Federal Pensions: Thrift Savings Plan Has Key Role in Retirement Benefits
(Letter Report, 10/19/95, GAO/HEHS-96-1).
Pursuant to a congressional request, GAO provided information on the
Federal Employee Retirement System (FERS), focusing on: (1) the extent
to which employees under FERS voluntarily contribute to the Thrift
Savings Plan (TSP); (2) how well TSP educational materials address the
importance of employee participation; and (3) whether there should be
additional TSP investment options.
GAO found that: (1) as of September 1994, about 76 percent of FERS
employees voluntarily contributed an average of 5.7 percent of their
salaries to TSP; (2) most of the non-contributing employees were in
lower pay grades; (3) lower paid workers contribute less of their
salaries to TSP because Social Security benefits are proportionally
higher for lower income workers; (4) these workers may not need to
contribute as much to TSP in order to achieve appropriate retirement
income levels; (5) mid- and high-pay level workers need to contribute at
least 5 percent of their salaries over their careers to achieve 60 to 80
percent of preretirement income; (6) although TSP educational materials
extensively discuss the plan's financial aspects, they do not explicitly
discuss the importance of employee participation in TSP; (7) the TSP
Board is seeking legislation that would add two stock fund options to
the three investment options it already offers in order to make these
options more similar to those available in private sector plans; and (8)
although they carry a higher investment risk, these new investment
options could potentially produce higher earnings for plan participants.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: HEHS-96-1
TITLE: Federal Pensions: Thrift Savings Plan Has Key Role in
Retirement Benefits
DATE: 10/19/95
SUBJECT: Federal employee retirement programs
Retirement pensions
Retirement benefits
Investments
Comparative analysis
Federal employees
Social security benefits
Financial analysis
Information dissemination operations
IDENTIFIER: Civil Service Retirement System
Federal Employees Retirement System
Federal Thrift Savings Plan
**************************************************************************
* This file contains an ASCII representation of the text of a GAO *
* report. Delineations within the text indicating chapter titles, *
* headings, and bullets are preserved. Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are *
* identified by double and single lines. The numbers on the right end *
* of these lines indicate the position of each of the subsections in the *
* document outline. These numbers do NOT correspond with the page *
* numbers of the printed product. *
* *
* No attempt has been made to display graphic images, although figure *
* captions are reproduced. Tables are included, but may not resemble *
* those in the printed version. *
* *
* A printed copy of this report may be obtained from the GAO Document *
* Distribution Facility by calling (202) 512-6000, by faxing your *
* request to (301) 258-4066, or by writing to P.O. Box 6015, *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time. *
**************************************************************************
Cover
================================================================ COVER
Report to Congressional Requesters
October 1995
FEDERAL PENSIONS - THRIFT SAVINGS
PLAN HAS KEY ROLE IN RETIREMENT
BENEFITS
GAO/HEHS-96-1
Thrift Savings Plan
(105680)
Abbreviations
=============================================================== ABBREV
CSRS - Civil Service Retirement System
FERS - Federal Employees Retirement System
OPM - Office of Personnel Management
TSP - Thrift Savings Plan
Letter
=============================================================== LETTER
B-260594
October 19, 1995
The Honorable Constance A. Morella
House of Representatives
The Honorable Eleanor Holmes Norton
House of Representatives
In the early 1980s, the Congress began to consider a new retirement
system for federal employees in part because it believed the existing
Civil Service Retirement System (CSRS) was too costly. As a result,
the Congress enacted legislation that extended Social Security
coverage to federal employees hired after December 31, 1983, and
established the Federal Employees Retirement System (FERS), which it
modeled after private sector retirement systems. Unlike CSRS
retirees, FERS retirees receive benefits from three sources: Social
Security, a federal government annuity, and Thrift Savings Plan (TSP)
distributions.\1
Congressional deliberators believed that this combination of benefits
would enable employees to maintain roughly their preretirement
standard of living in retirement.
Because of your concerns about whether TSP was fulfilling its role in
helping FERS-covered employees to save effectively for retirement, in
your capacities as ranking minority and chair of the former
Subcommittee on Compensation and Employee Benefits, Committee on Post
Office and Civil Service, you asked us to develop information on
the extent to which FERS-covered employees voluntarily contribute
to TSP and the percentage of their salaries contributed (the
deferral rate);
how well TSP's educational materials address the importance of
employee participation to meet certain retirement income goals;
and
the desirability of having additional TSP investment options.
