Federal Workforce: Payroll and Human Capital Changes During Downsizing
(Chapter Report, 08/13/1999, GAO/GGD-99-57).
Pursuant to a congressional request, GAO provided information on the
factors causing the federal payroll to increase while the number of
federal employees decreased during downsizing, focusing on the extent to
which each major factor contributed to the increase in the federal
payroll during fiscal years (FY) 1993 through 1997.
GAO noted that: (1) between the beginning of FY 1993 and the end of FY
1997, the total federal payroll grew by $8.7 billion to $102.4 billion,
while the size of the federal workforce decreased from 2.2 million
employees to 1.9 million; (2) in real terms, however, overall federal
payroll costs decreased because, in 1997 constant dollars, the payroll
declined by $2.4 billion during the 5-year downsizing period; (3)
because the decrease in the number of employees for the most part offset
actual aggregate payroll cost increases for those employees remaining on
the payroll, GAO calculated payroll costs for a constant workforce of
1.9 million employees over the 5-year period to isolate the payroll cost
increases and their causes; (4) on this basis, GAO estimated that,
between FY 1993 and FY 1997, payroll costs in nominal dollars increased
about $11,600 per employee and approximately $21.6 billion in total; (5)
for comparison, in terms of 1997 constant dollars, this payroll increase
was approximately $6,460 per employee and $12.0 billion for 1.9 million
employees; (6) the increased payroll costs were attributable to several
causes, but the predominant cause was the annual pay comparability
adjustment that is intended to keep federal pay competitive with that of
nonfederal employers; (7) the cost of employee benefits and changes in
the characteristics of the federal workforce also played a major role in
increasing the overall federal payroll cost; (8) employee benefits
increased due primarily to: (a) incentives paid to separating employees;
(b) the increasing proportion of employees in the Federal Employees'
Retirement System (FERS) and the increasing cost of the government's
required match for FERS employees' Thrift Savings Plan contributions;
and (c) increases in health insurance costs; (9) changes in the
characteristics of the federal workforce that increased payroll costs
included: (a) career step increases based on tenure and satisfactory
performance; (b) promotions; and (c) pay increases due to high quality
performance; and (10) the payroll cost increases that resulted from
these factors, however, were partially offset by the limited hiring of
staff, at grades below the governmentwide average, whose lower pay
levels helped dampen the overall average payroll and grade increases.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: GGD-99-57
TITLE: Federal Workforce: Payroll and Human Capital Changes
During Downsizing
DATE: 08/13/1999
SUBJECT: Compensation
Federal employees
Employee buyouts
Federal downsizing
Payroll records
Employee benefit plans
IDENTIFIER: Federal Thrift Savings Plan
Federal Employees Retirement System
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. This text was extracted from a PDF file. **
** Delineations within the text indicating chapter titles, **
** headings, and bullets have not been preserved, and in some **
** cases heading text has been incorrectly merged into **
** body text in the adjacent column. Graphic images have **
** not been reproduced, but figure captions are included. **
** Tables are included, but column deliniations have not been **
** preserved. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
United States General Accounting Office GAO Report to
Congressional Requesters August 1999 FEDERAL WORKFORCE
Payroll and Human Capital Changes During Downsizing GAO/GGD-99-57
United States General Accounting Office GAO Washington, D.C.
20548 B-280545 August 13, 1999 The Honorable Fred Thompson
Chairman, Committee on Governmental Affairs United States Senate
The Honorable Thad Cochran Chairman, Subcommittee on International
Security, Proliferation and Federal Services Committee on
Governmental Affairs United States Senate This report, prepared at
your request, provides information on the factors causing the
federal payroll to increase while the number of federal employees
decreased during downsizing and summarizes our prior studies on
the impact of downsizing on human capital. Specifically, the
report identifies the extent to which each major factor
contributed to the increase in the federal payroll during fiscal
years 1993 through 1997. The factors relate primarily to the
current system for compensating federal employees, a prominent
topic in the Office of Personnel Management's strategic plan and
an area that will need additional consideration as the government
evaluates the guiding principles and best practices for addressing
a wide range of federal human capital issues. In this regard, we
believe a strong link is needed between human capital planning and
agencies' strategic and programmatic approaches to accomplishing
their missions. Accordingly, in this report we are recommending
that Congress, in considering buyout legislation, continue to
require agencies to prepare and submit strategic buyout plans for
congressional review as a prerequisite for implementing buyout
authority and to implement downsizing consistent with the results
of their planning efforts. As agreed, unless you announce the
contents of this report earlier, we plan no further distribution
until 30 days from the date of this letter. At that time, we will
send copies of this report to Senator Joseph I. Lieberman, Ranking
Minority Member, Senate Committee on Governmental Affairs; Senator
Daniel Akaka, Ranking Minority Member, Subcommittee on
International Security, Proliferation and Federal Services, Senate
Committee on Governmental Affairs; Representative Dan Burton,
Chairman, and Representative Henry A. Waxman, Ranking Minority
Member, House Committee on Government Reform; and Representative
Joe Scarborough, Chairman, and Representative Elijah E. Cummings,
Ranking Minority Member, Subcommittee on Civil Service, House
Committee on Government Reform. We will also send copies to the
Honorable Janice R. Lachance, Director, Office of Personnel
Management. Copies will be made available to others upon request.
Page 1 GAO/GGD-99-57 Payroll and
Human Capital Changes During Downsizing B-280545 Please call me on
(202) 512-2700 if you have any questions concerning this report.
GAO contacts and major contributors to this report are
acknowledged in appendix III. Nancy R. Kingsbury Acting Assistant
Comptroller General Page 2 GAO/GGD-99-
57 Payroll and Human Capital Changes During Downsizing Page 3
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Executive Summary Since 1991, Congress and the Bush and Clinton
administrations have been Purpose reducing federal
employment levels as a means of restructuring the workforce and
reducing federal costs and budget deficits. The Department of
Defense (DOD) and non-DOD agencies have been using pay separation
incentives-commonly known as buyouts-to induce employees to
voluntarily leave federal service. During fiscal years 1993
through 1997, the federal civilian workforce (excluding the Postal
Service) was reduced by almost 300,000 employees, or 13.8 percent.
At the same time, the federal civilian payroll-basic pay, premium
pay, and benefits-increased by 9.3 percent, to $102.4 billion. In
addition, in analyzing this reduction in the federal workforce,
GAO and others have raised concerns about the implications of
downsizing on the employees-a human capital issue. In view of the
significance of these issues, the Chairmen of the Senate Committee
on Governmental Affairs and its Subcommittee on International
Security, Proliferation and Federal Services asked GAO to
determine why, during a period of downsizing, the federal payroll
increased while the number of federal employees decreased. In
response, GAO analyzed data from the Office of Personnel
Management's (OPM) Central Personnel Data File (CPDF) of federal
employees and reports on work years and personnel costs. GAO
identified the major factors and the extent to which each
contributed to the increase in the federal payroll during fiscal
years 1993 through 1997, and interviewed officials at selected
agencies to determine some of the reasons why the cost of these
factors increased. GAO also agreed to summarize prior GAO studies
on the impact of downsizing on human capital. Between the
beginning of fiscal year 1993 and the end of fiscal year 1997,
Results in Brief the total federal payroll grew by $8.7 billion
(9.3 percent) to $102.4 billion, while the size of the federal
workforce decreased from 2.2 million employees to 1.9 million.1 In
real terms, however, overall federal payroll costs decreased
because, in 1997 constant dollars, the payroll declined by $2.4
billion during the 5-year downsizing period.2 Because the decrease
in the number of employees for the most part offset actual
aggregate payroll cost increases for those employees remaining on
the payroll, GAO calculated payroll costs for a constant workforce
of 1.9 million employees over the 5-year period to isolate the
payroll cost increases and their 1 The numbers represent the
number of full-time equivalent employees or work years, which
generally consist of one or more employed individuals who
collectively complete 2,080 work hours in a given year. 2 GAO
normalized dollar amounts to 1997 inflation adjusted or constant
dollars for overall payroll comparisons. However, GAO primarily
used nominal dollars in the analyses because it allowed us to
illustrate all the causes of the payroll increase, some of which
were related to inflation and some not. Page 4
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Executive Summary causes. On this basis, GAO estimated that,
between fiscal years 1993 and 1997, payroll costs in nominal
dollars increased about $11,600 per employee and approximately
$21.6 billion in total.3 For comparison, in terms of 1997
constant dollars, this payroll increase was approximately $6,460
per employee and $12.0 billion for 1.9 million employees. The
increased payroll costs were attributable to several causes, but
the predominant cause was the annual pay comparability adjustment
that is intended to keep federal pay competitive with that of
nonfederal employers. (See figure 1.) The cost of employee
benefits and changes in the characteristics of the federal
workforce also played a major role in increasing the overall
federal payroll cost. Employee benefits increased due primarily
to (1) incentives paid to separating employees, (2) the increasing
proportion of employees in the Federal Employees' Retirement
System (FERS) and the increasing cost of the government's required
match for FERS employees' Thrift Savings Plan (TSP) contributions,
and (3) increases in health insurance costs. Changes in the
characteristics of the federal workforce that increased payroll
costs included (1) career step increases based on tenure and
satisfactory performance, (2) promotions, and (3) pay increases
due to high quality performance. The payroll cost increases that
resulted from these factors, however, were partially offset by the
limited hiring of staff, at grades below the governmentwide
average, whose lower pay levels helped dampen the overall average
payroll and grade increases. (See figure 1.) 3 GAO did not have
the data necessary to determine the exact payroll costs for the
1.9 million employees or for the net reduction of 300,000
employees during fiscal years 1993 through 1997. Therefore, GAO
used the difference between the average payroll cost per employee
for 2.2 million employees in 1993 and 1.9 million employees in
1997 to approximate a $21.6 billion payroll increase for a
constant workforce of 1.9 million employees. Page 5
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Executive Summary Figure 1: Percentage Effect of Various Factors
Contributing to Federal Payroll Increase, Fiscal Years 1993-1997
Source: GAO calculations based on OPM data. The various factors
that caused the payroll increase were not unique to the downsizing
period. The most significant factor contributing to the increased
payroll, the annual pay comparability adjustment, increased at a
somewhat lower rate during downsizing than in the preceding 5-year
period. Benefit cost increases, another significant contributor to
payroll cost increases during downsizing, increased about 29
percent less than in the 5-year period before downsizing. Career
step increases, promotions, performance pay, and hiring and
separation patterns all contributed to changes in the
governmentwide averages for employees' grade and payroll cost from
fiscal years 1993 through 1997. These changes in the
characteristics of the workforce resulted in an average grade
level increase of about one-half grade, from the beginning to
about the mid-point of GS-9. This increase of about one- half
grade during downsizing was similar to the average grade increase
in the 5 years before downsizing. The separation and hiring
practices employed by the agencies during downsizing also
contributed to substantially greater increases in Page 6
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Executive Summary employees' average age and service time during
downsizing than occurred during the previous 5-year period. The
implications of changes during downsizing on the government's
human capital will require continuing attention. GAO reviews have
found, for example, that a lack of adequate strategic and
workforce planning during initial rounds of downsizing by some
agencies may have affected their ability to achieve organizational
missions. Some agencies have reported that downsizing in general
led to such negative effects as the loss of institutional memory
and an increase in work backlogs. In a study of six agencies'
experiences in the later stages of the downsizing effort, GAO
found that most had planned and implemented buyout programs more
effectively than had agencies during the initial rounds of
downsizing and had generally linked buyouts to achieving specific
organizational objectives. This increase in buyout program
effectiveness was due in part to a statutory requirement (P.L.
