STUDY OF TOTAL COMPENSATION IN THE FEDERAL, STATE AND PRIVATE SECTORS
Document Type:
Collection:
Document Number (FOIA) /ESDN (CREST):
CIA-RDP89-00066R000800210002-2
Release Decision:
RIFPUB
Original Classification:
K
Document Page Count:
123
Document Creation Date:
December 23, 2016
Document Release Date:
November 21, 2012
Sequence Number:
2
Case Number:
Publication Date:
December 4, 1984
Content Type:
REPORT
File:
Attachment | Size |
---|---|
![]() | 4.34 MB |
Body:
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
COMMITTEE ON POST OFFICE
AND CIVIL SERVICE
US. HOUSE OF REPRESENTATIVES
Study of Total Compensation
in the
Federal, State and Private Sectors
December 4, 1984
Prepared by
Hay/Huggins Company
and
Hay Management Consultants
Atlanta ? Boston ? Charlotte ? Chicago ? Cincinnati ? Dallas ? Houston ? Kansas City ?
Los Angeles ? Minneapolis ? New York ? Philadelphia ? Phoenix ? Pittsburgh ? St. Louis ?
San Francisco ? San Jose ? Seattle ? Stamford ? Walnut Creek ? Washington, D.C.
HAY
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
COMMITTEE ON POST OFFICE
AND CIVIL SERVICE
US. HOUSE OF REPRESENTATIVES
Study of Total Compensation
in the
Federal, State and Private Sectors
December 4, 1984
Prepared by
Hay/Huggins Company
and
Hay Management Consultants
Atlanta ? Boston ? Charlotte ? Chicago ? Cincinnati ? Dallas ? Houston ? Kansas City ?
Los Angeles ? Minneapolis ? New York ? Philadelphia ? Phoenix ? Pittsburgh ? St. Louis
San Francisco ? San Jose ? Seattle ? Stamford ? Walnut Creek ? Washington, D.C.
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
TABLE OF CONTENTS
SECTION PAGE
I. SUMMARY OF REPORT AND FINDINGS 1
II. BASIS FOR TOTAL COMPENSATION COMPARISON ? 7
A. Total Compensation Perspective 8
B. Analytic and Comparative Methodology Used 8
C. Federal Data Base 9
D. Private Sector Data Bases 12
E. State Government Data Bases 14
III. METHOD OF COMPENSATION AND BENEFITS COMPARISON 16
A.
Cash Compensation Comparison
1. Components
2. Method
3. Data Collection
4. Data Preparation
5. Presentation
16
16
18
23
34
40
B.
Benefits Comparison
42
1. Components
42
2. Method
43
3. Data Collection
45
4. Presentation
46
IV.
COMPARISONS OF RETIREMENT SYSTEMS
47
A.
Cost Analysis
48
B.
Comparison of Private Sector Systems and CSRS
50
C.
Comparison of State Government Systems and CSRS
54
D.
Reasons for Difference
54
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
TABLE
OF CONTENTS (contd)
SECTION
PAGE
V.
FRINGE BENEFITS OTHER THAN RETIREMENT
62
A.
Life Insurance
64
B.
Disability Income Protection
68
C.
Health Care Benefits
74
D.
Holidays and Vacation
77
E.
Executive Perquisites
80
F.
Total Benefits
81
VI.
CASH COMPENSATION
85
A. Federal Sample Evaluation Results 85
B. Federal Compensation Practice 89
C. Comparison of Federal to Aggregate Private Sector
Systems 91
D. Comparison of Federal to State Cash Compensation 101
E. Comparison of Results to the 1984 PATC Survey 104
VII. TOTAL COMPENSATION 108
A. Comparison of Federal to Private Sector Total
Compensation 108
B. Comparison of Federal to State Total Compensation 112
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
List of Charts
Chart Page
Chart III-1: Illustrative Scattergram 22
Chart IV-1: Retirement Benefits Comparison
Federal to Private Sector 52
Chart IV-2: Retirement Benefits Comparison
Federal to Selected States 55
Chart IV-3: Replacement Rate Comparison
Federal to Private Sector
Age 62 - 20 Years Service 57
Chart IV-4: Replacement Rate Comparison
Federal to Private Sector
Age 62 - 40 Years Service 58
Chart IV-5: Replacement Rate Comparison
Federal to Private Sector
Age 65 - 20 Years Service 59
Chart IV-6: Replacement Rate Comparison
Federal to Private Sector
Age 65 - 40 Years of Service 60
Chart V-1: Death Benefits Comparison
Federal to Private Sector 65
Chart V-2: Death Benefits Comparison
Federal to Selected States 66
Chart V-3: Disability Income Comparison
Federal to Private Sector 70
Chart V-4: Disability Income Comparison
Federal to Selected States 71
Chart V-5: Health Care Benefits Comparison
Federal to Private Sector 75
Chart V-6: Health Care Benefits Comparison
Federal to Selected States 76
Chart V-7: Holidays and Vacations Comparison
Federal to Private Sector 78
Chart V-8: Holidays and Vacations Comparison
Federal to Selected States 79
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
List of Charts (contd)
Charts Page
Chart V-9: Executive Perquisites
Private Sector 82'
Chart V-10: Total Benefits Comparison
Federal to Private Sector 83
Chart V-11: Total Benefits Comparison
Federal to Selected States 84
Chart VI-1: Depiction of Evaluation Results
GS and Equivalent Grades 3-15 88
Chart VI-2: Federal Cash Compensation Practice
GS and Equivalent Grades 3-15 90
Chart VI-3: Private Sector Total Cash Compensation
GS and Equivalent Grades 3-15 93
Chart VI-4: Total Cash Compensation Comparison
Federal to Private Sector
GS and Equivalent Grades 3-6 95
Chart VI-5: Total Cash Compensation Comparison
Federal to Private Sector
GS and Equivalent Grades 7-15 97
Chart VI-6: Total Cash Compensation Comparison
Federal to Selected States
GS and Equivalent Grades 3-15 102
Chart VII-1: Total Compensation Comparison
Federal to Private Sector 111
Chart VII-2: Total Compensation Comparison
Federal to Selected States 114
-iv-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
List of Tables
Table Page
Table II-1: Size Distribution of Participants in
Hay's 1983 Benefits Comparison 13
Table 11-2: States that Participated in the Hay
Survey of Compensation and Benefits 15
Table III-1: Distribution of GS and Equivalent
Employee Population by Grade Level 1-18 25
Table 111-2: Distribution of GS and Equivalent Employee
Population by Selected Grade Level 26
Table 111-3: Sample Coverage of Employee Population by
GS and Equivalent Grade Level 28
Table 111-4: Weighted Average Step-in-Grade and
Interpolated Base Salary for Selected 35
GS and Equivalent Grades as of 1/2/84
Table 111-5: Weighted Average GS and Equivalent
Base Salaries Adjusted to 3/1/84 36
Table IV-1: Relative Value of Retirement System
Benefits 53
Table V-1: Differences between Federal Benefit
Plans and Average Private Sector and
State Benefit Plans as a Percent
of Pay 63
Table VI-1: Evaluation Results for the Sample of
392 Federal Positions by Grade Level 87
Table VI-2:
Table VI-3:
Aggregate U.S. Private Sector Average
Total Cash Compensation Results at 240
Points, for Exempt and Non-Exempt
Positions
92
Comparison of Federal to Aggregate
Private Sector Average Total Cash
Compensation for Grades 3, 4, 5 and 6 94
Table VI-4: Comparison of Federal to Aggregate
Private Sector Average Total Cash
Compensation for Grades 7-15 96
-'7-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
List of Tables (contd)
Tables Page
Table VI-5: Comparison of Federal to Aggregate
Private Sector Average Total Cash
Compensation, All Grades, and
Computation of Overall Average
Difference 100
Table VI-6: Comparison of Federal to Aggregate State
Average Total Cash Compensation, All Grades,
and Computation of Overall Difference 103
Table VII-1: Comparison of Federal to Aggregate Private
Sector Average Total Compensation, All Grades, and
Computation of Overall Average Difference 110
Table VII-2: Comparison of Federal to Aggregate State
Average Total Compensation, All Grades,
and Computation of Overall Average
Difference 113
-NT-i-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
List of Appendices
Appendices
A. Hay Cash Compensation Comparison Participant List
B. Hay/Huggins Benefits Comparison Participant List
C. Detailed Description of Hay Job Evaluation
Methodology
D. List of 392 GS and Equivalent Positions Selected
by Grade and Series, Including Population
E. Lists of 38 SES Jobs Evaluated
F. List of 24 Bureau of Labor Statistics Positions Included
G. Matrix of Jobs Matched by State
H. Prevalence of Benefit Practices
I. Characteristics of Participating Firms
J. Regression Statistics for Federal Cash Compensation Practice
k. Base Salary Comparisons to the Aggregate Private Sector
-vii-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
I. SUMMARY OF REPORT AND FINDINGS
The Committee on Post Office and Civil Service of the United
States House of Representatives has requested studies from The Hay
Group and the Congressional support agencies to assist with its
development of a new retirement system for Federal employees who are
subject to the social security system. The series of reports were
designed to provide a complete description of the key issues
affecting the development of a new system. This report compares
Federal total compensation practice to that of the private sector
and state governments so that retirement can be considered as part
of overall compensation policy.
It has been recognized that the Civil Service Retirement System
(CSRS) provides-benefits that, in some respects, are more valuable
than those provided to employees covered by other retirement
systems. However, much of the analysis of the CSRS and other
retirement systems which has been presented to the public is based
on a comparison of available sources which often are not consistent
in their treatment of retirement costs. Further, these assessments
seldom place the retirement benefits in the perspective of total
compensation. This report applies rigorous and consistent survey
and measurement methods that have long been used by The Hay Group
for private sector employers to compare Federal with non-Federal
compensation; including cash compensation, retirement and all fringe
benefits.
The retirement systems analysis includes the total package of
retirement benefits available to employees. While Federal employees
rely entirely on CSRS
employees in comparable
basic retirement system
for
retirement benefits, private sector
firms have social security as well as a
and, in the majority of cases, additional
deferred compensation plans which can be used to provide retirement
income. Examination of all these sources of income shows that the
retirement income available to the career employee in the private
sector comes close to that available to Federal employees covered by
-1-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
CSRS for a comparable career. However, since Federal employees can
retire earlier on full benefits, and since benefits are generally
indexed to the rate of inflation, CSRS is more valuable than private
sector systems by 6.4% of pay. For State systems, which contain
early retirement and indexing features closer to those of the
Federal program, the CSRS advantage is only 2.5% of pay.
While retirement for Federal employees is more valuable than
retirement for private sector employees, much of this difference
disappears when other fringe benefits are also analyzed.
Consideration of other fringe benefits reduces the difference
between the Federal and private sector to 2.8% of pay, and the
Federal and State difference to .5% of pay. This occurs for two
reasons. First, private sector employers have disability and life
insurance programs that provide protection which in the Federal
government is largely provided through the Civil Service Retirement
System. Second, other Federal benefit programs are often less
valuable than those found in both the private sector and State
governments. For instance, health benefits are worth 2.2% less, as
a percentage of pay, for the Federal employee than for the average
private sector employee.
Comparison of total cash compensation shows that General
Schedule pay would have to be increased by 10.3% to match private
sector pay levels for equivalent positions. The private sector
advantage is observed at all pay levels, and a special analysis of
Senior Executive Service (SES) positions shows that the equivalent
private sector executive is earning 58.4% more than the SES
incumbent. When compared to a sample of State government practices,
Federal pay is ahead by 7.8%.
As a result of the cash compensation difference, the Federal
fringe benefit advantage over the private sector disappears when all
elements of compensation are combined. The total compensation of
the Federal employee lags the private sector by 7.2%. Since the
-2-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Federal and State benefits are almost identical, the Federal total
compensation advantage of 7.1% is close to the cash compensation
difference.
In considering the retirement system to be developed for new
Federal employees to supplement social security coverage, one key
question is whether or not to provide benefits that cost
approximately the same as the current CSRS. If only retirement
systems are considered, and if comparability to other employers is
the goal, it might be suggested that the new supplemental system
should provide benefits that are less valuable than the current
system. However, the total compensation perspective shows that it
is only in the area of retirement that Federal employees are clearly
ahead of their private sector counterparts. Consideration of other
elements of compensation shows that, even with the current level of
retirement benefits, the average Federal employee will have a total
compensation package that lags the private sector.
Since compensation practices and levels change over time, i? is
important to bring data on all compensation practices to a single
point in time.. For this study, the date selected was March 1,
1984. Since the Committee will be developing a new retirement
system in 1985, it will be useful to update the comparison in 1985
to provide a current total compensation background. Accordingly,
Hay will update the analysis in 1985 to a date selected by the
Committee. Among other considerations, the update will incorporate
the upcoming Federal pay raise as well as the growth in private
sector salaries after March 1, 1984. If the Federal employees
receive a 3.5% pay increase, Hay anticipates that the difference
between Federal and private sector total compensation will increase
to 9% or more.
-3-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Summary of Sections
The report details the careful and consistent analytic approach
applied by Hay. Section II explains the basis for the total
compensation comparison. The selection of the data bases and the
methods used parallel Hay's total compensation analyses for major
employers in the United States. The approach is founded on the
thesis that meaningful findings must be based on the comparison of
"like things". Section II, therefore, presents the basis for the
comparison, noting that the Federal government's use of a single
national compensation system for the bulk of its civilian employees,
the size of its workforce and the variety of its activities make it
similar to a large and diverse corporation. While a significant
number of medium size employers (those with 100 to 1,000 employees)
are included in the data bases and the comparisons, it was not
considered appropriate to include small independent corporations
which have neither national compensation systems nor a similar
diversity of employment. This section also includes a discussion of
the data bases which are further detailed in the Appendices.
Section III presents the method of compensation and benefits
comparison. All significant elements of compensation (direct cash
and benefits) were reviewed in order to determine the total
compensation package available to private sector and State employees
as well as to Federal government employees. The section further
details the methods applied in this project, which h:ave been used
extensively in the analysis of private sector compensation
practices.
The results of the retirement plan analysis are presented in
Section IV. This section includes analysis of all retirement
programs available to employees, including defined benefit pension
plans and capital accumulation plans as well as social security.
This section demonstrates that while the replacement income at
retirement for a typical private sector retirement system is close
to that of the CSRS, there are features in the Federal system that
-4-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
make it more expensive. When all elements are considered, the Civil
Service Retirement System is worth 24.7% of pay .compared to 18.3%
for the average private sector retirement system and 22.2% for the
average State system.
Section V analyzes each of the other fringe benefits. These
include death benefits, disability income protection, health care
plans, holidays and vacations, and executive perquisites. While
executive perquisites are generally not available to Federal
employees, they comprise a significant part of the compensation
provided to private sector employees, particularly at higher pay
levels. Because many of the packages available to private sector
employees are not duplicated in the Federal government, and many of
those which are duplicated are more valuable in the private sector,
the total benefit package, exclusive of retirement, is worth 3.6% of
salary more in the private sector and 2.0% more in State governments
than in the Federal government. When combined with the advantage in
the retirement system, Federal benefits, as a whole, are worth 2.8%
of pay more than private sector employer benefits and .5% more than
State government benefit programs.