To develop this information, we reviewed the legislative history of
FERS and TSP, and discussed it with Congressional Research Service
staff. We also reviewed their studies of the program, updated
certain data in their 1986 report comparing CSRS and FERS retirement
benefits (see app. I), and analyzed surveys of private sector plans
established under Internal Revenue Code section 401(k) that are
similar to TSP. In addition, we interviewed TSP staff and officials
and obtained and analyzed TSP demographic data from 1987 to 1993,
TSP's educational materials, and information on the TSP Board's
consideration of additional investment options. Our work was
performed from January through August 1995 in accordance with
generally accepted government auditing standards.
--------------------
\1 CSRS employees can also participate in TSP.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :1
As of September 1994, some 942,000 FERS-covered employees (about 76
percent) were voluntarily contributing an average of 5.7 percent of
their salaries to TSP. Most of the remaining 300,000 FERS-covered
employees (24 percent) who were not contributing were in the lower
pay ranges. Lower-paid workers who were contributing were doing so
at lower rates than higher-paid workers--an average of about 4.4
percent of their salaries.
Although lower-paid workers are deferring at lower rates, we found
that it may be less necessary for them to contribute to TSP than it
is for workers with higher earnings. A general measure of retirement
income needs used by pension professionals is about 60 to 80 percent
of preretirement income. Thus, lower-paid workers may achieve
retirement income levels within this range even with low deferral
rates because Social Security benefits are proportionately greater
for them. On the other hand, mid- to higher-pay-level workers need
to defer at least 5 percent of their salaries throughout their
careers, if not more, to achieve retirement income levels within this
range.
Educating FERS workers can play a central role in their making wise
preretirement investment decisions. We found that although TSP's
educational materials extensively discuss the plan's financial
aspects, they are not explicit in discussing how TSP can enable
FERS-covered employees to achieve their retirement income goals.
A full range of options can also facilitate sound preretirement
investment decisions. More specifically, the TSP Board concluded
that a domestic small capitalization stock fund and an international
stock fund would give employees more diversity in investment options
and the opportunity for potentially higher earnings, although at
greater risk. In May 1995, the TSP Board decided to seek legislation
that would add these two investment options to the three it currently
has. We found that these additions would make TSP's investment
options more closely resemble those available in similar private
sector plans.
BACKGROUND
------------------------------------------------------------ Letter :2
In 1986, the Congress replaced CSRS with FERS\2 for federal employees
hired beginning January 1, 1984, in part to (1) recognize the
inclusion of federal employees under Social Security and (2) reduce
federal pension costs. Among the concerns of congressional
deliberators in crafting FERS were that its retirement benefits be
comparable with those under CSRS and enable employees to maintain
their standard of living in retirement. To accomplish these and
other goals, FERS provides a retirement benefit that comprises three
components: a basic FERS annuity, Social Security payments, and TSP
payments. The total income from these sources is meant to help
individuals to receive retirement benefits comparable with CSRS
benefits and commensurate with their retirement income goals.
The basic FERS annuity is similar to CSRS in that it guarantees a
specific monthly retirement benefit based on age, length of
creditable service, and the average of the highest 3 consecutive
years' salaries.\3 However, the FERS annuity is lower because its
benefit formula credits each year of service generally at 1 percent
while CSRS service credits range from 1.5 to 2 percent per year of
service. In addition, cost-of-living adjustments authorized by FERS
are lower and generally are not provided before age 62.
Unlike the FERS basic annuity, the benefit provided under Social
Security's benefit formula declines as a proportion of individuals'
preretirement earnings as their earnings increase. For example, a
person aged 62, with a certain lifetime earnings pattern and earnings
of $20,000 in his or her final year of employment, would receive
Social Security benefits that represent about 35 percent of those
earnings. In contrast, a person aged 62, with a certain lifetime
earnings pattern and earnings of $75,000 in his or her final year of
employment, would receive a benefit that represents just about 17
percent of those earnings.
Pension professionals believe that to maintain roughly the same
living standard in retirement, individuals' income needs generally
range from 60 to 80 percent of their preretirement annual pretax
earnings. Among other things, retirees typically pay less taxes, do
not have work-related expenses such as daily commuting costs and
clothing needs, may no longer have dependent children, and may have
their mortgages paid.
