate base
and locality pay, the base pay shown in figure 3 does not include locality pay percentages.

Figure 3: Nonexecutive Minimum and Maximum Pay Ranges and Average Actual
Pay for Mission-Critical Occupations by Regulator, 2006

^aThe figure shows the minimum pay of the lowest level and the maximum pay
of the highest level of an occupation for which agencies have an
established occupation. Agencies may not currently have incumbents in each
level of the occupation. The base pay ranges exclude locality pay except
for the Federal Reserve Board and OFHEO.

^bThe actual average base pay data in CPDF did not segregate CFTC's
examiners and investigators or OFHEO's examiners and financial analysts
because these occupations are assigned the same OPM job series number.
Therefore, we did not separately present the actual average pay for these
four occupations at the two agencies. The combined actual average pay for
examiners and investigators is $83,501 at CFTC; and $131,294 for examiners
and financial analysts at OFHEO.

^cThe base pay ranges shown for the Federal Reserve Board and OFHEO
reflect the agencies' total pay ranges because Washington D.C., is the
agencies' only duty station.

^dThe actual average base pay data are provided by the Federal Reserve
Board because CPDF does not contain data for the Board.

Because each financial regulator sets its own locality pay percentage
based on its respective policies, locality pay percentages often differ
from those that OPM sets for General Schedule employees (with the
exception of CFTC) and vary among agencies for the same duty station. For
example, in New York City, the OPM locality pay percentage is 22.97
percent but the regulators' locality pay percentages range from 21.19 at
FDIC and FHFB to 33.20 percent at OTS. Table 2 shows the locality pay
percentages for OPM and for the eight financial regulators that have
locality pay percentages for selected cities.

Table 2: Selected OPM Locality Pay Percentages Compared to Financial
Regulators' Locality Pay Percentages, Fiscal Year 2006

                                        

       Locality        OPM  CFTC   OCC   OTS  FDIC     FCA  FHFB  NCUA    SEC 
Atlanta, Ga.     15.10%   N/A   N/A 6.50% 7.43%     N/A 7.43% 8.06% 11.13% 
Boston, Mass.     19.99   N/A 13.00 17.50 22.67     N/A 22.67 18.55  24.29 
                                                ^a                            
Chicago, Ill.     21.15 21.15  8.00 11.20 16.20     N/A 16.20 22.26  21.62 
Dallas, Tex.      16.39   N/A  3.00   N/A  9.20    6.50  9.20 11.25  12.92 
Denver, Colo.     19.49   N/A  8.00  2.30   N/A   10.50   N/A 18.27  17.72 
Los Angeles,      23.18   N/A 18.00 18.10 19.28     N/A   N/A 25.70  24.35 
Calif.                                                                     
Miami, Fla.       17.84   N/A  3.00  6.80 11.33     N/A   N/A 15.97  15.02 
                                                ^b                            
Minneapolis,      17.31 17.31  3.00  7.30 11.76  8.20^c   N/A 13.23    N/A 
Minn.                                                                      
New York City,    22.97 22.97 23.00 33.20 21.19     N/A 21.19 23.61  24.92 
N.Y.                                                                       
San Francisco,    28.68   N/A 28.00 46.70 32.41     N/A 32.41 36.43  25.14 
Calif.                                                                     
Seattle, Wash.    17.93   N/A  8.00 15.80 13.93     N/A 13.93 14.52    N/A 
Washington, D.C.  17.50 17.50  8.00 25.90 13.30 15.20^d 16.30 12.87  17.50 

Source: GAO analysis of agency data.

Note: Washington D.C. is the Federal Reserve Board's and OFHEO's only duty
station, thus the agencies do not have separate locality pay percentages.
For the other agencies, N/A indicates localities where they had no office.

^a Locality includes Braintree, Mass.

^b Locality includes Sunrise, Fla.

^c Locality includes Bloomington, Minn.

^d Locality includes McLean, Va.

The benefits that the 10 financial regulators offered also varied, which
we discuss in detail in appendix IV. For example, half of the regulators
offer their employees 401(k) retirement savings plans with varying
employer contributions in addition to offering the governmentwide Federal
Thrift Saving Plan (except for the Federal Reserve Board). According to
agency officials, factors such as the year an agency first became subject
to comparability provisions, budget constraints, the needs and preferences
of different workforces, and ways to attract and retain workforces play a
role in compensation decisions and contribute to the variations in pay
ranges and benefits. Moreover, agency officials emphasized that it was not
their goal to have identical pay and benefits packages; rather, they
considered pay and benefits as a total package when seeking to maintain
pay and benefits comparability and when setting pay policies aimed at
recruiting and retaining employees.

Few Employees Have Moved among the Financial Regulators, Most Resigned
from Federal Employment

While the total number of financial regulatory employees resigning from
federal employment between fiscal years 1990 and 2006 generally declined,
there was no clear trend among the number who moved to another financial
regulator. As shown in figure 4, the number of employees leaving one
federal regulator for another declined from the previous fiscal year in 10
of the 16 years and increased from the previous fiscal year in the other 6
years. Figure 4 also shows the percentage of financial regulatory
employees who went to another financial regulator, went to other federal
agencies, and resigned from federal employment, and the total number of
financial regulatory employees during this period. Of all the employees
who left their financial regulatory agency voluntarily (moved to another
financial regulator or executive branch agency, or resigned) from fiscal
year 1990 through fiscal year 2006, the vast majority--86 percent
(13,433)--of the 15,627 employees leaving the regulators voluntarily
(i.e., moved or resigned), resigned from the federal government. The
number of employees who moved to another financial regulator ranged from a
low of 16 of 1,362 who moved or resigned in fiscal year 1997 to a high of
97 of 1,229 who moved or resigned in fiscal year 1991. The total number of
financial regulator employees was 15,400 and 19,796 during those 2 years,
respectively.^29 Similar lows were also experienced in 1996 and 2003. Some
agency officials told us that they believe that the FIRREA comparability
provision and similar provisions in subsequent laws applicable to
financial regulators have been effective in ensuring that regulators' pay
and benefits are generally comparable among the 10 agencies, which
probably helps minimize employee movement among financial regulatory
agencies. Of the financial regulator employees who moved or resigned, the
percentage of those who resigned from federal employment fluctuated
slightly over the period, ranging from a low of 73.7 percent in fiscal
year 2003 to a high of 94.8 percent in fiscal year 1996.

^29The analysis of employee movement does not include data for the Federal Reserve Board,
because the CPDF does not include data for the agency.

Figure 4: Attrition among Financial Regulators, Fiscal Years 1990-2006

Note: This figure does not include data for the Federal Reserve Board.

The movement of mission-critical employees among financial regulators also
did not reveal a discernible trend. For the number of employees who moved
to another financial regulator from fiscal year 1990 through fiscal year
2006, see table 10 in appendix V. The numbers ranged from no movement for
7 of the 11 occupational categories (accountants, auditors, business
specialists, economists, financial analysts, investigators, and
information technology (IT) specialists) in at least 1 of the fiscal years
we reviewed to a high of 37 employees (38.1 percent of all those who moved
that fiscal year) for the "all other" occupational category in fiscal year
1991.^30 During this period (fiscal years 1990 to 2006), some occupational
categories experienced very little movement. For example, fewer
accountants, auditors, business specialists, and investigators moved than
employees in the other categories. In contrast, examiners had the largest
number of employees moving among financial regulators in 8 of the 17
years, including the 3 most recent years for which data were available.
The average number of employees in mission-critical occupations moving
among the 9 financial regulators from fiscal year 1990 through fiscal year
2006 ranged from 0.1 for investigators to 11.7 for examiners. See appendix
V for additional data on employee movement.

For those employees that did not move to another financial regulator, we
could not determine in all cases where the employees moved because CPDF,
the most complete data set available with federal employment information,
does not include information on employment outside executive branch
agencies. We were able to identify those employees that went to another
federal agency. These numbers ranged from a low of 48 in fiscal year 1994
to a high of 128 in fiscal year 1991, higher than the number of employees
who moved to another financial regulator, which was 23 in fiscal year 1994
and 97 in fiscal year 1991. Officials from the 9 agencies told us that
they do not track the employment of their employees after the employees
leave their agencies. Further, they said that their employees generally
sought employment outside the federal government, including the private
sector and state and local government, but that their main competitors
were private-sector entities.

^30The ï¿½all otherï¿½ category combines specialists in occupations such as human resources
management, administration, clerical, management and program analysis, financial
administration, and paralegal work. The three occupational categories with employee
movement were attorney, examiner, and supervisor.

Conclusions

Like other federal agencies, the experiences of the financial regulators
illustrate the challenges inherent in establishing well-functioning,
performance-based pay systems and that these systems are works in progress
that are constantly evolving. These regulators have taken various
approaches to revise their performance management systems and introduce
performance-based pay. Although the regulators have incorporated many of
the key practices for effective performance management systems,
opportunities exist for a number of them to make improvements as they
continue to refine their systems. Specifically, some regulators have
opportunities in the areas of strengthening safeguards to enhance
transparency and fairness and making meaningful distinctions in
performance. As some regulators develop new systems or revise their
existing systems, they have an opportunity to build in aspects of the key
practices, such as improving transparency by communicating the overall
results of performance appraisal rating and performance-based pay increase
decisions to all employees to help employees understand how they performed
relative to other employees in their organization, while protecting
individual confidentiality. For regulators that negotiate with unions,
there are also opportunities to work together to accomplish this.

SEC has some additional opportunities to pursue improvements in specific
aspects of its performance management system, which it is in the process
of revamping. For example, SEC can establish and communicate to
nonexecutive employees using the new system clear criteria for making
performance rating and pay increase decisions. Finally, while it may have
been an isolated incident, by not completing performance assessments of
senior officers in the 2005 performance appraisal cycle we reviewed, SEC
missed an opportunity for two-way feedback and assessments of individual
and organizational progress toward organizational goals. While funding
circumstances specific to that appraisal cycle contributed to this
situation, in the future it will be important to complete assessments
regardless of the availability of funding for increases.

The agencies have taken a variety of actions in seeking to maintain pay
and benefits comparability. While we did find some variation in base pay
ranges, locality pay percentages, actual average pay, and benefits among
the agencies, we found that a number of reasons could contribute to the
variation, including the following: regulators were granted flexibility
under Title V and subject to comparability requirements at varying times,
pay and benefits are considered comprehensively in seeking comparability,
the average length of service of employees, and where employees are
located. While pay and benefits comparability cannot be precisely
determined, all the agencies are working to maintain comparability in pay
and benefits. One recent initiative--studying the feasibility of
conducting a common survey on pay and benefits--should help to increase
the efficiency of this effort. In addition, given the relatively small
amount of employee movement among federal regulators, the variation in
pay, benefits, and locality pay percentages in some locations across the
regulators does not appear to be encouraging large numbers of employees to
move among financial regulators. This may be an indication that the
comparability provisions of FIRREA and other pertinent legislation have
been working as intended. Moreover, from fiscal years 1990 through 2006,
the agencies' attrition rates have trended downward indicating that a
smaller percentage of employees were leaving.

Recommendations for Executive Action

The Chairman of the Board and Chief Executive Officer of the Farm Credit
Administration, the Chairman of the Federal Housing Finance Board, and the
Director of the Office of Federal Housing Enterprise Oversight should

ocommunicate the overall results of the performance appraisal and pay
increase decisions to all employees agencywide while protecting individual
confidentiality.

The Chairman of the National Credit Union Administration and the Chairman
of the Commodity Futures Trading Commission should

owork with unions to communicate the overall results of the performance
appraisal and pay increase decisions to all employees agencywide while
protecting individual confidentiality.

The Chairman of the Securities and Exchange Commission should

ocommunicate clearly the criteria for making performance rating and pay
increase decisions to nonexecutive employees

owork with the union to communicate the overall results of the performance
appraisal and pay increase decisions to all employees agencywide while
protecting individual confidentiality and

oassess senior executives' performance at the end of the performance
appraisal cycle regardless of the amount of funding available for
performance-based pay increases.

Agency Comments and Our Evaluation

We provided drafts of this report to the Chairman, Commodity Futures
Trading Commission; Chairman of the Board and Chief Executive Officer,
Farm Credit Administration; Chairman, Federal Deposit Insurance
Corporation; Chairman, Federal Housing Finance Board; Chairman, Board of
Governors of the Federal Reserve System; Chairman, National Credit Union
Administration; Comptroller of the Currency, Office of the Comptroller of
the Currency; Director, Office of Federal Housing Enterprise Oversight;
Director, Office of Thrift Supervision; and Chairman, Securities and
Exchange Commission; for review and comment.

We received written comments from six of the agencies. See appendixes VI,
VII, VIII, IX, X, and XI for letters received from CFTC, the Federal
Reserve Board, FHFB, NCUA, OFHEO, and SEC. These six, along with the other
four agencies, also provided clarifying and technical comments, which we
incorporated as appropriate.

The agencies generally agreed with our recommendations. With respect to
the recommendation to communicate the overall results of the performance
appraisal and pay increase decisions on an agency-wide basis, CFTC, FCA,
FHFB, NCUA, OFHEO, and SEC indicated that they plan to implement the
recommendation. In describing specific actions, the executive director of
CFTC explained that the agency has already discussed working with the
unions to communicate overall results of performance appraisal and pay
decisions across the agency as part of the development of their new
performance management and pay-for-performance systems. The Chief Human
Capital Officer of FCA stated that the agency plans to communicate the
overall results of the 2006 performance appraisal and 2007 pay increase
decisions to FCA employees by the end of June 2007. The Executive Director
of NCUA explained that sharing overall information on ratings and pay
increase decisions with all employees is one of the issues being
negotiated as part of the ongoing negotiations over pay and benefits with
the National Treasury Employees Union, and stated that the agency's
proposal to the union provides for this type of transparency. The
Executive Director of SEC agreed with the report findings and stated that
SEC has established a new branch within the Office of Human Resources to
oversee performance-related issues and has launched a new pilot
performance management system that will address the recommendations.
Finally, the Acting Director of FHFB and the Chief Human Capital Officer
of OFHEO also stated that their respective agencies will implement the
recommendation.

We will send copies of this report to the appropriate congressional
committees; the Chairman, Commodity Futures Trading Commission; Chairman
of the Board and Chief Executive Officer, Farm Credit Administration;
Chairman, Federal Deposit Insurance Corporation; Chairman, Federal Housing
Finance Board; Chairman, Board of Governors of the Federal Reserve System;
Chairman, National Credit Union Administration; Comptroller of the
Currency, Office of the Comptroller of the Currency; Director, Office of
Federal Housing Enterprise Oversight; Director, Office of Thrift
Supervision; Chairman, Securities and Exchange Commission; and other
interested parties. We will make copies available to others upon request.
The report will also be available at no charge on our Web site at h 
ttp://www.gao.gov.

If you or your staff have any questions regarding this report, please
contact Orice M. Williams at (202) 512-8678 or [email protected] or
Brenda Farrell at (202)512-5140 or [email protected] . Contact
points for our Offices of Congressional Relations and Public Affairs may
be found on the last page of this report. GAO staff who made major
contributions to this report are listed in appendix XII.