104-208) that agencies prepare and submit strategic buyout plans
for congressional review prior to implementing their buyout
authority. Most of the recent and pending legislation to provide
buyout authorities requires specified agencies to prepare and
submit strategic buyout plans for congressional review prior to
implementing buyouts. GAO is recommending that Congress continue
these requirements. At the end of fiscal year 1997, the federal
workforce was composed of Background about 1.9 million
employees, excluding postal employees. About 1.6 million employees
were in white-collar positions, most of which are covered by the
General Schedule (GS). The Federal Wage System covered the
remaining trade, craft, and labor employees. GS and wage system
employees have separate pay schedules consisting of individual
grades and steps within grades. The federal payroll has three
primary components-basic pay, premium pay, and the cost of
benefits provided to employees. Basic pay consists of salaries and
wages paid directly to employees for duties performed during the
regular workweek. Premium pay is supplemental pay for overtime
work in excess of the regular workweek, which is generally an 8-
hour day and 5-day week, and for work at night and on Sundays and
holidays, as well as for availability duty. Benefit costs include
health and life insurance premiums, retirement and Thrift Savings
Plan contributions, and separation pay. Page 7
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Executive Summary Pay adjustments for the federal workforce have
traditionally occurred on an annual basis. Under the Federal
Employees Pay Comparability Act of 1990, pay adjustments for the
majority of the workforce are to consist of two components: (1) a
nationwide basic pay increase linked to the Employment Cost Index
(ECI) and (2) individual locality pay adjustments that apply to
specific geographical areas where nonfederal pay exceeds federal
pay by more than 5 percent. Beginning in 1994, the locality pay
component of the pay adjustment was to be phased in over a 9-year
period to reduce the then estimated pay gap of about 28 percent. A
different statutory pay system, based on comparability with local
prevailing wage rates, applies to wage system employees. Principal
Findings Between the beginning of fiscal year 1993 and the end of
fiscal year 1997, Overview of Federal Payroll the total federal
payroll grew by $8.7 billion (9.3 percent) to $102.4 billion,
Changes while the size of the federal
workforce decreased from 2.2 million employees to 1.9 million. In
real terms, however, overall federal payroll costs decreased
because in 1997 constant dollars the payroll declined by $2.4
billion during the 5-year downsizing period. The payroll cost
could have been about 16 percent higher in nominal dollars had
there been no employment reductions and had the average pay per
employee increased annually at the rate experienced during
downsizing. However, this estimate cannot be used as a downsizing
savings estimate because it does not consider all downsizing
costs, such as the amount of separation payments and payments to
contractors to provide services previously provided by downsized
employees. For example, a prior GAO review found that 5 of the 24
agencies reviewed were contracting for work previously done by
employees who had taken buyouts, to compensate for their reduced
workforces. Because the decrease in the number of employees offset
actual aggregate payroll cost increases for those employees
remaining on the payroll, GAO calculated payroll costs for a
constant workforce of 1.9 million employees over the 5-year period
to isolate the payroll cost increases and their causes. On this
basis, GAO estimated that, between fiscal years 1993 and 1997,
payroll costs increased about $11,600 and $6,460 per employee in
nominal and constant dollars, respectively, and approximately
$21.6 billion in total nominal dollars. In terms of 1997 constant
dollars, the total payroll increase was $12.0 billion. Page 8
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Executive Summary The increased payroll costs were attributable
primarily to the annual pay comparability adjustments that the
President recommended and that were agreed to by Congress. These
adjustments are intended to keep federal pay competitive with that
of nonfederal employers. Other factors that directly increased the
compensation of federal employees were increases in the costs of
benefits, premium pay, promotions and other performance-based pay
increases, and tenure-based career step increases. The latter
three pay- related factors, offset to some extent by limited
hiring at lower grade levels, pushed the average grade level, and
hence pay, of federal employees higher during the downsizing
period than they were before. Annual Pay Comparability
The annual pay comparability adjustments were responsible for an
Adjustments Were the Primary estimated 58.9 percent, or
$12.7 billion, of the estimated overall increase in Cause of
Increased Payroll Costs federal payroll costs for a constant
workforce of 1.9 million employees. This estimate includes the
effect of comparability adjustments not only on the basic pay of
federal employees, but also on the increased costs of certain
federal pay and benefits, like retirement benefits, whose costs
are set as a percentage of employees' basic pay. Increases in
basic pay totaled about $10.1 billion due to annual comparability
adjustments, and associated increases in premium pay and benefits
added about $2.6 billion to this total. Overall, the comparability
adjustments amounted to about a 17-percent increase in employees'
pay and benefits. Comparability adjustments were approximately 19
percent during the preceding 5-year period. Nonfederal employers
increased their employees' pay and benefits approximately 16
percent, as measured by the ECI, during the federal downsizing
period. In most years before and during downsizing, the President
and Congress acted to reduce the adjustments that otherwise would
have been provided under the statutory pay formulas. During the
5-year period, for example, the combined nationwide and locality
pay adjustments were about 9 percentage points less than the 24.5
percent adjustment that was calculated, but not fully provided,
under the pay formulas. Benefits Cost Increases Some
of the benefits that are calculated on the basis of employees'
basic Accounted for Over 10 Percent pay increased more than
the amount that was due to basic pay increases. of the Payroll
Cost Increase In addition, other benefits, such as health
insurance premiums, are not calculated on the basis of employees'
basic pay, and some of these benefits increased in cost during
fiscal years 1993 through 1997. In total, increases in the cost of
benefits, exclusive of those attributable to increases in Page 9
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Executive Summary employees' basic pay, accounted for an estimated
13.6 percent, or $2.9 billion, of the overall increase in federal
payroll costs. The average annual increase in benefit costs per
employee during downsizing was about 29 percent less than the
increase in benefit costs during the previous 5-year period.
According to OPM officials, the primary reasons for the increases
in benefit costs that did occur were agencies' costs for the
buyouts offered to encourage employees to leave, matching
contributions to the Thrift Savings Plan that agencies are
required to make for the increasing number of employees' in the
Federal Employees Retirement System (FERS), and the continuing
increase in the cost of health insurance premiums of which
agencies pay a portion. Premium Pay Increases Were Premium
pay, exclusive of the effect of comparability and other pay
Minimal increases, increased during
fiscal years 1993 through 1997, accounting for about 0.3 percent,
or $0.1 billion, of the overall increase in federal payroll costs.
This increase in the average amount of premium pay during
downsizing was about 91 percent less than during the previous 5-
year period. Agencies cited overtime payments as the primary
reason for the premium pay increases. Pay Actions, Combined With
Career step increases, promotions, performance pay, and the
combined Changes in Workforce effect of hiring and
separation patterns together resulted in an overall Composition,
Contributed to payroll increase and an increase in the
workforce's average grade. The Over One-Fourth of the Payroll
combination of the pay actions and the hiring and separation
patterns increased the average employee's payroll cost by about
$3,180 and the Increase total workforce's
payroll by about 29 percent, or $5.9 billion of the $21.6 billion
increase for a constant workforce of 1.9 million employees during
fiscal years 1993 through 1997. These changes in the
characteristics of the workforce also resulted in an increase in
the average grade level from GS- 9.1 to GS-9.5. This almost one-
half grade increase was similar to the rate of increase during the
previous 5-year period.4 The agencies' separation and hiring
practices used during downsizing also contributed to 1.7 and 1.6
year increases in employees' average age and service time,
respectively, during downsizing. These rates of increase for
average age and length of service were 55 and 100 percent greater,
4 The average grade levels cannot be used to compute the dollar
impact on the federal payroll because the composition of the
average may vary depending on the step levels of the employees,
and even higher grade and step levels do not necessarily equate to
higher pay. For example, if an average grade of GS-9, step 1, was
composed of a GS-8, step 8, and a GS-9, step 4, the average pay of
the two employees would actually be much higher than GS-9, step 1,
pay since both GS-8, step 8, and a GS-9, step 4, are paid at rates
higher than GS-9, step 1. Page 10
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Executive Summary respectively, than the rates that occurred
during the previous 5-year period. This aging of the workforce
could pose problems if knowledgeable and experienced workers begin
to retire in greater numbers. The administration's goal of
increasing the ratio of staff to supervisors to 15 to 1 during
downsizing was only partially met, as the ratio increased from 6.6
to 7.7 employees per supervisor. Principally because most of the
former supervisors remained with the government at the same grade
and pay, the change in the supervisory ratio had little impact on
the federal payroll. When federal agencies' allocate operating
resources, the largest share Impact of Downsizing on often is
devoted to their workforces. Thus, agencies' effective management
Human Capital of their human capital is critical to
both accomplishment of their missions and to efficient, effective,
and economical agency operations. Agencies' ability to provide
quality service while downsizing its human capital can be an
extremely challenging effort without adequate planning. GAO
reviews have found that a lack of adequate strategic and workforce
planning during downsizing by some agencies may have affected
their ability to provide quality service. Also, the use of
agencywide hiring freezes made the replacement of employees with
the types of skills needed to perform agencies' missions
problematic. Some agencies acknowledged that the loss of critical
employees could have been avoided had the agencies done meaningful
planning and granted their buyouts consistent with those plans.
Agencies have also reported that downsizing in general led to such
negative effects as the loss of institutional memory and an
increase in work backlogs. A more recent GAO study of six agencies
found that most of these agencies had more effectively planned and
implemented buyout programs after Congress began to require that
strategic buyout plans be submitted to Congress. These agencies
also had generally linked buyouts to achieving specific
organizational objectives. These agencies were required by statute
(P.L. 104-208) to submit these plans to the appropriate
congressional appropriations and oversight committees prior to
implementing their buyout authority. The plans were required to
outline the intended use of the authority, including a description
of how the agency would operate without the eliminated positions
and functions. Recent and pending legislation to provide buyout
authorities generally, but not always, require the agencies to
prepare and submit strategic buyout plans for congressional review
prior to implementing buyouts. Page 11 GAO/GGD-99-57
Payroll and Human Capital Changes During Downsizing Executive
Summary Strategic buyout plans are especially important given the
general lack of attention to human capital issues in federal
agencies' annual performance plans that are prepared under the
Government Performance and Results Act of 1993. In this regard, we
recently reported that most of the fiscal year 2000 annual
performance plans do not sufficiently address how the agencies
will use their human capital to achieve results.5 This general
lack of attention to human capital issues is a very serious
omission because only when the right employees are on board and
provided the training, tools, structure, incentives, and
accountability to work effectively is organizational success
possible. Human capital issues will require continuing attention
to minimize possible adverse effects on government performance. In
this regard, GAO continues to believe that a strong link is needed
between human capital planning and agencies' strategic and
programmatic approaches to accomplishing their missions. To ensure
the most cost-effective use of any future buyouts and to help
Recommendations to mitigate the adverse effects that can result
from poorly planned Congress downsizing, GAO
recommends that Congress, in considering buyout legislation,
continue to require agencies to prepare strategic buyout plans as
a prerequisite for implementing buyout authority and to implement
downsizing consistent with the results of their planning efforts.