Section VI presents the analysis of cash compensation. The
significant finding is that Federal cash compensation would have to
be increased by 10.3% in order to equal aggregate private sector
average total cash compensation. The advantage of the private
sector over the Federal government exists at all levels of pay
within the General Schedule, ranging from 2% to 25%. When executive
positions are considered, it is found that Federal total cash
compensation would have to be increased 58.4% to equal aggregate
private sector total cash compensation. On the other hand, State
salaries lag behind Federal salaries by 7.8% of Federal pay.
-5-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Section VII combines the benefits and cash compensation into a
total compensation comparison.. Since cash compensation and fringe
benefits other than retirement are more valuable in the private
sector, the total compensation perspective shows that Federal
employees' total compensation is 7.2% behind the private sector on
average. As a result, even if a supplemental retirement system, is
linked with social security to produce benefits that are comparable
to those
1984, the
available to Federal employees hired before January 1,
total compensation available to new Federal employees will
also lag the private sector. Federal employees' total compensation
is 7.1% ahead of the total compensation of State employees. When
the two data bases are combined, the total compensation of Federal
employees lags the total compensation of other employees by 6.2%.
It is expect that the 1985 update of this analysis will increase the
advantage of private sector total compensation to 9% or more.
-6-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
II. BASIS FOR TOTAL COMPENSATION COMPARISON
The purpose of this study is to support the design of a new
Civil Service Retirement System (CSRS) by providing a total
compensation perspective. The Federal government is treated as if
it were a large private sector employer, and the same methods of
quality assurance, analysis, and comparison Hay has successfully
employed during its forty years of consulting to major American
employers are applied.
This approach is founded on the thesis that meaningful findings
must be based on comparisons of "like things". And as later
discussed in detail, the Federal government's use of a single
national compensation system for the bulk of its civilian employees
(the General Schedule), the size of its workforce, and the variety
of activities undertaken, make it most similar to a large and
diverse corporation. Hence, it is treated as such and compared to
primarily medium size (100 to 1,000 employees) and large (over 1,000
employees), diverse corporations to ensure the most relevant study
findings. In addition to private sector comparisons, Federal
government compensation is also compared to that of a sample of
State governments.
The critical features of this study are as follows:
? Total compensation perspective;
? Analytic and comparative methodology;
? Federal data base;
? Private sector data bases; and
? State data base.
Each of these features is discussed below.
-7-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
A. Total Compensation Perspective
The operative thesis of this study is comparability of total
compensation: specifically, the premise is that an equitable
compensation system will provide the same levels of total
compensation for positions of similar difficulty.
Total compensation is defined as the sum of base salary,
supplemental cash compensation, and the monetary value of benefits;
i.e., the value of all compensation, regardless of how provided.
Position difficulty, or content, is measured in terms of the
requirements of the specific position. Position requirements are
expressed in terms of factors which (1) are common to all positions;
and (2) exhaustively describe positions. The Hay Guide
Chart-Profile Method of Job Evaluation is used to measure position
difficulty. Section III provides a more detailed description of the
elements of total compensation and the Hay Method of Job Evaluation.
B. Analytic and Comparative Methodology Used
The approach taken throughout this study is the same as tHat
employed by Hay for major American companies which regularly compare
their overall compensation practices to those of other, comparable
organizations nationwide. In particular, the analyses are
formulated in terms of overall organizational compensation
practices, which describe relationships between compensation and
evaluated job difficulty. This approach allows organizations to
compare their compensation practices despite variations in employee
populations.
The analyses therefore consist of determining the overall
compensation practice of each included organization. This in turn
requires that, at a
selected, that each
properly evaluated
minimum, a representative sample of positions is
is accurately described, and that each is then
in terms of its content (difficulty).
-8-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Thereafter, the resulting set of combinations of job difficulty
score and compensation developed for each position must be analyzed
to develop an accurate representation of the organization's overall
compensation practice.
The client organization's practice is then compared to the
practices of other relevant organizations, each similarly
constructed. In this study, the Federal government compensation
practice is compared to those extant in the aggregate private
sector, as represented by the Hay compensation data bases.
Comparisons are provided for four types of compensation: (1) base
salary; (2) total cash compensation (base salary plus supplemental
cash compensation); (3) benefits value; and (4) total compensation.
Auxiliary comparisons to a sample of State governments and to
subsets of the private sector data base are included.
Section III presents a detailed description of the analytic and
comparative methodologies employed. The remainder of this section
discusses the nature of the data bases used for analytic and
comparative purposes. The relationships between this study and the
Federal government's annual Professional, Administrative, Technical
and Clerical Survey (PATC Survey) are discussed in Section VI, Cash
Compensation.
C. Federal Data Base
The Civil Service Retirement System covers three large
categories of employee: (1) General Schedule and equivalent;*
(2) Wage Grade; and (3) Postal Service. However, the U.S. Congress
directly controls cash compensation only for the first group. Wage
*Six other pay schedules are equivalent to the General Schedule,
according to OPM: each typically was designed to respond to a
specific need (e.g., employees in the Panama Canal Zone).
-9-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Grade compensation is based upon locally prevailing rates, while
Postal Service compensation is determined through collective
bargaining. Therefore, it is appropriate to exclude Postal Service
and Wage Grade employees from this total compensation comparability
study because of the fundamental and structurally distinct ways in
which their base salaries are established.
An additional reason for focusing on General Schedule positions
is to use comparable, homogeneous groups. The study focuses on the
white collar compensation practices of the Federal government and of
medium and large size employers in the United States. To maintain
this homogeneity, the study does not include pension plans available
to specific population groups in the private sector (such as
multiple-employer collective bargaining plans), specific categories
in State government, (such as teachers, police, and state police),
and Federal employees subject to special retirement provisions (such
as law enforcement officers and air traffic controllers). It is
assumed that Congress will first determine a CSRS system which best
fits the needs of the majority of Federal employees and then deal
with each divergent type of group individually. In that instance,
it might be appropriate to examine the total compensation packages
of some of the larger of these groups before extending retirement
policy in general.
The resulting group, General Schedule (GS) and equivalent
employees, numbered 1,442,510 full-time, permanent, civilian
members, as of January 2, 1984, and accounted for half of all
employees covered by the CSRS.* A sound study of Federal total
compensation requires that the compensation practice for this group
of GS and equivalent employees be comprehensively and accurately
represented. To this end, the study is based on a sample of 392
such positions which are covered by the CSRS.
*Members of the Senior Executive Service, although not on the
General Schedule or an equivalent, are also covered by CSRS.
-10-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The Federal sample satisfies the criteria entailed by this
total compensation comparability study: a comprehensive and
representative sample of positions from which to determine an
accurate representation of the Federal compensation practice for the
most relevant employee population.
The 392 positions are drawn from each grade level with at least
2.5% of the employee population, and constitute the most populous
positions in each grade level selected. At each selected grade
level the positions selected encompassed the majority of incumbents;
typically, over two-thirds of all employees at a grade are accounted
for by the positions selected.
While SES positions are not on the General Schedule, they were
analyzed as an additional self-standing category since these
positions have effectively replaced GS grades 16, 17 and 18. The
range of SES compensation is relatively narrow, while the span of
job difficulty is rather large. Hence, a sample of the most
difficult SES positions, paid at the highest (ES-6) level, is
included for illustrative purposes only, and not in any calculations
used to perform compensation comparisons.
The uniform benefit package extended to General Schedule
employees was used to develop cash equivalent values for the
employee benefit plans. In the health care area, where employees
may choose from among a number of options, the plan most frequently
chosen by enrollees was selected for the analysis. The programs
considered in the, analysis included:
? Civil Service Retirement System;
? Federal Employees Group Life Insurance Program (FEGLI);
? Federal Employees Health Benefits Program (FEHBP);
? Annual leave and scheduled holidays;
-11-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
? Sick leave; and
? Statutory programs such as Medicare, Worker's Compensation
and Unemployment Compensation.
Section III describes the Federal sampling methodology and
results in greater detail.
D. Private Sector Data Bases
Hay's extensive private sector data bases permit comparisons of
Federal base salary and total cash compensation to private sector
data. The Hay/Huggins Benefits Comparison (HHBC) allows comparison
of benefit values and practices followed by a broad spectrum of the
employer community. The combination of data on firms included in
the Hay cash and benefits data bases allows Hay to construct a
consistent, comprehensive data set for total compensation analysis
purposes.
The Hay cash data bases include information on exempt* jobs
from 1,249 medium and large size companies in the U.S. The
Hay/Huggins benefits data base contains data on 854 such
organizations.
The comparison of Federal compensation to that of primarily
medium and large private sector employers is particularly
appropriate because of inherent, fundamental characteristics which
distinguish small companies from large employers, including the
Federal government: (1) the nature of the organization's processes
and structure; and (2) the nature of the labor market in which the
organization competes for exempt employees.
*By exempt, Hay specifically refers to those jobs which are exempt
from the overtime provisions of the Fair Labor Standards Act.
-12-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Medium and large organizations are fundamentally different from
small independent ones, typically, in terms of their processes and
structures, e.g., in the emphasis on formal rules, regulations and
procedures; number of hierarchical levels of review and approval
required; and internal culture. Each of these factors significantly
affects the nature of jobs within an organization, and therefore the
comparability of compensation provided by an organization.
Further, small, independent companies tend to compete in local
or regional labor markets, and their compensation practices reflect
such tendencies. On the other
national practices for exempt
appropriate comparisons in view of
single national compensation
and its recruitment
hand, large companies reflect
positions, which provide more
the Federal government's use of a
system for GS and equivalent positions
of personnel on a nationwide basis.
However, some small and medium-sized companies install the Hay
system and participate in the Hay compensation surveys. For
example, nearly 30% of the 854 companies participating in the
Hay/Huggins Benefits Comparison in 1983 employ less than 1,000
employees, as indicated in Table II-1 below.
Table II-1
Size Distribution of Participants
in Hay's 1983 Benefits Comparison
Size Category
Number
Percent*
Under 100 employees
28
3.5%
100-499 employees
105
13.1%
500-999 employees
104
13.0%
1000+ employees
564
70.4%
Unreported
53
Total
854
100.0%
*Percent of the 801 reporting employment.
-13-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The Hay data bases therefore comprise a group of employers which
are comparable to the Federal government. Data on exempt positions,
which are analogous to the median evaluated job difficulty at grade
7 and above, were derived from the Hay data bases. Hay
traditionally supplements its exempt data bases by using data from
the Bureau of Labor Statistics (BLS) Area Wage Surveys. Hay applies
its standard position evaluation methodology to the data collected
by BLS for selected nonexempt positions. Hay has applied the same
approach to develop aggregate (national) results for the purposes of
this study, which are relevant to the grades below 7.
It should be noted that the number of companies included in the
Hay data base is not directly comparable to a number of
establishments, as defined by the BLS. Specifically, an
establishment may be viewed as a physical location where a
commercial activity is performed. Therefore, a company may in fact
(and often does) consist of numerous establishments. In the case of
the Hay Cash Compensation Comparison surveys, each company
represents a voluntary survey participant which has autonomous
control over its compensation system: such firms may legally be
owned or controlled by others; and each probably comprises numerous
establishments. Appendix A presents the 1,249 companies
participating, by subsector. Appendix B presents the 854 companies
in the Hay/Huggins Benefits Comparison.
E. State Government Data Bases
No comparable data bases on State government compensation and
benefit practices existed; hence, Hay conducted a special survey of
thirteen key States. The sample of States was drawn as follows.
The eight States included in a recent General Accounting Office
(GAO) report* on retirement systems was adopted as an initial
reference point, on the premise that comparable information would be
*Federal Employee Demographics and Integration of State Retirement
Plans with Social Security (GAO/FPCD-83-38), July 27, 1983.
-14-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
particularly valuable. Hawaii, however, was excluded from
consideration because of its unique organization (the State provides
a great many services not provided by other State governments). The
resulting sample of seven States was considered too small to be
representative; therefore, an additional six States were included in
order to provide enriched geographic coverage, to reflect varying
regional practices and economic conditions within the context of
selecting large states most comparable to the Federal government.
Table 11-2 presents the total list of thirteen States that
participated in the survey.
model
type,
Table 11-2
States that Participated in the Hay Survey
of Compensation and Benefits
California
Connecticut
Florida
Maryland
Massachusetts
Michigan
Nebraska
New Jersey
New York
North Carolina
Oregon
Pennsylvania
Texas
The State survey was founded upon the development of a generic
representative of State government employment patterns by job
and the collection of the
relevant to each position included.
compensation and benefits data
Section III presents a detailed
discussion of the methodology employed.
-15-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
III. METHOD OF COMPENSATION AND BENEFITS COMPARISON
This section describes the methodology employed in the study.
The first segment deals with cash compensation comparisons, the
second with benefits comparisons.
A. Cash Compensation Comparison
In order to document the approach taken during each stage of the
analytical process, the following elements are presented in sequence:
? Components of Cash Compensation
? Method
? Data Collection
? Data Preparation
? Presentation
1. Components
Two types of cash compensation are analyzed in this study, as
appropriate: base salary, and total annual cash compensation. The
definitions used are identical to those employed throughout all of
Hay's private sector cash compensation surveys, as follows.
? Base Salary is the straight-time annualized amount (usually
based on monthly payroll) paid for work performed. It
excludes performance bonus payments of all kinds and
differentials. It includes a year-end bonus of a uniform
percentage if it doesn't exceed two weeks pay and is
distributed to all personnel regardless of performance.
? Total Annual Cash Compensation combines the base salary and
supplemental cash compensation reported. Supplemental cash
compensation includes bonus, incentive, commission, etc.
paid in cash or with the unlimited option to be taken in
cash, earned over the past year and based on individual,
unit, or company performance. More specifically:
-16-
Declassified and Approved ForRelease2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
? Bonus compensation plans are based on
accomplishment, the award being determined after
the fact and/or on a discretionary basis.
? Incentive compensation plans are based on
pre-established objectives with awards based on
attainment of the objectives.
? Cash-equivalent compensation is included, if
provided in lieu of direct cash compensation
(stock bonus plans). stock options, performance
shares, and like programs are included under
benefits.
? Deferred compensation is included in the year
earned regardless of when paid. For example, if
a person at $30,000 base salary were awarded a
bonus of $10,000 for the current year, payable
$4,000 now and $6,000 deferred over the next two
years, the report would show $30,000 base salary,
plus $10,000 bonus, for a total annual cash
compensation of $40,000. Compensation deferred
for or until retirement without unlimited current
option is included in the benefits analysis.
Supplemental cash compensation is generally not provided to
Federal employees on the General Schedule (GS) or equivalent pay
systems. In aggregate, incumbents of Merit Pay positions receive
annual increases in base salary equal to the general and
step-in-grade increases received by GS or equivalent employees,
although the increase for any individual is related to job
performance (merit). There is provision for a cash award system
which operates like the bonus compensation plans discussed earlier;
however, such awards constitute a relatively small sum in comparison
to base salary, and a minute proportion of the total GS and
equivalent pay system payrolls. Therefore, the GS and equivalent
-17-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
pay systems are essentially base salary systems without significant
supplemental cash compensation features.