Orice M. Williams, Director
Financial Markets and Community Investment

Brenda Farrell, Acting Director
Strategic Issues

List of Congressional Requesters

The Honorable Barney Frank
Chairman
The Honorable Spencer Bachus
Ranking Member
Committee on Financial Services
House of Representatives

The Honorable Luis V. Gutierrez
Chairman
Subcommittee on Domestic and International Monetary Policy, Trade,
and Technology
Committee on Financial Services
House of Representatives

The Honorable Paul E. Kanjorski
Chairman
Subcommittee on Capital Markets, Insurance, and Government
Sponsored Enterprises
Committee on Financial Services
House of Representatives

The Honorable Bernard Sanders
United States Senate

Appendix I:  Objectives, Scope, and Methodology

The objectives of this report were to (1) review how the performance-based
pay systems of 10 federal financial regulatory agencies are aligned with
six key practices for effective performance management systems, (2) review
actions these 10 agencies have taken to assess and implement comparability
in compensation, and (3) review the extent to which individuals in
selected occupations have moved between or left any of the agencies. These
agencies are the Commodity Futures Trading Commission (CFTC), the Farm
Credit Administration (FCA), the Federal Deposit Insurance Corporation
(FDIC), the Federal Housing Finance Board (FHFB), the Board of Governors
of the Federal Reserve System (the Federal Reserve Board), the National
Credit Union Administration (NCUA), the Office of the Comptroller of the
Currency (OCC), the Office of Federal Housing Enterprise Oversight
(OFHEO), the Office of Thrift Supervision (OTS), and the Securities and
Exchange Commission (SEC).

To address our first objective, we analyzed documents on the regulators'
performance management and pay systems, including guidance, policies, and
procedures on the systems; performance planning and appraisal forms; union
contracts and agreements; training materials; internal evaluations of
systems; and materials used to communicate with employees about the
systems. We also reviewed documents assessing the agencies' systems,
including results from the 2006 Federal Human Capital Survey conducted by
the Office of Personnel Management (OPM), recent human resources
operations audits performed by OPM, and relevant material from agencies'
offices of inspector general.

We also interviewed key human resources officials at each agency, as well
as officials from other functional areas knowledgeable about each agency's
performance-based pay practices. In addition, we interviewed employees at
the agencies who served as members of employee groups. At six of the
agencies, the employees we spoke with were union representatives.
Specifically, employees at FDIC, OCC, NCUA, and SEC are represented by the
National Treasury Employees Union, and OTS headquarters staff and CFTC
staff at two regional offices are represented by the American Federation
of Government Employees. Employees at FCA, the Federal Reserve Board,
FHFB, and OFHEO did not have a union; at these agencies we spoke with
employees who served on employee committees or working groups.

In addition, we examined small, select sets of individual performance
plans for employees, which outline the annual performance expectations for
employees. The selection of these performance plans was not intended to
allow us to make generalizations about all performance plans at the
agencies, and we have used information from the plans for illustrative
purposes only. The performance plans we received were selected by agency
officials based on our request for a mix of occupations and grade levels
at each agency. The smallest number of performance plans we examined from
an agency was one, in a case where the performance plans for all employees
are completely standardized. The largest number of plans we reviewed from
an agency was 32. The individual performance plans we examined pertained
to each agency's last completed performance appraisal cycle when we began
this review. Table 3 shows the appraisal cycle by agency.

Table 3: Performance Appraisal Cycle by Agency

                                        

               Agency                     Performance appraisal cycle         
CFTC                           July 1, 2005 to June 30, 2006               
FCA                            October 1, 2004 to September 30, 2005       
FDIC                           January 1, 2005 to December 31, 2005        
Federal Reserve Board          October 1, 2004 to September 30, 2005       
FHFB                           October 1, 2005 to September 30, 2006       
NCUA                           January 1, 2005 to December 31, 2005        
OCC                            October 1, 2004 to September 30, 2005       
OFHEO                          April 1, 2005 to March 31, 2006             
OTS                            January 1, 2005 to December 31, 2005        
SEC:                           
for senior officers            October 1, 2004 to September 30, 2005       
for all other employees        May 1, 2005 to April 30, 2006               

Source: Various regulators.

Finally, we analyzed data from each agency on performance ratings and
performance-based pay awarded to employees as well as aggregate data on
all types of pay increases at each agency not linked to performance
ratings. We used these data to calculate the Spearman rank correlation
coefficient to show the strength of the relationship between employee
performance ratings and the associated performance-based percentage pay
increases at
each agency.^1 In computing the correlation coefficients, we noted that a
few agencies used a table or procedure that specified particular pay
increases corresponding to specific ratings. Taken in isolation, the use
of the table or procedure would be expected to produce a perfect
correlation, i.e., +1.0. However, other aspects of these agencies' systems
contributed to the resulting coefficients being less than +1.0. For
example, at one agency, employees with rating scores below a certain
threshold were not eligible for any pay increase. While these employees
may have had different rating scores, none of them received a pay
increase, which contributed to a coefficient that was less than perfect.
Other mechanistic factors in these agencies' systems, such as adjusting or
changing the specified percentage pay increase based on the grade level or
current salary of the employee, also had the effect of producing a less
than perfect coefficient at these agencies. Given the influence that these
procedural but nondiscretionary variations may have had on the resulting
coefficients at these agencies, the coefficients are primarily useful in
their overall demonstration of the positive linkage between ratings and
pay increases at all the agencies, and the range of coefficients that
occurs. The magnitude of the coefficients, however, is not sufficient for
ranking the agencies or making other types of comparisons.

^1The performance appraisal and pay systems at FDIC and SEC were sufficiently different
from those of the other eight agencies that a correlation coefficient would not be a useful
descriptor of the relationship between ratings and pay increases at these two agencies. Both
FDIC and SEC use two-phase performance management systems that preclude the
meaningful use of a single correlation coefficient. Neither the first or second processes in
their systems are sufficiently comparable to the multilevel performance rating scales and
pay increase determinations used at the other financial regulators. In the first process at
FDIC and SEC, almost all employees were rated as ï¿½acceptableï¿½ on what was essentially a
two-level, ï¿½acceptable/unacceptableï¿½ rating scale. The lack of variation in performance
ratings given in the first process would show little association with pay increases, and
computing a correlation for this step would yield a coefficient at or around zero. In the
second process, which was not considered to be a performance rating at these agencies,
employees were assigned to performance groups at FDIC or awarded steps at SEC that were
associated with specified percentage increases in pay. Computing a correlation between
group assignment or steps awarded and percentage of pay increase would yield a perfect
correlation since group or step assignment was synonymous with a specific percentage
increase in pay.

We also analyzed agency data on performance ratings to determine the
distribution of employee performance ratings at each agency. All data were
provided to us by agency officials, and pertained to the performance
appraisal cycles noted in table 3.

To address our second objective, we first analyzed the content of
compensation comparability provisions in the agencies' laws and related
legislative histories. We reviewed the most recent pay and benefits
surveys conducted by external compensation consultants for these agencies,
obtained agency pay and benefits data, and analyzed actual pay data from
CPDF. In addition, we interviewed agency officials about their experience
with these surveys and the agencies' informal interactions to assess pay
comparability and to determine the feasibility of conducting a common
survey.

To report on the pay ranges for non-executive employees in selected
occupations, we analyzed the base pay data provided to us for mission-
critical occupations at nine of the agencies in our review. We selected
the mission critical occupations by: (1) identifying nonclerical and
nonblue-collar occupations with 45 or more employees in at least one
financial regulatory agency and (2) vetting this list with the 10
agencies.^2 The agencies provided us with pay range information as set
forth in each agency's pay policies as of September 2006 for every job
title under each occupational category, including jobs with no incumbents
at the time the agencies reported the data to us.

^2OPMï¿½s six occupational categories are Professional, Administrative, Technical, Clerical,
Other White-Collar, and Blue Collar, collectively known as ï¿½PATCOB.ï¿½ When we sent the list
of mission critical occupations to the agencies, we listed the occupational title and asked
the agency to identify the OPM job series number they used for that occupation. For some
occupations, all agencies used the same job series (for example, 0905 for attorneys). For
other occupations, some agencies used different job series for the same or similar
occupation (for example, agencies used several agency-unique job series for ï¿½examinersï¿½
such as 1831, 0580, and 0570). Two agencies (CFTC and OFHEO) placed employees in the
same job series (1801 for CFTC and 0501 for OFHEO) into two different occupations
(examiners and investigators, and examiners and financial analysts, respectively). In both
cases, we placed all employees in the examiner occupation because the CPDF does not
facilitate separating employees with the same job series into separate occupations. For a
few occupations, some agencies used an OPM job series number that we did not expect (for
example, OFHEO used 0301 for information technology specialists and FCA used 1101 for
examiners). When we analyzed CPDF data, we used the occupational titles and job series
numbers the agencies provided to us. Executives were not included in the pay analyses. We
excluded executives from the analysis of average actual pay and pay scale, because we
wanted to focus on mid-level management.

To report on the actual average base pay of employees in the selected
occupations, we analyzed actual pay data from CPDF for fiscal year 2006.
Because the CPDF does not include data for all agencies, the Federal
Reserve Board provided us with actual pay data for our analysis of its
employees' actual average pay for fiscal year 2006. To show the financial
regulators' locality pay percentages and general schedule employees'
locality pay percentages, we selected the cities where four or more
financial regulators had duty stations in fiscal year 2006. We obtained
fiscal year 2006 locality pay percentages information from the financial
regulators and general schedule locality pay percentages from the OPM Web
site. To report on the benefits offered by the agencies, we obtained and
analyzed data from each agency that included a list of benefits the
financial regulators offered as of September 2006 and brief descriptions
of each benefit. We also interviewed agency officials about the factors
that affect the actual average base pay, and how each agency sets its
locality pay percentage.

To address our third objective, we analyzed movement data from CPDF for
fiscal years 1990 to 2006, the most recent available data as of December
2006. For each fiscal year, we identified the number of employees in
selected mission-critical occupations at a financial regulator who (1)
moved to another financial regulator, (2) moved to other federal agencies,
and (3) resigned from federal employment. We identified those who moved
from one financial regulatory agency to another by identifying employees
who had a CPDF separation code for a voluntary transfer and who also had a
CPDF accession code from another financial regulatory agency within 25
days of the transfer out. Also, for each mission-critical occupation, we
examined the number of financial regulator employees who moved to another
financial regulator in each fiscal year and the average number of
employees who moved among the nine financial regulators over the 16 years
of our review. Our analysis of supervisors included executives, who
constituted 1 to 2 percent of all supervisors who moved to another
financial regulator. We also included all other agency occupations that
were not classified as "mission-critical occupations" in an "all other"
category, which includes occupations such as specialists in human
resources management, administration, clerical, management and program
analysis, blue collar occupations, financial administration, and paralegal
work.

We did not include the Federal Reserve Board in our analysis of the
movement of financial regulator employees because CPDF does not include
data on the Federal Reserve Board. Federal Reserve Board officials told us
that data on employee movement for fiscal years 1990 to 1996 are not
readily accessible. The agency provided us some data for fiscal years 1997
to 2005, including data on employees who transferred, resigned, were
fired, were subject to a reduction in force, or otherwise separated, and
the agency's total number of employees, but was unable to identify whether
their employees left for another financial regulator. Because the data the
agency provided were not comparable with the CPDF data we used for the
other financial regulators, we did not include the Federal Reserve Board
in our analysis. We also did not include information on the employment of
financial regulatory employees after they left federal employment because
CPDF does not include data on employment outside some agencies and
officials told us that they do not track the employment of their employees
after the employees leave their agencies.^3

^3We were not able to always find a transfer-in personnel action to match a transfer-out
personnel action within the time frames stipulated for transfers. Some transfer-in personnel
actions that we did not find in fiscal year 1990 could be due to the fact that the transfer-out
personnel action occurred in fiscal year 1989. Similarly, some of the transfer-out personnel
actions in fiscal year 2006 for which we did not find a matching transfer-in personnel action
could be due to the fact that they occurred in fiscal year 2007.

We assessed the reliability of the various sets of data used in our study.
To assess the reliability of the performance and pay increase data
provided by the agencies, we conducted various inspections and electronic
testing of the data for reasonableness and the presence of any obvious or
potential errors in accuracy and completeness. We also reviewed related
agency documentation, interviewed agency officials knowledgeable about the
data, and brought to the attention of these officials any concerns or
discrepancies we found with the data for correction or updating. Based on
the results of these procedures, we believe the data are sufficiently
reliable for use in the analyses presented in this report. We did not
independently verify the pay and benefits data we received from the
agencies but consider these data sufficiently reliable for the
illustrative purpose of our review. Based on our data reliability testing
of CPDF data, we believe the CPDF data are sufficiently reliable for this
review. When analyzing employee movement using CPDF data, we found
exceptions from standard personnel procedures, such as employees with a
transfer-out code but with an accession code in the hiring agency that was
not a transfer-in code, or employee records with transfer-out and
transfer-in dates that exceeded 3 calendar days. We also found duplicate
separation or accession records for the same individual on the same day.
We deleted one of the duplicate records. We also found cases where an
individual had two separation actions on the same day but they were
different types of actions (e.g., a transfer out and a resignation).
Because we could not determine which separation action was the correct
one, we deleted both records. However, these types of data problems
represented less than one-tenth of 1 percent of the data used. As a
result, we concluded that the data were sufficiently reliable to show the
magnitude of movement between financial regulatory agencies, to other
federal agencies, and to nonfederal employers.

We conducted our work from February 2006 through June 2007 in accordance
with generally accepted government auditing standards.

Appendix II:  Information on Agencies 

Table 4: Agency Information

                                        

