oversight

Federal Retirement: Comparison of High-3, 4, and 5 Salary Factors

Published by the Government Accountability Office on 1997-04-25.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

United States
General Accounting Office
‘Wdtington, D.C. 20548

General    Government   Division


B-276730

April 25, 1997

The Honorable John L. Mica
Chairman, Subcommittee on Civil Service
Committee on Government Reform and Oversight
House of Representatives

Subject:     Federal Retirement:   Comnarison   of High-3. 4. and 5 Salarv Factors

Dear Mr. Chairman:
 .
This letter responds to your February 20, 1997, request for information on the
effects of changing the high-3 salary factor in the formulas that are currently
used to compute Civil Service Retirement System (CSRS) and Federal
Employees Retirement System (FERS) pension benefits. Congressional
consideration of modifications to high 3 has led to the impression that
employees would need to work a number of years longer in order to earn
annuities under high 4 or high 5 that would be comparable to their high-3
annuities. Our objective was to determine how much longer retiring employees
would need to work to earn basic annuities under a high 4 or a high 5 that
would be comparable to the annuities they would have received under a high 3,
had they retired before the computation factor were changed.’

To meet this objective, we created illustrative CSRS and FERS general schedule
employees who had met the age and service requirements for retirement and
computed the basic annuities they would have received under the high-3, high-4,
and high-5 average salaries. In making these computations, we took into
account that the basic annuity a retiring employee receives is a function of the
employee’s high-average salary, creditable service, and the applicable accrual
rate. An accrual rate is an annual benefit percentage that is multiplied by a
high-average salary in computing an annuity. CSRS and FERS accrual rates
differ. The CSRS accrual rate provides greater benefits because CSRS is a
stand-alone pension program, while the FERS program includes pension
benefits from Social Security and the Thrift Savings Plan, in addition to a basic



‘In this report we use the term basic annuity to refer to the annuity a retiring
employee would receive before any reductions for the retiree’s share of survivor
and/or health benefits.
                                       GAO/GGD-97-84R High 3,4, and 5 Salary Factors
B-276730


annuity.” In computing CSRS and FERS annuities, we developed salary histories
for general schedule employees that we believed generally would capture the
least and most number of months retiring employees might have to work under
a high 4 or a high 5 to earn annuities comparable to those they would have
received under a high 3, had they retired before any change went into effect.
These salary histories were for two groups of employees at five different grade
levels. The first group included employees with rising salary histories (i.e.,
those who, over their last 5 years of service, had promotions and/or in-grade
step increases). The second group included employees with stable salary
histories (i.e., those who, over their last 5 years of service, received the
maximum salary for their grade level).

We did our work in Washington, D-C., between February and April 1997 in
accordance with generally accepted government auditing standards. A more
detailed description of the assumptions and methods used in our analysis is
provided in enclosure I to this letter.

RESULTS

Employees retiring under either CSRS or FERS would need to work longer to
receive annuities under a high 4 or a high 5 that would be comparable to the
annuities they would have received under a high 3, before any change in the
annuity computation factor.3 The amount of extra time, however, is measured
in months rather than years. Reasons why include the fact that an employee’s


“The formula for CSRS annuities is: (a) 1.5 percent x high-3 average salary x
the first 5 years of service, plus (b) 1.75 percent x high-3 average salary x the
next 5 years of service, plus (c) 2.0 percent x high-3 average salary x all service
over 10 years. For example, an employee retiring at age 60 with exactly 20
years of service would receive a basic annuity of 36.25 percent of his or her
high-3 average salary. The accrual rate for FERS annuities is either 1.1 percent
x high-3 average salary x years and months of service, if the employee is at
least age 62, and has 20 or more years of service or 1.0 percent x high-3 average
salary x years and months of service, if the employee is under age 62 or age 62
 or older and has less than 20 years of service. For example, an employee
 retiring at age 62 with 20 years of service would receive a basic annuity of 22
 percent of his or her high-3 average salary.

 3As discussed in greater detail in enclosure I, we used the term comparable to
 mean an annuity that under a high 4 or a high 5 had become equal to or slightly
 larger than the high-3 annuity.