Similar to what was done for buyouts authorized by P.L. 104-208,
Congress should also require agencies to submit their plans to
appropriate congressional committees prior to implementing their
buyout authority. DOD and OPM provided written comments on a draft
of this report. Each Agency Comments agency concurred with
the report's findings. OPM's additional comments are included at
the ends of chapters 2, 3, and 4. Subsequent to obtaining these
agencies' comments, GAO revised the report to provide additional
context on the impact of downsizing on human capital. Since
comments had been previously requested from the appropriate
agencies regarding the human capital issues discussed in its prior
reports, GAO did not request additional comments. 5 Managing for
Results: Opportunities for Continued Improvements in Agencies'
Performance Plans (GAO/GGD/AIMD-99-215, July 20, 1999). Page 12
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Page 13 GAO/GGD-99-57 Payroll and Human Capital Changes During
Downsizing Contents 4 Executive Summary 18 Chapter 1
Origins of Federal Downsizing
18 Introduction Composition of the Federal Workforce
19 Components of the Federal Payroll
20 Employee Pay Adjustments
20 Objectives, Scope, and Methodology
22 27 Chapter 2 Payroll Increases for Remaining
Employees Were Larger 29 Impact of
Downsizing Than Payroll Decreases Resulting From
Downsizing Factors that Increased the Federal Payroll During
30 on Federal Payroll and Downsizing Human Capital
Impact of Downsizing on Human Capital
31 Conclusions
32 Recommendations to Congress
32 Agency Comments
33 34 Chapter 3 Annual Pay Comparability
Adjustments Were the Major 34 Annual Pay
Cause of the Increases in Payroll Costs Benefits Cost Increases
Accounted for Over 13 Percent of 35
Comparability the Payroll Cost Increase Adjustments
Were the Premium Pay Increases Were Minimal
37 Agency Comments
39 Primary Cause of Increased Payroll Costs 40 Chapter 4
Impact of Career Steps, Promotions, and Performance
40 Career Steps, Other Pay Impact of Hiring and
Separation Patterns 43 Pay
Actions, and Workforce Decrease in Number of
Supervisors and Managers Had 47 Limited
Payroll Impact Composition Changes Contributed About One-Fourth of
the Payroll Increase Page 14 GAO/GGD-99-57 Payroll and
Human Capital Changes During Downsizing Contents Agency Comments
49 Appendix I: Comparison of Benefit Costs and Premium
50 Appendixes Pays per Employee for the 5-Year Periods Before
and During Downsizing Appendix II: Comments From OPM
52 Appendix III: Major Contributors to This Report
53 Table 2.1: Factors Causing Increased Cost per Employee
31 Tables and Federal Payroll to Change, Fiscal Years
1993 Through 1997 Table 3.1: Changes in Cost of Benefits per
Employee and 36 Governmentwide During
Downsizing Table 3.2: Changes in Cost of Premium Pay per
37 Employee and Governmentwide During Downsizing Table 4.1:
Number and Amount of Personnel Actions
41 Affecting Payroll During Downsizing Period Table 4.2:
Proportion of Employees in, Separated From,
46 and Newly Hired Into Occupational Categories During Downsizing
Table 4.3: Effect of Acquiring and Separating or
49 Reclassifying Supervisors and Managers Governmentwide During
Downsizing Table I.1: Changes in Annual Average Cost of Benefits
per 50 Employee During Downsizing Due to
Changes in Benefits and Comparability and Other Pay Actions Table
I.2: Changes in Annual Average Cost of Premium
51 Pays per Employee During Downsizing Due to Changes in Premium
Pays and Comparability and Other Pay Actions Figure 1: Percentage
Effect of Various Factors 6
Figures Contributing to Federal Payroll Increase, Fiscal
Years 1993-1997 Figure 2.1: Percentage Increase Contributed by
Various 28 Factors to Federal Payroll
Costs, Fiscal Years 1993- 1997 Figure 2.2: Actual Versus
Projected Fiscal Year 1993-1997 30 Changes in
the Federal Payroll Amounts Figure 4.1: Annualized Average Number
of Career Step 41 Increases, Promotions,
and Performance Pay Increases per 1,000 Employees During
Downsizing Compared With the 5-year Period Before Downsizing Page
15 GAO/GGD-99-57 Payroll and Human Capital Changes
During Downsizing Contents Figure 4.2: Dollar Value of Career
Step Increases, 42 Promotions, and
Performance Pay Increases per Pay Action During Downsizing
Compared With the 5-year Period Before Downsizing Abbreviations
COLA cost of living adjustment CPDF
Central Personnel Data File DOD Department of Defense
ECI Employment Cost Index FAA Federal
Aviation Administration FEPCA Federal Employees Pay
Comparability Act of 1990 FERS Federal Employees
Retirement System FTE full-time equivalent FWS
Federal Wage System GS General Schedule OPM
Office of Personnel Management SSA Social Security
Administration Page 16 GAO/GGD-99-57 Payroll and Human
Capital Changes During Downsizing Page 17 GAO/GGD-99-57 Payroll
and Human Capital Changes During Downsizing Chapter 1 Introduction
Since 1991, Congress and the Bush and Clinton administrations have
been reducing federal employment levels as a means of reducing
federal costs and controlling deficits. For much of the period
since January 1, 1993, Department of Defense (DOD), and since
March 30, 1994, non-DOD, agencies have been paying separation
incentives-commonly known as buyouts-to induce employees to
voluntarily leave federal service. However, the federal payroll
increased by $8.7 billion, or 9.3 percent, from $93.7 billion to
$102.4 billion, while the federal civilian workforce was reduced
by about 297,000 employees, or 13.8 percent, to 1.9 million
employees during fiscal years 1993 through 1997. To determine why,
we identified and analyzed the major factors that resulted in an
increased federal payroll at the end of this period of downsizing.
We also summarized our prior studies on the impact of downsizing
on human capital. The administration and Congress began to reduce
the number of Origins of Federal employees in DOD agencies in
1991-primarily through military base Downsizing
closings-to reduce the federal payroll. Beginning in 1993,
Congress and the administration, through a series of legislative
acts and executive orders, established authorities and goals for
reducing federal staffing levels governmentwide. In this regard,
three key pieces of legislation were the National Defense
Authorization Act for Fiscal Year 1993; the Federal Workforce
Restructuring Act of 1994; and section 663 of the Treasury, Postal
Service, and General Government Appropriations Act of 1997, each
of which authorized agencies to pay separation incentives-commonly
known as buyouts-of as much as $25,000 to eligible employees as
inducements to leave federal service voluntarily. Buyouts provided
agencies a tool to avoid or reduce the need for reductions-in-
force. The National Defense Authorization Act authorized buyouts
for civilian employees at DOD agencies from January 1, 1993,
through September 30, 1997. This closing date was later extended
to September 30, 1999.1 The Federal Workforce Restructuring Act of
1994, in addition to authorizing buyouts, mandated governmentwide
reductions of 272,900 full-time equivalent (FTE) positions through
fiscal year 1999.2 The Workforce Restructuring Act generally gave
non-DOD agencies authority to offer buyouts from March 30, 1994,
through March 31, 1995. The 1997 1P.L. 102-484 authorized DOD
buyouts through September 30, 1997; P.L. 103-337 extended DOD
buyouts through September 30, 1999. 2According to OMB guidance, an
FTE or work year generally consists of one or more employed
individuals who collectively complete 2,080 work hours in a given
year. Therefore, either one full-time employee or two half-time
employees equal one FTE or work year. These hours include
straight-time hours only and exclude overtime and holiday hours.
Page 18 GAO/GGD-99-57 Payroll and Human
Capital Changes During Downsizing Chapter 1 Introduction
appropriations act gave most non-DOD agencies the authority to
offer buyouts from October 1, 1996, through December 30, 1997.
More recent legislation has extended the authority for offering
buyouts for DOD agencies to September 30, 2001, and for some non-
DOD agencies until September 30, 1999, with some specific
agencies, such as the Internal Revenue Service, having authority
until as late as January 1, 2003. The federal workforce was
composed of about 1.9 million FTE non-postal Composition of the
employees at the end of fiscal year 1997. Most "white-collar"
workers, Federal Workforce about 1.6 million, are covered
for pay and classification purposes by the General Schedule (GS),
which consists of 15 grades, each broadly defined in law in terms
of difficulty and responsibility of the work and the
qualifications required for its performance. Other white-collar
employees not covered under the GS system include the Foreign
Service, certain Veterans Health Administration medical personnel,
and the Senior Executive Service, which includes most employees
above the GS-15 level. The Office of Personnel Management (OPM)
categorizes each white-collar occupation on the basis of the
general subject matter of work, level of difficulty or
responsibility, and educational requirements. A description of the
five categories follows. * Professional occupations require
incumbents to use discretion and judgment in applying knowledge
acquired through education or training equivalent to a bachelor's
degree in a specialized field. Professional occupations include
statistician, accountant, and architect. * Administrative
occupations involve the exercise of analytical ability, judgment,
and the application of a substantial body of knowledge of
administrative or management principles and practices. While these
occupations do not require specialized educational majors, they do
involve the type of skills typically gained through a general
college education or progressively responsible work. These
occupations include air traffic controller, criminal investigator,
and financial analyst. * Technical occupations consist of
nonroutine work that is learned on-the- job or from specialized
training less than that represented by college graduation, to
support professional or administrative fields. Technical
occupations include practical nurse, economics assistant, and
engineering technician. * Clerical occupations require incumbents
to do structured work according to established policies, which are
learned through training or work experience, to support office
operations. Clerical occupations include secretary, clerk-typist,
and customs aide. Page 19 GAO/GGD-99-57 Payroll and Human
Capital Changes During Downsizing Chapter 1 Introduction * Other
white-collar occupations include those miscellaneous occupations
that are not included in one of the four other categories. Other
white-collar occupations include corrections officers and police.
Professional and administrative occupations generally have higher
entry levels and average grade levels than do clerical
occupations. As of September 30, 1997, federal blue-collar
employees numbered about 250,000. They are covered by the Federal
Wage System (FWS), the major pay system covering trade, craft, and
labor occupations. These occupations include machine tool work,
printing, and plumbing and pipefitting. Wage schedules consist of
15 grades, covering most nonsupervisory employees. Schedules for
supervisors and leaders are based on the nonsupervisory schedules,
but are separate from them. The federal payroll has three primary
components-basic pay, premium Components of the pay, and the
cost of benefits provided to the employees. Basic pay consists
Federal Payroll of the salaries and wages paid directly to
employees for duties performed during the employees' regular
workweek. Basic pay also includes amounts paid for annual and sick
leave. Premium pay is supplemental pay, such as overtime, for work
in excess of the regular workweek, which is generally an 8-hour
day and a 5-day week that does not include Sunday. Benefits
provided to employees at a cost to the government include
agencies' shares of health and life insurance premiums and
retirement contributions. The benefits cost category also includes
severance pay or buyout payments made to employees leaving the
federal government. Most federal employees receive annual pay
comparability adjustments Employee Pay under the Federal
Employees Pay Comparability Act of 1990 (FEPCA). Adjustments
FEPCA introduced a new annual pay-setting process that, beginning
in 1994, was to gradually raise federal pay rates to within 5
percent of nonfederal rates by the year 2002. Annual pay
comparability adjustments consist of two components: a single
nationwide basic pay schedule percentage adjustment and varying
adjustments in specific localities. The nationwide adjustment is
based on the Employment Cost Index (ECI), an index prepared by the
Bureau of Labor Statistics that measures the change in nonfederal
employers' wages and salaries. The ECI adjustment generally
applies to GS-that is, most- federal employees; Foreign Service
pay schedules; and pay schedules established under title 38 for
certain Veterans Health Administration employees. It may also be
applied to other pay systems, such as the Senior Executive
Service. The locality pay adjustments apply to specific Page 20
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 1 Introduction geographical areas within the continental
United States in which surveyed nonfederal pay rates exceed GS pay
rates by 5 percent. As with the ECI adjustment, the locality pay
adjustment may also be applied to other pay systems up to
established statutory pay ceilings. The locality and nationwide
pay comparability adjustments were intended to deal with the pay
gap as well as any annual increases in nonfederal pay rates. At
the time of enactment, the estimated pay gap was about 28 percent,
and FEPCA required that 20 percent of the gap be closed in 1994,
with locality pay adjustments each year thereafter until the pay
gap was reduced to 5 percent.3 FEPCA also authorizes the President
to propose lower alternative pay comparability adjustments under
certain specified conditions. Since 1994, either the adjustment
based on the ECI or the locality pay surveys have been limited by
alternative adjustments by the President and Congress. Employees'
pay also varies at a given grade level, depending on which of 10
steps the employee occupies within that grade. Each higher step
increases the employee's rate of basic pay within that grade.
Employees' advancement to the next step within a grade (career
step increases) is to be based on performance at an acceptable
level of competence and established waiting periods for the steps.
The waiting periods for steps 1 through 4, 5 through 7, and 8
through 10 are 52, 104, and 156 weeks, respectively. Employees
demonstrating "high quality performance" may advance more rapidly
through the steps by being granted additional performance-based
step increases (quality step increases). An employee may receive
only one such increase during any 52-week period. Pay under FWS is
to be based on what private industry is paying for the same kind
of work in local wage areas. Federal blue-collar employees are to
be paid the full prevailing rate at step 2 of their grade. At step
5, the highest step, they may be paid 12 percent above the
prevailing rate. Since fiscal year 1979, separate legislation has
limited or delayed annual FWS wage adjustments. Blue-collar
employees also receive career step increases and, based on
creditable service, may advance to step 5 in about 6 years.