    Agency        Mission       Number of      Funding^b        Employee    Performance-based 
                              employees^a                    representation        pay        
Commodity      To protect             446  Appropriated      The American   In October 2006,  
Futures        market users                                  Federation of  CFTC began        
Trading        and the public                                Government     transitioning to  
Commission     from fraud,                                   Employees      a new             
(CFTC)         manipulation,                                 represents     performance-based 
               and abusive                                   employees in   pay system that   
               practices                                     CFTC's Chicago will be fully     
               related to the                                and New York   implemented       
               sale of                                       offices only.  beginning in July 
               commodity and                                                2007.^c           
               financial                                                                      
               futures and                                                                    
               options, and                                                                   
               to foster                                                                      
               open,                                                                          
               competitive,                                                                   
               and                                                                            
               financially                                                                    
               sound                                                                          
               commodity                                                                      
               futures and                                                                    
               option                                                                         
               markets.                                                                       
Farm Credit    To promote a           248  Nonappropriated^d FCA is not     Performance-based 
Administration safe, sound,                                  unionized. FCA pay began in      
(FCA)          and dependable                                has an         1993; the system  
               source of                                     employee       has had some      
               credit and                                    council.       revisions since   
               related                                                      its inception.    
               services for                                                                   
               agriculture                                                                    
               and rural                                                                      
               America.                                                                       
Federal        To maintain           4328  Nonappropriated   The National   Performance-based 
Deposit        stability and                                 Treasury       pay began in      
Insurance      public                                        Employees      1998. FDIC has    
Corporation    confidence in                                 Union has the  essentially two   
(FDIC)         the nation's                                  right to       pay and           
               financial                                     bargain over   performance       
               system by                                     employees' pay management        
               insuring                                      and benefits   systems -- one    
               deposits,                                     at FDIC.       that applied to   
               examining and                                                senior managers,  
               supervising                                                  and one that      
               financial                                                    applied to        
               institutions,                                                bargaining unit   
               and managing                                                 employees as well 
               receiverships.                                               as nonbargaining  
                                                                            unit employees.^e 
Federal        To ensure that         122  Nonappropriated   FHFB is not    The               
Housing        the 12 federal                                unionized.     performance-based 
Finance Board  home loan                                     FHFB has an    pay system began  
(FHFB)         banks are safe                                employee       in the mid 1990s; 
               and sound so                                  working group. the system has    
               they serve as                                                been slightly     
               a reliable                                                   revised since its 
               source of                                                    inception.        
               liquidity and                                                                  
               funding for                                                                    
               the nation's                                                                   
               housing                                                                        
               finance and                                                                    
               community                                                                      
               investment                                                                     
               needs.                                                                         
Board of       To foster the         1855  Nonappropriated   The Board is   The               
Governors of   stability,                                    not unionized. performance-based 
the Federal    integrity, and                                The Board has  pay system began  
Reserve System efficiency of                                 an employee    in 1989; the      
               the nation's                                  representative system has been   
               monetary,                                     committee.     slightly revised  
               financial, and                                               since its         
               payment                                                      inception.        
               systems so as                                                                  
               to promote                                                                     
               optimal                                                                        
               macroeconomic                                                                  
               performance.                                                                   
National       To facilitate          919  Nonappropriated   The National   The               
Credit Union   the                                           Treasury       performance-based 
Administration availability                                  Employees      pay system began  
(NCUA)         of credit                                     Union has the  in 1991.          
               union services                                right to                         
               to all                                        bargain over                     
               eligible                                      the impact and                   
               consumers,                                    implementation                   
               especially                                    of changes                       
               those of                                      made to NCUA's                   
               modest means,                                 performance                      
               through a                                     management                       
               regulatory                                    system.                          
               environment                                                                    
               that fosters a                                                                 
               safe and sound                                                                 
               federally                                                                      
               insured credit                                                                 
               union system.                                                                  
Office of the  The OCC was           2908  Nonappropriated   The National   Performance-based 
Comptroller    created by                                    Treasury       pay began in      
of  the        Congress to                                   Employees      1981; the current 
Currency (OCC) charter                                       Union has the  performance       
               national                                      right to       management system 
U.S.           banks, to                                     bargain over   began in 2001,    
Department of  oversee a                                     the impact and although some     
the Treasury   nationwide                                    implementation revisions have    
               system of                                     of changes     been made since   
               banking                                       made to OCC's  then.             
               institutions,                                 performance                      
               and to assure                                 management                       
               that national                                 system, but                      
               banks are safe                                not over                         
               and sound,                                    employee pay                     
               competitive                                   and benefits.                    
               and                                                                            
               profitable,                                                                    
               and capable of                                                                 
               serving in the                                                                 
               best possible                                                                  
               manner the                                                                     
               banking needs                                                                  
               of their                                                                       
               customers.                                                                     
Office Of      To promote             204  Nonappropriated^d OFHEO is not   Performance-based 
Federal        housing and a                                 unionized.     pay has existed   
Housing        strong economy                                OFHEO has      since the         
Enterprise     by ensuring                                   established an agency's          
Oversight      the safety and                                ad hoc         inception in      
(OFHEO)        soundness of                                  employee       1992.             
               Fannie Mae and                                working group.                   
               Freddie Mac                                                                    
               and fostering                                                                  
               the strength                                                                   
               and vitality                                                                   
               of the                                                                         
               nation's                                                                       
               housing                                                                        
               finance                                                                        
               system.                                                                        
Office of      To supervise           956  Nonappropriated   The American   Performance-based 
Thrift         savings                                       Federation of  pay began in      
Supervision    associations                                  Government     1991.             
(OTS) U.S.     and their                                     Employees                        
Department of  holding                                       represents                       
the Treasury   companies in                                  some employees                   
               order to                                      in OTS's                         
               maintain their                                Washington,                      
               safety and                                    D.C., office                     
               soundness and                                 only.                            
               compliance                                                                     
               with consumer                                                                  
               laws, and to                                                                   
               encourage a                                                                    
               competitive                                                                    
               industry that                                                                  
               meets                                                                          
               America's                                                                      
               financial                                                                      
               services                                                                       
               needs.                                                                         
Securities and To maintain           3488  Appropriated      The National   Performance-based 
Exchange       fair, orderly,                                Treasury       pay began in      
Commission     and efficient                                 Employees      2002.             
(SEC)          securities                                    Union                            
               markets,                                      represents                       
               facilitate                                    two-thirds of                    
               capital                                       SEC employees.                   
               formation, and                                                                 
               protect                                                                        
               investors.                                                                     

Source: Various regulators

^aEmployment figure for the Federal Reserve Board is from December 2006
and includes all regular employees; employment figures for all other
agencies are for career employees and come from the CPDF as of September
2006.

^bSome financial regulators receive funding through appropriations from
Congress while others are funded from fees collected from members or
assessed on regulated entities.

^cCFTC did not have a performance-based pay system prior to October 2006.

^dThe business operations of FCA and OFHEO are not financed by taxpayer
funds. Their annual operating budgets, however, undergo the federal
budgetary and appropriations process and are constrained by the amount
approved by Congress and signed into law by the President.

^eThe nonbargaining unit employees are never "covered" by the compensation
agreement with the union, but rather, the FDIC Board proactively decides
annually what performance management and performance-based pay standards
will apply to this population. During the performance appraisal cycle we
reviewed, the same system that applied to bargaining unit employees was
applied to nonbargaining unit employees.

Appendix III:  Financial Regulators Have Implemented Key Practices in Varying
Ways

High-performing organizations have recognized that a critical success
factor in fostering a results-oriented culture is a performance management
system that creates a "line of sight" showing how team, unit, and
individual performance can contribute to overall organizational goals and
helping employees understand the connection between their daily activities
and the organization's success. Effective performance management systems
are essential for successfully implementing performance-based pay. In the
letter, we addressed important aspects of how 10 financial regulatory
agencies have implemented two key practices: (1) linking pay to
performance and (2) making meaningful distinctions in performance. This
appendix provides detailed information on the financial regulators'
implementation of four additional key practices important for effective
performance management systems, as well as some additional material
pertaining to the linking pay to performance practice covered in the
letter. The four additional practices are:

oAlign individual performance expectations with organizational goals.

oConnect performance expectations to crosscutting goals.

oUse competencies to provide a fuller assessment of performance.

oInvolve employees and stakeholders to gain ownership of performance
management systems.

The 10 financial regulatory agencies have implemented these four key
practices for effective performance management systems in various ways,
reflecting the unique needs of their organizational cultures and
structures.

Agencies Have Aligned Individual Performance Expectations with
Organizational Goals in Different Ways

The 10 federal financial regulatory agencies have implemented the practice
of alignment in a variety of ways. An explicit alignment of daily
activities with broader results is a key feature of effective performance
management systems in high-performing organizations. These organizations
use their performance management systems to improve performance by helping
individuals see the connection between their daily activities and
organizational goals and encouraging individuals to focus on their roles
and responsibilities in helping to achieve these goals. The financial
regulators reinforced alignment of individual performance expectations to
organizational goals in policy and guidance documents for their
performance management systems, used standardized performance elements or
standards for employees in their performance plans, used customized
individual performance expectations that contributed to organizational
goals in individual performance plans, and included the corresponding
organizational goals directly on the individual performance plan forms.^1

Agencies Have Reinforced Alignment in Policies and Guidance for
Performance Management Systems

Several of the financial regulatory agencies, including FDIC, OCC, FHFB,
and OFHEO, have reinforced alignment by including language on linking
individual performance expectations to organizational goals in policy and
guidance materials for the performance management systems. The following
are examples of how selected agencies have reinforced alignment through
policies and guidance.

oA key objective of FDIC's performance management program as stated in a
policy directive is to "establish fair and equitable performance
expectations and goals for individuals that are tied to accomplishing the
organization's mission and objectives."^2 The directive further states
that employees at FDIC are assessed against performance criteria, which
are defined as "the major goals, objectives, and/or primary
responsibilities of a position which contribute toward accomplishing
overall organizational goals and objectives" (as found in FDIC's strategic
plan and annual performance plan).

^1According to OPM, performance elements identify the activities, skills, or responsibilities
that the employee is expected to achieve during the year and performance standards
identify how well the employee must meet each performance element to receive a specific
performance rating.

^2Federal Deposit Insurance Corporation, Performance Management Program, Directive
System Circular 2430.1, Mar. 28, 2002.

oAt OCC, the Policies and Procedures Manual for the performance management
system states that the system is designed to align employee performance
expectations with organizational objectives and priorities. The manual
also explains that the starting point for identifying individual
performance expectations should be unit objectives established at the
executive committee, district, field office, or division level.

oThe handbook and guide for FHFB's and OFHEO's performance management
systems, respectively, contain several references to alignment of
individual expectations to organizational goals.

Agencies Have Included Alignment in Standardized Performance Elements for
Employees

Several of the financial regulators, including FCA, CFTC, FHFB, OCC and
OTS, have reinforced alignment by including standardized performance
elements or performance standards that link performance expectations to
organizational goals in employees' performance plans. We have previously
reported that results-oriented performance agreements can be effective
mechanisms to define accountability for specific goals and to align daily
activities with results.^3 Individuals from the agencies with standardized
performance elements in their individual performance plans are assessed
against the same set of performance elements and standards at the end of
the appraisal cycle, as the following examples illustrate.

^3GAO-04-614; GAO, Managing for Results: Emerging Benefits From Selected Agenciesï¿½ Use
of Performance Agreements, GAO-01-115, Washington, D.C.: (Oct. 30, 2000).

oFCA has included a requirement to contribute to the achievement of
organizational goals in standardized performance elements for all
employees in their individual performance plans. Specifically, FCA has
developed a set of standardized performance elements for each of its four
occupational groups and in some of these elements, requires individuals to
contribute to achieving organizational goals and objectives. For the
senior manager's occupational group, individuals have a standardized
performance element--"Leadership and Motivation Skills"--in their
individual performance plans that measures the employees' ability to
accomplish the agency's goals and objectives. For the other three
occupational groups, individuals have a standardized performance
element--"Teamwork and Interpersonal Skills"--in their individual
performance plans that measures the extent to which the employee places
emphasis on achieving organizational and team goals. In this way, all
employees at FCA are assessed on the extent to which they contribute to
organizational objectives through a standardized performance element.

oWhile not requiring a standardized performance element related to
alignment in the individual performance plans for all employees, CFTC has
reinforced alignment through the performance standards used for rating all
employees at the end of the performance appraisal cycle. Specifically, in
order for all employees to achieve the highest summary performance rating,
individuals must "achieve element objectives with extensive impact on
organizational mission," which reinforces the line of sight between
individual performance and organizational results. In this way, for all
employees at CFTC, the individual's contributions to organizational goals
affect his or her ability to achieve the highest possible performance
rating. Alignment is further reinforced for managerial employees at CFTC
because they are also assessed on the standardized performance element of
"Effective Leadership," which requires them to, among other things,
accomplish the mission and organizational goals of the work unit, and
communicate organizational goals to subordinates.

oFHFB has reinforced alignment in standardized performance elements for
several occupational groups. Standardized elements for executives,
managers/supervisors, staff attorneys, and professional positions contain
references to aligning with or contributing to organizational goals.

oOCC has applied an alignment focus in a generic performance standard for
four occupational groups at the agency. Executives, managers, commissioned
examiners, and specialists are all rated against a standardized
performance standard that requires them to contribute to organizational
goals in order to get the highest rating level of 4 for a particular
performance element. For example, managers have a standardized performance
element called "leadership skills," for which the highest level
performance standard includes language on meeting OCC goals and
objectives. Commissioned examiners and specialists have a standardized
performance element in their individual performance plans called
"organizational skills," with an accompanying performance standard that
requires individuals' work products to be closely aligned with OCC's
goals, objectives, and priorities in order to receive the highest rating
level.

oOTS has reinforced alignment in a standardized performance element for
managers and senior managers. Under the "Leadership Skills" standardized
performance element, managers are assessed on accomplishing the agency's
goals and objectives, taking initiative and incorporating organizational
objectives into the organization, and scheduling work assignments. In
addition, senior managers have a supplemental performance element that
holds them responsible for supporting the achievement of OTS's strategic
plan. An OTS official stated that the agency is considering expanding the
requirement for alignment as it makes future changes to the performance
management system.

Agencies Have Strengthened Alignment by Linking Customized Individual
Performance Expectations to Organizational Goals

Several financial regulatory agencies, including SEC, OCC, and the Federal
Reserve Board, have reinforced alignment for some individual employees
through customized performance expectations specific to individuals that
link to higher organizational goals. We have reported that high-performing
organizations use their performance management systems to improve
performance by helping individuals see the connection between their daily
activities and organizational goals and encouraging individuals to focus
on their roles and responsibilities to help achieve these goals. One way
to encourage this is to align performance expectations of individual
employees with organizational goals in individual performance plans. We
reviewed a small, select set of individual performance plans from each
agency, and identified the following examples of individual performance
expectations that linked to higher organizational goals.

oThe performance plan for a senior officer at SEC included the performance
expectation "Plans and Coordinates Inspection Programs and Ensures that
Internal Management Controls Exist and Operate Effectively" that supports
SEC's strategic goal to "Maximize the Use of SEC Resources."

oIn individual performance plans, OCC has used customized performance
expectations unique to the individual in addition to standardized
performance elements to appraise employees. Specifically, the performance
plan for an information technology (IT) specialist included a customized
expectation to provide timely, professional, and quality IT support to
promote efficient utilization of OCC resources. This expectation supported
the annual OCC objective--"OCC reflects an efficient and effective
organization."

oAt the Federal Reserve Board, a performance plan for an economist
contained a performance expectation to produce a weekly monitoring report
on Japan and cover Japanese banking and financial issues, which
contributed to one of the Board's annual performance objectives in the
area of monetary policy function: "contribute to the development of U.S.
international policies and procedures, in cooperation with the U.S.
Department of the Treasury and other agencies."

Agencies Have Strengthened Alignment by Stating Organizational Goals in
Individual Performance Plans

FHFB and OCC have reinforced the linkage between the individual's
performance expectations and organizational goals by including the
corresponding organizational goals directly on the individual performance
plan forms. This helps make clear the line of sight between the employee's
work and agency goals, as the following examples illustrate.

oFHFB has included the agency mission statement and office mission
statement to which an employee is contributing at the top of the first
page of the performance plan form.

oIn many of the individual performance plans we examined from OCC, the
annual OCC objective to which each customized performance element
contributed was listed on the form, along with performance measures.
According to an official, while OCC's performance management policy does
not specifically require that the higher organizational objective to which
each customized performance element contributes be listed on the
employee's performance evaluation form, managers are advised to include
the organizational goals and the majority of forms do include them. The
official stated that it was an oversight not to include this requirement
in the policy, and they plan to revise the performance evaluation form to
include space for the corresponding organizational objectives. Figure 5
shows an example of how a customized performance element on an individual
performance plan is linked to an agency goal, clarifying the relationship
between individual and organizational performance.

Figure 5: Excerpt from an OCC Commissioned Examiner's Individual
Performance Plan

Agencies Have Connected Performance Expectations to Crosscutting Goals in
Different Ways

The financial regulatory agencies have connected performance expectations
to crosscutting goals in several ways. As public sector organizations
shift their focus of accountability from outputs to results, they have
recognized that the activities needed to achieve those results often
transcend specific organizational boundaries. We reported that key
characteristics of high-performing organizations are collaboration,
interaction, and teamwork across organizational boundaries.^4
High-performing organizations use their performance management systems to
strengthen accountability for results, specifically by placing greater
emphasis on those characteristics fostering the necessary collaboration,
both within and across organizational boundaries, to achieve results.

The specific ways in which the financial regulatory agencies have
connected performance expectations to crosscutting goals vary. In our
review of a small, select set of performance plans from some of the
agencies, we identified some examples of customized individual performance
plans that identified crosscutting goals that would require collaboration
to achieve, as well as either the internal or external organizations with
which the individuals would collaborate to achieve those goals. All of the
agencies recognized the importance of collaboration by including
performance elements for collaboration or teamwork within and across
organizational boundaries in individual performance plans for at least
some employees. Several agencies applied standardized performance elements
related to teamwork or collaboration to employees.