TSP is administered by the Federal Retirement Thrift Investment
Board, which is an independent agency. The Board consists of five
part-time members who are appointed by the President. TSP's daily
activities are carried out by a staff headed by an executive director
selected by the Board. Retirement benefits from TSP are the flexible
component of FERS because they depend on the amount that is in each
employee's account at retirement. Thus, TSP can help FERS-covered
employees to save toward a total retirement benefit that is
commensurate with their retirement income goals.
Employees under FERS are automatically enrolled in TSP because
federal agencies are required to contribute an amount equal to 1
percent of their employees' salaries to the plan. In addition,
employees can make voluntary contributions up to 10 percent of their
salaries: agencies match the first 3 percent on a dollar-for-dollar
basis and the next 2 percent at 50 cents to a dollar, for a 5 percent
total agency contribution; additional employee contributions are not
matched, but all contributions and earnings thereon are tax deferred.
CSRS employees may also participate in TSP by contributing up to 5
percent of their salaries; while there is no agency match, the
contributions and earnings are tax deferred. However, all employee
contributions are limited to a statutory inflation-adjusted cap,
which was $8,994 in 1993.
TSP contributions can be invested in a federal government securities
fund (G fund), a commercial bond fund (F fund), and a commercial
large capitalization stock fund (C fund). The C and F funds are
passively managed index funds that track changes in a certain body of
securities in the stock and bond markets. These investment options
were specified in TSP's statute, which also provided for adding
investment options, via amendments, at the request of TSP's Board.
In addition, TSP's law restricted the amounts that could be invested
in the C and F funds through 1990.\4 With the lifting of the
restriction in 1991, employees have increased their contributions to
the C and F funds. For example, in January 1991 about 5 percent and
2 percent of contributions were going into the C and F funds,
respectively, while in August 1994 the comparable rates were 35
percent and 10 percent. In January 1995, TSP contributions and
earnings were invested as shown in table 1.
Table 1
TSP Investments as of January 1995
(Dollars in Billions)
Percen
Investment Amount t
------------------------------------------------------ ------ ------
G fund $18.9 70
C fund 6.4 24
F fund 1.6 6
======================================================================
Total $26.9\ 100
a
----------------------------------------------------------------------
\a Separate data for CSRS- and FERS-covered employees are not
available.
TSP's three funds have had different average annual rates of return
since 1987. The C fund has averaged 12.5 percent, a higher return
than the F and G funds' average earnings of about 8.0 percent each
over the 7 years of plan experience. The C and F funds also have
been more volatile than the G fund as shown in figure 1.
Figure 1: Growth of a $1,000
Investment in Each TSP Fund,
1987-1994
(See figure in printed
edition.)
Figure 1 shows that a $1,000 investment in the C fund on January 1,
1987, would grow to $2,452 over the following 7 years based on actual
annual rates of return. Similarly, $1,000 investments in the F and G
funds would grow to $1,836 and $1,868, respectively, over the same
period. The higher returns available from the C fund also connote
the somewhat higher risks inherent in a stock portfolio. Thus, the
retirement income TSP ultimately provides a participant will depend
on how much the individual has contributed and on the rates of return
earned on those contributions. Since returns and risks are related,
the ability to diversify investments among stocks and bonds is an
important factor for participants in a program such as TSP because it
allows them to tailor their investment portfolios to reflect the
level of risk they are willing to assume.
--------------------
\2 Employees under CSRS had the option to transfer to FERS.
\3 A detailed discussion of FERS and CSRS benefits is included in the
Congressional Research Service's report, Federal Civil Service
Retirement: Comparing the Generosity of Federal and Private Sector
Retirement Systems (95-687 EPW, June 5, 1995).
\4 TSP's statute required all CSRS-covered employees' contributions
to be invested in the G fund through 1990. Similarly, all
FERS-covered employees' contributions had to be invested in the G
fund in 1987; this requirement decreased each year by 20 percent
through 1990 and was eliminated in 1991. Also, beginning in 1991,
employees could reallocate their past contributions among the three
funds.