However, there are no provisions for their receiving quality step
increases. 3Prior to FEPCA, employees' pay was adjusted annually
in accordance with the Federal Pay Comparability Act of 1970. The
1970 Act also had the principle of pay comparability, requiring
annual adjustments to make federal pay rates comparable with
private sector pay rates for the same levels of work. Page 21
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 1 Introduction The Chairmen of the Senate Committee on
Governmental Affairs and its Objectives, Scope, and Subcommittee
on International Security, Proliferation and Federal Methodology
Services asked us to analyze changes in the federal workforce's
pay and composition since passage of the National Defense
Authorization Act for Fiscal Year 1993 and the Federal Workforce
Restructuring Act of 1994, to determine why the federal payroll
increased while the number of federal employees was decreasing.
Our specific objectives were to determine * to what extent and
for what reasons the amounts of basic pay, costs of benefits, and
premium pay changed during downsizing, and * how, and to the
extent possible why, grade and pay levels, the proportion of
employees in occupations, and other demographics of the workforce
changed in DOD and non-DOD agencies during downsizing and what
impact these changes had on the federal payroll. We also answered
the question of to what extent the number and salaries of
supervisors and managers and their ratio to employees changed
during downsizing and how those changes affected the payroll. In
addition, we summarized our prior findings on human capital issues
related to downsizing. To determine the extent to which and for
what reasons basic pay, benefits costs, and premium pay changed
during downsizing, we identified the factors that affected these
payroll costs-annual comparability adjustments, changes in the
grades and pay levels of federal employees, and changes in
agencies' use of benefits and premium pay. We obtained data on pay
comparability adjustments from reports prepared by the President's
Pay Agent4 and OPM's annual reports entitled Pay Structure of the
Federal Civil Service. We obtained data on the number of
employees, basic pay, benefits costs, and premium pay and the
average grade of full- time federal employees from OPM's annual
reports entitled Work Years and Personnel Costs and The Fact Book:
Federal Civilian Workforce Statistics. We used the difference
between the average payroll cost per employee for the 2.2 million
employees at the beginning of fiscal year 1993 and for the 1.9
million employees at the end of fiscal year 1997 to approximate a
$21.6 billion payroll increase for a constant workforce of 1.9
million employees. We did not have the data necessary to determine
the exact payroll cost for 4The President's Pay Agent is composed
of the Secretary of Labor and the Directors of OPM and the Office
of Management and Budget. Page 22
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 1 Introduction the employees who separated during fiscal
years 1993 through 1997. However, we were able to determine that
the average payroll cost per employee for the 2.2 million
employees included subsequently separated employees, who had lower
average payroll costs than the remaining employees. We calculated
the impact of annual comparability adjustments on the basic pay
portion of the federal payroll and on the average employee's basic
pay during fiscal years 1988 through 1997. We used a single
average for each of the years to represent the nationwide and
locality pay adjustments. To facilitate our analysis, we applied
these increases to the entire payroll for all federal civilian
employees, including blue-collar employees, who received pay
adjustments under another system.5 After determining the amount
that basic pay increased as a result of annual comparability
adjustments, we calculated the amounts that basic pay increased
each year due to other pay actions and separation and hiring
trends that resulted in changes in employees' grade and pay
levels. We then determined the proportionate relationship each
year between the impact on basic pay of comparability adjustments
and the impact on basic pay of other pay actions and of separation
and hiring trends. To calculate the effect comparability
adjustments and other pay actions had on the costs of benefits and
premium pay, we analyzed the extent to which the cost of each
benefit and premium pay changed during fiscal years 1988 through
1997 due to basic pay increases--as distinct from changes in usage
of the benefits and premium pays, which may also increase their
costs. Since most benefits and premium pays are based on
employees' basic pay, we calculated the proportions of the costs
of benefits and premium pays that increased annually due to pay
comparability adjustments. We then used these proportionate cost
increases due to comparability adjustments, and the proportionate
relationship of comparability adjustments to grade and pay level
changes, to calculate the increases in the costs of benefits and
premium pay due to grade and pay level changes. One exception was
that we did not attribute any increases in the government's share
of health insurance premiums to pay comparability adjustments or
other pay actions, since health insurance premiums are not based
on employees' pay. We considered the changes in costs of benefits
and premium pay to be the difference between the amounts due to
comparability adjustments and grade and pay changes and the total
changes in the annual costs of benefits and premium pays. 5Blue-
collar employees are covered by a different pay system; however,
their annual pay adjustments have been similar to those received
by white-collar employees. Page 23
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 1 Introduction Our calculations of the impact of
comparability and other pay increases do not provide an exact
amount but rather an estimate, because the government's costs for
some benefits and premium pay, such as overtime, are also affected
by other factors, such as a cap on the amounts that can be paid.
In addition, per employee costs were calculated on the basis of
all employees because data were not readily available on the
specific number of employees who received each type of benefit or
premium pay. Further, although temporary employees generally do
not receive benefits, we did not determine and adjust for the
number of such temporary employees in our calculations. Temporary
employees comprised about 10 percent of the federal civilian
workforce in fiscal year 1997. We totaled the change due to usage
in the costs of various benefits and premium pays to determine
whether, in aggregate, the annual costs of benefits and premium
pays increased during the downsizing period and thus contributed
to an increase in the federal payroll.6 In addition, we determined
the extent to which the costs of benefits and premium pays due to
usage increased or decreased during fiscal years 1993 through
1997. To determine how, and to the extent possible why, the
proportion of employees in occupations and grade and pay levels
changed governmentwide, as well as in DOD and non-DOD agencies, we
used OPM's Central Personnel Data File (CPDF) to analyze data on
the numbers of employees in occupations and general occupational
categories in each agency and governmentwide.7 We calculated the
extent to which the percentage of employees in occupations with
500 or more employees, and in general occupational categories,
changed in proportion to the federal workforce between October 1,
1987, and December 31, 1992, compared with the period January 1,
1993, to March 31, 1998. We also did the same analysis for the
periods from October 1, 1987, through March 31, 1994, and April 1,
1994, through March 31, 1998, for non-DOD agencies. We ranked the
five agencies with the highest increases and decreases in
individual occupations and in each general occupational category.
We then selected eight agencies that were among those ranked
highest in a number of these categories and that also had payrolls
that represented over 3 percent of the total federal payroll. Two
agencies with smaller payrolls-the Federal Emergency Management
Agency and the Department of Housing and 6The entire amount of
annual changes in the cost of health insurance premiums was
included in the total amount of benefit costs attributed to usage.
7The CPDF contains computerized records of most federal civilian
employees, including employees' grade levels and pay, as well as
their personnel actions, such as promotions and performance pay
increases. These records are maintained and updated quarterly by
OPM from data provided by federal agencies. Page 24
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 1 Introduction Urban Development-were selected because
they had the highest number of changes in several occupations and
occupational groups. We interviewed officials at these 10 agencies
to determine reasons why the changes had occurred. To determine
how and why employees' grade and pay levels changed and the impact
of those changes, we identified the primary factors that increased
employees' pay. We obtained and analyzed CPDF data on the types of
pay actions and determined that career steps, promotions, and
performance pay increases together played the largest part in
increasing employees' pay. We analyzed the CPDF data to determine
the total number and amounts of these actions for the two 5-year
periods and adjusted the amounts to exclude the sums attributable
to annual pay comparability adjustments. We had previously
estimated the effect of comparability adjustments on the costs of
benefits and premium pay. We used these estimates and assumed that
the pay actions' impact was in the same proportional relationship
to the comparability pay adjustments' effect on the costs of
benefits and premium pay as was the proportional relationship of
comparability and pay actions for basic pay. To determine the
impact of separation and hiring patterns on the average grade
levels, we also analyzed the CPDF database to determine the
average grade for employees who were separated from, were hired
into, and who remained in the government in fiscal years 1988
through 1997. To determine whether the average grade increased at
a different rate during downsizing than before downsizing, we
calculated the difference in employees' average grade for the
periods before downsizing-fiscal years 1988 through 1992-and
during downsizing-fiscal years 1993 through 1997. We recognize
that some of the change in the average grade and pay would occur
as a result of downsizing if the distribution of separated
employees were uneven across federal grade levels. To determine
the extent to which the number and salaries of supervisors and
managers, and their span-of-control, changed, and the effect of
such changes on the federal payroll, we analyzed the CPDF data on
GS and blue collar employees to identify the number of supervisors
and managers, and their ratio to staff, on September 30, 1987;
September 30, 1992; and September 30, 1997, for all agencies
governmentwide.8 For fiscal years 1993 through 1997, we also
determined whether employees who had been 8We excluded blue collar
employees in calculating the number of employees in grades GS-12
through GS-15 on September 30, 1992, and September 30, 1997,
because blue collar employees do not have GS grades. Page 25
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 1 Introduction identified as supervisors or managers after
downsizing began, but were no longer identified as supervisors or
managers at subsequent dates, had a code indicating the reason for
the change, such as reclassification or transfer to another
position. For these former supervisors and managers, we reviewed
their salaries to determine the amount of any salary change at the
time they changed to nonsupervisory or nonmanagement positions.
For employees who became supervisors or managers, and for
supervisors and managers who left the federal government during
downsizing, we determined the amount of their salaries at those
times. Our estimate of the total pay of new supervisors and
managers assumed that 86 percent of those who were not hired from
outside government were either replaced in their previous
nonsupervisory positions by new hires or that, somewhere in the
line of succession caused by the vacancy in their former position,
an employee from outside government was hired. We assumed that 14
percent of the vacated positions were not filled, to reflect the
overall 14 percent staffing reduction during downsizing. We could
not verify our assumptions because the CPDF does not have
information on whether newly appointed supervisors or managers or
other employees were replaced in their former positions by new
hires or existing employees. To provide contextual information on
the impact of downsizing on human capital, we reviewed and
summarized the relevant portions of our prior reports. These
reports, issued during fiscal years 1996 through 1999, are cited
where the relevant discussion occurs in chapter 2. We did our
audit work in Washington, D.C., from September 1998 through July
1999 in accordance with generally accepted government auditing
standards. We requested comments on a draft of this report from
the Director of OPM and the Secretary of DOD or their designees.
OPM and DOD provided written comments. OPM's additional comments
are presented at the ends of chapters 2, 3, and 4, and are
reprinted in appendix II. OPM and DOD also provided technical
comments, which we incorporated in the report where appropriate.
Subsequent to obtaining these agencies' comments, we revised the
report to provide additional context on the impact of downsizing
on human capital. Since comments had been previously requested
from the appropriate agencies regarding the human capital issues
discussed in our prior reports, we did not request additional
comments. Page 26 GAO/GGD-99-57 Payroll and Human Capital
Changes During Downsizing Chapter 2 Impact of Downsizing on
Federal Payroll and Human Capital Between the beginning of fiscal
year 1993 and the end of fiscal year 1997, the total federal
payroll cost grew by $8.7 billion to $102.4 billion, even though
the number of employees decreased by about 300,000, because the
payroll cost reductions attributable to the employees who left
were less than the increased payroll costs for those employees who
remained with the federal government. In terms of 1997 constant
dollars, however, the payroll declined by $2.4 billion during the
5-year downsizing period. Based on the change in the average cost
per employee during fiscal years 1993 through 1997, the payroll
cost could have been about $16.4 billion, or 16 percent, higher
had there been no employee reductions. This estimate, however,
does not consider costs resulting from downsizing, such as the
cost of an increased number of separation payments and the use of
contractors to provide services previously provided by downsized
employees. Because the $8.7 billion increase was based on 2.2
million employees in 1993 and 1.9 million employees in 1997, we
recalculated the payroll increase of about $11,600 per employee as
it related to only 1.9 million employees so that we could isolate
the factors contributing to the increase. We developed a payroll-
increase approximation of about $21.6 billion.1 For comparison,
in terms of 1997 constant dollars, the payroll increase was
approximately $6,460 per employee and $12.0 billion for 1.9
million employees. The increased payroll costs for those employed
by the federal government were attributable to several causes, but
the predominant cause was the annual, statutorily-based pay
adjustment meant to make federal pay competitive with that of
nonfederal employers. Other factors that directly increased the
compensation of federal employees were increases in the cost of
employee benefits and a small increase in the amount of premium
pays that employees earned. These factors are discussed in more
detail in chapter 3. Certain changes in the characteristics of the
federal workforce also increased the overall federal payroll cost.