Agencies Have Identified Crosscutting Goals and Organizations for
Collaboration in Individual Performance Plans

We found examples of performance plans customized to individuals at OCC,
FCA, the Federal Reserve Board, and SEC that identified crosscutting
goals, as well as either the internal or external organizations with which
the individuals would collaborate to achieve these goals. We have reported
that more progress is needed to foster the necessary collaboration both
within and across organizational boundaries to achieve results.^5 One
strategy for fostering collaboration is identifying in individual
performance plans specific programmatic crosscutting goals that would
require collaboration to achieve. Another strategy for fostering
collaboration is identifying the relevant internal or external
organizations with which individuals would collaborate to reinforce a
focus across organizational boundaries in individuals' performance plans,
as the following examples illustrate.

^4GAO, Human Capital: Managing Human Capital in the 21st Century, GAO/T-GGD-00-77
(Washington, D.C.: Mar. 9, 2000).

^5GAO-04-614.

oAt OCC, an employee had an expectation in his individual performance plan
to enhance the division's ability to work cooperatively and effectively
together with other operational risk divisions, as well as enhance
coordination with federal and state agencies and outside banking groups to
promote consistency and to advance OCC viewpoints, while contributing to
OCC's objective for U.S. and international financial supervisory
authorities to cooperate on common interests.

oA senior manager at FCA had a customized expectation in his individual
performance plan to work closely with and coordinate Office of Examination
initiatives with other offices, notably the Office of General Counsel and
Office of Public Affairs, to support the FCA Chairman and Chief Executive
Officer's three strategic goals, which are (1) improving communications
and relationships with the Farm Credit System, (2) gaining greater
efficiency and effectiveness of the agency, and (3) promoting the Farm
Credit System to become the Premier Financier of Agriculture and Rural
America.

oAn executive at the Federal Reserve Board had an expectation in his
individual performance plan to undertake expanded discussions with SEC on
information-sharing, cooperation, and coordination with the aim of
strengthening consolidated supervision and achieving consistency in the
implementation of Basel II.^6

^6Basel II is a set of proposed changes to the original set of risk-based capital rules based on
an internationally adopted framework developed by the Basel Committee. In the United
States, Basel II rules are intended to apply primarily to the largest and most internationally
active banking organizations.

oAt SEC, a senior officer in the market regulation division had an
expectation in his individual performance plan to advance market
regulation objectives through cooperative efforts by coordinating with
other SEC offices, other U.S. agencies, self-regulatory organizations,
international regulators, and the securities industry.

Agencies Have Included Performance Elements Related to Collaboration or
Teamwork in Individual Performance Plans

All of the financial regulators included performance elements related to
collaboration or teamwork within and across organizational boundaries in
individual performance plans for at least some of their employees.
Performance elements related to collaboration or teamwork in individual
performance plans can help reinforce behaviors and actions that support
crosscutting goals and provide a consistent message to all employees about
how they are expected to achieve results. CFTC, FHFB, NCUA, and the
Federal Reserve Board provide examples of how standardized performance
elements pertaining to teamwork or collaboration have been applied to
employees.

oCFTC has established a standardized performance element for all employees
that emphasizes collaboration or teamwork, called "Professional Behavior,"
which requires employees to behave in a professional and cooperative
manner when interacting with coworkers or the public and willingly
initiate and respond to collaborative efforts with coworkers, among other
things.

oAt FHFB, all employees have performance elements or standards related to
collaboration or teamwork in the standardized performance plans for their
occupational groups. For example, the standardized performance plan for
executives includes a performance element for "teamwork" that requires
executives to collaborate effectively with associates and promote positive
and credible relations with associates, among other things. The
standardized performance plan for administrative positions also includes a
"teamwork" performance element. For the other three occupational groups,
collaboration or teamwork is captured in a performance standard. For
example, the standardized performance plans for professional positions and
managers/supervisors have a performance element that emphasizes
collaboration or teamwork, called "Professionalism," which requires the
employee to develop and maintain effective working relationships with all
employees at all levels throughout the agency and external to the agency
and foster effective internal and external communication, among other
things.

oNCUA has performance elements related to collaboration or teamwork in the
standardized individual performance plans for some occupational groups,
such as examiners. For example, in the standardized performance plan for
some examiners, there is a performance element for "customer service and
teamwork" that requires the individual to demonstrate initiative,
responsibility, and accountability to both internal and external customers
and work in collaboration with coworkers and others toward common goals.
NCUA officials stated that a collaboration/teamwork performance element
may not be applicable to all positions. They also said that, to the extent
that this is an appropriate performance element on which an employee
should be rated, the agency has or will include it in that employee's
performance plan.

oAccording to Federal Reserve Board officials, the performance plans for
some occupations at the agency, such as security and administrative
positions, include teamwork as a standard element. Officials also said
that customized performance plans for other occupations typically include
teamwork or collaboration as a competency.

Agencies Have Used Competencies in Various Ways to Provide a Fuller
Assessment of Performance

All 10 of the financial regulatory agencies have used competencies, which
define the skills and supporting behaviors that individuals are expected
to demonstrate to carry out their work effectively. High-performing
organizations use competencies to examine individual contributions to
organizational results. We have reported that core competencies applied
organizationwide can help reinforce behaviors and actions that support the
organization's mission, goals, and values and can provide a consistent
message about how employees are expected to achieve results.^7 As
previously discussed, while some of the financial regulatory agencies have
included customized performance expectations specific to individuals in
performance plans, we found that all of the agencies have used
competencies. There are some variations in the ways in which the agencies
have structured and applied competencies to evaluate employee performance.
One of these variations concerns whether or not the agency has assigned
different weights to competencies when determining overall summary ratings
for individuals.

^7GAO-04-614.

Agencies Have Applied Competencies Organizationwide

With the exception of the Federal Reserve Board, all of the federal
financial regulatory agencies have developed sets of core competencies
that apply to groups of employees, and assess employee performance using
those competencies as part of the annual performance appraisal process.
Using competencies can help strengthen the line of sight between
individual performance and organizational success by reinforcing
performance expectations that support achievement of the agency's goals,
as the following examples illustrate.

oFCA has a different standardized performance plan for each of four
occupational groups of employees--senior managers, supervisors,
examiners/attorneys/analysts/other specialists (non-supervisory), and
administrative/technicians. Each of the plans includes a standard set of
competencies, called critical elements, which applies to all employees in
that group. Specifically, the performance plan for employees in the
examiners/attorneys/analysts/other specialists group contains the
following competencies--technical and analytical skills; organizational
and project management skills; teamwork and interpersonal skills; written
and oral communication skills; and equal employment opportunity (EEO),
diversity and other agency initiatives. A few sentences are included on
the performance plan form to describe what each element measures in terms
of the employee's knowledge, skills, and behavior, as shown in figure 6.

Figure 6: Excerpt from an FCA Individual Performance Plan

oFor the July 2005 to June 2006 performance appraisal cycle we reviewed at
CFTC, all employees were assessed on a set of five competencies, called
critical elements. Managerial employees were also assessed on three
additional competencies having to do with leadership, developing staff,
and supporting diversity and EEO programs.

oFDIC has 27 different performance plans with corresponding sets of
competencies, called performance criteria, to cover all employees.
According to agency officials, FDIC has learned from experience that
having a performance management system that is based on standardized sets
of competencies has allowed employees' performance to be compared more
easily to the standards from period to period. In addition, FDIC's system
bases merit pay increases for individuals at least partly on corporate
contributions (defined as contributions to corporate, division, or
unit-level goals). Officials said that this type of system really enhances
employee line of sight and has helped employees focus on how their
contributions align with the achievement of organizational goals. In their
view, this type of system promotes alignment and consistency more
effectively than a system of individual contracts between supervisors and
their employees.

oNCUA has approximately 240 detailed performance plans that are tailored
to specific occupations and grade levels of employees and that include
competencies, which are called elements. All of the employees to whom a
particular performance plan applies are assessed on the same set of
elements and performance standards. Elements for some employees within the
same occupation are universal, but standards can differ by grade level.
For example, the performance plans for examiners in grades 7, 11, and 12
all include basically the same elements, but some of the performance
standards upon which individuals are to be appraised for each element vary
by grade level.

oThe Federal Reserve Board differs from the other financial regulatory
agencies in the way it uses competencies. The agency does not have sets of
core competencies that apply to specified groups of employees across the
agency. Instead, divisions have latitude to vary the design and
implementation of the performance plan form and process. According to
agency officials, divisions select competencies that best suit
occupational types and the divisions' goals, because the Board has
multiple responsibilities dealing with monetary policy and financial
institution regulation. It is possible for employees in the same
occupational group, but in different divisions, to be rated against
different sets of competencies. Agency officials said that they have not
heard complaints from similar occupational groups that they may be
assessed against different competencies. Further, all officers, managers,
and supervisors are rated against the same four management objectives of
communications, staff development, effective planning and administration
of financial resources, and equal employment opportunity.

Agencies Have Assigned Weights to Competencies

A few of the agencies, such as OFHEO, FCA, and NCUA, allow differing
weights to be assigned to specific competencies when determining overall
summary performance ratings for individuals. Using weights enables the
organization to place more emphasis on selected competencies that are
deemed to be more important in assessing the overall performance of
individuals in particular positions. Other agencies, including OCC, OTS,
FDIC, CFTC, and FHFB, do not assign differing weights to competencies, as
the following examples illustrate.

oAt OFHEO, the rating official for each employee assigns a weight to each
of the competencies (called performance elements) included in the
individual's performance plan, in consultation with the reviewing
official. Each competency must have a minimum weight of at least 5, with
the total weight of all the competencies in an individual performance plan
equaling 100. Any competency with a weight of 20 or higher is considered
to be critical. Each competency element is weighted and scored (see figure
7), and then the weighted ratings for the competencies are summed to
derive the total summary rating for the individual.^8

^8According to an agency official, OFHEOï¿½s weighting system is expected to be modified as
revisions are made to the performance management system.

Figure 7: Example of OFHEO's Worksheet for Weighting Performance Elements

oFCA also permits supervisors to assign different weights to competencies
for individual employees, within the standardized performance plans, at
the beginning of the appraisal period. No competency can be weighted less
than 5 percent or more than 40 percent.

oAt NCUA, the elements for the various occupations and grade levels have
different weights assigned to them, depending on the priorities and skills
pertaining to the positions. The weights are specified on the performance
plan form for each position.

oSome of the financial regulatory agencies, including OCC, OTS, FDIC,
CFTC, and FHFB, do not assign different weights to competencies when
appraising employee performance. Instead, all of the competencies in an
employee's performance plan are equally considered during the appraisal.
For example, at OCC, all of the competencies (which are called skill-based
performance elements) that are contained in an individual's performance
plan are considered to be critical, so they receive equal weight when
determining the overall summary rating for that individual, according to
an official.

Agencies Have Involved Employees and Stakeholders in Various Ways to Gain
Ownership of Performance Management Systems

The financial regulatory agencies have used several strategies to involve
employees in their systems, including (1) soliciting or considering input
from employees on developing or refining their performance management
systems, (2) offering employees opportunities to participate in the
performance planning and appraisal process, and (3) ensuring that
employees were adequately trained on the performance management system
when rolling out the system and when changes were made to the system.
Overall, the 10 agencies have employed these strategies differently.
Effective performance management systems depend on individuals', their
supervisors', and management's common understanding, support, and use of
these systems to reinforce the connection between performance management
and organizational results. Employee involvement improves the quality of
the system by providing a front-line perspective and helping to create
organizationwide understanding and ownership.

Agencies Have Considered Employee Input for Developing or Refining
Performance Management Systems

All of the financial regulatory agencies, in some way, solicited or
considered employee input for developing or refining their performance
management systems by working with unions or employee groups to gather
employee opinions or conducting employee surveys or focus groups. An
important step to ensure the success of a new performance management
system is to consult a wide range of stakeholders and to do so early in
the process. High-performing organizations have found that actively
involving employees and stakeholders, such as unions or other employee
groups that represent employee views, when developing results-oriented
performance management systems helps to improve employees' confidence and
belief in the fairness of the system and increase their understanding and
ownership of organizational goals and objectives. Feedback obtained from
these sources is also important when creating or refining competencies and
performance standards used in performance plans. However, in order for
employees to gain ownership of the system, employee input must receive
adequate acknowledgement and consideration from management.

Agencies Have Involved Employee Groups in the Performance Management
System Process

Unions and employee groups had some role in providing comments or input
into the performance management systems at some of the financial
regulators. Six of the regulators (CFTC, FDIC, NCUA, OCC, OTS, and SEC)
had active union chapters, and four agencies (FCA, Federal Reserve Board,
FHFB, and OFHEO) had employee groups.^9 We have previously reported that
obtaining union cooperation and support through effective labor-management
relations can help achieve consensus on planned changes to a system, avoid
misunderstandings, and more expeditiously resolve problems that occur.^10
The degree to which unions and employee groups were involved in providing
comments or input into the development or implementation of performance
management systems varied from agency to agency. A few of the agencies
with unions have to negotiate over compensation. Unions at some agencies
were involved in participating in negotiations, entering into formal
agreements such as contracts and memoranda of understanding, and
initiating litigation concerning the development or implementation of
performance management systems. At other regulators, employee groups were
invited to comment on aspects of the performance management system, as the
following examples illustrate.

^9Employees at FDIC, NCUA, OCC, and SEC are represented by the National Treasury
Employees Union. OTS employees in Washington, D.C., and CFTC staff at two offices are
represented by the American Federation of Government Employees.

^10GAO, Human Capital: Practices that Empowered and Involved Employees, GAO-01-1070
(Washington, D.C.: Sept. 14, 2001).

oOFHEO has used ad hoc employee working groups to study different human
capital issues and advise management on recommendations for changes.
Specifically, OFHEO established a working group to look at teamwork and
communication in the agency and the group recommended changes to the
individual performance plans relevant to teamwork and communications. As a
result of the group's recommendation, OFHEO included additional language
for the agency's performance plans in the performance elements of teamwork
and communication.

oAt FDIC, the union participated with management in formal negotiations
regarding the establishment of the agency's performance management and pay
for performance systems and how the systems would work. Both parties are
bound by the terms of the formal agreements that resulted.

oAt NCUA, union representatives together with management issued a
memorandum of understanding in June 2006 detailing how supervisors are
supposed to introduce new performance plans for specified examiner
positions. The agreement set the timing of the introduction of new
performance standards, required training for rating officials, required
supervisors to give progress reviews to their employees on achievements to
date, and required supervisors and employees to discuss the new standards.

oSEC will implement a new compensation and benefits system as a result of
an October 2006 ruling from the Federal Service Impasses Panel (Panel).^11
The Panel became involved when SEC and union negotiations over a
compensation and benefits agreement reached an impasse. SEC management
told us that they have formed a labor-management working committee to
discuss how to implement the terms of the new Compensation and Benefits
Agreement as provided for under the Panel ruling.

^11The Federal Service Impasses Panel is part of the Federal Labor Relations Authority and
resolves impasses between federal agencies and labor unions representing federal
employees arising from negotiations over conditions of employment. The Panel may make
recommendations to the parties on how to overcome the impasse, if bargaining and
mediation are unsuccessful.