Supplemental cash compensation does become an increasingly
greater proportion of total potential private sector cash
compensation
Hence, the
compensation
identical to
total cash
salaries are
supplemental
levels of job
as a position's level of responsibility increases.
most accurate and relevant comparison of cash
is that between Federal base salary (essentially
Federal total cash compensation) and private sector
compensation. Comparisons to private sector base
provided only to indicate the relative importance of
cash compensation in the private sector at relevant
difficulty.
2. Method
Comparisons of cash compensation are valid only to the extent
that they control for differences between jobs. The most valid
comparison is based upon using the compensation for identical jobs.
Three methods are commonly applied in order to control for
variations among jobs: (1) title matching; (2) job content matching;
and (3) point-factor evaluation. Each is briefly described below.
The title matching method, as its name implies, merely attempts
to control for variation among jobs by comparing only jobs which are
similarly titled. As such, it is a rather imprecise and unreliable
approach because it provides only the most superficial level of
control. Common sense as well as empirical research clearly
indicate the substantial amount of variance among jobs which
identical titles often
the fact that two jobs
be somewhat defensible
obscure. Further, different titles may hide
are in fact the same. While the approach may
in application to the simplest of positions,
or to industry-specific jobs where the nature of the position is
dictated by technology and/or process, it cannot reliably be used as
a general methodology for numerous types of positions which vary
-18-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
significantly by subject matter, activities t and responsibility. It
may too often result in invalid comparisons.
Job content matching (job matching) improves upon title matching
by striving to consider the nature of the position. It is, however,
a subjective and rather limiting process. Its subjectivity derives
from the need to develop and implement the concept of a match.
Typically, similarity of activity, by type of activity and subject
matter, is selected as a joint criterion. Subjectivity enters into
the process at several points: arbitrary decisions about the
similarity of activity and subject matter; and whether the degree of
similarity is sufficient to warrant defining the job as a match.
The process is fundamentally restrictive due to the inherent
approach of considering only matches, however defined.
Point-factor evaluation systems take an entirely different
approach by applying a different type of yardstick to determine job
similarity. Rather than apply measures which depend upon the
specific nature of the position (e.g., title, subject matter, type
of activity), point-factor evaluation methods measure factors which
are common to all jobs. Such methods are based upon the assignment
of a point score to a factor (such as level of knowledge required to
perform the job competently) which is independent of the subject
matter of such knowledge (e.g. computer science, law). They are
therefore relatively precise, in that point scores are used to
distinguish degrees of difficulty, and universal in that the
dimensions (factors) are common to all jobs. The latter aspect
eliminates the problem of restrictiveness inherent in the other
approaches -- but with no sacrifice in control for similarity.
The Hay Method assigns points which indicate the relative
difficulty of jobs on three primary factors: the knowledge
required; the difficulty of the problems requiring solution; and the
level of responsibility involved. The sum of the points assigned to
each primary factor constitutes the job evaluation score -- the
job's relative difficulty or content.
-19-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Each primary factor .consists of at least two dimensions. The
knowledge required is a three-dimensional factor, consisting of the
level of technical knowledge needed, the supervisory or managerial
skill to be exercised, and the required human relations skills. The
problem-solving factor is evaluated
along two interactive
dimensions: the degree of challenge posed and the degree to which
the environment constrains the thought process involved. The level
of responsibility -- accountability -- is measured along three
dimensions: the freedom to act: the job's impact on its most
important goal, and the magnitude of the goal. Appendix C contains
a more detailed description of the system.
This study therefore provides comparisons of cash compensation
for positions of the same overall difficulty, or content, regardless
of the nature of the position. It thus applies a precise,
quantified, valid basis for comparison which is not restrictive. In
fact, the issue of restrictiveness is significant. Clearly, the Hay
Method does not limit the amount of comparative data that can be
collected for any given position of interest. It does not constrain
the nature of the job which is to be compared. In other words, the
fact that an organization may have specific positions which are rare
(or even nonexistent) in other organizations is unimportant. Hence
an accurate comparison can readily be performed even in such cases,
due to the application of the Hay Method.
The nature of the process used to assign a reliable and valid
set of Hay job evaluation point scores (Hay Points, or HP, or
points) to positions is presented in subsection 3, below.
Development of Compensation Practices
Given the existence of a set of positions representative of an
organization's employee population, its compensation practice at a
point in time can be depicted as the relationship between the total
points for each job and the average compensation for that job at
that time. Plotting compensation on the vertical (y) axis and
-20-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
points on the horizontal (x) axis creates a scattergram, a pictorial
representation of the relationship. To the extent that cash
compensation is commensurate with job content, as would be expected,
the scattergram should indicate the existence of a pattern of dots
which displays consistently increasing compensation as job content
(HP) increases.
The pattern can be summarized in a linear fashion by calculating
the weighted linear regression line which best fits the data. The
statistics for such a regression indicate the degree to which a
linear pattern exists and the expected deviation about the line as
well as the formula of the line itself. To the extent that the line
fits the data well (i.e., the measurement points all fall on or near
the line), the line represents the organization's practice relative
to paying for job content.
Chart III-1 depicts a scattergram and a weighted regression line
reflecting the compensation practice evidence.
Comparison of Compensation Practices
One organization's pay practice may therefore be compared to
another's by means of the regression lines computed for each, which
constitutes one comparative method. In addition, an overall summary
comparison may be performed by calculating the percentage difference
between cash compensation at selected HP values, and computing a
measure of central tendency (e.g., an average, or the median) for
such differences.
Both types of comparisons are provided in this study. There are
graphic presentations comparing the regression lines calculated for
the Federal sample -- GS and equivalent positions -- to the
equal-weighted average (and selected percentiles) of the regression
lines for the comparator sample (e.g., 1,249 private sector firms)
calculated at relevant HP values. Numerical comparisons are
presented in tables which are based upon the difference between the
-21-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
slu!od lualuoo got,
o
o
Annua 1 Total Cash
(Thousands)
N)
o
(A
o
-Ft,
o o 0
- 2 2 -
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
equal-weighted averages for the comparator lines and the actual
weighted average compensation for the Federal sample at that HP
level. The HP levels used in the tabular comparisons correspond to
the median HP for all incumbents in all sample positions in a
specific grade (e.g., GS-5 and equivalent). The overall summary
numerical comparison consists of the average of such differences,
weighted by the total Federal employee population in each grade
included. Such comparisons are presented for both base salary and
total cash, as appropriate. The details of the process are
presented in the next sections.
3. Data Collection
The cash compensation comparison-related data collection
activities conducted specifically for this study encompassed only
the Federal sample and the State sample. The private sector data
were already available by virtue of the existence of comprehensive
annual Hay survey data bases. Since each study element is
significant to the outcome of this activity, this section describes
the data collection procedures relevant to the Federal sample, the
private sector sample, and the sample of States.
Federal Sample Data Collection Procedures
The development of a comprehensive, accurate representation of
the General Schedule and equivalent compensation practice is a
cornerstone of this study. Therefore, the sample must satisfy a
number of criteria:
? Represent the bulk of the Federal employee population to
ensure that the results constitute an accurate
representation of the overall compensation practice.
? Represent the employee population in terms of apparent,
pragmatically significant variation in overall job
difficulty (i.e., by grade level).
-23-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
? Represent the employee population in terms of variation in
the substantive content of positions (i.e., by position
type within grade level).
? Provide reliable measures of job difficulty and average
compensation at key levels in order to facilitate
development of an overall summary measure of the
differences between the resultant practice and other
compensation practices.
These criteria were applied within a conceptual framework
consisting of a three-dimensional hierarchical matrix: (1) a
structure of 18 GS and equivalent grades; (2) a set of positions
classified at each grade; and (3) a group of incumbents in each
position at a grade level. The criteria were implemented as follows
to develop the Federal job sample:
1. Select each GS and equivalent grade level which has a
significant proportion of employees in jobs assigned at
that grade relative to the total population -- to reflect
apparent variation in overall job difficulty.
2. Within each selected grade level select the most populous
positions in sequence and select enough positions to
represent the majority of incumbents in each grade -- to
represent substantive variation among positions and to
represent the bulk of the Federal employee population.
The first stage of this procedure was applied to the sampling
universe defined in Table III-1.
-24-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Table III-1
Distribution of GS and Equivalent Employee Population
by Grade Level 1-18*
Grade Level Population % of Total Population
1
2,858
0.19%
2
15,462
1.04
3
82,329
5.56
4
171,185
11.56
5
198,685
13.42
6
92,329
6.23
7
136,107
9.19
8
31,400
2.12
9
162,560
10.98
10
28,911
1.95
11
170,224
11.49
12
173,026
11.68
13
115,603
7.81
14
61,401
4.15
15
37,803
2.55
16-18
1,165
0.08
Total
1,481,048
100.00%
*Data derived from Occupations of Federal White-Collar and
Blue-Collar Workers, U.S. Office of Personnel Management, October
31, 1981, Table D-1 (Full-Time Civilian White Collar Employment by
Occupation, General Schedule and Equivalent Grade, Median Grade, and
Average Grade, All Areas, October 31, 1981). This was the most
recent data set available at the time the sample was drawn.
-25-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Inspection of this data revealed that selection of eleven
grades, which each contained at least 2.5% of the population, would
allow thorough coverage of the entire structure in terms of the
presumed relationship between job difficulty and grade level, for
the majority of the Federal employee population. Specifically, the
table indicates that a very small proportion of employees hold jobs
at grades 1, 2, 8, 10, 16, 17 and 18. Therefore, the Federal
compensation practice can be comprehensively represented by analysis
of eleven grades, which cover 94.6% of the entire population at
grades 1-18, as shown in Table 111-2 below.
Table 111-2
Distribution of GS and Equivalent Employee Population
by Selected Grade Level
Selected Grade
Level
Population
% of Total
Population
3
82,329
5.56%
4
171,185
11.56
5
198,685
13.42
6
92,329
6.23
7
136,107
9.19
9
162,560
10.98
11
170,224
11.49
12
173,026
11.68
13
115,603
7.81
14
61,401
4.15
15
37,803
2.55
Total
1,401,252
94.62%
-26-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The second stage of the sampling process consisted of selecting
the most populous positions in each grade level sufficient to
represent the bulk of the population at each grade level. The
resultant sample was selected, as noted earlier, from all positions
covered by the general formula of the Civil Service Retirement
System and the OPM series of classification and qualification
standards for General Schedule and equivalent positions. Due to the
differential degree to which the employee population was
concentrated in specific positions at each grade, the final sample
includes different numbers of positions at each grade, and provides
different levels of coverage.
The final sample of 392 positions selected covers nearly 71% of
the population in the eleven grades selected, and over 67% of the
total population (all 18 grades): it provides coverage of almost
one million Federal employees. Further, enough positions were
selected at each grade included to cover the majority of incumbents
in the grade (see Table 111-3 following).
Appendix D contains a complete list of the 392 positions
included in the sample, by grade and occupational series.
In addition to the 392 positions selected at the eleven GS and
equivalent grade levels, a sample of Senior Executive Service (SES)
positions was also drawn. Although SES positions are not
compensated on the GS or equivalent pay schedules, such positions
are covered by the CSRS and were formerly classified at GS and
equivalent grades 16-18, accounting for the relatively low
population total in those grades currently. Therefore, for
illustrative purposes only (in order to estimate comparability for
the most difficult jobs) a sample of SES positions was included.
-27-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
01
I-,
I-,
I
CO
0
42..
Is.) I-,
0
VD CO
CT 0 42.
1-,
4t.
CO
(a
01
I-,
-4-4
N
I-.
01
La
1-.
(a
1-, 1-,
VD lIZ ...4
CO
1-,
I-,
-.1
1-,
Ln
4...) o
cc
r..)
1--
Ch
N 00 I-,
N
0
N
.
.?
.
?? *0
??
???
.0
.1.
..1 .. ..I.
..i
.0
.?
/...
CO
.14.
CI
CD K.)
k0
Lri
.I:.
I-,
ta Ch 1-,
La
4.
CO
CI1
0
0
0
N t.?)
1-,
01
0
0
N CO CO
isj
CT
0
CO
0IG11-.(.4014t.1-.00-.41/4000t0NCO
Z
I-4
1.4 PG
Z
Z
PG 1.4
I.4
Z
Z
0
(D
M CD
0
(D
0
(D
(D
(D
0
0
D*2
Cn En Cn
Ci)
Cn
CA ca CA
CO
LI
Lo
I
N
0) 0
I
41.
I
.1=.?
1?..) LI (.4
I-,
I
I
N
I
0
.1=.
01 C
ItvI
L..ecno?NL41/401
_I-..
I-.
42.
Isj
l..)
....,j N.) I-.
0
CO
01 PAb 1446
CI
I
.
.
. . .
.
.
. . .
.?
LI
01
crN
I
I
01
l.0
i-,
0)
N.)
I-,
0
(A
CO 0 I-,
0 CD I-,
.c.. L..1 1/4.0
an -.I 01
I
I
I
ON
k7
I--.
CI1
I
I
I
...1
Ln
CO
01
k.0 to (...1
LO tv (NJ
( . 4 T .P.
...1 ...I CO
k,C,
co
k.0
CO
I
1
I
I
1
I
I
0
?
.tu
?
1--, L...$ co
? ? ?
1
L...)
.
I
4:.
?
tn az. 0
? ? ?
0
?
I
I
cOP
0
1--.
co I.-. t...)
co
co
co 1.-. ui
a>
c112
-28-
uoneTndod
<
Pi
CL CuDI
el
CI 1.0
(D 4 (D
r? 0
(D <
a TduieS al-11
tu papnToui
SUOT4TS0c1
Jo laquitiN
JO quaolad
DI
tTi
CD
7 'CI
rr
0
0 64
SD 0
M nzi
0
(D
<
0 7
rr
0
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
E-III aTqej,
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
In order to develop estimates of the job content and comparable
compensation levels for the most difficult SES positions, eleven of
the largest Federal agencies were requested to select 3 or 4 of
their most difficult positions at the highest SES level (level 6,
denoted as ES-6) and to provide position descriptions for each.
Appendix E contains a list of the 38 positions evaluated.
Federal Sample Position Evaluation Process
Published Office of Personnel Management classification
standards
relative difficulty of the 392
positions selected. Specific job
standards) were intentionally not
constituted the information base for the evaluation of the
GS and equivalent grade 1-15
descriptions (as contrasted with
evaluated in order to provide a
consistent, accurate and unbiased basis. This approach was founded,
in part, upon Hay's knowledge of and experience in the process used
by OPM to produce such standards, and in part on Hay's similarly
extensive familiarity with the use of individual job descriptions in
the Federal and private sector contexts.