EXTENT TO WHICH FERS-COVERED
EMPLOYEES CONTRIBUTE TO TSP
------------------------------------------------------------ Letter :3
The proportion of FERS-covered employees contributing to TSP has
steadily increased. For example, in September 1987 some 219,000
FERS-covered employees (about 38 percent) were making voluntary
contributions to TSP; whereas, in September 1994 about 942,000 (76
percent) were doing so. However, the degree of voluntary
participation in TSP has varied considerably among salary ranges as
shown in table 2.
Table 2
Percent of FERS-Covered Employees Making
Voluntary Contributions to TSP by Salary
Ranges, 1987-1993
All
$10- $20- $30- $40- $50- $60- salary
Year 19 29 39 49 59 69 $70+ levels
---------------- ------ ------ ------ ------ ------ ------ ------ ------
1987 27% 42% 53% 72% 78% 78% 80% 44%
1988 28 45 63 77 84 85 85 49
1989 30 48 67 78 86 85 86 52
1990 35 54 71 80 85 87 86 57
1991 37 58 75 84 89 89 88 63
1992 41 63 78 87 90 92 90 68
1993 45 69 81 89 93 93 96 73
--------------------------------------------------------------------------------
Source: Latest TSP demographic data available.
Most of the 300,000 (24 percent) FERS-covered employees who did not
make any voluntary contributions were lower-paid workers.
Historically, such employees have been less likely to make voluntary
contributions than have employees in the middle and higher salary
ranges. However, as the table shows, the lower salary ranges have
shown the greater increase over time in the percentage of individuals
who make contributions.
SALARY DEFERRAL RATES HAVE
INCREASED, BUT LOWER-WAGE
EMPLOYEES' RATES REMAIN
LOWER THAN OTHERS
---------------------------------------------------------- Letter :3.1
Overall, in 1993 FERS-covered employees making voluntary
contributions were deferring an average of 5.7 percent of their
salaries compared with 3.7 percent in 1987. The deferral rates
varied from 4.4 percent of their salaries for low-wage employees to
7.2 percent for the highest-wage employees as table 3 shows.
Table 3
FERS-Covered Employees' Overall Deferral
Rates by Salary Ranges, 1987-1993
All
$10- $20- $30- $40- $50- $60- salary
Year 19 29 39 49 59 69 $70+ levels
---------------- ------ ------ ------ ------ ------ ------ ------ ------
1987 3.3% 3.6% 3.8% 4.1% 4.4% 4.7% 5.1% 3.7%
1988 3.9 4.6 5.2 5.7 6.4 6.8 7.3 4.9
1989 4.0 4.7 5.3 5.8 6.5 6.9 7.4 5.0
1990 4.1 4.8 5.4 5.8 6.4 6.8 7.4 5.1
1991 4.1 4.9 5.6 6.1 6.5 6.9 7.3 5.3
1992 4.2 5.1 5.8 6.4 6.7 7.0 7.4 5.6
1993 4.4 5.2 6.0 6.5 6.9 7.2 7.2 5.7
--------------------------------------------------------------------------------
Source: Latest TSP demographic data available.
Also, as with the percentage of employees making contributions,
deferral rates vary among salary groups. The deferral rate among
employees in the lower salary range has also increased the least
compared with the rates of the other employees since 1987--about 27
percent compared with over 50 percent for all but the highest salary
range. The 41-percent increase in the highest salary range may be
partly due to the statutory inflation-adjusted cap on annual
contributions, which was $8,994 for 1993.
TSP CONTRIBUTIONS NEEDED TO
REACH TARGET RETIREMENT
REPLACEMENT RATES
---------------------------------------------------------- Letter :3.2
Our analysis showed a disparity in the extent to which higher- and
lower-paid employees under FERS may need to contribute to TSP to
achieve total FERS retirement benefits that would be commensurate
with their preretirement standard of living. In general, lower-paid
workers may achieve retirement income goals, or total benefits that
are in the range of 60 to 80 percent of final annual earnings, with
minimal TSP deferral rates, while higher-paid workers need to defer
at correspondingly higher rates.
A July 1986 Congressional Research Service report\5 included
illustrative comparisons of the replacement rates\6 under FERS and
CSRS for various retirement assumptions and TSP benefits from (1)
just the mandatory agency 1-percent contribution and (2) employee
voluntary contributions of 5 percent. In our analysis, we updated
the Congressional Research Service's illustration for employees
retiring after 30 years of service at age 62. Our analysis showed
that such employees with earnings in the lower salary ranges might
achieve a level of FERS benefits that would be within 60 to 80
percent of final annual earnings with just their agencies' mandatory
1-percent contribution but that employees in the higher salary ranges
would not. However, using conservative assumptions of TSP returns of
6.1 percent, contributions of 5 percent throughout their careers
would also provide higher-paid employees with an overall FERS
replacement rate within this range as shown in table 4.