Agencies Have Directly Engaged Employees in Consultations about the
Performance Management System

The financial regulatory agencies involved employees in different ways
when developing their performance management systems. This process can
involve directly engaging individual employees and collecting opinions
from all employees through focus groups, surveys, or other forms of
feedback to develop a successful performance management system. Further,
soliciting employee input is also important when developing or revising
competencies or performance elements and related performance standards in
a performance management system in order to ensure that the competencies
and standards reflect skills and behaviors that are relevant to employee
tasks and responsibilities. While all of the financial regulators involved
employees to some degree, as the following examples illustrate, NCUA did
not consistently solicit input on developing or revising the competencies
and standards.

oIn 2003-2004, when the Federal Reserve Board sought to revamp its
performance management system, the agency hired an outside consultant to
conduct focus groups with the intent of identifying issues raised by
employees and making recommendations to address any concerns. Some focus
group participants said that the agency's recommended rating distribution
guidelines might prevent some employees from achieving a rating in the
highest category. Furthermore, some employees were concerned about
possible unfairness in ratings and wanted to see the distribution of the
performance ratings for all employees published. As a result of this
feedback, management began publishing the agency's ratings distributions,
and added information on the system's process to the agency's internal Web
site on the performance management system.

oWhen developing its first performance-based pay system in 2006, CFTC
solicited employee input through a variety of methods. The agency hired a
contractor to conduct focus groups and to survey employees about
transitioning to a performance-based pay system and the administration of
a performance management system. The contractor also hosted a Webinar, a
Web-based interactive seminar that allows for the submission of anonymous
questions and comments, to present the results of the employee survey.
Additionally, CFTC conducted town hall meetings to inform employees about
development of the system. As a result of employee feedback, management
decided to delay the first phase of implementation of the system from July
2006 until October 2006 in order to allow additional time for employees to
learn about the system and make the transition. Union representatives at
CFTC (Chicago and New York) told us that prior to CFTC's transition to
performance-based pay, the agency's management communicated frequently
with the union and provided appropriate notice prior to implementing
changes.

oThrough internal surveys, OFHEO received feedback on employee concerns
regarding opportunities for promotion and the frequency of progress
reviews. According to an agency official, feedback from an employee survey
indicated that employees wanted more opportunities for promotion than the
prior six pay-band system allowed. On the basis of this employee feedback,
OFHEO made the decision to switch to 18 pay grades and created career
ladders. Further, employees commented through the survey that they wanted
more feedback on their performance during the year. As a result, OFHEO
increased the number of progress review meetings from two to four per
year. An agency official stated that the Office of Human Resources
Management monitors these meetings to ensure that they have been held.

oSEC has analyzed data on SEC responses to OPM's governmentwide Federal
Human Capital Survey. According to agency officials, SEC has tracked
employee responses to questions on, for example, how well the agency
rewards good performers and deals with poor performers. In addition, SEC
has created a mailbox for anonymous employee comments and constructive
criticism on the performance management system.

oFCA circulated a draft of its proposed performance management system in
2002, and solicited comments from employees. As a result of employee
comments, FCA revised the descriptions of performance elements in the
performance plans, changed the weight of an element dealing with equal
opportunity employment, eliminated one element, and provided additional
guidance and training. To show how employee feedback was addressed, FCA
management presented a briefing to employees, which listed some of the
employee comments about the individual performance plans with accompanying
responses from management.

oAccording to an NCUA official present at the time when the agency
originally developed its performance elements and standards, NCUA
conducted job analysis studies for all positions, which involved employees
and supervisors in identifying specific duties, skills, and competencies
needed to accomplish different jobs. In addition to the studies, she said
that NCUA surveyed employees and conducted an assessment to identify any
gaps in the performance elements and standards. In 2006, when NCUA revised
the elements and standards for some examiner positions, NCUA used a
committee consisting of managers, supervisors, and one employee to develop
the new elements and standards. Union representatives told us they were
briefed on the final version of the elements and standards, but were not
asked for input. NCUA is currently revising individual performance plans
for other positions and the process does not include provisions for
soliciting and incorporating employee input. In comments on the draft of
this report, NCUA officials stated that NCUA sought to solicit input from
employees for certain positions, but that it was not necessary for
positions that are common across the government, since NCUA usually adopts
the competencies established by OPM for those positions.

Some union and employee group representatives we spoke with did not think
that management gave adequate consideration to employee input. For
example, the Employees' Committee at the Federal Reserve Board, which
provides advice to the Management Division on a variety of issues, was
asked to provide comments during the latest revision of the performance
management system. According to committee members, the committee submitted
a paper containing recommendations in response to this management request.
The committee, however, did not receive a written response from management
acknowledging their recommended changes. Committee members told us they
are now hesitant to submit input during the current strategic planning
process because they are concerned about the usefulness of putting time
and energy into developing recommendations that may not be considered.
According to agency officials, the responses from the Employees' Committee
and other employee focus groups held on this topic were summarized by the
consultant hired for the project and the consultant presented the summary
comments to management through the executive oversight committee. In
addition, management officials stated that they met with other committee
members (i.e., the heads of special interest groups) to discuss their
input. The Federal Reserve Board's Administrative Governor has also held
monthly meetings with randomly selected employees as an opportunity for
employees to voice their concerns about the performance management
program, among other topics.

Agencies Have Encouraged Employee Participation in Performance Planning
and Appraisals

All of the agencies, including FCA and FHFB, required or encouraged
employee participation in developing individual performance plans or
writing self assessments, contribution statements, or reports summarizing
accomplishments at the end of the appraisal cycle. In high-performing
organizations, employees and supervisors share the responsibility for
individual performance management and both should be actively involved in
identifying how individuals can contribute to organizational results and
be held accountable for their contributions. By actively participating,
employees are not just recipients of performance expectations and ratings,
but rather, have their ideas heard and considered in work planning and
assessment decisions. However, employee representatives from some
agencies, such as FDIC, OTS, and OCC, expressed concern that employees
were not actively involved in the performance planning and appraisal
processes even when the agency required or encouraged such
participation.^12

^12We did not assess whether meetings involving employees in performance planning or
appraisals were conducted in accordance with agency policy or the quality of the
interactions between supervisors and employees. We relied on representatives from
employee groups and unions to describe their perceptions of actual employee participation
in expectation-setting meetings and their perceptions of managementï¿½s consideration of
employee input into appraisals.

oAt FCA, employees could participate in performance planning by working
with their rating officials to identify accomplishments expected to be
achieved during the appraisal period. In addition to participating in an
official mid-year performance review, at the end of the appraisal cycle,
employees and supervisors could meet for a pre-appraisal interview to
discuss the employees' accomplishments during the previous year.
Additionally, employees could submit an optional self assessment of their
performance. This input was supposed to be considered when the supervisor
evaluated the employee, according to FCA policy.

oEmployees at FHFB had several options for participating in developing
their performance plans--working with the supervisor to develop the plan,
providing the supervisor with a draft plan, or commenting on a plan
prepared by the supervisor.

Although FDIC, OTS, and OCC provided some opportunities for employee
participation in the planning and appraisal processes, we heard from union
representatives at these agencies that this participation did not always
occur, as the following examples illustrate.

oFDIC's performance management directive requires that the employee and
the supervisor have a meeting to discuss all performance criteria included
in the employee's performance plan and any expectations regarding the
quality, quantity or timeliness of work assignments. The policy also
encourages the employee to submit an accomplishment report and to submit
written comments on his or her supervisor's draft assessment of the
employee's "Total Performance" before it is forwarded to higher levels of
review within a pay pool. However, union representatives told us that
expectation-setting meetings have not been consistently conducted;
instead, sometimes employees have simply signed a form to acknowledge
receipt of their performance plans. Additionally, employee comments on the
appraisal form have not been taken into account by supervisors, according
to union representatives. FDIC officials stated that the rating official
and employee are required to meet to discuss expectations at the beginning
of the rating period or whenever there is a change in performance
criteria. Officials also noted that the performance management program is
a collaborative process that relies on communication between a manager and
his or her employees, and that the employee is supposed to seek
clarification on performance criteria or expectations from the supervisor
if necessary, as is explained in the directive.

oAn employee union representative at OTS maintained that employees have
not been very involved in setting their own performance expectations;
instead, supervisors have informed them about what they should do at the
beginning of the performance appraisal cycle. The representative told us
that supervisors may discuss changing expectations with employees during
the year, but these discussions have not always occurred. According to an
agency official, OTS has encouraged managers to regularly meet with their
employees and provide a clear picture of what is expected of employees for
the year in terms of their individual roles and responsibilities for the
standardized performance expectations and what will be considered in
appraising the employees' performance.

oAlthough OCC provided opportunities for employee participation in the
performance planning and appraisal processes, union representatives told
us that this participation did not always occur. At OCC, employees may
participate in developing their individual performance plans and are
supposed to submit accomplishment reports. Further, officials explained
that many employees at OCC have secondary objectives in their performance
plans. Because secondary objectives are customized, there should be a
discussion between the supervisor and the employee. According to an
official, if an employee has customized secondary objectives included in
his or her individual performance plan, the employee and supervisor are
supposed to have a discussion about it.^13 However, representatives from
the union at OCC told us that performance plans are pretty generic and are
distributed to individuals based on their grade levels. They said that
some employees do not sit with their managers to tailor the plans;
instead, employees just sign the forms to acknowledge receipt of the
plans.

^13According to OCC officials, generic performance plans are established for examiners
because they all perform similar functions. The primary objectives in these standardized
plans are generic. Any additional primary objectives or secondary objectives included in the
examiner performance plans are tailored to individuals. According to officials,
commissioned examiners should have at least one secondary objective in their performance
plans so that they have the potential to receive a level 4 (highest) performance rating.
Precommissioned examiners, who are focused on completing a rigorous training program,
are not required to have secondary performance objectives, although they are not
prohibited from having them.

Agencies Have Provided Training on Performance Management Systems

All of the financial regulatory agencies have conducted some form of
training or information dissemination on topics related to performance
management. Asking employees to provide feedback should not be a one-time
process, but an ongoing process that occurs through the training of
employees at all levels of the organization to ensure common understanding
of the evaluation, implementation, and results of the systems. Providing
training  when changes are made to a performance management system can
help ensure that employees stay connected to the system and reinforce the
importance of connecting individual performance expectations to
organizational goals. At some agencies, such as SEC and FHFB, training has
been mainly directed at supervisors, while at FDIC training has been given
to nonmanagers as well. Formal training for nonsupervisors at the agencies
has typically been directed at new employees or has occurred when
significant changes were being made to a performance management system.
Some agencies have distributed materials through the agency intranet,
memos, emails, or other written documents, as the following examples
illustrate.

oSEC has offered several opportunities for supervisors to learn the
mechanics and skills necessary for administering the performance
management system. Specifically, new supervisors have received general
training on supervisory roles and responsibilities, including performance
management. For supervisors, SEC has offered two levels of classes on
managing performance and communicating expectations. Supervisors have also
had the opportunity to receive training on managing labor relations, which
has included discussions of SEC's agreement with the union, and the
performance-based pay and award systems. Supervisors could also attend a
briefing on performance management concepts and processes. In addition to
offering supervisor training, SEC informs new employees about the
performance management system during the orientation program. Performance
management information is also available to employees through the agency's
intranet web site. Finally, supervisors are supposed to brief new
employees on the performance management system at the beginning of the
rating cycle, during discussions of individual performance standards.

oMost employees at FHFB have not received training on performance
management since the late 1990s, and are expected to learn about the
system from their supervisors. However, FHFB offered training for managers
and supervisors in 2004 on the performance management system and how to
conduct performance appraisals.

oFDIC has conducted several training sessions and disseminated information
to managers and employees related to its performance management and pay
for performance programs. This has included in-person training sessions,
taped sessions made available for viewing on IPTV, and "question and
answer" documents and policy directives available on the agency intranet.
FDIC provided specific training for nonsupervisors in 2006 when management
and union representatives jointly conducted training sessions on the
agency's new compensation agreement. Training was intended for
non-management employees, including bargaining unit and non-bargaining
unit employees, and was conducted in a variety of formats. Sessions
included discussions of employees' roles and responsibilities in the
performance management and pay for performance systems.

Agencies Generally Have Linked Pay to Performance and Built in Safeguards

As discussed, the 10 financial regulatory agencies linked pay to
performance and built safeguards into their performance management systems
but could make improvements to ensure that poor performers do not receive
pay increases and to improve the communication of performance standards
and transparency of performance results. This section provides more
detailed information on the different ways in which the agencies
translated performance ratings into pay increases and used different
budgeting strategies for performance-based pay. The section also discusses
how the agencies awarded pay increases that considered performance but
were not dependent on ratings. Finally, information is presented on agency
implementation of two additional safeguards: higher-level reviews of
performance rating decisions and establishing appeals processes for
performance rating decisions.

Agencies Used Differing Methods to Translate Performance Ratings into Pay
Increases

For increases that were linked to performance ratings, the financial
regulatory agencies used different methods to translate employee
performance ratings into pay increases. These methods included
establishing ranges for increases, using formulas, and considering current
salaries when making decisions on the amounts of performance-based pay
increases for individuals.

Several agencies established ranges of potential pay increases
corresponding to the various performance rating levels. These systems gave
managers the discretion to determine the exact pay increase amounts for
individuals, within those ranges, as the following examples illustrate.

oAt OTS, employees who received a rating of 5 (on a 5-level scale)
received between a 5.5 percent and 7.5 percent pay increase, while
employees who received a rating of 3 received between a 1.5 percent and
3.25 percent pay increase during the appraisal cycle we reviewed.
Employees who received a rating of 1 or 2 did not receive any pay
increase. OTS gave managers the flexibility to determine the specific pay
amount each employee would receive within the range of possible pay
increases corresponding to that performance rating.

oOCC established ranges of potential pay increases that corresponded to
different performance rating levels and gave managers the flexibility to
decide on the exact amount of pay increase that each individual would
receive within the range that corresponded to that employee's rating
level. Each year OCC adopts a merit pay matrix that defines a range of
allowable percentage increases that may be paid for performance rating
levels 3 and 4 (the two highest rating categories). During the appraisal
cycle we reviewed, individuals with a level 3 performance rating were
eligible to receive a merit increase between 2.1 percent and 5.5 percent,
and individuals with a level 4 rating could receive a merit increase
between 5 percent and 9 percent. The rating official recommended the
percentage of merit pay that each employee with a summary rating of 3 or 4
should receive. Agency officials told us that it can be challenging for
managers to determine the pay increase amount for each employee within
those preestablished pay increase ranges. Managers want to ensure
consistency among employees with similar levels of performance and often
consult with other managers or human resources staff for advice when
making these pay increase decisions. Employee representatives expressed
some concern about the overlapping ranges for pay increases, and a
representative said that employees are unclear about what performance
behaviors are needed to achieve merit increases.

Other agencies used formulas for determining the amounts of pay increases
linked to performance ratings to be awarded, as the following example
illustrates.

oNCUA used a pay matrix tied to employees' performance rating scores
(which could range from 0 to 300) to calculate the pay increase
percentages. All employees in the same pay pool that received the same
performance rating would receive the same pay increase percentage.
Specifically, an employee who received a performance rating score of 234
fell within the "fully successful" performance rating range and received a
pay increase of 3.066 percent. Another employee who received a performance
rating score of 235 fell within the "highly successful" performance rating
range and received a pay increase of 3.076 percent. Employees who received
a performance rating score below 165 fell within the "unsatisfactory" or
"minimally successful" performance rating ranges and did not receive any
pay increases.