However, while career steps, promotions, and performance pay
increased the overall grade and payroll cost averages
governmentwide, hiring and separation patterns had the combined
effect of lessening the 1 We did not have the data necessary to
determine the exact payroll costs for the 1.9 million employees or
for the net reduction of 300,000 employees during fiscal years
1993 through 1997. Therefore, we used the difference between the
average payroll cost per employee for 2.2 million employees in
1993 and 1.9 million employees in 1997 to approximate a $21.6
billion payroll increase for a constant workforce of 1.9 million
employees. Page 27 GAO/GGD-99-57
Payroll and Human Capital Changes During Downsizing Chapter 2
Impact of Downsizing on Federal Payroll and Human Capital growth
in the governmentwide average grade and payroll cost. These
factors are discussed in more detail in chapter 4. Figure 2.1
shows the percentage each of these factors contributed to the
increase in payroll costs for those employed in 1997. The factors
that affected the federal payroll but not the characteristics of
the workforce are unshaded, while the factors that affected both
the payroll and the characteristics of the workforce are shaded.
Figure 2.1: Percentage Increase Contributed by Various Factors to
Federal Payroll Costs, Fiscal Years 1993-1997 Source: GAO
calculations based on OPM data. While our primary objective was to
identify and analyze reasons for the payroll increase, another was
to summarize our past findings on human capital issues related to
downsizing. Our previous reviews of downsizing found that a lack
of adequate strategic and workforce planning by some agencies may
have affected their ability to achieve organizational missions
with a reduced workforce. In a subsequent study of six agencies'
experiences in the later stages of their downsizing efforts, we
found that Page 28 GAO/GGD-99-57 Payroll and Human
Capital Changes During Downsizing Chapter 2 Impact of Downsizing
on Federal Payroll and Human Capital most of these agencies had
more effectively planned and implemented buyout programs and had
generally linked buyouts to achieving specific organizational
objectives. These latter agencies were required to submit
strategic buyout plans to Congress prior to implementing buyouts.
The actual payroll at the beginning of fiscal year 1993 was $93.7
billion, Payroll Increases for and it increased by $8.7
billion to $102.4 billion as of September 30,1997, Remaining
Employees the end of that fiscal year. During that same time
period, federal employment was reduced from 2.2 million employees
to about 1.9 million. Were Larger Than The net
increase of $8.7 billion in the payroll occurred because
employees' Payroll Decreases payroll costs increased
in excess of the payroll cost reductions resulting Resulting From
from the separation of about 300,000 employees during the period.
In Downsizing terms of 1997 constant dollars,
however, the payroll declined by $2.4 billion during the 5-year
downsizing period. At the same time, the average payroll cost per
employee increased in nominal dollars by $11,600, from about
$43,400 to $55,000, and by approximately $21.6 billion for a
constant workforce of 1.9 million employees. For comparison, this
increase in terms of 1997 constant dollars was about $6,460 per
employee and about $12.0 billion for 1.9 million employees. Had
the number of employees remained at 2.2 million, and had the
average cost of pay and benefits per employee still increased by
$11,600, the federal payroll would have increased from $93.7
billion to $118.8 billion in fiscal year 1997. Although the
federal payroll is less than it would have been without
downsizing, it nevertheless increased in relation to what it would
have been had it been held constant since the beginning of fiscal
year 1993. In effect, the retrospective 1993 payroll for the 1.9
million employees on the rolls as of September 30, 1997, was an
estimated $80.8 billion, and it increased by approximately $21.6
billion to $102.4 billion based on the $11,600 increase in the
average cost of pay and benefits per employee. Figure 2.2 presents
the actual and projected federal payrolls for fiscal years 1993
through 1997 for the scenarios discussed in the preceding
paragraph. The figure shows the projected federal payrolls if the
number of employees had remained at 2.2 million during downsizing
while the average pay per employee increased annually at the rate
experienced during downsizing, as well as showing the actual
federal payrolls during downsizing. Page 29 GAO/GGD-99-
57 Payroll and Human Capital Changes During Downsizing Chapter 2
Impact of Downsizing on Federal Payroll and Human Capital Figure
2.2: Actual Versus Projected Fiscal Year 1993-1997 Changes in the
Federal Payroll Amounts Source: GAO calculations based on OPM
data. The difference of approximately 16 percent between the
projected federal payroll if 2.2 million employees had remained in
the federal government and the actual payroll in fiscal year 1997
does not accurately represent the amount saved by downsizing. This
estimate does not consider the costs resulting from downsizing,
such as the cost of an increased number of separation payments and
the use of contractors to provide services previously provided by
downsized employees. In a previous report, we noted that 5 of 24
agencies reviewed acknowledged contracting out some work, formerly
done by federal employees who had taken buyouts, to compensate for
having a smaller workforce.2 Table 2.1 shows the factors and the
extent to which each contributed to Factors that Increased the
changes in employees' average pay and benefits and to total
federal the Federal Payroll payroll costs.
These factors are discussed in more detail in chapters 3 and 4.
During Downsizing 2 Federal Downsizing: Better Workforce and
Strategic Planning Could Have Made Buyouts More Effective
(GAO/GGD-96-62, Aug. 26, 1996). Page 30
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 2 Impact of Downsizing on Federal Payroll and Human
Capital Table 2.1: Factors Causing Increased Cost per Employee and
Federal Payroll to Change, Fiscal Years 1993 Through 1997
Employee's Impact on federal payroll (in
Percentage of increase caused Status/change factor
average pay
billions) by
factor Fiscal year 1993
$43,400 Increase due to annual pay comparability adjustmentsa
6,840 $12.7
58.9 Increases in employees' benefitsa
1,580 2.9
13.6 Increases in employees' premium paya
30 0.1
0.3 Increases due to career steps, promotions, performance pay
increases, and the combination of hiring and separation patternsb
3,180 5.9
27.3 Fiscal year 1997c
$55,000d
$21.6 100.0d
aAddressed in chapter 3. bAddressed in chapter 4. cEmployees'
average pay and the extent to which the factors increased the
federal payroll as of fiscal year 1997. dNumbers do not total due
to rounding. Source: GAO calculations based on OPM data. When
federal agencies allocate operating resources, the largest share
often Impact of Downsizing is devoted to
their workforces, that is, human capital. Thus, effectively on
Human Capital managing their human
capital is critical to both effective accomplishment of agencies'
missions and to efficient, effective, and economical operations.
Providing quality service while agencies downsize their human
capital can be extremely challenging without adequate planning.
Our reviews have found that a lack of adequate strategic and
workforce planning during downsizing by some agencies may have
affected their ability to provide quality service.3 Also, the use
of agencywide hiring freezes made the replacement of employees
with the types of skills needed to perform agencies' missions
problematic. Some agencies acknowledged that the loss of critical
employees could have been avoided had the agencies done meaningful
planning and granted their buyouts consistent with those plans.
Agencies have also reported that downsizing in general had
negative effects. For example, 11 or more of 24 agencies in the
1996 review cited a loss of institutional memory, an increase in
work backlogs, or a somewhat of a or great hindrance in performing
their mission. In a subsequent study of six agencies' experiences
in the later stages of their downsizing efforts, we found that
most of these agencies had more effectively planned and
implemented buyout programs and had generally 3 Federal
Downsizing: Better Workforce and Strategic Planning Could Have
Made Buyouts More Effective (GAO/GGD-96-62, Aug. 26, 1996) and
Federal Downsizing: Buyouts at the Farm Service Agency (GAO/GGD-
97-133, July 23, 1997). Page 31
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 2 Impact of Downsizing on Federal Payroll and Human
Capital linked buyouts to achieving specific organizational
objectives.4 This increase in buyout program effectiveness was
due in part to a statutory requirement (P.L. 104-208) that
directed agencies to submit strategic buyout plans to the
appropriate congressional appropriation and oversight committees
prior to implementing buyout authority. The plans were required to
contain an outline of the intended use of the authority, including
such information as the number and amounts of buyouts to be
offered, the positions and functions to be reduced or eliminated,
and a description of how the agency would operate without the
eliminated positions and functions. Recent and pending legislation
to provide buyout authorities generally, but not always, require
the agencies to submit similar plans prior to or after
implementation of the buyout program. Strategic buyout plans are
especially important given the general lack of attention to human
capital issues in federal agencies' annual performance plans that
were prepared under the Government Performance and Results Act of
1993. In this regard, we recently reported that most of the fiscal
year 2000 annual performance plans do not sufficiently address how
the agencies will use their human capital to achieve results.5
Specifically, few of the plans relate how the agency will build,
marshal, and maintain the human capital needed to achieve its
performance goals. This general lack of attention to human capital
issues is a serious omission because only when the right employees
are on board and provided the training, tools, structure,
incentives, and accountability to work effectively is
organizational success possible. Improved planning by some
agencies during the more recent downsizing Conclusions
period should help to minimize those agencies' future problems.
Problems such as skills imbalances and work backlogs, however, may
take some time to resolve. When human capital planning is linked
to agencies' strategic planning, problems such as skills
imbalances are more easily avoided. Thus, we continue to believe
that a strong link is needed between human capital planning and
agencies' strategic and programmatic approaches to accomplishing
their missions. To ensure the most cost-effective use of any
future buyouts and to help Recommendations to mitigate the
adverse effects that can result from poorly planned Congress
downsizing, we recommend that Congress, in considering buyout 4
Federal Downsizing: Effective Buyout Practices and Their Use in FY
1997 (GAO/GGD-97-124, June 30, 1997). 5 Managing for Results:
Opportunities for Continued Improvements in Agencies' Performance
Plans (GAO/GGD/AIMD-99-215, July 20, 1999). Page 32
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 2 Impact of Downsizing on Federal Payroll and Human
Capital legislation, continue to require agencies to prepare
strategic buyout plans as a prerequisite for implementing buyout
authority and to implement downsizing consistent with the results
of their planning efforts. Similar to what was done for buyouts
authorized by P.L. 104-208, Congress should also require agencies
to submit their plans to appropriate congressional committees
prior to implementing their buyout authority. DOD and OPM
concurred with our findings. OPM agreed that much of the Agency
Comments increase in payroll cost during downsizing was due to
pay raises, with smaller increases due to changes in the
workforce's composition and in benefits costs. OPM also said that
it believes the report reflects the success of the
administration's efforts to slow the rate of growth in federal
employment and payroll costs, noting that the growth would have
been much greater had the federal workforce not been reduced by
about 300,000 employees during the period we reviewed. We agree
that the federal payroll could have been larger in the absence of
downsizing and recognize the importance of the administration's
role in downsizing the government. However, as noted in chapter
1, both Congress and the administration played roles in the
downsizing effort. Subsequent to obtaining the agencies' comments,
we revised the report to provide additional context on the impact
of downsizing on human capital. Since comments had been previously
requested from the appropriate agencies regarding the human
capital issues discussed in our prior reports, we did not request
additional comments. Page 33 GAO/GGD-99-57 Payroll and
Human Capital Changes During Downsizing Chapter 3 Annual Pay
Comparability Adjustments Were the Primary Cause of Increased
Payroll Costs The increased payroll costs for the 1.9 million
federal employees in fiscal year 1997 were due primarily to
employees' annual pay comparability adjustments. The annual pay
comparability adjustments were responsible for an estimated 59
percent of the overall increase in federal payroll costs. This
estimate includes the effect of comparability adjustments not only
on the basic pay of federal employees, but also on the costs of
other types of pay and employee benefits, such as premium pay, the
costs of which are based on percentages of employees' basic pay.
Basic federal pay rose about $5,440 per employee due to annual
comparability increases during fiscal years 1993 through 1997, and
associated pay and benefits increased about $1,400 per employee.