Table 4
Illustrative Final Annual Pay
Replacement Rates for FERS-Covered
Employees for Retirement at Age 62 With
30 Years of Service
$20,00 $30,00 $45,00 $75,00 $100,0
FERS income source 0 0 0 0 00
---------------------------------------- ------ ------ ------ ------ ------
FERS annuity\a 32 31 32 31 31
Social Security 35 26 24 17 14
TSP at agency 1%\a 3 2 3 2 2
TSP at 5% match 29 24 25 20 19
================================================================================
Total replacement rate with TSP at 1% 70 59 59 50 47
================================================================================
Total replacement rate with TSP at 5% 96 81 81 68 64
match
--------------------------------------------------------------------------------
Note: We developed this table to illustrate the Social Security
formula's effects as well as TSP's pivotal role in FERS. It should
not be viewed as definitive of specific outcomes. For example, the
replacement rates are greater for service over 30 years.
\a The annuity varies generally because of differences in salary
growth rates and rounding. See appendix I for the details of our
methodology and the assumptions used for this illustration.
Source: GAO analysis.
Again, the disparity in the total replacement rates largely results
from the varying level of benefits that Social Security provides to
individuals in different earnings brackets. As table 4 shows, the
Social Security replacement rate is just 14 percent for an employee
aged 62 with final pretax wages of $100,000 but over twice as high
(35 percent) for someone with final wages of $20,000.
Furthermore, table 4 shows that a 5-percent deferral provides a total
FERS replacement rate for higher-paid workers that is in the lower
end of the range that pension professionals believe is needed (that
is, the 64 and 68 percent shown in table 4). These lower replacement
rates may not reflect such individuals' retirement income goals and,
consequently, these employees would need to contribute more than 5
percent to TSP to achieve a higher level of total FERS benefits.
In general, the lower TSP's investment earnings are the more an
individual would need to contribute in order to reach a certain total
FERS replacement rate goal; conversely, higher TSP returns would
provide individuals with a higher retirement income than they
projected as their goal at a given deferral rate. For example, using
TSP's actual average rate of return of 8.95 percent for the period
1988 to 1994 produces TSP replacement rates that are about 50 percent
higher than those shown in table 4.
--------------------
\5 A Retirement Plan for Federal Workers Covered by Social Security:
An Analysis of the Federal Employees Retirement System (P.L. 99-335)
(86-137 EPW, July 21, 1986).
\6 The replacement rate measures the proportion of preretirement
earnings that is provided by retirement benefits.
EFFECT OF PRERETIREMENT
CONTRIBUTIONS NOT EXPLAINED IN
TSP EDUCATIONAL MATERIALS
------------------------------------------------------------ Letter :4
The TSP Board produces and provides to federal agencies a variety of
educational materials for their employees. Among other things, these
leaflets, pamphlets, and brochures emphasize the monetary benefits of
TSP, such as the advantages of tax deferral, the effects of
compounding, and the higher returns possible from beginning to make
contributions early in one's career. In addition, these materials
inform employees about the pros and cons, including potential risks,
of investing in each of TSP's three funds and the earnings history of
each fund.
However, TSP's educational materials are not explicit in discussing
the importance of employee TSP contributions in achieving total FERS
retirement benefits that would be commensurate with preretirement
living standards, that is, benefits in the range of 60 to 80 percent
of earnings. For example, the materials do not include illustrative
examples of FERS replacement rates at varying TSP deferral rates and
their effect on total FERS benefits. Private sector plans have such
examples in their educational materials. Were TSP's Board to revise
its materials to include that type of example, it would need to do so
in collaboration with the federal Office of Personnel Management
(OPM), which has some responsibility for overall FERS education,
including establishing training programs for agency retirement
counselors.
TSP BOARD TO SEEK TWO
ADDITIONAL INVESTMENT OPTIONS
------------------------------------------------------------ Letter :5
In May 1995, TSP's Board decided to seek legislation that would add
two investment options: an indexed domestic small capitalization
equity fund and an indexed international equity fund. The Board
selected these funds because they add diversity and provide the
opportunity for greater returns than the current options though at
somewhat increased risk. Adding the two funds would make TSP's
number of investment options and mix more like those provided under
private sector section 401(k) plans.