Some agencies considered employees' current salaries when deciding on the
amounts for pay increases linked to performance ratings, as the following
example illustrates.

oAt FCA, the percentage pay increase an employee received depended on
where the employee's current salary fell within the pay band. FCA used a
merit matrix to calculate merit pay increases. The matrix considered an
employee's existing salary position within the relevant pay band (with
position defined in terms of one of five possible quintiles), as well as
the employee's performance rating, and determined the percentage pay
increase corresponding to those factors. For example, for the performance
appraisal cycle we reviewed at FCA, the percentage increase in pay that an
employee who received a fully successful performance rating could receive
ranged from 3.5 percent (for an individual whose salary was in the bottom
quintile of the pay band) to 2.0 percent (for an individual whose salary
was in the top quintile of the pay band). For employees with the same
performance rating, an employee whose salary was considered to be below
market rate at the bottom of the pay band would receive a larger
percentage pay increase than an employee whose salary was considered to be
at or above market rate. FCA provided pay increases only to employees who
performed above a minimally successful rating level.

At many of the agencies, as an employee's salary approached the top of the
pay range for a position, increases linked to performance ratings could be
received as a combination of permanent salary increase and a one-time,
lump sum cash payment, as the following example illustrates.

oAt FHFB, for an employee in a position with a pay range of
$70,000-$90,000, if the individual's salary was near the top of the pay
range, he or she would receive a performance-based merit increase to take
his or her salary to the top of the salary range and then receive a lump
sum payment.

Across the various methods used to translate performance ratings into pay
increases, the expectation would be that larger pay increases are
associated with higher performance ratings. As a means of providing a
quantified descriptor of how strongly increases in ratings were associated
with increases in pay linked to those ratings at each of the agencies, we
computed a Spearman rank correlation coefficient between employees'
performance ratings and the percentage increases in pay that were linked
to performance ratings.^14 Although the correlation coefficients for the
eight agencies varied from +0.63 and +0.94, they all demonstrated a strong
positive association between higher performance ratings and higher
ratings-linked pay increases (expressed as a percentage increase in
salary).^15

^14A correlation coefficient is a measure of association (strength) of the relationship between
two variables; in this case, ratings and percentage increases in pay. A positive correlation
coefficient would mean that the two variables tend to increase (or decrease) together. A
positive coefficient would indicate that as the rating goes up, so does the percentage
increase in pay. A negative coefficient would mean that the two factors have an inverse
relationshipï¿½as one variable increases, the other decreases. Values of the coefficient may
range from -1.0 to +1.0. The percentage increase in ratings-linked pay was based on the
combined value of both increases in base pay and lump sum (one time) payments, relative
to prior annual salary. The performance appraisal and performance-based pay systems at
FDIC and SEC were sufficiently different from the other eight agencies that we do not
include correlation coefficients for these two agencies. See appendix I, Objectives, Scope,
and Methodology, for additional information concerning the exclusion of FDIC and SEC
from this analysis.

^15In those instances where agency rating scales used lower numeric values to indicate higher
performance levels, the values were reversed so that all coefficients would reflect the
relationship of an increase in ratings with an increase in pay as a positive coefficient.

While the correlation coefficients provide some additional perspective on
the linkage between performance ratings and pay increases at the financial
regulatory agencies, they should be viewed as a rough gauge of the overall
strength of the relationship across the agencies and are not sufficient
for ranking or making other comparisons between agencies.^16 In reviewing
the coefficients, we noted that agencies with some of the lowest
correlations were using a four-level rating system that produced rather
constrained ratings distributions. In one instance, for example, employees
rated at the two lowest performance levels (called levels 1 and 2) were
not eligible for pay increases, and over two-thirds of all employees
received a level 3 rating. Both the base pay increases and bonus amounts
that could be awarded for level 3 performance overlapped with those for
level 4 (the highest level), such that some employees rated at level 3
realized a percentage increase in pay that was twice the amount obtained
by other level-3-rated employees, as well as even some level-4-rated
employees.

^16See appendix I for additional information on the limitations of the correlation coefficient.

Agencies Used Different Budgeting Strategies for Performance-Based Pay

The federal financial agencies also varied in their strategies to budget
for pay increases directly linked to performance ratings. Many of the
agencies set aside funds each year for performance-based pay increases. At
some agencies, these funds were treated as an agencywide funding pool or
pools for performance-based pay increases, as the following examples
illustrate.

oAccording to agency officials, NCUA established two agencywide merit
funding pools for different employee grade-level groups because higher
graded employees usually received higher ratings and consequently, higher
merit pay increases. Officials stated that the establishment of two merit
funding pools was more advantageous to lower graded employees and
increased the amount of funds available for their merit pay.

oSEC established one pool of funds for performance bonuses and quality
step increases available for senior officers, and another pool for all
other employees.

At some agencies the performance-based pay increases budget was divided
into separate pay pools by suborganizational unit, and the responsibility
for distributing merit pay increases was delegated to management at the
subunit level, as the following examples illustrate.^17

^17Officials at one agency told us that setting separate budget amounts for merit pay
increases for each unit helped to ensure that merit pay resources were more fairly
distributed across the units.

oFor the "Pay for Performance" program at FDIC that covers bargaining unit
and nonbargaining unit employees, the agency established pay pools at the
division level (and at the regional level for the large Division of
Supervision and Consumer Protection), and allocated funds for
performance-based pay increases to the pools. Funds were allocated through
pay pools to each division and office, with subsequent separations of each
division or office into separate populations for bargaining unit and
nonbargaining unit employees. (Corporate managers and executives at FDIC
are covered by a separate pay-at-risk compensation system.)

oFHFB provided each office with a pay pool for performance-based annual
pay increases. The merit increase pool amounts were determined based on
the approved governmentwide general increase plus 2.5 percent of the total
base salaries for all employees in the office. An FHFB official stated
that the reason each office was provided with a pool of funds was to avoid
comparing individuals with different functions and responsibilities to
each other, and this official believed that FHFB had greater control when
pay decisions were made at the office level. For example, an office
director could decide to assess all his staff at the outstanding level,
but less performance-based pay would be available for each office
employee. Office directors were responsible for determining the sum of all
merit increases and lump sum payments for their offices, while not
exceeding their offices' merit increase pool allocations.

Agencies Provided Other Increases That Considered Performance

In addition to providing ratings-based pay increases, the financial
regulatory agencies awarded pay increases that considered individual
performance in some way without being directly linked to employees'
performance ratings. The following are additional examples of these types
of pay increases at the agencies to supplement the material presented in
the body of the report.

oThe Federal Reserve Board offered a cash awards program, which accounted
for about 2.5 percent of the total agency salary budget, to reward
employees who sustained exceptional performance or made significant
contributions to successful projects, according to officials. According to
the Federal Reserve Board's criteria for this awards program, cash awards
could be given to employees who initiated, recommended, or accomplished
actions that achieved important Federal Reserve Board goals, realized
significant cost reductions, or improved the productivity or quality of
Board services. These awards could be made in any amount up to a maximum
of 10 percent of an employee's base pay within the same performance cycle.
The 10 percent maximum did not apply to variable pay awards, which are
given instead of cash awards to economists, attorneys, or Federal Reserve
Board officers.

oFor some regulators, these types of pay increases were sizeable. For
example, at OCC, approximately 10 percent of employees were awarded a
special increase during the completed appraisal cycle we reviewed. The
awards represented a 5 percent raise for those individuals. According to
OCC policy, special increases are to be awarded to recognize increased
value an employee contributes to his or her job by applying desirable
skills over a significant period of time or by assuming higher-level
responsibilities within his or her pay band. OCC also provided some pay
increases for competitive and noncompetitive promotions during the
appraisal cycle we reviewed. Interestingly, of the eight financial
regulators that participated in OPM's 2006 Federal Human Capital Survey,
OCC had the largest percentage of employees agreeing with the view that
awards in their work units depended on how well employees performed their
jobs. At OCC, 55.7 percent of employees agreed with this view.
Governmentwide the corresponding figure was 39.8 percent of employees. Two
other agencies, FCA and NCUA, also had slightly over 50 percent of their
employees agreeing with this statement.

Results from the 2006 OPM Federal Human Capital Survey suggest that the
financial regulatory agencies have done relatively better than many
agencies governmentwide in linking pay to performance. All eight of the
financial regulators that participated in the 2006 survey had percentages
of positive responses from their employees that were about the same as or
better than the governmentwide percentage of 21.7 positive responses to an
item asking employees whether they agreed or disagreed with the statement
that pay raises depended on how well employees performed their jobs at
their agencies. The percentage of employees giving a positive response to
this item was at least twice as high as the governmentwide value for a
majority of the eight agencies participating in the survey.

Agencies Built in Safeguards

While the financial regulatory agencies built safeguards into their
performance management systems, the agencies established and communicated
standards for differentiating among performance rating categories and
criteria for performance-based pay decisions to varying degrees. The
agencies also built in additional safeguards of establishing higher-level
reviews of performance rating decisions by either higher-level officials
or oversight groups, and all have established appeals processes for
employees to request reconsiderations of performance rating decisions. It
is important for agencies to have modern, effective, credible, and, as
appropriate, validated performance management systems in place with
adequate safeguards to ensure fairness and prevent politicization and
abuse. We have reported that a common concern that employees express about
any performance-based pay system is whether supervisors have the ability
and willingness to assess employees' performance fairly.^18 Using
safeguards can help to allay these concerns and build a fair and credible
system.

^18GAO-06-142T.

Agencies Implemented Higher-Level Reviews of Performance Rating Decisions

Although they have used different approaches, all of the federal financial
regulatory agencies have provided higher-level reviews of individual
performance rating decisions to help ensure that performance standards
were consistently and equitably applied across the agency. All of the
agencies have established at least one level of review of employees'
performance ratings to help ensure that performance standards were applied
appropriately. At some agencies, this oversight process has involved a
second-line supervisor or higher-level official reviewing the employee's
performance rating to ensure that the rating was appropriate and
consistent with any narrative describing the employee's performance. Some
agencies also have offices outside of the employee's team/office, such as
the Human Capital Office, review employee performance ratings to ensure
that rating decisions for groups of employees (agencywide, or by division
or region) were fair and equitable, as the following examples illustrate.

oOCC officials indicated that at the end of every appraisal cycle, they
have evaluated the results of the performance management and pay system by
looking, for example, at the differentiation in ratings and pay decisions
and how the pay ranges were used. The human resources officials have
discussed these results with managers to show them how their employees'
performance ratings and pay decisions influenced OCC's overall results.
For example, OCC introduced merit bonuses for the first time in the 2005
performance appraisal cycle. Upon reviewing the results of the merit bonus
decisions, OCC officials found that the percentage of employees in each
organizational unit that received a merit bonus varied widely among the
units--ranging from a high of over 80 percent of employees receiving a
bonus in one unit to 30 percent in another unit. As a result, according to
agency officials, OCC decided to recommend a minimum amount for bonuses
and restrict the percentage of staff who can receive a bonus to 50 percent
within each organizational unit. Agency officials also indicated that they
have identified areas of future training on the system based on the
results of reviews and subsequent discussions with managers, in order to
improve implementation of the system.

oAt NCUA, an employee's performance rating was completed and signed by the
rating official, and then a reviewing official (an office or regional
director) reviewed the employee's performance rating to ensure that the
rating was supported. Reviewers also look for consistency throughout the
rating process. For example, an Associate Regional Director will look
across all examiners' ratings in the region for consistency.

oFHFB provided a supervisory review of performance ratings to help ensure
that an employee's recommended rating was justified as well as consistent
with other ratings in the employee's work group. Once the rating official
(usually a first-line supervisor) recommended an initial summary rating,
the rating official would forward the rating to a second-line supervisory
reviewer (usually the division director or deputy director), called a
reviewing official. According to FHFB officials, the reviewing official
was usually knowledgeable about the employee's performance and could
discuss the rating narrative and final rating decision with the rating
official before the rating was shared with the employee. In addition, the
reviewing official checked whether performance rating narratives supported
individual performance elements, summary ratings were properly calculated
and appropriately signed, and there was consistency of ratings across the
work group. FHFB employee representatives with whom we spoke stated a
belief that the rating review process was effective and that supervisors
did not give ratings unless they first reviewed their decisions with
management. Employee representatives noted that there is a commitment in
the agency to be fair and equitable in assigning ratings.

oFCA provided multiple levels of reviews of ratings to ensure the
appropriateness of rating scores and consistency in applying performance
standards across FCA offices. After the rating official completed an
initial rating, a second-line reviewer was assigned to review each
employee's rating against the standards. Before final ratings were issued,
FCA's Office of Management Services provided a check to ensure that
offices were appropriately and consistently applying performance standards
and to look for any significant outliers. Employee performance assessments
rated as outstanding and as less than fully successful would be reviewed
to determine whether rating scores matched the narrative discussions. Any
potential issues identified would be brought to the attention of the
rating official for discussion and resolution. Management officials told
us that the Chief Human Capital Officer would meet with division
management to discuss whether the rating criteria were appropriately
applied and then division managers would determine whether to change any
performance ratings. In addition, the Office of Management Services
performed a post-rating distribution audit to review final rating
distributions to help inform future rating practices.

Establish Appeals Processes for Performance Rating Decisions

As mentioned previously, all of the federal financial regulatory agencies
have established appeals processes for employees to request
reconsiderations of performance rating decisions to help ensure accuracy
and fairness in the process. Providing mechanisms for employees to dispute
rating decisions when they believe decisions are unfair can help employees
gain more trust in the system, as the following examples illustrate.

oEmployees at CFTC could ask for an appeal of their overall rating through
the agency's reconsideration process. An employee could first appeal his
or her rating to the manager who reviewed the rating (called the reviewing
official) by defending his or her position orally or in writing. This
Reviewing Official then considered the employee's justification as well as
the original rater's opinion and provided a final decision on the matter.
According to CFTC officials, employees sometimes wanted to change the
wording in their performance evaluations.

oOTS has defined a grievance policy for employees who are dissatisfied
with their performance ratings. Employees covered by the bargaining unit
agreement may file a grievance under the negotiated agreement while
employees not covered by the agreement may request a grievance (within 10
days of receiving their ratings) under the agency's administrative
grievance procedures. OTS' union representative reported that in the past,
management and union representatives had resolved many cases of rating
disputes prior to employees filing formal grievances.

oThe Federal Reserve Board has established an appeals process so that an
employee can appeal the fairness of an overall rating decision, the rating
on an individual element, or any adverse comments appearing on the
performance assessment form. Employee representatives we spoke with said
that they believe that employees understand the appeals process, but
thought that more employees could take advantage of this opportunity. An
employee may first appeal his or her performance rating to a division
director, who in turn will notify the appropriate supervisor who submitted
the rating. Then, the division director will determine whether the rating
is appropriate based upon a review of documentation provided by both the
employee and supervisor. If the employee is not satisfied with the
first-level appeal decision, the employee may make a second-level appeal
to the Associate Director of Human Resources and specify areas of
disagreement with the performance assessment. The Associate Director for
the second-level appeal will then determine whether the division has
reasonably followed procedures and whether performance assessment
guidelines were applied consistently to other employees reporting to the
same supervisor. This supporting documentation submitted by the division
will be shared with the employee, except in cases where doing so infringes
on the confidentiality of other employees. As a result, first- or second-
level appeal decisions may result in changes to an overall rating, changes
to the rating of an individual element, or changes in the language in the
employee's performance assessment.

oOFHEO has established a three-level appeals process to ensure that
employees can dispute rating decisions when they disagree with rating
decisions. Employees can appeal the overall performance rating or
individual performance elements within the rating. For the first-level
appeal, the employee can submit a request with supporting documentation to
the performance rating official for reconsideration. If an appeal is not
resolved at the first level, the employee can request that the
second-level supervisor review the performance rating and supporting
documentation. Finally, the employee can request a third-level appeal by
the third-level supervisor, if necessary.