 2                                      GAOIGGD-97-34B   High 3,4, and 5 Salary Factors
B-276730


pay normally increases when he or she works longer, thus, so does the
employee’s annuity at retirement. Employees who work into the next calendar
year in order to earn comparable annuities can receive general schedule pay
increases early in the calendar year as well as step increases. In addition, the
extra time employees work is added to their years of creditable service, which
also increases the value of their annuities at retirement. For our illustrative
employees, under a high 4, the least amount of extra time a retiring employee
would need to work would be about 3 months, the most about 5. Under a high
5, the least number of extra months would be about 5, the most about 9.
Tables 1 and 2 show these results in greater detail and provide comparisons of
our illustrative employees. In the tables, employees with rising salary histories
are those having service that culminated at step 3 or 4 of the grade; employees
with stable salary histories are those having service that culminated at step 10
of the grade.

Table 1 shows that if a high 4 were used there would be a small difference in
the length of time employees would need to work to eaSn annuities comparable
to their high-3 annuities. The least number of extra months would be 3, the
most 5. Table 1 also shows that, overall, those employees with rising salary
histories would need to work somewhat longer than those with stable salary
histories to earn annuities comparable to their high-3 annuities. For example,
CSRS employees with 30 years of service and stable salary histories would need
to work an extra 4 months, compared to 5 months for those with rising salary
histories.  Similarly, FERS employees with 15 years of service and stable salary
histories would need to work an extra 3 months, compared to 4 months for
those with rising salary histories.




3                                      GAO/GGD-97-84R   High 3,4, and 5 Salary Factors
B-276730


Table 1: Additional Number of Months Emnlovees at Selected Grade/Stex,
Levels Would Need to Work Under a High 4 to Earn Annuities That Would Be
Comnarable to Those Available If a High-3 Formula Were Used to Calculate
Basic Retirement Annuities


     ‘ension Plan/
     Years of Service        3 Months           4 Months                  5 Months
     XRS with 30                          GS-5/10                    GS-5/4
     fears of service                     G S-7/1 0                  GS-713
                                          GS-9/10                    GS-913
                                          GS-1200                    GS-1213
                                          GS-1 5/l 0                 GS-1513
     CSRS with 20       GS-5110            GS-514                    GS-713
     years of service   GS-7/10            GS-913
                        GS-9/10            GS-12l3
                        GS-12/10           GS-1513
                        GS-15/10
     FERS with 20       GS-5110            G S-514                   GS-713
     years of service   GS-7/10.           GS-12/3                   G S-913
                        GS-g/IO            GS-1513
                        GS-12/10
                        GS-15/10
     FERS with 15       GS-5/10            GS-5/4
     years of service   GS-7/10            GS-713
                        GS-9/10            GS-913
                        GS-12/i 0          GS-I 2l3
                        GS-I 5/l 0         GS-1513

 Source: GAO calculations.

 Table 2 shows that if a high 5 were used there would be a larger difference in
 the length of time employees would need to work to earn annuities comparable
 to their high-3 annuities. They would need to work between 5 and 9 extra
 months. Also, under a high 5, employees with rising salary histories would
 continue to need to work longer than those with stable salary histories to earn
 annuities comparable to their high-3 annuities. However, the effects of salary
 history would become more pronounced under a high 5. For example, table 2
 shows that CSRS employees with 30 years of service and stable salary histories

 4                                      GAO/GGD-97-34R   High 3,4, and 5 Salary Factors
B-276730


would need to work about 7 extra months, compared to 9 extra months for
those with rising salary histories. Similarly, FERS employees with 20 years of
service and stable salary histories would need to work an extra 5 months, while
those with rising sa.lary histories would need to work as many as 9 extra
months.

Table 2: Additional Number of Months Emnlovees at Selected Grade&ten
Levels Would Need to Work Under a High 5 to Earn Annuities That Would Be
Comnarable to Those Available If a High-3 Formula Were Used to Calculate
Basic Retirement Annuities


 Pension
 Plan/Years
 of Service        5 Months        7 Months             8 Months             9 Months
 CSRS with                       GS-5/10                                  GS-514
 30 years of                     GS-700                                   GS-7/3
 service                         GS-9/10                                  GS-9/3
                                 GS-12/10                                 GS-12/3
                                 GS-15/10                                 GS-15/3
 CSRS with      GS-5/10                             GS-5/4
 20 years of    GS-7/10                             GS-713
 service        GS-9/10                             GS-913
                GS-12/10                            GS-1213
                GS-15/10                            GS-1513
 FERS with      GS-5/10                             GS-514                GS-7f3
 20 years of    GS-7110                             G S-913               GS-15/3
 service        GS-9/10                             GS-I 2/3
                GS-12/10
                GS-15/10
 FERS with      GS-5/10          GS-5/4             G S-7/3
 15 years of    GS-7/10          GS-12/3            GS-9/3
 service        GS-9/10          GS-15/3
                GS-12/i 0
                GS-15/10




                                       GAO/GGD-97-84R   High 3,4, and 5 Salary Factors
B-276730


AGENCY COMMENTS

We shared a draft of this letter with the Director of the Office of Personnel
Management (OPM) for review and comment. In a letter dated April 11, 1997,
the Director agreed that the scenarios presented in the letter accurately show
how CSRS and FERS benefits would be reduced if an average salary change
were implemented. He also stated that the Administration has not sought
legislation to change benefit computations because it favors stability in long-
term benefit programs for federal employees. OPM’s written comments are
reprinted as enclosure II to this letter.