Classification standards are more appropriate for use in this
study because of the care with which they are generally developed by
OPM:
? Extensive background research is conducted;
? A large number of actual job descriptions is collected;
? A series of interviews (as well as other data collection
procedures) are conducted in order to gather voluminous
information about the jobs from supervisors, incumbents and
personnel specialists;
? The data is then analyzed by an objective third party (an
OPM employee) in order to distill the essence of the work
done and the job requirements;
? A tentative standard is published in order to allow
interested parties to comment; and
? A revised standard is ultimately published for operational
use.
-29--
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
This process therefore provides, in essence, a meticulously
prepared generic job description. It eliminates possible problems
(which may be severe) in the use of certain specific job
descriptions due to the age of the description, its relative
atypicality, or potential inaccuracies. In view of the intent to
apply the evaluation results to all incumbents of a position, use, of
the OPM classification standards is particularly appropriate.
However, this is not the case regarding the SES positions in the
sample. At the SES level, and especially at ES-6, there exist few
if any multiple-incumbent positions. Rather, each position is
unique, and specific job descriptions must be used.
The Hay Guide Chart-Profile Method of Job Evaluation was applied
to all 430 positions* by two senior Hay consultants assigned to the
task in view of their extensive experience in evaluating both public
and private sector positions. Each consultant independently
analyzed and evaluated each position. The consultants then jointly
reviewed their evaluations, and resolved the few differences between
them. This phase of the process ensured that a reliable evaluation
was assigned to each position on an internally consistent basis, and
in the same way that private sector positions are evaluated.
Thereafter, an additional quality control process was applied by
a third Hay consultant, independent of the initial evaluation
process. As is the case for all Hay evaluation work, evaluations
are not considered final until reviewed by specialists in Hay's
centralized Job Measurement Quality Assurance (JMQA) group. The
members of this group are senior consultants with demonstrated
expertise in job evaluation. Their task is to review evaluations
performed by Hay consultants in order to ensure consistent
application of the measurement instruments (the Guide Charts). This
review process led to the development of a set of final evaluations
*392 GS and equivalent positions plus 38 SES positions.
-30-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
which are internally consistent and valid, and consistent with the
intent of the Guide Charts. The latter attribute ensures the
validity of subsequent compensation comparisons on the basis of
evaluated job difficulty, since it ensures that all evaluations are
directly comparable.
State Sample Data Collection Procedures
The sampling and data collection approach appropriate to the
State government survey necessarily differs from that applied to the
Federal sample for two reasons: (1) many States cannot easily
provide data required for sampling purposes in a timely and/or
readily usable fashion; and (2) the nature of the classification
standards used by each State vary considerably. It was therefore
not possible to attempt to implement the same approach as used for
the Federal sample.
The State sample data is based on an empirically-based
conceptual model of a State government organization because State
governments are similar in terms of the services provided, and hence
the nature of their positions. The model is designed to satisfy two
criteria: represent the spectrum of positions by job difficulty; and
represent the employee population. The conceptual model was founded
on two data sets: (1) Hay knowledge of State government
organization, processes and structure, derived from numerous
applications of the Hay Method of job evaluation in the State
government context; and (2) the data submissions the sample of
States provided relevant to the development of representative
samples (i.e., provision of data on the distribution of employees by
job and grade).
This information led Hay to develop a generic model of the types
of jobs, and levels, typically found in a State government, for
positions relevant to the Federal sample. The result is a series of
job structures for the various areas of activity common to State
government (e.g., auditing, clerical, personnel).
-31-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The survey itself consisted of a series of 104 synthesized job
descriptions. The States were requested to provide compensation
data for each State position which matched a survey position. The
job descriptions were designed to balance the need for specificity
so as to distinguish between job levels and difficulty, and
generality so as to ensure the development of a sufficiently large
data base. Additional compensation-related data were collected so
as to allow date adjustments to be made, as described below.
Each of the 104 synthesized job descriptions was evaluated using
the same process and procedures as were applied to the Federal job
data. Therefore, the resulting job evaluations are internally
consistent, and externally consistent with the Federal and private
sector job data bases.
As an additional element of quality control, each State was
asked to submit a sample of at least ten (10) actual classification
standards for jobs matched in the survey. These standards were
reviewed by Hay consultants to assess the reliability and validity
of the matching process. This review indicated that the process was
reliable and valid: in no case did a participant appear to yield an
error rate greater than 10%. Because each State was able to
reliably and validly match 77% to 98% of the 104 jobs surveyed, (the
overall average is 90%), the ultimate results are representative.*
Private Sector Sample Data Collection Procedures
Hay meticulously conducts annual surveys of the cash
compensation provided by U.S. private sector firms in the
industrial, financial and service subsectors. This study utilizes
data from 1,249 such firms collected as of May 1, 1983. The data
*The 104 jobs include law enforcement positions in order to allow
development of useful participant feedback reports. However, all
such positions, which typically are covered by special retirement
programs, are excluded from the comparisons presented in this report.
-32-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
submitted by each voluntary participant virtually excludes
non-exempt positions. Moreover, the data submissions are no
included unless the firm's data meets Hay's multiple quality
assurance, review and control requirements. The existence of the
Hay data base on private sector cash compensation eliminates the
need to collect such data for positions exempt from the Fair Labor
Standards Act.
One of the major strengths of these data bases is the meticulous
attention paid to quality control and validity: the processes used
assure clients that compensation comparisons are meaningful and
valid because the underlying job evaluations are consistent and
comparable. The validity of the Hay data bases derives from two
sources: the process initially used to develop job evaluations for
an organization; and the continuous quality review and assurance
applied to them subsequently.
Typically, the initial Hay job evaluation process is conducted
as follows. Position descriptions are prepared and approved by
incumbents as well as their supervisors. Then, a committee of
client organization employees is trained in the Hay Method of job
evaluation. The committee consists of individuals who rank higher
than the incumbents of the jobs to be evaluated, and who
collectively represent a broad sample of different activity areas
within the organization. A Hay consultant guides this committee in
evaluating each position. The nature of the committee's
constitution facilitates its development of evaluations which
represent relative difficulty in that organization within the
confines of proper application of the measurement technology. At
the end of the process, all evaluations are reviewed by the
committee and revisions made as necessary to ensure that the
methodology has been consistently applied. Thereafter, the results
are reviewed by a member of Hay's JMQA staff, and additional
revisions are made as needed.
-33-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The JMQA staff also provides an additional level of quality
control as part of its ongoing review activities. A client
organization must have had its evaluations reviewed and approved by
a JMQA staff member within the preceding three years in order to
have its data entered into a Hay data base. These activities thus
ensure that the data base contains only reliable information.
Hay relies on raw data collected by the Bureau of Labor
Statistics (BLS) in order to augment its data base for positions not
exempt from the overtime provisions of the Fair Labor Standards
Act. Hay evaluates all 24 white collar positions selected from the
jobs included in the Area Wage Surveys, and uses the BLS-collected
compensation data.
4. Data Preparation
Federal Cash Compensation Data Preparation
All compensation data were date-adjusted to March 1, 1984 in
order to provide a consistent basis for comparisons. Different but
consistent procedures were applied in preparing the Federal, private
sector, and State data, and these are next discussed.
The appropriate requirement to develop cash compensation
comparisons on a grade level basis dictated the following process.
The OPM maintains readily-available compensation data for Federal
employees in the form of current grade and step-in-grade.
Therefore, Hay obtained data on the weighted average step-in-grade
for all positions in each of the eleven selected grades, excluding
positions not covered by the CSRS.
According to OPM, there are six (6) other pay systems which are
equivalent to the General Schedule (GS): CZ (for Canal Area GS type
positions); GG (for grades similar to the GS); GH (for GG employees
converted to a merit pay plan system); GM (for Merit Pay system
employees; GW (for student trainees classified under and paid at a
Declassified and Approved For Release 2012/11/21: CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
GS rate); and LG (Liquidation Graded, for FDIC employees). The
weighted average step-in-grade for all relevant employees in each of
the seven relevant pay systems as of the first pay period in
January, 1984 was used to compute, by linear interpolation, the
weighted average compensation by grade. The following table
presents those results.
Table III-4
Weighted Average Step-In-Grade
and Interpolated Base Salary for Selected GS
and Equivalent Grades as of 1/2/84
(Including the 3.5% General Increase)
Selected Grade Level
Weighted Average
Step in Grade
Interpolated Base Salary
3
3.00
$ 11,751
4
4.18
13,677
5
4.82
15,598
6
5.45
17,710
7
4.79
19,302
9
4.48
23,398
11
4.70
28,496
12
5.17
34,626
13
2.78
38,297
14
1.86
43,947
15
1.32
50,788
In
order to adjust the base salary figures to 3/1/84, two
factors
were applied. The first adjustment
factor derived from the
need to
reflect the step-in-grade movement that occurs on a regular
basis.
OPM-provided data were used to estimate an overall annual
average movement of approximately 0.098 steps, at an average rate of
increase of 3% per step. The base salary data was thus adjusted to
include two months of increase at this annual rate. The second
factor derived from the retroactive 0.5% increase granted in April,
-35- '
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21: CIA-RDP89-00066R000800210002-2
1984. Compounded together, these factors resulted in a 0.549%
upward adjustment to the interpolated base salary data, the results
of which are presented in the following table.
Table 111-5
Weighted Average GS and Equivalent
Base Salaries Adjusted to
Selected
Grade Level
3/1/84
Adjusted
Base Salary
3
$11,816
4
13,752
5
15,684
6
17,807
7
19,408
9
23,527
11
28,653
12
34,816
13
38,507
14
44,188
15
51,067
These figures are used throughout this report.
The SES schedule does not provide steps, and hence only the
0.5% retroactive increase was applied to the salary data to adjust
it to 3/1/84. The resultant base salary figure at the ES-6 level is
$69,900. For total cash compensation purposes, OPM data on the
average size of an SES bonus was used to estimate a value of 2.9% of
base salary, resulting in a figure of $71,927.
-36-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Private Sector Cash Compensation Data Preparation and Analysis
The two elements of the private sector data base (data
collected and analyzed by Hay; and data on non-exempt positions
collected by BLS but modified and analyzed by Hay) are described in
turn.
The most recently available annual Hay Cash Compensation survey
collected data effective May 1, 1983. In order to provide clients
interim information, Hay also conducts a mid-year update survey
every November to determine actual increases granted since the
preceding May 1, and projections to May 1 of the following year.
This survey collects data on both base salaries and total cash
compensation separately for the three major annual Hay surveys of
industrial firms, financial firms, and service firms. Over 300
firms responded to the 1983 update survey.
The results of this survey were used to date adjust the Hay
data for the ten months intervening between May 1, 1983 and March 1,
1984. The base salary adjustment rates varied from about 5-7%,
while the total cash compensation adjustment rates varied from about
8-14%. The annual rate appropriate to each type of firm and
compensation element was applied to the sub-sectoral results to
adjust them on a pro-rated basis.
The Hay cash compensation data used in this study are produced
separately for base salary and total cash compensation, as follows.
Each firm's data submission is inspected for reasonableness, and
questions are clarified. Then the data submitted, which typically
cover the entire exempt employee population, are subjected to a
weighted linear regression analysis. Piecewise linear curves are
developed as necessary in order to ensure that the resultant
representation of the organization's compensation practice relative
to evaluated job content is accurate.
-37-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Then, separately for each subgroup of firms of interest (i.e.,
Industrial, Financial, Service) and overall, the data are aggregated
to develop summary statistics at selected point values. Each firm
in a group is weighted equally in the aggregation process, so that
the results indicate, for example, how much employers pay, on
average, for a job of a certain difficulty -- not the average par of
incumbents in jobs of a certain difficulty. This approach therefore
vitiates the impact that extreme results for specific jobs, or very
large firms, may have on the overall results, and provides each
employer with the information on organizational compensation
practices required to administer its own overall practice.
The actual process computes, at each selected point value
(e.g., 100 points), the summary statistics for the set of
compensation values derived from the set of participant compensation
practice lines on an equal-weighted basis, including percentile as
well as average figures. The charts presented in Section VI are
then prepared by connecting the relevant data points. Linear
interpolation is used to compute compensation values between
adjacent selected point values. The size of the data base, together
with the linearity of the underlying regression results, ensures
that interpolated results are almost identical to exact
calculations, as Hay's empirical research has demonstrated.
The Bureau of Labor Statistics collects compensation data in
over 65 geographic areas throughout the United States (Area Wage
Surveys). These surveys focus on positions which are not exempt
from the overtime provisions of the Fair Labor Standards Act.
Because the surveys collect data on more routine types of positions,
the matching process employed by BLS is reliable, and Hay has used
the BLS data as a base from which to select data in order to develop
the series of area pay surveys it has published for the last 10
years. Consistent with the focus of this study, Hay publishes
results based on twenty-four white collar positions (see Appendix F
for a list of the twenty-four positions covered).
-38-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Each of the positions is evaluated using the Hay Method, and
then compensation practice lines are developed .for each area. The
results are then aggregated on the same basis as the other Hay data
to produce an aggregate nation-wide result.
The data were updated to March 1, 1984, by area for this
study. This was necessary because the BLS surveys are not conducted
simultaneously as of March 1. The Hay approach to updating the BLS
data is to collect information on national rates of movement from a
variety of sources, and to adjust the results for local conditions
by using available data on recent labor contract settlements,
historical rates of movement, and economic conditions for each area
in the decision process. These adjustments have typically shown
error rates of not more than a few tenths of a percent.
State Cash Compensation Data Preparation and Analysis
None of the thirteen participating States provides supplemental
cash compensation to incumbents of positions relevant to this
study. Hence, all analyses were performed in terms of base salary.
The State data were collected prior to March 1, 1984; moreover,
the effective dates of the data varied by State. Therefore, a
telephone survey was undertaken during August 1984 to collect the
information needed to adjust the data to March 1. As in the Federal
context, it was necessary to consider two types of adjustments: one
for general increases; and the other for merit increases
(step-in-grade movements). Such data was collected for every State,
and the salary data provided initially were accordingly adjusted
pro-rata to reflect the best possible estimates as of March 1, 1984.
The specific process used to prepare and analyze the State data
paralleled that used for the private sector. Initial data
submissions were reviewed for apparent accuracy and reasonableness,
and clarifications were obtained as needed. Similarly, the ten or
more actual classification standards submitted by each State were
-39-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
reviewed for accuracy in terms of similar job content; and
additional information was .gathered as needed. Where potential
matching problems were identified, further discussions were
undertaken with the State representatives. Several questionable
matches were deleted. Appendix G presents a detailed listing of the
number of matches included, by State and overall.
After the data for each State were thus prepared, the
compensation data were adjusted to March 1, 1984. Thereafter, the
results were analyzed in the same fashion as the Hay private sector
data base. A weighted regression-based piecewise linear curve was
developed to represent each State's compensation practice as a
function of evaluated job content. At each relevant point value,
the compensation provided by the States (using the practice lines)
was averaged to produce an average state practice line. The average
value at each evaluation point level representing a GS and
equivalent grade was then used in the subsequent comparisons.