Appendix IV:  Actions Taken by Financial Regulators to Seek to Maintain Pay and Benefits
Comparability and Pay and Benefits Data

The federal financial regulatory agencies have made an effort to meet the
comparability requirements as required by the Financial Institutions
Reform, Recovery, and Enforcement Act of 1989 (FIRREA) and subsequent
legislation.^1 However, we found that factors such as funding constraints,
when the agency was granted flexibility under Title V of the U.S. Code,
the needs or preferences of their respective workforces, and each agency's
pay and benefits policies can result in some variation in their pay and
benefits.^2 They have also taken steps to explore a common survey that
would enable them to more efficiently collect information for pay and
benefit comparability purposes.

^1For the six FIRREA financial regulatory agencies (FCA, FDIC, FHFB, NCUA, OCC, and
OTS), see sections 301, 702, 120-3, 1206, and 1210 of the Financial Institutions Reform,
Recovery, and Enforcement Act of 1989, Pub. L. No. 101-73 (1989); for OFHEO, section 1315
of Pub. L. No. 102-550 (1992); for SEC, section 8(a) of Pub. L. No. 107-123 (2002), and for
CFTC, section 10702 (a) of Pub. L. No. 107-171 (2002).

^2While many of the financial regulatory agencies received increased flexibility under
FIRREA in 1989, such flexibility was afforded to OFHEO in 1992 and to CFTC and SEC in
2002.

Financial Regulators Have Conducted Individual Pay and Benefits
Comparability Surveys and Regularly Consult with Each Other, but Noted
Some Inefficiencies in the Process

To seek to maintain pay and benefits comparability, the majority of the 10
federal financial regulators have hired external compensation consultants
to conduct individual formal pay and benefits comparability surveys that
have included the other financial regulators. As shown in table 5, 7 of
the 10 financial regulators conducted pay and benefits comparability
surveys. Of the 7, 5 agencies also have included benefits in their formal
surveys. According to agency officials, because some of the 10 agencies
perceive the private sector as their main competitor for skilled
employees, they have included private-sector entities in their pay and
benefits surveys or have obtained additional private-sector data through
the Bureau of Labor Statistics and private vendors to complement their pay
and benefits surveys.

Table 5: Pay and Benefits Surveys That Federal Financial Regulators
Conducted through External Compensation Consultants, 1991-2006

                                        

Agencies    Hired external      Included      Years in        Agencies     
                compensation      benefits in      which     participating in 
               consultants to    comparability agencies have     surveys      
                conduct pay         surveys      conducted                    
               comparability                      surveys                     
                  surveys                         through                     
                                                 external                     
                                               compensation                   
                                                consultants                   
CFTC     No. In 2003 and      N/A           N/A           Used FDIC and    
            2005, used                                       OCC 2002 survey  
            consultant to review                             results in 2003; 
            existing surveys.                                used interagency 
                                                             group data in    
                                                             2005.            
FCA      Conducts a pay       No            1991, 1993,   The 2002 survey  
            survey once every                  1996, 1999,   included FDIC,   
            2-3 years.                         2002          FHFB, the        
                                                             Federal Reserve  
                                                             Board, NCUA,     
                                                             OCC, OFHEO, and  
                                                             OTS.             
FDIC     Conducts pay and     Yes           1996, 1999,   The 2005 pay     
            benefits survey                    2002, 2005    survey included  
            about once every 3                               all FIRREA       
            years.                                           agencies as well 
                                                             as the Federal   
                                                             Reserve Board,   
                                                             CFTC, OFHEO,     
                                                             SEC, and several 
                                                             Federal Reserve  
                                                             Banks.           
                                                                              
                                                             The 2005         
                                                             benefits survey  
                                                             included all     
                                                             FIRREA agencies  
                                                             and the Federal  
                                                             Reserve Board.   
FHFB     Conducted a pay and  Yes           2002          The 2003 survey  
            benefits survey                                  compared pay and 
            once. Also uses                                  benefits with    
            FDIC's pay and                                   FCA, FDIC, OCC,  
            benefits surveys as                              NCUA, OTS, and   
            a guide.                                         the Federal      
                                                             Reserve Board.   
Federal  Conducts pay and     Yes           Annual        The 2005 survey  
Reserve  benefits survey                    surveys       included OCC,    
Board    every 1-2 years.                   conducted     FDIC, NCUA,      
                                               between 1994  OFHEO, OTS, SEC, 
                                               and 2005,     as well as the   
                                               excluding     private-sector   
                                               2003.         entities and     
                                                             academia.        
NCUA     Conducted pay and    Yes           2000, 2004    The 2004 pay and 
            benefits surveys                                 benefits survey  
            twice.                                           included FDIC,   
                                                             OCC, OTS, FHFB,  
                                                             FCA, SEC, OFHEO, 
                                                             and the private  
                                                             sector.          
OCC      Conducted a pay      No            1999, 2001,   The 2006 pay     
            survey every year,                 2002, 2003,   survey included  
            alternating between                2006          FCA, FDIC, FHFB, 
            a full survey that                               NCUA, OTS,       
            covered many                                     OFHEO, SEC, and  
            benchmarked jobs and                             the Federal      
            a simplified survey                              Reserve Board.   
            that covered a few                                                
            of the benchmarked                                                
            jobs.                                                             
OFHEO    Conducted pay and    Yes           2000, 2005    The 2005 pay and 
            benefits surveys                                 benefits survey  
            every 3 to 5 years.                              included OCC,    
                                                             the Federal      
                                                             Reserve Board,   
                                                             FDIC, OTS, FHFB, 
                                                             and FCA.         
OTS      No. Conducts         N/A           N/A           N/A              
            informal benchmark                                                
            surveys as needed.                                                
SEC      No                   N/A           N/A           N/A              

Source: GAO summary of agency data.

The remaining three regulators (CFTC, OTS, and SEC) have participated in
the pay and benefits surveys of other agencies, and officials from these
agencies said that they have used the results of these surveys, but have
not conducted their own. For example, an SEC official told us that his
agency often uses FDIC's data because, like SEC, FDIC has a large number
of compliance examiners and must negotiate pay and benefits with the same
union as SEC. In 2002 and 2003, CFTC has also hired consultants to review
existing surveys from FDIC and OCC as well as from information gathered
from other regulators.

The agencies hired external compensation consultants to conduct the
surveys because, according to officials from FCA and FDIC, these
consultants provide an objective view of their agencies' pay and benefits.
And, because they have often worked with other FIRREA agencies, the
consultants can provide insights and perspectives based on information
from other agencies. For pay comparability surveys, external compensation
consultants compare base pay ranges for a given occupation, locality pay
percentages and, to a lesser extent, annual bonus and other cash award
policies. To compare pay across agencies, consultants send questionnaires
on behalf of the sponsoring agency and ask participating agencies to match
the jobs based on the job descriptions provided. The job descriptions
usually contain information on duties, scope of responsibilities, and
educational requirements.  External compensation consultants also have
used various methods to assess the comparability of benefits. For example,
the consultant for FDIC did a side-by-side comparison of benefits offered
at other agencies, and also calculated the total cost of benefits per
employee.

In addition to conducting comparability surveys, agency officials told us
that human capital officials at the 10 regulators have formed an
interagency Financial Regulatory Agency Group. The members regularly
consult with each other on pay and benefits issues, and as they prepare
their budgets for the coming year, they meet to exchange information on
potential and actual changes to pay and benefits. For example, the group
has exchanged information on updates in merit pay ranges, bonuses, salary
pay caps, and benefits such as flexible work schedules. Agency officials
also have taken turns to update a spreadsheet that lists the pay ranges
and benefits for all 10 financial regulators, a key document the agencies
use to compare pay and especially benefits informally across agencies.

However, in consulting with each other to meet comparability requirements,
agency officials told us that because many of the financial regulators
conduct comparability surveys, their staffs have had to respond to
numerous and often overlapping inquiries, which can be burdensome and
inefficient. This is especially the case for smaller agencies, such as FCA
and FHFB, which tend to have smaller human capital (personnel) departments
than larger agencies that may have pay and benefits specialists who can
handle comparability issues full time, including filling out and
processing various comparability surveys. According to officials from a
few regulators, partly as a result of the substantial investment of time
and resources, some agencies have not been timely or forthcoming in
sharing their pay and benefits information.

Regulators Are Exploring the Feasibility of a Common Survey

According to several agency officials, in response to renewed interest of
upper management from several agencies in consolidating pay and benefits
surveys, the regulators are studying the feasibility of such a method. In
December 2006, the regulators formed a subcommittee within the Financial
Regulatory Agency Group to study the feasibility of a common survey.
Agency officials are exploring whether consolidating the various
comparability surveys into a common survey will improve the process for
job matching and result in more efficient use of resources. They also told
us that the subcommittee also has discussed the feasibility of
establishing a Web-based data system to make the most current pay and
benefits information available to participating agencies. The subcommittee
is working on the details of allocating costs of a common survey among the
agencies, but has suggested that costs might be prorated based on the size
of each regulatory agency. As of March 2007, agency officials had not yet
received cost figures from potential consultants.

Agency officials who attended the first subcommittee meeting told us that
implementation of a common survey would require collaboration and
agreement on a number of matters, such as

ochoice of external compensation consultant to conduct the common survey,
since different consultants have different approaches to carry out the
common survey;

ogroup of jobs to be benchmarked for the common survey and best approach
for job matching, as some jobs are unique to certain agencies;

otiming and frequency of the common survey to meet everyone's needs since
agencies determine pay and benefits at different times of the year and
would need the updated information when the need arises;

onumber and types of organizations to include in the common survey because
while all agencies would want to include the financial regulators, some
may need information from certain private-sector entities; and,

ocost of the common survey may be substantial, which according to some
agencies, is a potential concern.

By forming the subcommittee to explore issues associated with developing a
common survey, agency officials have adopted some of the practices that we
identified that would enhance and sustain collaborative efforts. These
practices include defining and articulating a common outcome, establishing
means to operate across agency boundaries, and leveraging resources.^3

^3GAO, Results-Oriented Government: Practices That Can Help Enhance and Sustain
Collaboration among Federal Agencies, GAO-06-15 (Washington, D.C.: Oct. 21, 2005).

Agency officials who are members of the subcommittee told us that the
officials have sent a formal request for information to several consultant
candidates. The request inquired about the consultants' ability to plan
and execute a common survey that will provide customizable reports for
each agency and also create a secure, centralized data source on pay and
benefits. In addition, agency officials asked how the consultants would
approach job matching, a complicated task. For example, officials from
FDIC, OFHEO, and SEC told us that the use of different pay plans and
grades among agencies and the location of field offices in cities with
different employment market conditions contributed to the difficulty in
matching jobs across regulators. In addition, some agency officials said
that it is difficult to match jobs because agencies have different job
requirements that may differ even when a job title is the same. The
subcommittee received responses from various consultants and as of March
2007 was in the process of contacting the consultants to gather more
details and to discuss the options available to them.

Most Regulators Used Benchmarks Developed from Surveys and Other Data to
Assess Comparability and Make Adjustments to Pay and Benefits

In the absence of a legislative definition, agency officials told us that
agencies have used various benchmarks, as shown in table 6, to assess pay
and benefits comparability. For example, FDIC has sought to set its total
pay ranges (base pay plus locality pay) for specific occupations and grade
levels within 10 percent of the average of FIRREA agencies, a benchmark
that pay and benefits consultants have used in their comparability
surveys. FCA uses benchmarks, including average market rate paid by other
financial regulators. CFTC uses average payroll and salary structure
relative to other regulators. FHFB, NCUA, OTS, and SEC told us that they
have not used specific benchmarks, and OTS uses informal benchmarks as
needed.

Table 6: Selected Examples of Benchmarks Agencies Have Used to Assess Pay
and Benefits Comparability

                                        

      Agencies        Benchmark examples used to     Benchmark examples used  
                       assess pay comparability        to assess benefits     
CFTC            Average payroll, salary          Benefits being of similar 
                   structure, etc., relative to     types, and relative       
                   other regulators.                benefits costs as         
                                                    percentage of payroll.    
FCA             Pay at average market rate paid  No specific benchmarks.   
                   by other financial regulators.                             
                   The goal is to be in the middle                            
                   range of the financial                                     
                   regulators.                                                
FDIC            Since 1999, total pay (base pay  Benefits being of similar 
                   plus locality pay) within 10     types and having          
                   percent of the average of FIRREA equivalent or similar     
                   agencies. Prior to 1999, no      overall value, without    
                   specific benchmarks were used.   necessarily being         
                                                    identical.                
FHFB            No specific benchmarks.          No specific benchmarks.   
Federal Reserve Pay at the average of the        Benefits being similar    
Board           market, including regulators and types.                    
                   other appropriate competitors.                             
NCUA            No specific benchmarks.          No specific benchmarks.   
OCC             Pay within 10 percent of the     Benefits being similar    
                   other agencies' (or the          types. OCC does not       
                   market's) median base pay plus   consider the cost of      
                   locality pay for each occupation benefits.                 
                   and grade.                                                 
OFHEO           Pay within 10 percent of average Benefits being of similar 
                   salary of other agencies or      types and having          
                   market average for each          equivalent or similar     
                   occupation and grade.            overall value, without    
                                                    necessarily being         
                                                    identical.                
OTS             No specific benchmarks. Informal No specific benchmarks.   
                   benchmarks as needed.            Informal benchmarks as    
                                                    needed.                   
SEC             No specific benchmarks.          No specific benchmarks.   

Source: GAO summary of agency information.

Agency officials told us that all agencies, including the three agencies
that have not conducted formal benefits surveys, have assessed their
benefits comparability by comparing individual benefit items as well as
agency contributions to specific benefits. They added that most agencies
have used the interagency group spreadsheet that lists all the benefits
and agency contributions offered.

According to agency officials, the financial regulators have used
information from the pay and benefits comparability surveys and
discussions among the agencies in their efforts to seek to maintain
comparability. Table 7 provides some recent examples of these efforts.

Table 7: Selected Examples of Recent Pay and Benefits Adjustments
Resulting from Agencies' Comparability Assessments

                                        

Year                Action taken to adjust pay and benefits                
2002 FCA adjusted its pay ranges based on its comparability survey, which  
        stated that FCA's pay was lower than other FIRREA agencies.           
2002 In response to recently enacted comparability requirements, SEC       
        substantially increased its pay ranges to be comparable to those      
        offered at other FIRREA agencies.                                     
2003 As a result of gaining pay flexibilities, CFTC implemented new pay    
        ranges for its 2003 pay schedule; CFTC increased base pay by 20       
        percent for all eligible employees to partially close the 25 percent  
        gap between CFTC and FIRREA agencies.                                 
2006 OFHEO increased its pay ranges across the board based on the findings 
        from its 2005 pay and benefits survey.                                
2006 The FDIC increased its pay scale minimums by 1.5 percent and pay      
        scale maximums by 6 percent, effective February 18, 2006, based on    
        FDIC's consultant's analysis of its pay and benefits survey.          

Source: GAO analysis of agency information.

Agency Pay and Benefits Policies and Several Other Factors Contribute to
Variations in Pay and Benefits

Although the financial regulators have adjusted their pay and benefits to
seek to maintain comparability, several factors influence compensation
decisions that lead to some variations in pay ranges and benefit packages.
As shown in figure 3 in the report, with the exception of the Federal
Reserve Board and OFHEO, the financial regulators' total pay ranges
consist of base pay and locality pay percentages that are calculated based
on the employees' duty station. The Federal Reserve Board and OFHEO do not
have separate locality pay percentages because Washington, D.C., is their
only duty station. Figure 3 also shows that, for examiners, FDIC and NCUA
pay ranges generally have lower minimum base pay than other agencies, and
FDIC and OCC have higher maximum base pay for examiners. In addition, for
economists, CFTC and FDIC pay ranges have
lower minimum base pay than other agencies, and the CFTC and OCC pay
ranges have higher maximum base pay.^4

^4We did not include the Federal Reserve Board and OFHEO in our example on differences in
agenciesï¿½ base pay because the Federal Reserve Board and OFHEOï¿½s base pay include the
element equivalent to other agenciesï¿½ locality pay for Washington, D.C.