As we arranged with the Subcommittee, unless you publicly announce this
letter’s contents earlier, we plan no further distribution of it until 30 days after
the date of this letter. We will then send copies to the Director of OPM, the
Ranking Minority .Member of your Subcommittee, and the Chairman and
Ranking Minority Member of the Subcommittee on International Security,
Proliferation, and Federal Services, Senate Committee on Governmental Affairs.
We will also make copies available to others upon request.

The major contributors to this letter were Margaret ,Wrightson and Tyra
DiPalma. Please contact me on (202) 512-9039 if you have any questions
regarding this letter.

Sincerely yours,




Michael Brostek
Associate Director
Federal Management
   and Workforce Issues

 Enclosures - 2




                                          GAOIGGD-97-84R   High 3,4, and 5 Salary Factors
ENCLOSURE I                                                                ENCLOSURE I


SCOPE AND METHODOLOGY

To determine how much longer employees would need to work to receive
annuities under a high 4 or a high 5 that would be comparable to the annuities
they would have received under a high 3, we created salary histories at five
different grade levels and used them to compute annuities under each of the
three high-average salaries. In creating the sa3ary histories, we used a rising
step level (step 2 or 3) as well as a stable step level (step 10). Table 3 shows
the starting and ending points of these histories.’ Specifically, we created salary
histories for employees retiring in December 1996 under a high 3 at grades 5, 7,
9, 12, and 15 with step levels of either 2 or 3, and 1O.5 We also calculated
annuities using these salary histories under a high 4 and a high 5. The salary
histories were created using the actual general schedule rates for the years 1992
through 1997.6 Our analysis included 1997 because the illustrative employees
would have had to work for a period of months into the future (i.e., into 1997)
to earn annuities under a high 4 or a high 5 that would be comparable to the
high-3 annuities they would have received, had they retired in December 1996.
We selected the various grade and step levels because logically these histories
would provide the greatest contrast in the number of additional months retiring
general schedule employees might have to work. They were logical because in
averaging the salaries an employee might earn over a 3, 4, or 5-year period, the
reduction in his or her initial annuity from shifting to a high 4 or a high 5 would
be the smallest when the averaged salaries are equal-except for any general pay
adjustments. The reductions in initial annuities would be larger when the
salaries to be averaged increased each year-over and above any general pay
adjustments. Other factors equal, the larger the increases in salary over the
averaging period, the greater the reduction in the initial annuity.




4While step 2 was one of the step-level starting points, it was not an ending
point for any of the illustrative employees. In our analysis, none of the
employee’s service culminated at step-level 2 because of the way in which
actual step increases are timed.

5The rising step-level histories included a promotion    in 1995 from grades 4, 5, 7,
11, and 14, respectively.

‘For the years 1994 through 1997, the general schedule rates used incorporated
the locality pay area of Washington and Baltimore.

7                                       GAOIGGD-97-84R   High 3,4, and 5 Salary Factors
ENCLOSURE I                                                                ENCLOSURE I


Table 3: Grade and Steu Levels in 1994, 1995. and 1996


                                                      Grade/Step If Retired in
                                                             ber 1996 Under




IIGS-7/lO           I Not applicable                 I GS-7/10
IIGS-9110           I Not applicable                 I GS-g/10
lr   GS-12/10       I Not applicable                 I GS-12/10
     GS-15/10         Not applicable                      GS-15/10

Source: GAO calculations.

We created all salary histories under the assumption that the employees had
met certain age eligibility requirements, under CSRS (age 55 or 60) and FERS
(age 62).7 We calculated CSRS retirement annuities for the various grade and
step levels, assuming that the employees had either 20 or 30 years of service at
the time of their retirement. We calculated FERS retirement annuities under
the assumption that the employees retired at age 62 with either 15 or 20 years
of service.* Based on the salary histories, we calculated the annuities for


 7The FERS annuity formula applies to the basic annuity (e.g., defined benefit)
 portion administered by OPM. It does not include the Social Security or Thrift
 Savings Plan portions.