5. Presentation
Section VI presents findings relative to the Hay evaluations of
the Federal sample of positions, as well as the compensation
comparisons. The evaluation findings present an analysis of the GS
and equivalent classification structures in terms of job
difficulty. As the basis for all relevant comparisons, they
constitute a significant element of the entire study.
The cash compensation comparisons presented are two-fold in
nature. Graphic presentations using the regression line for the
Federal government are provided for illustrative purposes. However,
the tabular comparisons and all comparative calculations are based
on the exact Federal average compensation values presented
previously. As an illustration, the actual average Federal
-40-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
compensation figure at GS-13 and equivalent is used in the table and
all comparative calculations, although the regression analysis, by
its very nature, would predict a somewhat different value.*
The cash compensation practice for the Federal sample is
founded on the grade levels used. Specifically, it is necessary to
develop a point estimate for each relevant grade level in order to
calculate differences by grade, and overall. The method for
developing a cash compensation figure has already been presented.
The estimate of Hay evaluation points at each grade is the median
value for all incumbents (not jobs) in the Federal sample. Due to
the highly variable distribution of employee population by job type
(and, therefore, evaluation points) at a grade, an incumbent-based
approach is required to represent the entire employee population. A
median rather than a weighted average was used in order to
conservatively estimate evaluation points per grade, since medians
are less sensitive to skewing caused by extreme results.
The visual cash compensation comparisons are therefore based,
for the Federal sample, on the calculation of a linear regression
line on eleven points (one for each of the selected grade levels).
The median evaluation points for each of the grades were used as the
values at which to calculate the average of the State practice
lines, and at which to interpolate the private sector results.
However, the exact Federal compensation averages are presented in
the tabular analyses which underlie the calculation of differences,
rather than results derived from the Federal regression analysis.
*Specifically, the regression analysis constitutes an approximation
which, as discussed in Section VI, is somewhat misleading at grades
13-15. Hence, the exact grade-based Federal compensation values are
used for all comparative calculations.
-
41-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
B. Benefits Comparisons
1. Components
The Hay/Huggins analysis focuses on benefits that meet three
general criteria:
? Benefits that are offered by a majority of employers;
? Benefits that employees have come to rely on the work
environment to provide; and
? Benefits that represent a substantial dollar obligation to
the employer.
The benefits that satisfy these criteria are retirement
systems, life insurance, disability income protection, health care
plans, time-off-with-pay, and executive perquisites.
Employers who want to be competitive in both recruiting and
retaining talented personnel recognize that protection in each of
these key areas must be a part of the benefits package offered.- An
individual choosing between two job alternatives that are otherwise
similar will be influenced by the extent of the employer-provided
benefits. Additionally, employers must be sensitive to the
objectives of organized labor. Starting with the era of wage
freezes as early as the 1940's, organized labor has worked steadily
for extension of the type of employee benefits plans and/or
enhancements to those already existing. Therefore, competition and
labor-management relations both have encouraged employers to offer
coverage in each of these areas.
As employers have come to offer benefit plans in each of these
areas, employees have grown to depend on the work environment for
their provision. For example, in the area of health insurance,
employees rely on the employment relationship to provide group
-42-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
protection to not only the worker but dependent family members as
well. Thus, by virtue of the employer-provided group plan, the
employee is insulated against the financial hazard that unforeseen
medical expenses can generate.
In addition
employee reliance
there is a third
afford to ignore
to the dual pressures of the marketplace and
that demand an employer's attention to benefits,
compelling factor, namely cost. Employers cannot
the sizeable portion of the total compensation
budget that benefits represent. The 1983 Hay/Huggins Benefits
Comparison showed that the value of the overall benefits package
ranges from one-third to one-half of salary.
2. Method
Direct comparison of the cost of employee benefit programs
considers the dollar amount or
employer pays for such benefits.
of the substantial differences in
workforce as well as the economic
percentage of payroll that each
This method is misleading because
the specific demographics of each
environment and funding procedures
used by individual employers. For instance, identical retirement
plans that provide exactly the same level of benefits could cost one
employer more than another employer as a result of any or all of the
following factors:
? A lower level of past funding patterns;
? More stringent actuarial methods used in determining future
funding patterns;
? Less favorable assumed future economic conditions;
? Less favorable assumed future mortality, rates of
retirement, and other decrements;
-43-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
? Higher ratio of males to females in the organization; and
? Older age or longer service characteristics of the
employees.
Hay/Huggins applies a standard yardstick to measure the cost, of
all benefit systems. This method, first developed 15 years ago,
has been modified each year since to reflect changing conditions and
technology. By assigning a standard value to each important element
of the benefits package, equal benefit programs will have the same
value. For instance, if one employer provides a pension plan
formula of 1% times pay times service as a retirement benefit and a
second employer provides 2% times pay times service, the second
employer's plan will be valued at twice the first employer's plan if
all other elements of the pension plan package are identical.
The Hay/Huggins method is a computerized system called the
Benefit Value Comparison (BVC). Each key element of cost is
assigned a relative value based on the provision itself as well as
its' relation to other provisions. For instance, in the pension
plan area, a 2% accrual rate is assigned an initial value of two
times a 1% accrual rate. - Then the rate is adjusted for any other
factor that adds value such as vested, early retirement, disability,
and survivor benefits. The base plan factors are increased to
reflect the practice of the employer in providing post-retirement
increases.
Hay/Huggins applied this procedure to the Federal government's
benefit program, the systems of the 854 private sector employers in
the 1983 Benefits Comparison, and the benefits provided by the
thirteen selected States.
To make the retirement plan comparison most relevant to the
development of a new supplementary Civil Service Retirement plan,
the BVC method was calibrated to the demographic experience and
actuarial method used by the Office of Personnel Management for the
-44-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Civil Service Retirement System. Further, the standard set of
assumptions that is being used by the Congressional Research Service
throughout their analysis of CSRS was used to predict the future
costs in the BVC retirement model. This set of assUmptions is the
1984 social security II-B set of economic assumptions and future
longevity improvements coupled with the underlying behavior patterns
of Federal employees.
Similar analyses were performed in each other area of
benefits. For instance, the life insurance and health insurance
costs were developed based on the current cost of the FEGLI and FEHB
programs. Benefit items not available to Federal employees, such as
provision of executive retirement systems, were evaluated using
standard Hay/Huggins methodology.
3. Data Collection
To augment its existing data base on private sector benefit
practices, Hay/Huggins collected information on benefits from both
the Federal government and selected State governments.
To ensure that the public sector data analysis would be
comparable with that from the private sector, the same data
collection process used to develop the Hay/Huggins Benefits
Comparison of private sector firms was applied for the Federal and
State information. Public sector participants were asked to submit
booklets descriptive of the various benefit packages they offered
that covered the majoeity of their white-collar workforce. For the
Federal government, the benefit programs covering the General
Schedule employees were used. For State governments, the set of
benefits covering the bulk of their professional, technical and
administrative staffs was identified and analyzed.
In addition to descriptive plan documents, the participants
completed questionnaires that addressed issues generally not
included in summary plan descriptions, e.g., the method of funding a
given benefit plan.
-45-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Once the data were received, experienced analysts reviewed the
submissions, clarified unclear issues with the responsible benefit
managers, and processed the verified data through the Hay/Huggins
Benefits Value Comparison model to derive cash equivalent values for
the benefit packages.
4. Presentation
The Federal government's benefits have been analyzed by use of
the Hay/Huggins actuarial model which develops a cash equivalent
value for each benefit at varying salary levels. The value of the
Federal government's benefit plans are compared with similarly
constructed values of plans offered by private sector employers and
States. The private sector and State practices are displayed at the
following brackets:
10th Percentile (P10): 90% of all employers offer plans that
are more valuable than this line.
Mean:
the average practice line.
90th Percentile (P90): only 10% of all employers offer plans
that are more valuable than this line.
By contrasting the value of
plan against this backdrop, it
government practices are more or
employer in the comparison group
within the range.
the Federal government's benefit
can be determined whether the
less liberal than the average
as well as the relative position
To complement the benefit-specific value charts we have
included in Appendix H a tabular summary of the benefit practices
reported by the 13 State governments surveyed and by the 854
employers in the 1983 Hay/Huggins Benefits Comparison. The benefit
practices are displayed by major benefit category and by specific
provisions within those categories. The Federal government
practices are also indicated.
-46-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
IV. COMPARISONS OF RETIREMENT SYSTEMS
This section presents the findings of the retirement
analysis and compares the current Civil Service Retirement
(CSRS) to the retirement systems of employers in the private
and State data bases.
system
System
sector
While CSRS is the sole source of retirement income for Federal
employees, the non-Federal employers rely on a set of plans to
provide benefits. These include:
? Social security;
? Defined benefit pension plans;
? Defined contribution pension plans;
? Capital accumulation plans.
The Civil Service Retirement System provides about the same
level of replacement income at retirement as do the total retirement
systems of private sector employers and State governments. However,
CSRS has features which cause it to be more expensive than the
Systems in the private sector. The primary reasons for the cost
difference are that the employee can retire
of service, or age 60 with 20 years of
private sector employees must
wait until
benefits. CSRS also includes more
benefits then are typically found
sector employers. However, many of
the private sector through insurance
retirement system is not
at age 55 with 30 years
service, while typical
62 or 65 to retire on full
valuable survivor and disability
in systems provided by private
these benefits are provided in
plans so that this part of the
strictly comparable.
The most important reason for the difference between CSRS and
the private sector is the fact that CSRS benefits are indexed to the
rate of inflation. A large part of the private sector employees'
-47--
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
package, social security, is also fully indexed and the rest of the
package is often indexed on an ad hoc basis but the full indexing of
the CSRS system makes a substantial difference in the value of
benefits.
When all of these elements are considered the Civil Service
Retirement System is worth 24.7% of pay compared to 18.3% for the
average private sector retirement system. However, the most
valuable plans provided by private sector employers are at least as
valuable as the CSRS. For instance, the top 10% of employer plans
in the private sector are worth 25.1% of pay or more.
State retirement systems contain features that are more
comparable to CSRS. State pension systems are often indexed to at
least a part of inflation and permit early retirement. As a result,
the average State plan system is worth more than the average private
sector system. The average plan in the thirteen State sample is
worth 22.2% of pay compared to 24.7% of pay for the Federal plan.
A. Cost Analysis
The economic and demographic assumptions described in Section
III have been used by the Congressional Research Service, with the
assistance of Hay/Huggins, to determine that the normal cost of the
Civil Service Retirement System is 32.2% of payroll or an employer
cost of 25.2% after removing the 7% employee contribution. For this
report, the cost was adjusted by removing .05% of the payroll for
administrative costs and .42% of payroll for benefits payable to
categories of members of the Civil Service Retirement System who are
eligible for special benefits. The administrative costs were
removed because the Hay/Huggins model measures the value of benefits
of other systems without the administrative costs. The cost of the
benefits of special groups was removed to provide a comparison of
only the benefits available under CSRS to the employees who are not
entitled to a special benefit formula or special eligibility
conditions. This adjustment produced a total cost of 31.7% of
-48-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
payroll. After removing the 7% employee contribution, the
employer-provided cost of the Civil Service Retirement System for
the standard level of benefits was determined to be 24.7% of payroll.
The 1983 actuarial report of the Office of Personnel Management
shows the normal cost of a Civil Service Retirement System to be
36.5% of payroll. Most of the difference between the OPM figure and
the Hay/Huggins analysis is explained by the fact that the OPM
figure is the total cost of the system and the Hay's analysis deals
appropriately with the employer cost of the system. With this
adjustment the OPM cost is 29.5% of pay (36.5% less the 7% employee
contribution). However, there is still a significant remaining
difference that is attributable to different actuarial assumptions.
The major factor in the actuarial evaluation of retirement
systems is the use of appropriate economic assumptions. These
assumptions reflect the actuaries' judgment as to the future rate of
inflation, salary growth, and return on investment. Particularly in
systems such as CSRS, that are indexed to inflation, all three of
these assumptions have a strong effect on final value. For
instance, a change of one percent in the interest rate can change
the actuarial cost by 25%. A change in 1% in either of the other
two assumptions can change the actuarial cost by 10% to 15%.
The set of actuarial assumptions in the BVC analysis was
developed by the Congressional Research Service with the assistance
of Hay/Huggins. The comparison of the economic assumptions used in
the BVC analysis to those in the current Office of Personnel
Management valuation is as follows:
Hay/Huggins OPM
Interest return 6.1% 6.0%
Salary Growth 5.5 5.5
Inflation 4.0 5.0
-49-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
While these factors may appear to be similar, the leverage
effect of the small relative differences on the actuarial cost is
significant. For instance, the real interest rate for the OPM set
of assumptions, 1%, is only half of that of the CRS set of
assumptions, 2.1%. Other than the changes in the economic
assumptions, the key differences were to assume that mortality will
continue to improve in the future as it has in the past and that the
internal rate of salary growth will not be as high as in the OPM
valuation.
Both the OPM and Hay/Huggins sets of assumptions are supported
by actuarial analysis; Hay/Huggins is not suggesting that one set is
more appropriate than another. However, the OPM Board of Actuaries
set of assumptions were set seven years ago for the specific purpose
of financing the Civil Service Retirement System within current
law. Not only have actuarial assumptions and trends changed in the
last seven years but also the new system will require coordination
with social security. Therefore, for this comparison, it appears to
be more appropriate to use the most recent set of social security
assumptions as the basis.
Hay/Huggins understands that the Board of Actuaries is
currently reviewing recent experience and that they will select a
new set of assumptions for their 1984 report. It is expected that
some of the changes in the new set may bring the OPM results closer
to an employer cost of 24.7% of payroll.
B. Comparison of Private Sector Systems and CSRS
Chart IV-1 compares the value of the Civil Service Retirement
System to the retirement systems of the 854 HHBC firms. Appendix I
summarizes the characteristics of these firms which are largely
corporations with white collar workforces similar to the General
Schedule workforce of the Federal government. The HHBC employers
are drawn from all geographic regions, industry categories and
workforce sizes.
-50-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The BVC system calculates the value of the employer-paid
portion of the retirement system. Employees contribute to social
security and are permitted, increasingly so under the recent 401(k)
provision of the Internal Revenue Service code, to contribute
substantial amounts to deferred compensation systems. However, the
concern of this analysis is the cost to the Federal government as, an
employer, and, therefore, the amount the taxpayer must pay for the
retirement system. Employee contributions, whether voluntary or
mandatory, are reasonably treated as savings that the employees
contribute toward their individual retirement income and, thus, are
not included in the BVC retirement plan values.
Employer retirement systems include all plans that can be used
to produce retirement income. In the private sector this includes
(1) the employer-paid cost of a pension plan, (2) the employer
contribution to a capital accumulation plan, and (3) the employer
cost of social security.
Chart IV-1 shows the values of the Civil Service Retirement
System against the range of the Hay/Huggins survey private sector
systems at $10,000 current salary increments. For instance, for
employees at $30,000 current salary, the average private sector
retirement system has a relative value of $5,617 compared to the
CSRS value of $7,422. At the extremes of the distribution, ten
percent of the employers provide retirement systems valued at $7,616
or more and ten percent provide a value of $3,670 or less.