TSP's Board began looking into the possibility of increasing the
number of investment options in 1992 after the statutory restrictions
on C and F fund investments expired. Among other things, the Board
reviewed the investment options generally available under section
401(k) plans and the returns and risks associated with them. On
average, most private sector section 401(k) plans offer four or more
investment options that include a number of bond and stock funds of
varying risk.
The Board's actions to broaden TSP's investment options are
consistent with pension professionals' beliefs that employees should
have a variety of investment options encompassing a range of risks
and returns to provide the opportunity for higher earnings that would
increase their retirement nest eggs. The new options would allow TSP
participants to diversify their investments. The new funds would
complement the C fund, which has historically outperformed the G and
F funds by an average of about 4.5 percentage points since 1987.
Proposed legislation to add the options was introduced in the Senate
on July 27, 1995, and in the House of Representatives on September
12, 1995.
CONCLUSIONS
------------------------------------------------------------ Letter :6
TSP was designed to provide one source of retirement income for
FERS-covered employees. However, unlike the two other FERS
components whose benefits are determined by formula and are constant
for individuals with the same work histories, TSP's benefits will
vary according to the amounts that employees have contributed and the
investment returns on those contributions. Because of the effects of
Social Security's benefit formula, higher-paid workers will be more
dependent on TSP income than lower-paid workers in maintaining their
standard of living in retirement. TSP's educational materials,
however, are not explicit in making this distinction. These
materials should explain and provide examples of contribution rates
and their relationship to preretirement earnings and potential
retirement income.
TSP was also designed to be a retirement savings vehicle for federal
employees that is similar to section 401(k) plans for workers in the
private sector. The addition of the indexed domestic small
capitalization equity and indexed international equity funds will
provide federal employees the same opportunity that those in the
private sector have for tailoring their investment portfolios to
reflect the returns they seek and the risks they are willing to
undertake.
RECOMMENDATION TO THE CONGRESS
------------------------------------------------------------ Letter :7
We recommend that to help ensure that TSP participants have
investment opportunities similar to those available under comparable
private sector plans, the Congress enact legislation adding the two
investment options sought by TSP's Board.
RECOMMENDATION TO THE BOARD
------------------------------------------------------------ Letter :8
We recommend that the Board, in collaboration with OPM, include in
TSP's educational materials (1) an explanation of TSP's pivotal role
in enabling employees under FERS to achieve their retirement income
goals and (2) explicit illustrations of the effects of TSP deferral
rates on total FERS benefits.
AGENCY COMMENTS AND OUR
EVALUATION
------------------------------------------------------------ Letter :9
The Federal Retirement Thrift Investment Board provided written
comments on a draft of this report (see app. II). The Board
disagreed with our recommendation that TSP's educational materials
include an explanation of TSP's role in FERS and explicit examples of
the effect of TSP deferral rates on total FERS benefits. The Board
stated that such actions by TSP would constitute employee education
about FERS, which is an OPM responsibility under the FERS statute.
The Board noted that its educational materials are replete with
illustrations that show the dramatic effect of contributions and
investment earnings on the size of an employee's TSP account.
However, the Board added that the materials do not analyze or explain
the impact that employee TSP accounts will have on total FERS
retirement income because FERS legislation gave that responsibility
to OPM. Also, the Board provided some technical comments that we
incorporated in the report as appropriate.
While OPM has some responsibility for FERS education, such as
establishing training programs for agency retirement counselors, we
do not agree that authority to educate employees on the effects of
TSP investments on their total FERS benefits is vested exclusively in
OPM. We continue to believe that the Board is in a better position
to develop educational materials that include explicit examples of
TSP's potential effects on FERS retirement income. Such examples
would demonstrate TSP's pivotal role in the context of FERS,
particularly given the effect of Social Security's benefit formula.
For example, an OPM booklet on FERS includes examples of replacement
rates for four individuals retiring at various ages, with differing
work histories of federal and nonfederal service, and with TSP
deferral rates of 3 and 5 percent. However, while the examples are
helpful in showing the increased benefits derived from contributions
at 5 percent compared with 3 percent, they are not explicit in
demonstrating TSP's significance in overall FERS benefits at
retirement.