Actual average base pay figures that we obtained from the Central
Personnel Data File and from the Federal Reserve Board also vary among the
10 agencies in relation to the agencies' respective base pay ranges, as
shown in figure 3 in the report. For example, the actual average base pay
for examiners at OCC ($92,371) is 52 percent of the maximum pay range of
$177,600. However, actual average base pay as a percentage of maximum pay
can vary considerably, as in the case of SEC attorneys. Their actual
average base pay ($124,379) is 98 percent of the maximum pay range of
$126,987.

According to agency officials, two factors affect where actual average
base pay falls within an agency's pay range. One is the distribution of
the length of service among employees. For example, the actual average
base pay for agencies with a higher proportion of long-tenured employees
would be closer to the maximum of its pay range. Conversely, actual
average base pay for agencies with a higher proportion of new hires would
fall closer to the minimum of the pay scale. An OCC official told us that
despite the fact that OCC also has a large number of experienced
examiners, the actual average pay for OCC examiners may seem low compared
to other agencies because OCC has hired a large number of examiners during
the last 2 years. Officials from several federal regulators also told us
that they rarely hire at the lower grade level for some occupations. For
example, FHFB tends to hire mid-level employees because its relatively
small office cannot afford a long training period for new hires.

As shown in table 2 in the report, locality pay percentages vary among
agencies for the same duty station. Table 8 shows the methods that
agencies are currently using to determine their respective locality pay
percentages and adjustments.

Table 8: Agencies' Current Methods for Determining Locality Pay
Percentages and Adjustments

                                        

          Agency         Methods for determining locality pay percentages and 
                                             adjustments                      
CFTC                  Uses OPM locality percentages.                       
FCA                   Uses the average rate of 5 FIRREA agencies: FDIC,    
                         FHFB, NCUA, OCC, and OTS.                            
FDIC                  Uses a formula jointly developed by the National     
                         Treasury Employees Union and FDIC primarily based on 
                         the Bureau of Labor Statistics National Compensation 
                         Survey Cost of Labor data for federally defined      
                         locality pay areas, with Runzheimer International    
                         cost of living data influence for a very few areas   
                         where there are extreme differences between cost of  
                         living and cost of labor.                            
FHFB                  For Washington, D.C., considers other federal bank   
                         regulatory agencies per FHFB's comparability         
                         statute. For other locations, uses FDIC locality pay 
                         percentages.                                         
Federal Reserve Board Not Applicable. Washington, D.C., is the only duty   
                         station.                                             
NCUA                  Considers other FIRREA agencies.                     
OCC                   Uses primarily cost of labor data from the Economic  
                         Research Institute.                                  
OFHEO                 Not Applicable. Washington, D.C., is the only duty   
                         station.                                             
OTS                   Uses Runzheimer International cost of living data.   
SEC                   Currently adjusts yearly locality pay percentage     
                         increases by at least the minimum recommended OPM    
                         locality incremental adjustment.                     

Source: GAO analysis of agency information.

The benefits that the 10 financial regulators offered also varied.
Although all of the agencies offer standard federal government benefits,
there are variations in the extent of agency contributions and types of
additional benefits these agencies offer. For example, all financial
regulators offer the Federal Employees Health Benefits program, but agency
contributions differ. Some agencies pay for a percentage of the health
premium (e.g. 70 percent at FCA and 90 percent at OFHEO). CFTC contributes
100 percent for reservists called to active duty. The following are
selected examples of the additional benefits that some financial
regulators offer as of September 2006 unless noted otherwise:

oFive of the 10 regulators-- FDIC, FHFB, the Federal Reserve Board, OCC,
and OTS--offer their employees 401(k) retirement savings plans with
varying employer contributions. In addition, all agencies except the
Federal Reserve Board offer the federal Thrift Savings Plan.^5

^5In addition to these five agencies, FCA offered a 401(k) plan but discontinued it in
December 2006.

oThe Federal Reserve Board and OCC offer domestic partner benefits for
some types of plans.

oFCA and SEC offer child care subsidies, and FDIC and OTS offer on-site
day care.

oFCA, FDIC, FHFB, and OCC reimburse employee expenses related to items
such as fitness, recreation, and adoption in their wellness accounts.
Amounts differ from $250 per year at FDIC to $700 per year at FHFB.

According to agency officials, a number of factors have influenced their
pay and benefits policies and could have contributed to the variations in
their pay ranges and benefits. For example, the length of time an agency
has been under the comparability requirements and related compensation
flexibility provisions affected compensation. CFTC and SEC officials told
us that because their agencies received pay and benefits flexibilities and
were put under a comparability requirement much later (in 2002) than the
six FIRREA agencies (in 1989), CFTC and SEC have taken an incremental
approach to slowly increase their pay and benefits to close the gap with
the other financial regulators. According to a CFTC official, this would
allow time for employee input and acceptance while building agency
capacity to manage the authority. Budgetary constraints represent another
factor. OFHEO officials told us that OFHEO did not implement a new 401(k)
retirement savings plan recommended by its external compensation
consultant, Watson Wyatt, in its 2005 comparability survey because OFHEO
is working to control the growth of its personnel expenses and because
budget limitations resulting from being part of the appropriations process
has caused OFHEO to curtail new benefits programs. Furthermore, agency
officials said that an agency has to consider the particular needs and
preferences of the agency's workforce as well as ways to attract and
retain its workforce. For example, CFTC added a fully paid dental benefit
as a result of an online vote by employees on preferred benefits options.
FDIC officials indicated that its employees greatly value the matching
contribution FDIC provides on its 401(k) plan, and found that the matching
contribution is also an effective retention tool. Similarly, OCC added a
401(k) retirement savings plan in order to attract and retain employees.
According to an SEC official, SEC uses a student loan repayment benefit
because the benefit helps to attract and retain employees, many of whom
are recent law school graduates. Agency officials emphasized that it was
not their goal to have identical pay and benefits packages; rather, they
considered pay and benefits as a total package when seeking to maintain
pay and benefits comparability and when setting pay and benefits policies
aimed at recruiting and retaining employees. See table 9 for more detailed
information on the benefits that the 10 financial regulators offer.

The following table lists selected benefits identified by 10 financial
regulators as of September 2006, unless otherwise noted in the table. We
included the following categories of benefits: insurance, pre-tax
benefits, child care, leave, travel and relocation, educational and
professional expenses, retirement, work/life benefits, and other benefits
and payments.

Table 9: List of Benefits Offered by the 10 Financial Regulators

Source: GAO analysis of agency data.

Appendix V:  Analysis of Movement Data of Financial Regulator Employees from Fiscal
Years 1990 through 2006 

We reviewed the movement of financial regulator employees from fiscal year
1990 through 2006 using data from the Central Personnel Data File (CPDF).
We found that the movement of employees among the financial regulators was
very low and presented no discernible trend, but that 86 percent (13,433
of the 15,627) of employees leaving the regulators voluntarily (i.e.,
moving or resigning) resigned from the federal government. Our analysis
did not include the Federal Reserve Board of Governors because CPDF does
not contain data from the Federal Reserve Board. (For more detail on our
methodology, see app. I.) This appendix includes additional data for
fiscal years 1990 through 2006 on the average number of these employees
moving among the 9 financial regulators; the movement of employees among 9
of the 10 financial regulatory agencies by occupation; and employment by
occupation and employee movement agency snapshots.

Figure 8 shows the average number of employees in mission-critical and
other occupations moving among the 9 financial regulators for which we
have data from fiscal year 1990 through fiscal year 2006. On average, a
total of 919 employees per year moved or resigned. Movement ranged from an
average of less than 1 for investigators to an average of over 11 for
examiners.

Figure 8: Average Number of Employees in Mission-Critical and other
Occupations Moving among the 9 Financial Regulators, Fiscal Years
1990-2006

Note: This table does not include data for the Federal Reserve Board
because CPDF does not include data for the agency. The "all other"
category combines specialists in occupations such as human resources
management, administration, clerical, management and program analysis,
financial administration, and paralegal work.

Table 10 provides the actual number of financial regulator employees for
whom we had data, by mission-critical and other occupations, who moved to
another financial regulator from fiscal year 1990 through 2006.

Tables 11 through 19 provide employment by occupation and movement data
for 9 of the 10 agencies from fiscal year 1990 through 2006.^1

^1The data in these tables are for permanent employees only.

Table 10: Number of Financial Regulator Employees in Mission-Critical and
other Occupations Who Moved to another Financial Regulator, Fiscal Years
1990-2006

Source: GAO analysis of CPDF data.

Note: This table does not include data for the Federal Reserve Board.

^aSupervisors included executives, who constituted less than 1 percent of
all supervisors who moved to another financial regulator.

^bThe "all other" category, which combines specialists in occupations such
as human resources management, administration, clerical, management and
program analysis, financial administration, and paralegal work.

^cCFTC officials told us that some employees in the investigation job
series 1801 were examiners. Because the CPDF cannot distinguish employees
in the same job series but in different job titles, we called all the CFTC
employees classified as 1801 examiners and classified those in job series
1802 (compliance investigation and support) as investigators at CFTC.

^dOFHEO officials told us that some employees in the job series 0501
(financial administration and program) were examiners; because of
limitations of the CPDF we put all employees in job series 0501 into the
examiner job series.

Table 11: Commodity Futures Trading Commission Employment and Movement
Data, Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Table 12: Farm Credit Administration Employment and Movement Data, Fiscal
Years 1990-2006

Source: GAO analysis of CPDF data.

Table 13: Federal Deposit Insurance Corporation Employment and Movement
Data, Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Note: During some of the time under review, FDIC was actively downsizing
its workforce and achieved that in part by providing buyouts and other
incentives for employees to leave.

Table 14: Federal Housing Finance Board Employment and Movement Data,
Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Table 15: National Credit Union Administration Employment and Employee
Movement Data, Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Table 16: Office of the Comptroller of the Currency Employment and
Employee Movement Data, Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Table 17: Office of Federal Housing Enterprise Oversight Employment and
Employee Movement Data, Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Table 18: Office of Thrift Supervision Employment and Employee Movement
Data, Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Table 19: Securities and Exchange Commission Employment and Employee
Movement Data, Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Table 20: Other Federal Agencies Employment and Employee Movement Data,
Fiscal Years 1990-2006

Source: GAO analysis of CPDF data.

Appendix VI:  Comments from the U.S. Commodity Futures Trading Commission

Appendix VII:  Comments from the Board of Governors of the Federal Reserve System

Appendix VIII:  Comments from the Federal Housing Finance Board

Appendix IX:  Comments from the National Credit Union Administration

Appendix X:  Comments from the Office of Federal Housing Enterprise Oversight

Comments from the Securities and Exchange CommissionAppendix XI

Appendix XII:  GAO Contacts and 

GAO Contacts

Orice Williams, (202) 512-8678, or [email protected] Brenda Farrell,
(202) 512--5140 or [email protected]

Staff Acknowledgments

In addition to the contacts named above, Belva Martin and Karen Tremba,
Assistant Directors; Thomas Beall; Amy Friedlander; Robert Goldenkoff;
Eugene Gray; Simin Ho; Anne Inserra; Janice Latimer; Donna Miller; Marc
Molino; Jennifer Neer; Barbara Roesmann; Lou Smith; Tonya Walton; Lindsay
Welter; Gregory Wilmoth; and Robert Yetvin made major contributions.

(450460)

[65]transparent illustrator graphic

www.gao.gov/cgi-bin/getrpt?GAO-07-678.

To view the full product, including the scope
and methodology, click on the link above.

For more information, contact Orice Williams, 202-512-8678,
[email protected] or Brenda Farrell, 202-512-5140, [email protected].

Highlights of [67]GAO-07-678 , a report to congressional requesters

June 2007

FINANCIAL REGULATORS

Agencies Have Implemented Key Performance Management Practices, but
Opportunities for Improvement Exist

Congress granted financial regulators flexibility to establish their own
compensation systems and required certain agencies to seek to maintain
comparability with each other in pay and benefits to help the agencies
overcome impediments to recruiting and retaining employees and avoid
competing for the same employees. In response to a request, this report
reviews (1) how the performance-based pay systems of 10 financial
regulators are aligned with six key practices for effective performance
management systems, (2) the actions these agencies have taken to assess
and implement comparability in pay and benefits, and (3) the extent to
which employees in selected occupations have moved between or left any of
the agencies. GAO analyzed agency guidance and policies, agency data on
performance ratings and pay increases, agency pay and benefits surveys,
data from the Central Personnel Data File, and interviewed agency
officials.

[68]What GAO Recommends

GAO recommends that several regulators take steps to communicate the
overall results of appraisal and pay increase decisions to all employees
while protecting individual confidentiality. The regulators generally
agreed with the recommendations.

The 10 federal financial regulatory agencies have generally implemented
key practices for effective performance management but could improve
implementation of certain practices as they continue to refine their
systems. All of the financial regulators awarded some pay increases during
the appraisal cycles we reviewed that were linked to employees'
performance ratings, although two also provided across-the-board pay
adjustments, even to employees who had not received acceptable performance
ratings, weakening the linkage of pay to performance. Both agencies have
indicated in the future annual pay adjustments will not be awarded to
unsuccessful performers. The agencies have generally aligned individual
performance expectations and organizational goals, connected performance
expectations to crosscutting goals, used competencies to provide a fuller
assessment of performance, and involved employees and stakeholders in the
process. All of the agencies built safeguards into their performance
management systems to enhance credibility and fairness. However, the
extent to which the agencies communicated overall results of performance
rating and pay increase decisions to all employees varied, and some could
increase transparency by letting employees know where they stand relative
to their peers in the organization, while protecting individual
confidentiality.

Financial regulators have hired external compensation consultants to
conduct pay and benefits comparability surveys, exchanged pay and benefits
information, explored the feasibility of conducting a common survey, and
adjusted pay and benefits to seek to maintain comparability with each
other. Although financial regulators have adjusted pay and benefits partly
based on the results of their comparability efforts, there is some
variation in pay ranges and benefit packages among the agencies. According
to agency officials, factors such as the year the agencies first became
subject to comparability provisions, budget constraints, and the needs and
preferences of workforces play a role in compensation decisions and
contribute to this variation. Furthermore, agency officials emphasized
that it was not their goal to have identical pay and benefits packages;
rather, they considered pay and benefits as a total package when seeking
to maintain comparability and when setting pay policies aimed at
recruiting and retaining employees.

Between fiscal years 1990 and 2006, few employees moved among financial
regulators and the movement among these agencies presented no discernible
trend. Specifically, 86 percent (13,433) of the 15,627 employees that left
during this period (i.e., moving or resigning but not retiring), resigned
from federal employment. Annually, the percentage of employees who moved
to another financial regulator ranged from a low of 1 percent in fiscal
year 1997 (16 out of the 1,362 who moved or resigned) to a high of 8
percent in fiscal year 1991 (97 out of the 1,229 who moved or resigned).
The total number of financial regulatory employees was 15,400 and 19,796
during those 2 years, respectively.

References

Visible links
   67. http://www.gao.gov/cgi-bin/getrpt?GAO-07-678
*** End of document. ***