These pay comparability adjustments were comparable to the
increases provided to nonfederal employees during the same period,
as measured by the ECI. Some of the benefits that are calculated
on the basis of employees' basic pay increased by more than the
amounts that were due to basic pay increases. In addition, some
benefits costs, such as health insurance premiums, that are not
calculated on the basis of employees' basic pay also increased
during fiscal years 1993 through 1997. In total, increases in the
cost of benefits, exclusive of the increases directly attributable
to increases in employees' basic pay, accounted for an estimated
13.6 percent, or $1,580 per employee, of the overall increase in
the average payroll cost per employee. Premium pay, exclusive of
the effect of comparability and other pay increases, also
increased in cost during fiscal years 1993 through 1997. This
accounted for about 0.3 percent, or $30, of the overall $11,600
increase in federal payroll costs per employee. During fiscal
years 1993 through 1997, comparability adjustments Annual Pay
increased the average payroll cost per employee by about $6,800,
or an Comparability estimated $12.7 billion for the 1.9
million employees. These pay adjustments affected not only
employees' basic pay, but also the amounts Adjustments Were the
of premium pay and the cost of benefits provided by the
government, since Major Cause of the most of these costs are
based on percentages of an employee's basic pay. Increases in
Payroll Thus, the pay comparability adjustments resulted in
increases in the Costs average employee's basic
pay, benefits, and premium pay of about $5,400, $1,020, and $380,
respectively. The comparability adjustments increased pay and
benefits by about 17 percent, which was somewhat less than the
approximately 19-percent increase resulting from such adjustments
during the preceding 5-year period. Page 34 GAO/GGD-99-57
Payroll and Human Capital Changes During Downsizing Chapter 3
Annual Pay Comparability Adjustments Were the Primary Cause of
Increased Payroll Costs Annual pay comparability adjustments are
intended to make federal pay competitive with that of alternative
employers so that the federal government can attract and retain
employees. During this period, comparability adjustments to
employees' basic pay totaled about 17 percent. This was comparable
to the 16-percent increase provided by nonfederal employers, as
measured by the ECI. In most years before and during downsizing,
however, the President and Congress acted to reduce the
adjustments that otherwise would have been provided under the
statutory pay formulas. During the 5-year period, for example,
the combined nationwide and locality pay adjustments were about 9
percentage points less than the 24.5 percent adjustment that was
calculated but not provided under the pay formulas. During fiscal
years 1993 through 1997, the cost of benefits, exclusive of the
Benefits Cost effect attributable to
comparability and other pay increases, increased by Increases
Accounted $1,576 per employee. Thus, the cost of
benefits for 1.9 million employees contributed an estimated $2.9
billion, or 13.6 percent, of the $21.6 billion for Over 13 Percent
of increase in the total payroll. However, the average annual
increase per the Payroll Cost employee was about
29 percent less than during the previous 5-year Increase
period. Also, the total change in the cost of benefits, compared
with the 5- year period preceding downsizing, was $4.2 billion
less governmentwide, due primarily to the reduction in the number
of employees. The cost of benefits per employee increased during
the 5-year downsizing period, as well as between most of the years
during the period, due primarily to increases in the cost of five
benefits. The five benefits primarily responsible for the increase
were health insurance premiums; retirement contributions,
including agencies' contributions to employees' Thrift Savings
Plan accounts; separation and severance payments; Social Security
contributions; and other benefits, which included retirement and
other benefits charged to agencies under special plans for non-
U.S. citizens in foreign areas. Table 3.1 shows the change in the
cost of benefits, per employee and governmentwide, between the
beginning of fiscal year 1993 and the end of fiscal year 1997.
Page 35 GAO/GGD-99-57 Payroll and Human Capital
Changes During Downsizing Chapter 3 Annual Pay Comparability
Adjustments Were the Primary Cause of Increased Payroll Costs
Table 3.1: Changes in Cost of Benefits per Employee and
Governmentwide During Downsizing Change in cost of benefits per
employeeChange in cost of benefits projected to 1.9 Benefit
during downsizinga million employees (in
thousands) Health insurance
$249 $463,204 Life
insurance
(9) (16,287)
Retirement
520 968,102 Social
Security
232 430,946 Workers
compensation
89 165,272 Overseas
allowance
(70) (131,078)
Severance/separation pay
385 716,481 Other
benefits b
120 223,831
Miscellaneous benefits c
61 113,581 Total
$1,576d $2,934,051d
aPer employee costs are calculated on the basis of all employees
because data were not readily available on the specific number of
employees who received each type of benefit. bOther benefits
include agencies' costs for employees' retirement and other
benefits under special plans for non-U.S. citizens in foreign
areas. They also include relocation and other expenses related to
the movement of employees to new duty stations. cMiscellaneous
benefits include uniform allowances, nonforeign cost-of-living-
adjustments (COLA), retention allowances, and recruitment and
relocation bonuses. dNumbers do not total due to rounding. Source:
GAO calculations based on OPM data. Although benefits costs
increased during downsizing, the increases were at a lower rate
than during the previous 5-year period. Benefits costs per
employee increased by $315 annually during downsizing. This
increase was $129, or about 29 percent, less than the annual
increase per employee during the previous 5-year period. Appendix
I contains additional information comparing the downsizing period
with the previous 5-year period. Since the costs of benefits are
generally determined by governmentwide policies, we obtained
opinions from OPM officials who are knowledgeable of the effect of
these policies. According to OPM officials, the primary reasons
for the increases that did occur were agencies' costs for buyouts
and for their contributions to employees' Thrift Savings Plan
accounts. During downsizing, over 170,000 buyouts were paid,
totaling about $3.9 billion in separation pay. OPM officials said
that the increase in retirement and Social Security contributions
during downsizing was due primarily to the increasing number of
employees covered by the Federal Employees Retirement System
(FERS). As employees increased their contributions- now averaging
about 4 percent of annual pay-to the Thrift Savings Plan, the
agencies had to match those contributions under FERS. In addition,
under FERS, agencies are required to make Social Security
contributions for the increasing number of employees in FERS. Page
36 GAO/GGD-99-57 Payroll and Human Capital Changes
During Downsizing Chapter 3 Annual Pay Comparability Adjustments
Were the Primary Cause of Increased Payroll Costs During fiscal
years 1993 through 1997, the cost of premium pay, exclusive
Premium Pay of the effect of
comparability and other pay increases, increased by $30 per
Increases Were employee. Thus, the cost
of premium pay contributed to increasing the total payroll cost
for the 1.9 million employees by an estimated $0.1 billion,
Minimal or 0.3 percent, of the
$21.6 billion increase in the total payroll. The annual average
increase in the cost of premium pay per employee was about 91
percent less than during the previous 5-year period. Also, the
total cost of premium pay compared with the 5-year period
preceding downsizing was $1.4 billion less on a governmentwide
basis, primarily due to the reduction in the number of employees.
Premium pay costs on a per employee basis increased during the 5-
year downsizing period, as well as generally between years in the
period. Although most types of premium pay decreased during
downsizing, there was an overall increase, due primarily to
increases in the costs of overtime, Sunday premium pay, and
physicians' comparability allowances. Table 3.2 shows the change
in the cost of premium pay, per employee and governmentwide,
between the beginning of fiscal year 1993 and the end of fiscal
year 1997. Table 3.2: Changes in Cost of Premium Pay per Employee
and Governmentwide During Downsizing Change in cost of premium pay
per Change in cost of premium pay projected Premium pay
employee during downsizinga to 1.9 million employees
(in thousands) Overtime
$121 $225,506 Holiday
(3) (4,914) Sunday
8 14,426 Night
differential
(15) (28,758) Hazardous duty
(1) (2,308) Post
differential
(16) (30,043) Supervisory
differential
1 2,625 Physicians'
comparability allowances
30 55,339 Cash awards
(30) (55,413) Other premium
payb
(63) (117,862) Miscellaneous
premium payc
(2) (2,978) Total
$30 $55,618d aPer employee
costs are calculated on the basis of all employees because data
were not readily available on the specific number of employees who
received each type of premium pay. bIncludes payments above the
basic rates for any other premium pay, such as annual premium pay
for regularly scheduled standby duty. cIncludes staffing
differential and remote site allowance pay. dNumbers do not total
due to rounding. Source: GAO calculations based on OPM data. Page
37 GAO/GGD-99-57 Payroll and Human Capital Changes
During Downsizing Chapter 3 Annual Pay Comparability Adjustments
Were the Primary Cause of Increased Payroll Costs Premium pay
increased on a per employee basis during downsizing, although the
annual average increases were less than during the previous 5-year
period. The annual average increase of $6 per employee during
downsizing was about 91 percent less than the average annual
increase of $70 per employee during the previous 5-year period.
For additional information comparing the downsizing period with
the previous 5-year period, see appendix I. We contacted agencies
that experienced the highest increased costs for overtime, as well
as for several other premium pays, during downsizing to identify
the reasons for the cost increases. Department of Justice
officials attributed increases in overtime, Sunday, and other
premium pays to several factors, such as * a growth in the law
enforcement occupations that perform the majority of the
department's overtime, holiday, Sunday, and night work (for
example, the number of Border Patrol personnel increased by 33.2
percent), and * enactment of the Federal Law Enforcement Pay
Reform Act of 1990, which raised hourly and biweekly premium pay
caps, mandated higher special salary rates for certain law
enforcement employees, and provided pay increases for such
employees in specific large metropolitan areas.1 Justice officials
also attributed increases in the amount of physician comparability
allowances to a 55-percent increase in the Bureau of Prison's
correctional officer workforce, which includes physicians.
Department of Transportation officials said that their increases
in Sunday premium pay, holiday pay, and other premium pays during
downsizing were due to increased air traffic activity, which
resulted in higher numbers of Federal Aviation Administration
employees working hours that entitled them to premium pay. The
average amount of cash awards and other premium pay per employee
decreased annually during downsizing. DOD officials cited several
reasons for the decrease in the amount of most cash awards and
other premium pays, including a congressional mandate to reduce
the overall defense budget. 1While this authority was enacted
prior to the downsizing period, the effect continued during the
downsizing period. Page 38 GAO/GGD-99-
57 Payroll and Human Capital Changes During Downsizing Chapter 3
Annual Pay Comparability Adjustments Were the Primary Cause of
Increased Payroll Costs DOD and OPM agreed with our findings. OPM
also commented that the Agency Comments increased cost of
benefits was driven mainly by increases in direct agency payments
for retirement contributions and health insurance that would have
occurred in any case. Our review identified retirement
contributions and health insurance premiums as two of the primary
factors contributing to increased benefits costs, and they
probably would have occurred without downsizing. However, the
increase in severance pay, which contributed the second greatest
amount to the overall increase in benefits costs, was affected by
the buyouts and early retirements offered during the downsizing
period we reviewed. Page 39 GAO/GGD-99-57 Payroll and
Human Capital Changes During Downsizing Chapter 4 Career Steps,
Other Pay Actions, and Workforce Composition Changes Contributed
About One-Fourth of the Payroll Increase Pay actions other than
comparability payments and the combined effect of hiring and
separation patterns together resulted in an increase of about
$3,180 per employee, or about $5.9 billion of the payroll increase
for a constant workforce of 1.9 million employees during
downsizing. The combination of these factors also contributed to a
more highly paid workforce, with a higher proportion of employees
in occupations with higher pay governmentwide and an increase in
the average grade level from the beginning to about the mid-point
of GS-9. Although grade levels increased, the increases in the
average grade levels were about the same during the downsizing
period and the previous 5-year period. Our analysis of
governmentwide data also showed that reductions in the number of
supervisors and managers during downsizing increased the ratio of
supervisors to employees from 6.6 to 7.7 employees per supervisor
but had a limited effect on the federal payroll. Career step
increases, promotions, performance pay, and the combined Impact of
Career effect of hiring and separation patterns
contributed about $3,180 per Steps, Promotions, and employee and
$5.9 billion of the $21.6 billion payroll increase for 1.9 million
employees during downsizing. The resulting changes in the
Performance Pay characteristics of the workforce
caused the average grade level to increase from GS-9.1 to GS-9.5
during downsizing. Although employees received fewer promotions
during the downsizing period, this almost one-half grade increase
was similar to the rate-of-grade increase during the previous 5-
year period. Table 4.1 presents the total number and dollar
amounts of pay actions during downsizing for employees still in
the government in fiscal year 1997, exclusive of comparability pay
adjustments. The pay actions increased the federal payroll in the
years they were provided and in each succeeding year the employee
remained in the federal government. The effect of this $6.1
billion in pay increases on payroll cost was partially offset,
however, by the combined effect of hiring and separation patterns-
discussed in the next section-which served to lessen the payroll
increase to $5.9 billion during downsizing. Page 40
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 4 Career Steps, Other Pay Actions, and Workforce
Composition Changes Contributed About One-Fourth of the Payroll
Increase Table 4.1: Number and Amount of
Increase in basic pay due to Personnel Actions Affecting Payroll
Number of actions actions during downsizing During
Downsizing Period Type of action
during downsizing (in thousands)a Career step
increases 2,924,480
$2,733,829 Promotions
1,033,254 2,594,616 Performance based
increases 664,318
812,003 Total
4,622,052 $6,140,448 aAmounts exclude the
effect of comparability pay adjustments. Source: GAO calculations
based on data in OPM's CPDF. Figures 4.1 and 4.2 show that the
number and dollar amounts of promotions, career step increases,
and performance-based increases varied during the downsizing
period in comparison with the previous 5- year period. The average
number and amount of promotions and the average number of
performance pay awards decreased, while the average number and
dollar amounts of career steps and the average dollar amounts of
performance pay increased during downsizing. Figure 4.1:
Annualized Average Number of Career Step Increases, Promotions,
and Performance Pay Increases per 1,000 Employees During
Downsizing Compared With the 5-year Period Before Downsizing
Source: GAO calculations based on data in OPM's CPDF. Page 41
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 4 Career Steps, Other Pay Actions, and Workforce
Composition Changes Contributed About One-Fourth of the Payroll
Increase Figure 4.2: Dollar Value of Career Step Increases,
Promotions, and Performance Pay Increases per Pay Action During
Downsizing Compared With the 5-year Period Before Downsizing
Source: GAO calculations based on data in OPM's CPDF. During the
5-year downsizing period, 2.9 million career step increases for
Career Steps employees still in the
federal workforce in fiscal year 1997 accounted for an increase in
federal basic pay of about $2.7 billion, exclusive of
comparability pay increases. Career step increases are based on
the amount of time spent by an employee in a given grade and step
and an employee's demonstration of an acceptable level of
competence. Increases for GS employees may occur every 52, 104, or
156 weeks, depending on the employee's step. Exclusive of annual
pay comparability adjustments, the annual average number and
dollar amount of all career step increases during downsizing were
348 per 1,000 employees and $789 per career step, respectively.