Without its FERS context, we believe the value of TSP's educational
materials to the individual employee is greatly diminished.
Furthermore, TSP is the appropriate source for such information
because it periodically contacts all employees who participate in the
plan--including those not making any voluntary contributions.
Accordingly, we believe that TSP should prepare such educational
materials. OPM officials stated that the Board could do so in
collaboration with OPM.
---------------------------------------------------------- Letter :9.1
As arranged with your office, unless you publicly announce the
contents of this report earlier, we plan no further distribution
until 5 days after its issue date. At that time, we will send copies
of this report to other congressional committees and members with an
interest in this matter and to others upon request.
Our review was performed under the direction of Donald C. Snyder,
Assistant Director. Other contributors were Endel P. Kaseoru,
Evaluator-in-Charge, and evaluators Carolina M. Morgan and Gregory
Curtis. If you or your staff have any questions about this report,
please call me on (202) 512-7215 or Mr. Snyder on (202) 512-7204.
Jane L. Ross
Director, Income Security Issues
METHODOLOGY USED TO CALCULATE FERS
REPLACEMENT RATES
=========================================================== Appendix I
We calculated illustrative FERS replacement rates for each of the
program's three components--the basic FERS annuity, Social Security
benefits, and TSP--for employees retiring with 30 years of service at
age 62, the average federal retirement age in 1994 for regular
retirements. To make our calculations, we simulated the salary
histories of five hypothetical federal employees and estimated the
annuities they would receive under certain assumptions. The time
frame for our analysis was 1986 through 2015.
To produce the salary histories for our model, we used wage growth
rates that are consistent with federal General Schedule salaries.
The workers in our model began their federal careers in 1986 at
entry-level salaries for GS-2, -3, -5, -7, and -9 and retired in
January 2016 at age 62 with final annual salaries, as measured in
1995 dollars, of $20,000, $30,000, $45,000, $75,000, and $100,000.
We first created an inflation-adjusted earnings history for these
workers and then converted it to current year earnings using the
actual inflation rates from 1986 to 1995 and 3.4 percent thereafter.
To determine employees' FERS annuities, we used the basic FERS
annuity formula in the law. However, while the formula computes the
benefit at 33 percent of the average of the highest 3 consecutive
years' salaries, the replacement rate is less than 33 percent because
the estimated wages grow in each of the 3 years prior to retirement;
thus, the 3-year average used to calculate the annuity is lower than
the final year's wages. To calculate Social Security benefits, we
used the "ANYPIA" software program provided by the Social Security
Administration's Office of the Actuary. In applying this program, we
used the alternative I assumptions of future economic activity from
the 1994 report of the Board of Trustees of the Federal Old Age and
Survivors Insurance and Disability Insurance Trust Funds. The
alternative I assumptions are conservative, and thus they produced
replacement rates that were lower by 1 to 5 percentage points than
the rates produced by alternatives II and III.
To calculate the TSP replacement rates, we estimated the balance in
the individuals' accounts at retirement based on employee and agency
contributions of 5 percent each and the agency-only 1-percent
contribution. For our baseline analysis, we assumed that the
accounts earned a conservative return of 6.1 percent, the same rate
the Congressional Research Service used in its analysis. We also
calculated replacement rates using a weighted average of actual TSP
returns from 1988 to 1994 of 8.95 percent. This higher annual rate
of return produced TSP replacement rates that were about 50 percent
higher for each salary level. We then calculated an annuity for each
account balance using a worksheet in TSP's annuities booklet.\7 We
assumed an increasing single life annuity at 6-percent interest, the
rate used in TSP's worksheet. The replacement rates we computed,
shown in table 4, vary by final year wage because each had a
different growth rate over the 30 years we modeled. We also tested
different rates of wage growth, returns on TSP, and the FERS annuity
and found the results were consistent across the five final salaries
we modeled.
(See figure in printed edition.)APPENDIX II
--------------------
\7 Thrift Savings Plan Annuities, Federal Retirement Thrift
Investment Board (Washington, D.C.: Jan. 1994).
COMMENTS FROM THE FEDERAL
RETIREMENT THRIFT INVESTMENT BOARD
=========================================================== Appendix I
(See figure in printed edition.)
(See figure in printed edition.)
(See figure in printed edition.)