When measured on this basis, the Civil Service Retirement
System is 32% more valuable than the average private sector system
for employees at $30,000 pay but 3% less valuable than the benefits
provided by any of the top 10% of companies in the HHBC survey.
When expressed as a percentage of payroll, the CSRS is 6.0% more
valuable than the average private sector plan at $30,000. The
difference varies slightly at other salary levels primarily because
of the interaction of private sector plans with social security.
-51-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
1
Bonef it Valua ($)
(Thoueandea)
I
cii
cn
o
r?)
-52 -
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The preceding chart illustrated the average benefit at various
salary levels. Table IV-1 summarizes the dollar value of each of
the plans displayed in the chart. This type of comparison is
instructive because plans outside of the Federal government, and
proposed plans that the Congress will be dealing with, reflect the
tilt in benefits from the social security system and provide
different provisions at different levels of pay. By contrast the
current Civil Service Retirement System provides the same percentage
of pay at all levels.
Table IV-1
Relative Value of Retirement System Benefits
Civil Service Retirement
Current Salary
$10,000
$30,000
$50,000
System
$2,474
$7,422
$12,370
Hay/Huggins Benefits
Comparison:
90th Percentile
$2,426
$7,616
$12,415
Mean
1,751
5,617
9,059
10th Percentile
1,101
3,670
5,757
To provide one comparable figure, the spread of values by salary
were weighted by the number of employees at each pay level and
aggregated to provide an average cost. The resulting average values
across all pay levels are:
Civil Service Retirement System 24.7% of pay
HHBC Plans:
Highest 10%
Average
Lowest 10%
-53-
25.1% of pay
18.3
11.9
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
C. Comparison of State. Government Systems and CSRS
Chart IV-2 shows the values of the Civil Service Retirement
System against the average of the thirteen State retirement systems
at $10,000 current salary increments. For instance, for employees
at $30,000 current salary, the average State retirement system has a
relative value of $6,849 compared to the CSRS value of $7,422. When
measured on this basis, the Civil Service Retirement System is 8%
more valuable than the average State system for employees at $30,000
pay. As a percentage of payroll the CSRS is 1.9% more valuable than
the average State plan at $30,000. The difference varies slightly
at other salary levels primarily because of the interaction of most
State plans with social security. Only one of the thirteen States,
Massachusetts, is not covered by social security.
When weighted by the number of employees at each General
Schedule pay level, the resulting average values across all pay
levels are:
Civil Service Retirement System 24.7% of pay
State retirement systems 22.2% of pay
D. Reasons for Difference
The Civil Service Retirement System benefits are more valuable
than the benefits provided in the private sector through the
combination of social security, pension plans, and capital
accumulation plans. This difference occurs not because full career
Federal employees receive a higher annuity to begin. with than their
counterparts in the private sector but because of the more liberal
types of benefits provided and the indexing after retirement.
Charts IV-3 through IV-6 illustrate the replacement income
earned by retirees under the Civil Service Retirement System and
private sector systems who have served a full career and retired at
-54-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
14903 joloi Ionuuy
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
0
0
\
0
1
1
1
1
1
1
1
1
1
1
Banef it Val ua ($)
(Thous:m:31"de)
o '6;
r- T ? T
1
t N,?
.'o
\. \\
r' \\
\ 7(
\ 0
1
, 0
\ t
0 ?
c",,, \
0
0
0
0
\
\
?
Declassified and and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
NOS I elVd1A103
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
62 or 65. The charts show retirees with both 20 and 40 years of
service to illustrate the range of difference. These charts include
only the social security and defined benefit retirement systems.
Nevertheless, at age 65 the average private sector system provides
replacement income that is 18% higher than CSRS for 20 years service
and 3% lower at 40 years of service.
As in CSRS, private sector plans usually permit retirement after
age 55. However, the benefit is typically reduced below age 62 or
65. CSRS employees can retire on unreduced retirement benefits at
age 55 with 30 years of service or age 62 with 20 years of service.
This provision therefore adds substantial value to the CSRS system.
Employees who become disabled before full retirement age
(usually age 62 or 65) can receive disability benefits from most
employer pension plans and these are almost always supplemented by
insurance benefits. However, CSRS disability benefits are
substantially more valuable because of a more liberal definition of
disability.
The survivor benefits provided by CSRS are more liberal than
those provided by private sector plans alone. However, the generous
survivor benefits under social security, plus the retiree family
benefits,
increase the combined value of survivor and family
benefits close to the CSRS value.
The value of the vested benefits of the Civil Service Retirement
System is lower than for the private sector. Not only are private
sector employees entitled to social security, which is fully
portable, but CSRS employees often lose their vested rights by
taking a refund of contributions.
The main reason that the Civil Service Retirement System is more
valuable than the private sector is the full indexing of benefits.
In the private sector, social security is fully indexed but the
pension plans themselves are only about 30% indexed. Further, the
-56-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Z Of Final Salary
icS. n3
o w Ul co
0 4:.
0
o m
o -.i
o m
o
o I-
U)
o
^
-
-
m
E
a
m
1
I
I
I
/
/
/
/
I
1
I
1
i
I
1
I
I
i
i
I
r
I
I
r
1
r
I
I
i
i
I
r
i
I
r
r
I
1
r
I
1
i
I
1
-1Vd303d
-57-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
% Of Final Salary
N
1-(-7
0 CD (11 0) 03 (0
0 A
0
0
0 V
0
0
0 1(1..3
0
o
a
I 1
0
T 0
i0
ID;
is
L101.33S 3.1A/Anid Ol 1Vel0ad
Declassified and and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
,_0
0
ut
o
o)
C
.1
a
% Of Final Salary
18
0 (.0
0 A
o cri co
o a
o .1
o co
o
o
N ;
o
-
-
-
I -0
1 "
i 0
I
1
1
7
0
a
m
1
o
-
/
/
to
/
/
I
I
1
/
/
/
/ II
/ //
/
1
i
I 1
1
I
1
1
I
i
i
1
I -0
I 0
JO
r
r
I
r
/
//
t
//
/
//
/
/
d01.33S AlVAIdd 01 1id303_d
-59-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
31VH 1N31,133V1d3H
NOSIdVd1/103
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
5
N
ow
Li;
C
a)
C
C
0
I
-
-
-
-
-
-
-
_
C-;
1
I
% Of Final Salary
(n a)
oN 8 8 0 o
1
I
I
1
I
V
0
E
(0
I y I I I ,I
1,I
/
I I
,
,
I I
I
1 /
I
1
I
1
I1
f
I
i
I
II
I
I1
1
II
I
I
ri
7 1
E 1
0
1 I
0 I
....
P, I
I
1
;
I
;
;
I
;
;
NO133S 31VAI?Jd 01 7VN30ad
-6 6 -
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
indexing in the Civil Service Retirement System is relatively more
expensive than in social security because the benefit is typically
paid from an earlier age so the indexing operates over a longer
period.
The employee contribution under CSRS is somewhat higher than the
private sector employee's contribution to social security. Almost
all private sector pension' plans are non-contributory but voluntary
contributions are needed to take advantage of the capital
accumulation plan.
The value of the average State retirement system is closer to
that of CSRS than are the private sector systems because the early
retirement and indexing features of the State systems are similar to
those of CSRS. Also, disability and survivor benefits are closer to
CSRS practice. However, State employees tend to contribute to their
retirement system, as well as to social security, so the employer
paid value is lower, relative to the total benefit, than in CSRS or
the private sector.
-61-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
V. FRINGE BENEFITS OTHER THAN RETIREMENT
To be able to evaluate the entire compensation package an
employer provides for a firm's workforce, it is essential to examine
each major element that makes up that package from both a cash and
benefits perspective. Section IV addresses the important area of
retirement plans; Section VI analyzes the salary components of total
compensation. This section deals with benefit elements other than
the retirement system, namely death benefits, disability income
protection, health care plans, holidays and vacations, and executive
perquisites.
The Federal government provides a non-retirement benefit
package that is 3.6% of salary less valuable than the average
package offered by the private sector employer represented in the
Hay/Huggins data base. The average non-retirement State benefit
package is 2.0% more valuable than the Federal government's
offering. When added to the retirement system value, the Federal
benefits are worth 2.8% of pay more
.5% more than State plans.
than private sector plans and
The summary of findings in Table V-1 reflects the difference
between the value of the current Federal plans and the average value
of plans provided by private sector employers and States. The
differences are expressed as a percent of salary.
This section contrasts the value of the benefits now offered to
Federal employees with those offered by private sector employers and
State governments. Each major benefit segment is analyzed
separately and is annotated with charts that show Federal plan
values against the practices of public and private sector
employers. Appendix H is a tabular summary of benefit practices
reported by the States surveyed, by the private sector firms in the
-62-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Benefit Area
Table V-1
Differences between
Federal Benefit Plans and Average
Private Sector and State Benefit Plans
as a Percent of Pay
Federal vs. Average
Private Sector
Percent Difference
Federal vs. Average
State Percent
Difference
Death Benefits - .3% .5%
Disability Income Replacement - .7% -.2%
Health Benefits -2.2% -2.2%
Holidays and Vacations .8% -.3%
Executive Perquisites -1.2% 0
Statutory other than
Social Security 0 .2%
Sub Total -3.6% -2.0%
Retirement Plan 6.4% 2.5%
Total Fringe Benefits 2.8% .5%
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21: CIA-RECP89-00066R000800210002-2
Hay/Huggins data base, .and by the Federal government. The
prevalence of benefits practice table permits a comparison of
approaches to benefits used by each sector. The Appendix is broken
down into major benefit categories and provisions within those
categories.
A. Life Insurance
The value of the Federal Employees Group Life Insurance (FEGLI)
Program falls below the average value of the private sector employer
plans but above the State plan average value (see charts V-1 and
V-2).
Under the FEGLI Program, employees may elect basic coverage in
amounts that vary with age. The basic coverage amount is one times
annual basic pay, rounded to the next thousand dollars, plus
$2,000. Depending on the age of the employee, varying multiples of
the basic coverage amount are payable:
Multiple of Basic
Age Coverage Amount
35 or younger
36
37
38
39
40
41
42
43
44
45 or older
-64-
2.0
1.9
1.8
1.7
1.6
1.5
1.4
1.3
1.2
1.1
1.0
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
-
1 1 1 1 1 1
i
I
It
X
0
Bsnefit Value ($)
(Thousands)
1 1 1 1 I 1 1 1
1
sda
(it
0
0
11
0
1 1
0
,--. \\\
I 0
1 \-0
1
0
0
I
I
111111111
-65-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
',.1
0
0-.
Benefit Value ($)
(Thousands)
11111[111
-66-
N W
-1V130ad
H
o
0A133-1S
u)
H
>
H
m
U)
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The cost of the basic coverage is shared with the government
paying one-third and the employee paying two-thirds. The premium
reflects the cost of both current coverage as well as the cost of
extending coverage to retirees. This feature approximately doubles
the premium amount otherwise necessary.
Federal employees who elect basic coverage are also eligible for
three optional programs:
? Option A which provides a $10,000 life insurance amount and
a $10,000 accidental death and dismemberment benefit;
? Option B under which an employee may elect from one to five
times annual basic pay; and
? Option C which provides a $5,000 spouse benefit and a
$2,500 child's benefit.
Optional coverage is fully employee paid.
At retirement, the basic coverage continues at no cost to the
retiree. At age 65, the coverage reduces 2 percent per month to a
minimum of 25 percent of the basic insurance amount. Retirees can
opt to purchase, at their own expense, either coverage tha does not
reduce or that reduces only in part. In any case, accidental death
and dismemberment coverage stops at retirement.
Optional coverage continues at retirement. The retiree pays
full premium to age 65. At age 65 premiums are no longer payable
and coverage begins to reduce. Option A coverage reduces until it
reaches 25 percent of its face value; Options B and C reduce at 2
percent per month until coverage ends.
Private sector employers universally provide basic group life
insurance that typically is expressed as a multiple of earnings.
Most commonly this benefit is fully employer paid and is two times
-67-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
annual salary with a maximum ranging from $100,000 to $300,000.
Two-thirds of private sector firms continue coverage at retirement
in full or at a reduced amount at no cost to the retiree.
The basic benefit under State employer provided life insurance
tends to be a low multiple of pay (1 to 1.25 times salary) or a
modest uniform amount with one State reporting a $2,000 basic group
life benefit. The basic coverage is fully employer paid in about
one-half the cases; in the balance, the premium is either shared
with the employee or the employee pays for the coverage in full.
Over half of the States cancel coverage at retirement.
More than half of the private sector firms allow employees to
purchase supplemental group life insurance. The coverage is
generally an additional one or two times earnings. Three-quarters
of the supplemental plans are fully employee paid; 20 percent of the
employers share the cost of the coverage. Less than one-half of the
States offer supplemental coverage; two-thirds of those that do pay
nothing toward the premium.
Overall, the private sector's group life insurance coverage is
worth an average :3% of salary more than FEGLI. The typical private
sector plan offers a larger basic benefit at no cost to the
employee. FEGLI is more valuable than the average State plan by .5%
of salary because its basic benefit amount is larger than that of
many of the States and because FEGLI allows employees to purchase
supplemental coverage, a feature present in less than one-half of
the States.
B. Disability Income Protection
Employers provide for salary continuance in the event of either
short-term or long-term disability through a variety of methods.
Generally, private sector employers use a combination of formal sick
leave arrangements, accident and sickness insurance, and short-term
disability plans to cover instances of occasional illnesses or
-68-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
accidental injuries. State governments use short-term sick leave
plans based on accumulation of days, similar to the Federal
government's system. Chronic disability cases are covered under
long-term disability (LTD) plans or pension plan provisions. The
value of the Federal government's disability benefits is below the
average value of plans offered by both State and private sector
employers (see Charts V-3 and V-4).
Under the Federal government system employees accrue sick leave
at the rate of 13 days per year; there is no ceiling on the number
of hours of sick leave that an employee can accumulate. Once an
employee's sick leave balance is exhausted, agencies have the
discretionary authority to grant 30 days advance sick leave. While
an employee is on sick leave, salary is continued at 100%. Once
accrued and advance leave has been used, a disabled employee may
either go on leave-without-pay or apply for a disability retirement.
The typical State government sick leave is based on an
accumulation of days with no maximum on the number of days that may
be accrued. One-third of the States' accrual rates are greater than
the Federal government's. About one-half of the States report
employees accrue from 13 to 15 days per year.
In the private sector employers generally grant 10 or 12 days of
sick leave a year (28%) or provide either full or partial pay for a
scheduled number of weeks that varies with length of service (48%).
A well-designed employer plan based on accumulation of days, allows
employees to accumulate sick leave to the extent of the exclusionary
period imposed by the firm's LTD coverage, generally 90 or 180
days. Likewise, scheduled plans based on service generally express
the benefit as a number of weeks at full pay and a number of weeks
at partial pay up to a maximum of 26 weeks. The number of weeks at
full and partial pay are a function of the number of years the
employee has worked for the employer.