This was an annual average increase of 1 career step per 1,000
employees and $46 per employee compared with the average for the
previous 5-year period. During the downsizing period, employees
who were still in the federal Promotions
workforce in fiscal year 1997 received about 1.0 million
promotions to a higher grade level, which increased federal basic
pay by about $2.6 billion, exclusive of annual comparability
adjustments. The annual average number and dollar amount of all
promotions during downsizing was 122 per 1,000 employees and
$2,120 per promotion, respectively. Compared Page 42
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 4 Career Steps, Other Pay Actions, and Workforce
Composition Changes Contributed About One-Fourth of the Payroll
Increase with the previous 5-year period, this was an annual
average decrease of 48 promotions per 1,000 employees and about
$18 per promotion, exclusive of pay comparability adjustments.
During the downsizing period, about 700,000 performance pay
increases Performance Pay were awarded to employees still
in the federal workforce in fiscal year 1997, which increased
federal basic pay by about $0.8 billion, exclusive of
comparability pay increases. Performance pay increases were a
combination of GS quality step increases and merit pay increases.
Quality step increases are step increases granted by the head of
an agency in recognition of high quality performance and above
that ordinarily found in the type of position concerned. Unlike
career step increases, quality step increases can be granted
annually. GS quality step increases were awarded only to
nonsupervisory and nonmanagement employees during the 5-year
period preceding downsizing and through the end of fiscal year1993
during downsizing. Merit pay increases were awarded in the
Performance Management and Recognition System, which existed
between the beginning of fiscal year 1985 and the end of fiscal
year 1993. This program established step and partial step
increases for managers and supervisors in grades GS-13 through GS-
15 who received certain high level ratings. When this program was
terminated, these employees became eligible for the GS career and
quality step increases. The annual average number and dollar
amount of all performance step increases during downsizing were 80
per 1,000 employees and $1,035 per performance step, respectively,
exclusive of annual pay comparability adjustments. These figures
represented an annual average decrease of 10 quality performance
steps per 1,000 employees and an increase of $47 per step,
compared with the previous 5-year period. In addition to the pay
actions discussed in the previous section, hiring and Impact of
Hiring and separation patterns also affected the workforce
composition during Separation Patterns downsizing. The
combination of hiring and separation patterns lessened the extent
to which the average payroll cost and grade level increased during
downsizing and, to an even greater extent, during the previous 5-
year period. Hiring and separation patterns, however, caused
employees' average age and length of service to increase at a
greater rate during downsizing than in the previous 5-year period.
Hiring and separation patterns had a limited effect on the
proportion of employees in occupational categories. Page 43
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 4 Career Steps, Other Pay Actions, and Workforce
Composition Changes Contributed About One-Fourth of the Payroll
Increase While career steps, promotions, and performance pay
directly increased Hiring and Separation individual
employees' pay and grade level as well as the overall payroll
Patterns Lessened Payroll cost and grade averages, hiring and
separation patterns can also affect the and Grade Increases
governmentwide average grade and payroll cost. The pay actions
increased the total basic pay of the recipient employees still in
the government in fiscal year 1997 by $6.1 billion. However, due
to hiring and separation patterns, the total basic pay of all
employees in fiscal year 1997 was only $4.7 billion greater than
the total basic pay for a constant workforce of 1.9 million
employees in the beginning of fiscal year 1993. This only modest
increase was due to new hires with lower average basic pay
entering the government and employees with higher basic pay
separating from the government during downsizing, which had the
effect of lessening the increase in total basic pay by $1.4
billion, or about $775 per employee. As previously discussed, most
benefit costs and premium pay are proportions of basic pay, and
thus the increase in basic pay also increased benefit costs and
premium pay of the employees by $1.2 billion governmentwide. As a
result, the net average payroll increase was $5.9 billion, $4.7
billion in basic pay and $1.2 billion in premium pay and benefits.
New hires also were employed at lower grade levels, and thus
reduced the extent to which governmentwide average grade levels
increased. On the other hand, the separation of employees with
grade levels below the governmentwide average-which occurred in
both 5-year periods- increased average grade levels. Thus, as a
result of the combined effect of pay actions and hiring and
separation patterns during downsizing, the average grade level
increased from GS-9.1 to GS-9.5. The number of new hires during
downsizing was about 300,000 fewer than the number of separations,
compared with the previous 5-year period in which the number of
new hires exceeded the number of separations. From fiscal years
1993 through 1997, the combined effect of hiring fewer and
separating a larger proportion of staff at grade and pay levels
below the governmentwide average lessened, by about 6 steps, the
increases that would otherwise have resulted from the pay actions
related to individual employees.1 In comparison, hiring and
separation patterns lessened the effect of pay actions by 7 and
1/2 steps during the prior 5-year period. 1The average grade and
step levels cannot be used to compute the dollar impact on the
federal payroll because the composition of the average may vary
and each higher grade and step level does not necessarily equate
to higher pay. For example, if an average grade of GS-9, step 1,
was composed of a GS-8, step 8, and a GS-9, step 4, the average
pay of the two employees would actually be much higher than GS-9,
step 1, pay since both a GS-8, step 8, and a GS-9, step 4, receive
pay rates higher than GS-9, step 1. Page 44
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 4 Career Steps, Other Pay Actions, and Workforce
Composition Changes Contributed About One-Fourth of the Payroll
Increase To illustrate how hiring and separation patterns can
lessen grade level increases due to pay actions, consider two
employees: a GS-8, step 1, and a GS-10, step 1. In this case, the
average grade would be GS-9, step 1. Assume that during the next 5
years, the GS-10 received 4 promotions to GS-14, step 1, while the
GS-8 separated from the government and was replaced by a new hire
at GS-6, step 1. The resulting increase in average grade would be
one grade, from GS-9 to GS-10 (the average of a GS-6, step 1, and
GS-14, step 1). However, had the GS-8 stayed in the government and
not received any grade or step increases, the average grade would
have increased by two grades to GS-11 (the average of GS-8, step
1, and GS-14, step 1). During the downsizing period, the average
age and years of service of Employee's Average Age
federal employees increased by 1.7 and 1.6 years to 45.1 and 16.0
years, and Length of Service respectively, at the end
of fiscal year 1997. These increases were Increased at a Greater
Rate respectively 55 and 100 percent greater than the 1.1
and 0.8 year increases Due to Hiring and in the
average age and years of service experienced during the previous
5- year period. We did not analyze the ages and years of service
for the Separation Patterns employees leaving and
entering the federal government during downsizing. However, the
accelerated increase in the average age and service of federal
employees indicated that the combination of ages and years of
service of the employees separated and hired during downsizing,
and the reduced level of hiring, contributed to an older, more
experienced workforce. Although the total number of employees in
individual occupational series Hiring and Separation
sometimes changed fairly substantially, shifts in the proportion
of federal Patterns Had Limited employees in broader
occupational categories-another measure of the Impact on the
Proportion of composition of the workforce-were much less
pronounced. For example, Employees in Occupational the secretarial
occupation within the clerical category decreased by 28,721
employees. While this represented about a 31-percent decrease in
the Categories number of secretaries in the
government, the clerical category decreased overall by only 3.2
percentage points. Within broad occupational categories, the more
highly paid professional and administrative categories experienced
the greatest increases. As table 4.2 shows, the percentage of
employees governmentwide in the administrative and professional
categories increased by 3.6 and 1.8 percentage points,
respectively, from January 1, 1993, to March 31, 1998. These
increases were small primarily because, while the administrative
and professional occupations comprised about 25 and 22 percent of
employees governmentwide as of January 1, 1993, only 21 and 19
percent of these employees, respectively, separated during
downsizing. Page 45 GAO/GGD-99-57 Payroll and Human
Capital Changes During Downsizing Chapter 4 Career Steps, Other
Pay Actions, and Workforce Composition Changes Contributed About
One-Fourth of the Payroll Increase Table 4.2: Proportion of
Employees in, Separated From, and Newly Hired Into Occupational
Categories During Downsizing Percentage of
Percentage Percentage as of employees
Percentage of new Percentage as of difference between
Occupational category 1/01/93
separating hires 3/31/98 1/01/93 and
3/31/98 Professional 22.3
18.9 17.9 24.1
1.8 Administrative 25.0
21.0 10.6 28.6
3.6 Technical 18.6
19.2 22.2 19.0
0.4 Clerical 15.5
18.9 28.2 12.3
(3.2) Other 2.5
3.9 6.1 2.6 0.1
Blue collar 16.1
18.1 15.0 13.4
(2.7) Total 100.0
100.0 100.0 100.0
0.0 Source: GAO calculations based on data in OPM's CPDF. Non-DOD
agencies experienced most of the changes in the administrative and
professional occupational categories. Non-DOD agencies' employees
in the two categories increased by 3.1 and 1.8 percentage points
during the governmentwide downsizing period. DOD agencies'
employees in the administrative and professional occupations
increased by 0.5 and 0.1 percentage points, respectively, during
the same period. The Department of Transportation experienced the
highest percentage increase in administrative employees due in
part to the transfer of about 6,300 employees from a clerical
occupation to the newly established administrative occupation of
airway transportation systems specialist in the Federal Aviation
Administration (FAA). According to FAA officials, the new
occupational series more closely matched the transferred
employees' duties and responsibilities. The proportion of
employees in the blue collar and clerical occupations in the
overall workforce decreased by 2.7 and 3.2 percentage points,
respectively, from January 1, 1993, to March 31, 1998. Blue collar
and clerical occupations comprised 16.1 and 15.5 percent of the
workforce, respectively, as of January 1, 1993, but experienced
18.1 and 18.9 percent, respectively, of the separations during
downsizing. And while 15.0 and 28.2 percent of the new hires
entered these occupations, respectively, they were fewer in number
than the employees separating. For example, there were 74,446 new
hires in the clerical category, while 127,431 employees separated.