 *The FERS retirement formula is different depending on whether the employee
 is at least 62 years of age and has 20 years or more of creditable service at the
 time of separation. We used 15 years in the FERS formula for employees
 retiring with less than 20 years of service. Had we chosen another number
 below 20, the amounts of the FERS annuities we computed would have been

 8                                       GAO/GGD-97-84R    High 3,4, and 5 Salary Factors
ENCLOSURE I                                                                 ENCLOSURE I


employees retiring in December 1996 under a high 3 at the various grade and
step levels described above. In addition, we calculated the annuities for each
month thereafter under a high 4 and a high 5 until the basic annuities produced
by our calculations were comparable with the employee’s high-3 annuity of
December 1996. These high-4 and high-5 calculations showed the additional
length of time employees would need to work to receive an annuity comparable
to the one they would have received under a high 3. In this letter, we used the
term comparable to mean an annuity that under a high 4 or a high 5 had
become equal to or slightly larger than the high-3 annuity. Had we defined
comparable to mean an annuity slightly smaller than the high-3 annuity, our
estimate of the number of extra months employees would have needed to work
generally would have dropped by 1 month.

OPM’s Chief, Retirement Policy Division, agreed that our methodology was
reasonable and that it correctly demonstrated the range of extra months retiring
employees potentially would need to work under a high 4 or a high 5 to earn
annuities comparable to the ones they would have received under a high 3, had
they retired before the change. OPM also told us that most federal employees
retire with stable rather than rising salary histories.




different. However, the proportional reductions in initial annuities resulting
from a shift to a high 4 or a high 5 would have been similar to the ones we
computed for 15 years of service.

9                                       GAO/GGD-97-84R   High   3,4, and 5 Salary Factors
ENCLOSUREII                                                                          ENCLOSURE         II




                                            UNITED   STATES
                                OFFICE   OF PERSONNEL    MANAGEMENT
                                         WASHINGTON,DC 20415.0001

OFFICE OFTHE DIRECTOR




      Mr. Michael Brostek, AssociateDirector
      FederalManagementand Workforce Issues
      United StatesGeneralAccountingOffice
      GeneralGovernmentDivision
      Washington,D.C. 20548

      Dear Mr. Brostek:
      Thank you for the opportunityto review the draft correspondence
                                                                    on the comparisonof
      retirementbenefitsunderthe Civil ServiceRetirementSystem(CSRS)and the Federal
      EmployeesRetirementSystem(FERS)using an averageof the highest4 and 5 years of
      salary rather than the current highest3 yearsmethod.

      The scenariospresentedin the correspondence    accuratelyshow how CSRS and FERS
      benefitswould be reducedif an averagesalary changewere implemented. Reflecting our
      strong views in favor of stability in theselong-termbenefit programsfor Federalemployees.
      the Administrationhas not soughtlegislationto changebenefit computationsthat employees
      have reasonablyexpectedas a Governmentcommitment.

      Again, thank you for the opportunityto provide comments.


                                                         Sincerely,          r




 (410121)


 10                                               GAO/GGD-97-84R      High 3,4, and 5 Salary Factors
Ordering      Information

The first copy of each GAO,report     and testimony is free.
Additional   copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order
made out to the Superintendent     of Documents, when
necessary. VISA and MasterCard      credit cards are accepted, also.
Orders for 100 or more copies to be mailed to a single address
are discounted 25 percent.

Orders by mail:

U.S. General Accounting Office
P.O. Box 6015
Gaithersburg, MD 20884-6015

or visit:

Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting  Offrce
Washington, DC

Orders may also be placed by calling (202) 512-6000
or by using fax number (301) 258-4066, or TDD (301) 413-0006.

Each day, GAO issues a list of newly available reports and
testimony.   To receive facsimile copies of the daily list or any
list from the past 30 days, please call (202) 512-6000 using a
touchtone phone. A recorded menu will provide information         on   ,
how to obtain these lists.

For information on how to access GAO reports on the INTERNET,
send an e-mail message with~“info” in the body to:

info@www.gao.gov

or visit    GAO’s World     Wide Web Home Page at:

httpY/www.gao.gov




PRINTED ON G;&     RECYCLED   PAPER
United States
General Accounting 0ffice
Washington, D.C. 20548-0001

Ofiicial Business
Penalty for Private Use $300
Address Correction   Requested