-69-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
3
C
?10
0?-?
0
C
O0 4'
O Cc C.')
? 0
n
0
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Benef it Va 1 ue ($)
(Thousands)
Ui
AlIlIOVSIO
_l_VAIad
- 70 -
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
NOSIaVd1/103
24
J.
0
C
0
0
J
a
0
...,
LISpj ID401 IDnuuy
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
, 0
0
N
o
w
o
4:.
o
Ul
0
a)
o
..3
0
lp.111.
Benefit Value ($)
(Thous:lc:India)
N W A ul
I I I
1
I I I
S_J_ViS 0313313S 01 1Vei303d
-71-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
AlIlIOVSIO
N0SIHVd1/403 31A103N1
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
A less common approach used by employers to provide a regular
income to employees who are incapable of work on a short-term basis
is the purchase of accident and sickness insurance. Generally,
these plans are paid for fully by the employer and guarantee either
a stated percentage of salary or a flat dollar amount per week.
Typically, the maximum duration of benefits is 26 weeks. A well
designed accident and sickness policy will coincide with the
exclusionary period under the employer's LTD plan.
Federal employees have income protection in the event of a
long-term disability through the retirement system only. Under the
current Civil Service Retirement System (CSRS) employees are
eligible at any age for disability retirement if they have completed
5 years of creditable service and are unable, because of disease or
injury, to perform the duties of (1) their current position, or
(2)/a vacant position at the same grade or pay level for which they
are qualified for reassignment in the same agency and commuting
area. The amount of the benefit is usually 40 percent of high-three
average salary. Upon recovery the employee is eligible for
re-employment, but is not guaranteed that a position will be
available. The employee shares the cost of the coverage through the
7 percent contribution to the retirement system.
Typical of the private sector and one-third of the States
surveyed are long-term disability plans. These are generally
underwritten policies that provide 60 percent of a beneficiary's
salary until the employee reaches the normal retirement age under
the employer's pension plan. The amount of the benefit is directly
offset by 100 percent of any benefit payable from both social
security and any other employer sponsored plan. The majority of LTD
plans use a definition of disability similar to that used under the
current Civil Service Retirement System for the first two years of
benefits. If benefits continue to be payable after two years, a
beneficiary must satisfy a stricter disability definition that is
similar to that used by the social security system, namely, the
inability to engage in any substantive gainful activity. LTD
-72-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
benefits generally become payable after either a three or six month
exclusionary period.
Private sector LTD plans are generally coordinated with pension
plans in one of two ways:
? Approximately two-thirds of private sector employers allow
pension credits to accrue during a period of disability;
under such an arrangement the eventual pension benefit
payable at normal retirement age reflects full career
service; and
? Approximately one-third of the employer community provides
a disability retirement benefit upon expiration of the
short-term disability coverage. Typically a fully-accrued
benefit, unreduced for early retirement, is provided. If
the employer also provides an LTD plan, which is a typical
occurrence, a well designed package will offset the LTD
benefit by the amount of the disability pension benefit.
State governments that offer LTD plans tend to use the latter
approach in coordinating the benefits with pension plans.
About two-thirds of private sector employers pay the full cost
of LTD coverage; one-fifth share the cost with the employee; and the
balance look to the employee to pay the full premium. Among State
governments with LTD coverage, it is typical for the employee to pay
for the coverage either in part or fully. A recent change in the
tax code which treats benefits arising from employee contributions
much more favorably than those that are financed by employer
contributions, will undoubtedly prompt more employers to shift LTD
premiums to employees.
In the disability income area, the average private sector plan
is .7 percent of pay ahead of the Federal government; the average
State value is .2 percent of pay ahead of the Federal government.
-73-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
This is largely due to the fact that the Federal government does not
offer long-range disability income protection that is separate from
the pension plan. Finally, the average value of the State plans is
higher, in part, due to the higher average accrual rate under the
sick leave plans.
C. Health Care Benefits
The provision of health care benefits is virtually universal
among medium and large size employers in both the private sector and
State setting. Federal employees may choose coverage from a variety
of plans available under the Federal Employees Health Benefits
Program. The choices range from nationwide insurance-type plans
offered by major insurers or employee organizations to Health
Maintenance Organizations available in specific geographic areas.
For comparison purposes we used the plan elected by the most number
of Federal employees, namely, the High Option Service Benefit Plan
administered by Blue Cross/Blue Shield.
The value of the Federal government's provided health care plan
is below the average of comparable plans offered by private sector
and State employers (see Charts V-5 and V-6). Approximately
three-quarters of the private sector and State plans are more
liberal than the representative FEHBP plan.
The FEHBP plan covers the same type of hospital, medical and
prescription drug expenses as does the average private sector and
State plan. However, whereas nearly three-quarters of private
sector employers and half of the State plans provide significant
dental coverage, the FEHBP plan does not. Additionally, the amount
of the calendar year deductible, a feature common to all plans, is
twice as much under the FEHBP plan as under the typical private
sector or State arrangement ($200 versus $100).
-74-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
8
3
3
" C
?I a
0
C
00
0 rt. ?-?
2 0
CL
0
n
ID
a
7
c.n
?.1
I I I
1
Bansfit \talus ($)
(Thousands)
1 m
0
0.
1 0
I 1
I t...
0
I T Cl %
I 1 I)
T 0
I" it I \ '
I 0
0
10
I? ' 10
0
I
I I \
I I)
t
I I I
(JI
Declassified and Approved For Release 2012/11/21: CIA-RDP89-00066R000800210002-2
(spuoenoyi)
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
0
N
0
u)
C
0
o
o)
o
-4
o
-
-
-
-
-
-
_
Ill
Brancifit ValuGa ($)
(Thousands)
N W 4%. (Il
1
1
1
1
i
1
I
1
1
I
1
1
I I 1
1
0
I-
-Ti
1
a
o
rt
C)
o
I
I
I I I
-n
m
o
m
>
7
H
o
Oal_33-15
cn
H
>
H
in
cn
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The Federal government pays approximately 60% of the health care
premium for both employees and dependents. In the private sector
two-thirds of the employers pay for employee coverage in full and
39% pay for dependent coverage in full. Among the States surveyed,
approximately one-half pay for employee coverage in full and
one-quarter provide dependent coverage paid in full. The balance
share the premium cost with the employee paying generally 25% or
less of the cost.
Three factors -- dental coverage, deductible amounts, and
cost-sharing -- result in the average private sector plan being
worth about 2.2 percent of pay more than the representative Federal
government plan. The average State plan is also worth about 2.2
percent of pay more than the most frequently chosen Federal
alternative.
D. Holidays and Vacations
The value of the paid leave the Federal government provides its
employees tracks with the practices of the more liberal segment of
the private sector employer community and falls about on an average
with that provided by the States (see charts V-7 and V-8).
The Federal government in 1984 offers nine paid holidays while
the average private sector employer offers 10.2 days annually. The
most commonly reported number of holidays by State governments is 13
or more per year. The greater number of holidays, particularly for
State employees, increases the time-off-with-pay value for the
private and State sectors vis-a-vis the Federal government.
However, the private sector holiday advantage is offset by the
annual leave accrual pattern for Federal employees which is more
liberal for the short service employee than that provided by the
typical employer. State employees' vacation schedules are more
similar to that offered by the Federal government.
-77-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Benefit Value ($)
(Thousands)
p?-?N W UI 0) ?,) CO CO (1":3
11 I 1\\\1 1
1 1
X
11 1
1 1 1
111
11 1
1 1
ONV SAVOI10H
-78-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
\
Benefit Value ($)
(Thousaaricies)
1111111 III III III II I I I I II I
0
Oal_3373S
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
SAVOI1OH
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The combination of factors results in the Federal government's
leave plan being more valuable than the average private sector
employer by .8 percent of pay and less valuable than the average
State value by about .3 percent of pay.
E. Executive Perquisites
Private sector employers generally provide executive perquisites
to officers and other highly compensated employees. Commonly, plan
designs are geared either to improve benefits for executives within
the firm or to facilitate the recruitment of mature executives who
could suffer a loss in benefit expectation by making an employment
change. Generally, the executive benefits fall into three
categories:
? Executive benefit plans that are in addition to or parallel
the basic benefits program (e.g., executive vacation
schedules or executive group life insurance);
? Executive perquisites that provide key executives with
tax-free or tax favorable benefits which they normally
would be required to provide on their own with after tax
dollars (e.g., personal financial counseling or special
parking); and
? Long-term incentives that provide rewards for performance
or a vehicle for long-term capital accumulation (e.g., a
deferred compensation arrangement).
Employers tap a wide range of noncash avenues to add to
executive compensation packages. The most frequently offered
perquisites include employer-provided cars, special parking
privileges, payment for club memberships, directors' and officers'
liability insurance,. and a variety of deferred compensation and
stock arrangements.
_on_
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Chart V-9 shows the value spread of perquisites offered by the
private sector. Federal and state practice are not reflected since
there are generally no comparable benefits. The Federal government
trails the private sector by 1.2 percent of pay in this area.
F. Total Benefits
One of the difficulties in analyzing individual components of
the benefit package is the fact that many benefits interact and
often different types of benefits are used for the same purpose. In
this report, two of the key interactions are between the life
insurance and disability systems and the retirement systems. The
Federal government relies on the retirement system to provide all of
the long term disability income and substantial survivor income
plans. Thus, the fact that the life insurance and disability plans
of the Federal government lag behind those of the private sector is
explained, at least in part, by the fact that a large part of such
benefits are provided through the retirement system. Conversely,
part of the reason that the Federal retirement plan is ahead of the
private sector is that it includes substantial death and disability
benefits that are not commonly found in private sector retirement
systems.
If the private sector advantage in death benefits of .3% of pay
and in disability benefits of .7% of pay is deducted from the 6.4%
lag in retirement benefits, the difference is reduced to 5.4% of
pay. When all benefits are combined the total Federal benefit
package is 2.8% of pay ahead of the private sector and .5% of pay
ahead of the State governments. The range of these comparisons are
shown on Chart V-10 and Chart V-11.
-81-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
0.
o
0
J.
a)
o
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
p.111
IV
CO
1 v 1 1 1 1 1 1 1 1 1 r 1 1 1
\
-
-
-
-
-
-
\
\
\
\
\
\
\
\
Benef it Value ($)
(TInclutaczncla)
4. CJI a) V CO (0 8
I 1 1
1
1 I l\1 1 1 1 1 1 1 I 1 1 1 I
%
t
1
1
\
a
cn
C
\
I
t
\
t
\
t
i
1
1
1
1
1
i
I -0
1 0
to
i
1
I I I
Declassified and Approved For Release 2012/11/21: CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
8
0
3
3
-4c
0
7
0
7
UI
a)
0
Benefit Value ($)
(Thousands)
CJI 8
cn
0 41.
cn
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Benefit Value ($)
(Thousands)
1Wd303
H
o
0313313S
cn
H
>
H
m
u-)
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
NOSIelVd1/403 SlId3NAO
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
VI. CASH COMPENSATION
This section presents the evaluation results for the Federal
sample, and comparisons of Federal compensation to private sector,
and State compensation practices.
The major finding relative to compensation comparisons to
private sector employers as of March 1, 1984 is that the General
Schedule salaries would have to be increased by an average of 10.3%
in order to equal aggregate private sector total cash compensation.
By pay grade, average private sector total cash compensation exceeds
that for the Federal government by 2 to 25%.
For the Senior Executive Service (SES), 38 ES-6 positions were
included for illustrative purposes in order to depict results for
the most difficult positions analogous to those found on the General
Schedule. At the ES-6 level, Federal total cash compensation would
have to be increased by 58.4% to equal aggregate private sector
average total cash compensation for positions of the same level of
difficulty.
The major finding in the comparison to State employers is that
the General Schedule salaries exceed average State cash compensation
by 7.8% of average Federal compensation. Average Federal cash
compensation is below the State average at grades 3 and 4, but then
uniformly exceeds the State cash compensation average by 2 to 20%.
This finding is based on the cash compensation data provided by
thirteen States.
A. Federal Sample Evaluation Results
Table VI-1 presents the evaluation results for the 392 non-SES
positions included in the Federal sample. The following statistics
are presented for each selected grade: (1) incumbent-based median
evaluation (that evaluation level below which, and above which, 50%
-85-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
of the incumbent population is expected to fall); (2) evaluation
range; and (3) intergrade differential (the ratio of the median of
one grade to the next lower grade).
The median provides the most conservative, representative
measure of the central tendency of job content foi a grade level
because the median is less sensitive to skewing caused by extreme
results. The evaluation point range indicates the relative
variability of the evaluations in the grade. The position of the
median in the point range for a grade indicates the degree to which
the distribution of the employee population is skewed with respect
to evaluated job difficulty. Specifically, if the median falls
substantially toward the low or high end of the range, this
indicates that half of the population encumbers jobs within a narrow
range of difficulty (the range defined by the closer endpoint and
the median) while the remainder of the population encumbers
positions in a relatively wider band. The intergrade differential
indicates the progression of job difficulty by grade level.
The results indicate that, in general, there is a logical
progression in median evaluated job content as grade level
increases. However, as the intergrade differential results show,
there are relatively large jumps in median difficulty between grade
levels 3 and 4 (44%) and grade levels 12 and 13 (32%). These
factors may relate to the structure inherent in the classification
standards because, for example, many occupational series do not have
non-supervisory positions classified above GS-12.
The relationships between the median evaluation and the point
range for the grade level indicate the existence of some skewing
(i.e., there exist relatively low-population positions in a grade
which deviate from the median by more than 7%). However, the skew
is within the reasonably expected bound of plus or minus 10% except
for grade 3, for which the median is at the end of the point range.
-86-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The histogram (Chart VI-1) on the page following Table VI-1
depicts the degree of skew for each
between the evaluations for each grade.
does not exist. Nevertheless, the very
grade,
In an
strong
as well as overlap
ideal system, overlap
relationship apparent
between median job difficulty and grade level indicates that the
median job evaluations can be used to accurately represent the
overall GS and equivalent compensation practice. The tickmarks
denote the placement of each median within the evaluation range
depicted for each selected grade on the histogram.
Table VI-1
Evaluation Results for the Sample of
392 Federal Positions by Grade Level
Selected
Grade
Level
Evaluation
Point Median
Evaluation
Point Range
Intergrade
Differential
3
102
102-162
1.44 (grade 3-4)
4
147
117-194
1.20 (grade 4-5)
5
177
141-233
1.20 (grade 5-6)
6
213
169-298
1.18 (grade 6-7)
7
252
203-323
9
342
275-417
11
417
289-631
1.23 (grade 11-12)
12
511
363-677
1.32 (grade 12-13)
13
677
534-775
1.23 (grade 13-14)
14
830
654-860
1.24 (grade 14-15)
15
1028
800-1408
At the ES-6 level, the median evaluated job content was 1868
points, and the range was 1142 to 3536 points.