DOD agencies experienced the greatest change in the clerical and
blue collar occupations, with decreases of 1.9 and 2.6 percentage
points, respectively, while non-DOD agencies' employees in these
categories decreased by 1.3 and 0.1 percentage points,
respectively, from January 1, 1993, to March 31, 1998. Defense
Logistics Agency officials told us that the Page 46
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 4 Career Steps, Other Pay Actions, and Workforce
Composition Changes Contributed About One-Fourth of the Payroll
Increase reduction in the proportion of blue collar employees
occurred because of base closures of depot activities heavily
populated with blue collar employees. The Department of Navy also
cited base closures as a primary reason. During the downsizing
period, a number of DOD bases were closed, and the contracting out
of services increased. Changes in the occupational categories
continued a trend that existed during the 5-year period before
downsizing. The same occupational categories increased
(administrative, professional, technical, and other) and decreased
(clerical and blue collar) during the downsizing period as during
the preceding period, with the only variation being the extent of
the change. For example, the professional category increased by
3.3 percentage points before downsizing compared with 1.8
percentage points during downsizing. During downsizing, the number
of supervisors and managers decreased Decrease in Number of and
the ratio of staff to supervisors increased governmentwide, but
the Supervisors and effect on the federal payroll
was limited. While the staff-to- supervisor/manager ratio did not
meet the former National Performance Managers Had Limited Review's
goal of 15 to 1, the governmentwide ratio did increase from 6.6 to
Payroll Impact 7.7 employees per supervisor or
manager during downsizing.2 The number of supervisors and managers
declined by 88,162, or 26.5 percent, from 332,100 to 243,938
governmentwide from the beginning of fiscal year 1993 through
fiscal year 1997. The net decrease was composed of (1) 202,626
employees who left supervisory and management positions, and (2)
114,464 employees who became supervisors and managers through
hiring, promotion of other staff, and other personnel actions. Of
the 202,626 employees who left their supervisory and management
positions during downsizing, 74,757, or 36.9 percent, left the
federal government. The remaining 127,869 employees were
reclassified as nonsupervisory or nonmanagement, 59,080 of whom
were either transferred or downgraded to other positions; OPM data
did not indicate how the other 68,789 were reclassified. Many of
the supervisors and managers were reclassified as team leaders. An
OPM official said that, although our analysis indicated there were
2,295 such reclassifications, the number may be understated
because, through 2The National Performance Review was established
by the Clinton administration to reinvent the federal government.
The Review was replaced by the National Partnership for
Reinventing Government in 1998. Page 47
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Chapter 4 Career Steps, Other Pay Actions, and Workforce
Composition Changes Contributed About One-Fourth of the Payroll
Increase fiscal year 1997, many agencies were not yet using any
code for identifying team leaders when submitting data to OPM. In
several respects, team leaders perform duties similar to those of
supervisors and managers, as indicated by OPM's General Schedule
Leader Grade Evaluation Guide, which defines the duties an
employee must perform to be classified as a team leader. The guide
states that leadership and supervision may be thought of as points
along a continuum from nonsupervisory to managerial work. The
range of duties a team leader may be assigned is very flexible-for
example, duties may be just sufficient to meet the minimum for
classification as a team leader to almost sufficient to warrant a
supervisory classification. Team leaders are responsible for
coordinating and supporting the work of their assigned teams to
ensure that the work is completed. Their specific duties and
responsibilities include ensuring that the organization's mission
is communicated to the team and integrated into the team's work
plans and products, and leading the team in balancing the workload
among team members according to skill levels. The
reclassifications and separations of supervisors and managers,
combined with the addition of newly hired or designated
supervisors and managers, resulted in a small increase in payroll.
Over 50 percent, or 1 of every 2 supervisors and managers, was
replaced through new hires, promotions of other staff, and other
personnel actions. The basic pay of these replacements, combined
with the promotions received by some of those reclassified from
supervisory and managerial positions, exceeded the total basic pay
for supervisors and managers who left government plus the total
reduction in basic pay experienced by some of those who were
reclassified. Table 4.3 shows the estimated increase in the
federal payroll during downsizing due to the situation described
above. The estimate assumes that most of the new supervisors and
managers who were not hired from outside government were either
replaced in their previous positions by new hires or that,
somewhere in the line of succession caused by the vacancy in their
former position, an employee from outside the government was
hired. We assumed that, based on the governmentwide employee
reduction rate of about 14 percent, 14 percent of the new
supervisors and managers were not replaced in their former
positions. We could not verify our assumption because the CPDF did
not have information with which to determine whether newly
appointed supervisors or managers or other employees were replaced
in their former positions by new hires or existing employees. Page
48 GAO/GGD-99-57 Payroll and Human Capital Changes
During Downsizing Chapter 4 Career Steps, Other Pay Actions, and
Workforce Composition Changes Contributed About One-Fourth of the
Payroll Increase Table 4.3: Effect of Acquiring and Separating or
Reclassifying Supervisors and Managers Governmentwide During
Downsizing Total increase or Number of supervisors and
(decrease) in basic Personnel action
managers pay (in thousands) New
supervisors and managersa
114,464 $4,694,688 Former
supervisors and managers Separated supervisors and managers
(74,757) (4,533,121)
Reclassified supervisors and managers whose pay remained the same
(114,711) Reclassified supervisors and managers whose pay
decreasedb
(13,096) (53,496)
Reclassified supervisors and managers whose pay increasedc
(62) 229 Net
increase or (decrease)
(88,162) $108,300 aAmount
attributed to new supervisors' and managers' basic pay was
adjusted based on assumption that 14 percent were not replaced in
their former jobs. bSince reclassified supervisors and managers
remained in the government, only the amounts of decreases in their
basic pay were considered in computing reductions in the federal
payroll. cReclassified supervisors and managers who received
promotions within 2 weeks of having their position changed or of
being downgraded. Source: GAO calculations based on data in OPM's
CPDF. The impact on the federal payroll of the reduction in the
number and proportion of supervisors and managers can also be
considered from another perspective. While the number of
supervisors and managers decreased, the number of employees in the
grade and pay levels usually occupied by supervisors and managers-
GS-12 and above-increased by 3,580 employees, from 554,419 to
557,999, during downsizing. This indicates that, while the
supervisory and management positions decreased, similarly graded
and paid positions increased during downsizing. This was partially
due to the fact that most former supervisors and managers retained
their grade and pay levels. The results of our governmentwide
review are consistent with changes at the Social Security
Administration (SSA) during fiscal years 1994 through 1998.3 SSA,
while reducing its supervisor to employee ratio, at the same time
created 1,900 new nonsupervisory positions, most of which were
filled by former supervisors. Consistent with OPM guidance, the GS
grades and salaries of these former supervisors did not change.
The total number of SSA employees in grades GS-12 and above
increased by over 900 employees during this time. DOD and OPM
concurred with our findings. OPM noted that the changes in Agency
Comments the composition of the
federal workforce likely reflected the continuation of a long-term
trend away from a technical and clerical workforce toward a more
professional and administrative workforce. 3Social Security
Administration: Compliance With Presidential Directive to Reduce
Management-to- Staff Ratio (GAO/HEHS-99-43R, Jan. 22, 1999). Page
49 GAO/GGD-99-57 Payroll and Human
Capital Changes During Downsizing Appendix I Comparison of Benefit
Costs and Premium Pays per Employee for the 5-Year Periods Before
and During Downsizing Table I.1: Changes in Annual Average Cost of
Benefits per Employee During Downsizing Due to Changes in Benefits
and Comparability and Other Pay Actions Annual average
Annual average
Amount of change change per change per Amount
of the Amount of change per employee due to employee before
employee during change between per employee due
comparability and Benefit downsizing
downsizing the periods to benefits
other pay actions Health insurance $194.68
$49.77 $(144.91) $(144.91)
$0.00 Life insurance 1.51
1.46 (0.05) 0.25
(0.30) Retirement 290.04
275.39 (14.65) (44.99)
30.34 Social Security 144.13
116.53 (27.60) (43.07)
15.47 Workers' compensation 3.48
31.21 27.73 28.61
(0.88) Overseas allowance 20.79
(4.59) (25.38) (23.12)
(2.26) Severance/separation pay 4.90
92.12 87.22 74.43
12.79 Other benefitsa 19.81
36.49 16.68 17.69
(1.01) Miscellaneous benefitsb 6.64
16.47 9.83 5.92
3.91 Total $685.98
$614.85 $(71.13) $(129.19)
$58.06 aOther benefits include costs charged to the agency for
employee's retirement, and other benefits under special plans for
non-U.S. citizens in foreign areas. Also includes relocation and
other expenses related to permanent change of station.
bMiscellaneous benefits include uniform allowances, nonforeign
COLAs, retention allowances, and recruitment and relocation
bonuses. Source: GAO calculations based on OPM data. Page 50
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Appendix I Comparison of Benefit Costs and Premium Pays per
Employee for the 5-Year Periods Before and During Downsizing Table
I.2: Changes in Annual Average Cost of Premium Pays per Employee
During Downsizing Due to Changes in Premium Pays and Comparability
and Other Pay Actions Annual average Annual average
Amount of change change per change per Amount of
the Amount of change per employee due to employee before
employee during change between per employee due to
comparability and Premium pay downsizing
downsizing periods premium pays
other pay actions Overtime $76.58
$79.73 $3.15 $(3.89)
$7.04 Holiday 2.99
4.48 1.49 1.03
0.46 Sunday 4.63
6.26 1.63 0.90
0.73 Night differential 5.61
2.58 (3.03) (2.83)
(0.20) Hazardous duty 0.70
0.89 0.19 (0.20)
0.39 Post differential 2.09
(1.90) (3.99) (3.81)
(0.18) Supervisory differential 0.13
1.37 1.24 0.15
1.09 Physicians' comparability 30.55
13.87 (16.68) (24.61)
7.93 allowances Cash awards 35.42
11.34 (24.08) (26.11)
2.03 Other premium paya 7.59
(1.14) (8.73) (3.82)
(4.91) Miscellaneous premium payb 0.32
(0.28) (0.60) (0.64)
0.04 Total $166.61
$117.20 $(49.41) $(63.83)
$14.42 aOther premium pay includes payments above the basic rates
for any other premium pay, such as annual premium pay for
regularly scheduled standby duty. bMiscellaneous premium pays
include staffing differential and remote site allowance pay.
Source: GAO calculations based on OPM data. Page 51
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Appendix II Comments From OPM Page 52 GAO/GGD-99-57 Payroll and
Human Capital Changes During Downsizing Appendix III GAO Contacts
and Staff Acknowledgments Michael Brostek, (202) 512-8676 GAO
Contacts Larry H. Endy, (202) 512-8676 In addition to those
named above, Thomas C. Davies Jr., M. Wayne Barrett,
Acknowledgments Rebecca L. Shea, and Gregory H. Wilmoth made
key contributions to this report. Page 53 GAO/GGD-99-57
Payroll and Human Capital Changes During Downsizing Page 54
GAO/GGD-99-57 Payroll and Human Capital Changes During Downsizing
Page 55 GAO/GGD-99-57 Payroll and Human Capital Changes During
Downsizing Page 56 GAO/GGD-99-57 Payroll and Human Capital
Changes During Downsizing Ordering Information The first copy of
each GAO report and testimony is free. Additional copies are $2
each. Orders should be sent to the following address, accompanied
by a check or money order made out to the Superintendent of
Documents, when necessary. VISA and MasterCard credit cards are
accepted, also. Orders for 100 or more copies to be mailed to a
single address are discounted 25 percent. Order by mail: U.S.
General Accounting Office P.O. Box 37050 Washington, DC 20013 or
visit: Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S.
General Accounting Office Washington, DC Orders may also be placed
by calling (202) 512-6000 or by using fax number (202) 512-6061,
or TDD (202) 512-2537. Each day, GAO issues a list of newly
available reports and testimony. To receive facsimile copies of
the daily list or any list from the past 30 days, please call
(202) 512-6000 using a touch-tone phone. A recorded menu will
provide information on how to obtain these lists. For information
on how to access GAO reports on the INTERNET, send e-mail message
with "info" in the body to: [email protected] or visit GAO's World
Wide Web Home Page at: http://www.gao.gov United States General
Accounting Office Bulk Rate Washington, D.C. 20548-0001
Postage & Fees Paid GAO Permit No. G100 Official Business Penalty
for Private Use $300 Address Correction Requested (410320)
*** End of document. ***