-87-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
cri -
_
-
-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
10Aa1 apcup w0ion!nb3 pup so
Job Content Points
....? -.? ....! --.1. -1. - \
_.., IV C.,4 -F. Cil CD --4 CO CD 0 ..... 1\ ) 04 -F. 01
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1
-1.
tv
04
W
A
N
1
co
w
o
i
?-?
0
N
CD
S3CPM0 INI1VA11103 C1NV SO
-88-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
siinsA NOLLVMVAA AO NO1131dA0
C-)
7
0
1
ct
<
p-4
1
p-.
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
B. Federal Compensation Practice
Chart VI-2 on the following page illustrates the relationship
between the median evaluation and the weighted average base salary
at the selected grade levels. The straight line which appears is
the weighted linear regression result, serving to summarize the
relationship between job evaluation points and base salary which is
indicated by the pattern of dots representing the result for each
selected grade.
The regression statistics presented in Appendix J indicate that
this line fits the data very well. Inspection of the chart reveals
a tendency for the compensation at grades 13-15 to fall ever further
below the line, i.e., it suggests the possibility of a nonlinear
relationship (one where the line describing the relationship
flattens out to better reflect the different relationship between
evaluation points and compensation at grades 13-15). However, as
the statistical results in Appendix J clearly indicate for the case
of a quadratic representation of the data, very little explanatory
power is contributed to the linear result by the quadratic term in
points. This is attributable to the relatively low weight placed on
the data for grades 13-15 due to their relatively low populations.
Hence, the straight line represents the overall Federal compensation
practice quite well. the tabular analyses presented below are based
on the actual average Federal compensation values rather than values
derived from the regression result. Therefore, any lack of fit in
the regression is irrelevant to the ultimate findings.
The linearity of the result indicates that, except at grades
13-15, each unit of job content is rewarded at the same dollar rate
per unit. At the 13-15 levels, the rate tends to decline, on an
accelerating basis. A variety of factors may be responsible,
including higher proportions of relatively shorter-tenure employees
in those grades. Table 111-4 indicates that the average step-in
grade for grades 13-15 is substantially lower than at the other
grades.
-89-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
HO8VW AO SV VIVO)
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
slupd 4ualuo0 qor
o
o
Annua 1 Total Cash
(Thousands)
Iv
o
4s?
o
C)
C
GS AND EQUIVALENT GRADES 3-15
-90-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
FEDERAL CASH COMPENSATION PRACTICE
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Because SES level positions were included only for illustrative
purposes, they were excluded from the foregoing analyses.
C. Comparison of Federal to Aggregate Private Sector Systems
Supplemental cash compensation provided in the private sector
constitutes a significant element of total cash compensation.
Therefore, in the context of a total compensation comparability
study, total cash compensation is the basis to which benefit values
should be added to develop total compensation. Hence, this section
presents the results of total cash compensation comparisons for the
aggregate private sector. Base salary results are contained in
Appendix K. The industrial, financial and service subsector results
are discussed in the next section.
A prefatory note concerning the relationship between private
sector compensation practices for exempt versus nonexempt positions
is required. A priori, there is no reason to assume that any
organization will have the same practice for both groups. First,
the local nature of the nonexempt labor market may be different from
that of the national market for exempt positions, and compensation
ranges for nonexempt positions are typically much narrower. Second,
nonexempt positions are rarely eligible for supplemental cash
compensation, and when eligible, such awards are a negligibly small
percentage of total cash compensation. Therefore, in practice, base
salary is total cash compensation for nonexempt employees. Third,
employers vary in their desires to have their nonexempt and exempt
practices on the same continuum. Therefore, it is quite possible
that discontinuities between the two will be found, as is the case
here. The comparative results will therefore be discussed for
grades 3-6 analogous to the difficulty of nonexempt positions,
separately from grades 7-15 before the overall average comparison is
presented.
-91-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Chart VI-3, on the following page, presents the total cash
compensation result for the aggregate private sector. One line
represents the average compensation values. Above 240 points, two
lines represent the tenth and ninetieth company percentiles: i.e.,
the compensation values below which only 10% of companies pay (the
lower line) and the compensation values above which only 10%, of
companies pay (the upper line). These results are derived from the
data that Hay collects and analyzes for exempt positions. Below 240
points, the results are based on Hay's use of data collected by
BLS. In this case, the percentile lines represent the tenth and
ninetieth compensation percentiles for the areas represented in the
BLS surveys, rather than company compensation practices. The tenth
and ninetieth percentile lines are labeled as P10 and P90,
respectively.
As the chart indicates, the nature of the results for exempt
positions does in fact differ from those for nonexempt positions.
The variability in compensation practice across the 65 areas is much
smaller than across the 1200-plus companies. Further, there is a
slight discontinuity for average compensation at 240 points, with
the value for the nonexempt positions being somewhat greater than
the value for exempt positions. Table VI-2, below, summarizes the
relationships at 240 points.
TABLE VI-2
Aggregate U.S. Private Sector Average Total
Cash Compensation Results at 240
Points, for Exempt and Nonexempt Positions
Average
Total Cash Compensation Ratio
Cash Compensation Value (Nonexempt to
Indicator Nonexempt Exempt Exempt)
90th percentile $23,800 $26,400 90.2%
Average 21,800 21,200 102.8
10th percentile 19,200 16,300 117.8
-92-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
(17861 HalVW JO SV VIVO)
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
o
0 'El N
o
Annual Total Cash
(Thousands)
Wo 42. ?-.1 CO
o otri oci) CD
o o o
-
I
1
I
1
I
-
N
0 ?
0
2
c*
0
3
rt
-0
0
.3'
rt.
0
a)
o
o
-
_
-
-
^
-
-
-
co
0 ?
0
?
Oa
o
1-
-
_
-
-
I
1
\
1
1
1
1
1
I
I
g L-C s3avelo IN31VAI1103 ONV SO
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
EiVAIad
NOIlVSN3d1/103
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Therefore, Chart VI-4 on the next page depicts the private
sector comparison for the grades for which the median evaluation is
below 240 points (grades 3, 4, 5 and 6). Federal compensation
appears to approximate the average private sector compensation at
grade 3, but then continues to fall ever further below the average,
such that at grade 6, Federal pay is almost at the pay of the lowest
paying 10% of the areas.
The following tabular results present the companion comparison
of actual average Federal compensation by grade level to the average
private sector compensation at the median job difficulty level for
the grade.
Table VI-3
Comparison of Federal to Aggregate
Private Sector Average Total Cash Compensation for
Grades 3, 4, 5 and 6
Average Average Ratio (Private
Federal Private Sector Sector to
Grade Level Compensation Compensation Federal)
3
$11,816
$12,577
106.4%
4
13,752
15,309
111.3
5
15,684
17,270
110.1
6
17,807
19,497
109.5
-94-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
H38V11 JO SV VINO)
IT
o ? 0
? o
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
0
Annual Total Camh
CThoumanda)
Ui
0 0 0
-u
a
GS AND EQUIVALENT GRADES 3-6
1VAIeld 01
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
NOI1VSNII/103
NOSIWid1/403
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Chart VI-5, on the next page, illustrates the relationship
between Federal cash compensation and aggregate private sector
average total cash compensation for grades 7-15. Clearly, Federal
pay is consistently below the average, and tends to fall further
below the average as position difficulty increases. The following
table provides the actual cash compensation comparisons.
TABLE VI-4
Comparison of Federal to Aggregate
Private Sector Average Total Cash Compensation for
Grades 7 - 15
Average Average Ratio (Private
Federal Private Sector Sector to
Grade Level Compensation Compensation Federal)
7 $19,408 $21,830 112.48%
9 23,527 26,618 113.14
11 28,653 30,524 106.53
12 34,816 35,510 101.99
13 38,507 44,583 115.78
14 44,188 52,999 119.94
15 51,067 64,102 125.53
The relatively large size of the disparities between Federal and
private sector pay at the 13, 14 and 15 grade levels is due in part
to the lower average step-in-grade of employees at the 13, 14, and
15 levels. If this were not the case (i.e., if the average
step-in-grade at the 13, 14 and 15 levels were equal to the average
of approximately 4-5 for grades 3-7, 9, 11 and 12), the ratios would
be lower by six or more percent.
The preceding comparisons have indicated that private sector
average total cash compensation uniformly exceeds Federal total cash
compensation, by 2% to 25%. Further, the slope of the line
representing the Federal cash compensation practice is smaller than
the slope of the average private sector line, indicating that, on
average, private sector employers pay more per unit of job content
than does the Federal government.
-96-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
H38VW AO SV VIVO)
r
o
0 8
t 0
N
0
0
U
o
o
Ul
(10
00
0-
Cl
0
3
cr 01
0
3 63
(I.
Annual Total Cash
(Thounande)
N
W .1u
0 M
0
0
8 0 0
0 ,..1
0 M
0 W
0
-
?
-
-
-
-
0
0
C
o
.8
o
o
N
0
0
^
-
^
\ \ \
I \
1 \
%
\ \\
\
\
\ \
\
\
\
,
13
0
-1
130.
1 D
Q_
a a
cl? ,-
,..
C)
\
>
0
1
a
u2
m
\
-13
CD
o
0
? 9 7 ?
GS AND EQUIVALENT GRADES 7-15
-n
m
o
m
7:J
>
7
H
o
a0133S 3IVAIeld
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
The best single measure of the overall relationship between
Federal and aggregate private sector average total cash compensation
is one which weights the results by grade level in relation to the
number of incumbents in each grade level. The resulting figure then
indicates the percentage increase in the Federal payroll required to
provide overall average compensation equal to that paid by the
average of all employers in the data base.
Implementation of this procedure results in a weighted average
difference of 10.3%. That is, on a weighted average basis, for the
eleven grades selected, Federal pay would have to be increased by
10.3% in order to equal the total cash compensation paid on average
in the private sector, as of March 1, 1984. This result constitutes
a very reliable estimate of the overall relationship for the entire
GS (and equivalent) pay system for the following reasons:
? The eleven grades selected to represent the GS and
equivalent pay systems cover nearly 95% of all employees on
the GS and equivalent pay 17stems;
? At each grade, the positions evaluated are encumbered by
the majority of employees in jobs at the grade -- on
average, over 67% of the incumbents;
? The private sector data used for comparative purposes are
derived from the largest and most directly comparable data
base extant;
? The methodology used to control for job differences
constitutes a reliable, proven technique;
? The use of total cash compensation provides the most valid
measure of cash comparability; and
? Multiple levels of quality control and quality assurance
have been applied wherever applicable.
-98-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Table VI-5 on the next page provides the information from which
the overall average difference of 10.3% is obtained.
For the ES-6 level positions: the following results are
obtained.
Average ES-6
Total Cash
Compensation
$71,927
Average Aggregate
Private Sector Total
Cash Compensation
Ratio (Private
to ES-6 Cash
Compensation)
$113,968 158.4%
As a supplementary analysis, Hay reviewed the average private
sector total cash compensation for each of three subsectors
constituting the aggregate reported on heretofore: industrial,
financial and service. These analyses were based on inclusion of
the nationwide private sector results at grades 3-6 for each
subsector because no comparable data by subsector exists. The
results are summarized below:
? Industrial subsector average total cash compensation
uniformly exceeds that in the Federal government by 9% to
32% for grades 7-15. Except at grades 11 and 12, the
industrial average exceeds the Federal average by more than
20%. Including the nationwide private sector results at
grades 3-6, overall average Federal cash compensation
should be increased 14.8% to equal that in the industrial
subsector.
? Average financial subsector total cash compensation exceeds
that in the Federal government by 5% to 19% at grades 13,
14 and 15; however, average Federal compensation is 5% to
11% greater at grades 7, 9, 11 and 12. Including the
nationwide results for grades 3-6 results in the finding
that an overall average increase in Federal pay of 1.9%
would be required to equate it to financial subsector total
cash compensation.
-99-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Di CD CD CD ri
0 CD
rrCD &CD Cr)
0
? rr
P
aouaiajj-pa
P
Dl
0 Ca
Di
CD
1-1
Cu 0
rr
0
0
CL 0
1-h
0
CD
CD
CD
Cl-
I-.
CD
saaAoicluia reqoi
I-, I-. I-, l-
Ui
4a. (...) LID -.I 01 (.71 ? UJ
CDN
r
H
0
T
0 70
>
H
m
u)
m
0
H
x
-111-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
B. Comparison of Federal to State Total Compensation
Table VII-2 presents the comparison of Federal to State total
compensation.
Table VII-2 compares Federal and State government total
compensation and depicts that when the benefits are combined with
cash compensation the total compensation difference is 7.1% on
average. At grades 3 and 4, the values for State positions are
above the Federal values but in grades 5 to 15 the Federal positions
are ahead by 1% to 20%.
Even though benefits for comparable Federal salaries are
slightly higher than for State employees, the combination of
benefits and cash compensation reduces the cash compensation lag.
This is primarily a result of the fact that benefits are
proportionately higher at lower pay levels. Thus, the lower State
salaries often have a higher percentage of pay as benefits then in
the comparable Federal position.
When the private sector and State findings are weighted by the
number of employees in each category in the United States labor
force, the total compensation is 6.2% ahead of the total
compensation of the Federal employees. If Federal employees receive
a 3.5% salary increase in January 1985, it is expected that the
difference will grow to 8% or more.
-112-
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Grade Employee
Level Population'
3 70,146
4 156,361
5 184,813
6 88,382
7 127,096
9 137,396
11 142,112
12 159,660
13 100,798
14 50,410
15 25,232
1,242,406
Table VII -2
Comparison of Federal to Aggregate
State Average Total Compensation,
All Grades, and Computation of
Overall Average Difference
Average
Federal
Compensation
$19,360
22,202
24,946
27,854
29,973
35,821
43,023
51,716
56,886
64,856
74,452
Average
State
Compensation
$20,770
23,161
24,708
26,516
28,413
31,925
.36,799
41,607
50,087
57,836
66,497
Percentage
Difference to
Federal Base
7.28%
4.32
-0.95
-4.80
-5.20
-10.88
-14.47
-19.55
-11.95
-10.82
-10.69
Weighting
Factor 2
0.0565
0.1259
0.1488
0.0711
0.1023
0.1106
0.1144
0.1285
0.0811
0.0406
0.0203
1.0000
t'et
Weighted
Difference3
0.41%
0.54
-0.14
-0.34
-0.53
-1.20
-1.66
-2.51
-0.97
-0.44
-0.22
-7.06%
1/ Presents data obtained from OPM as of 1/2/84.
2/ Weighting Factor = total employees in the grade as a proportion of total employees in the eleven
selected grades.
3/ Weighted Difference = Private Sector to Federal Compensation difference multiplied by the weighting
factor.
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
100
90
80
70
o
c
60
CO
(,)
50
CID
E ?
40
-0
?
30
20
?
?
-
?
?
?
1 0 ?
0
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
Chart, V I I-2
TOTAL COMPENSATION COMPARISON
FEDERAL TO SELECTED STATES
3 4 5 6 7 9 11
12
13
GS and Equivalent Grade Level
0 Federal Comp State Comp
14
Declassified and Approved For Release 2012/11/21 : CIA-RDP89-00066R000800210002-2
15