oversight

Federal Employees' Compensation Act: Benefits for Retirement-Age Beneficiaries

Published by the Government Accountability Office on 2012-02-06.

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

United States Government Accountability Office
Washington, DC 20548




         February 6, 2012

         Congressional Committees

         Subject: Federal Employees’ Compensation Act: Benefits for Retirement-Age Beneficiaries

         In 2010, the Federal Employees’ Compensation Act (FECA) 1 program paid approximately $2.8 billion
         in total cash and medical benefits to federal employees who sustained injuries or illnesses while
         performing federal duties; about $1.9 billion of that was for cash benefits. 2 U.S. Postal Service
         (USPS) has the largest number of FECA beneficiaries. The Department of Labor (Labor), which
         oversees the program, categorizes FECA beneficiaries into groups based on their ability to work,
         length of time receiving benefits, and type of injury. There are some beneficiaries who have been
         receiving benefits for longer than 90 days, and are completely unable to work. We refer to this group
         as long-term, full-time beneficiaries. Because there are no time or age limits for receiving FECA
         benefits, long-term, full-time beneficiaries include people at or older than retirement age.

         We examined (1) the characteristics and associated compensation costs of long-term, full-time FECA
         beneficiaries, for USPS and non-USPS employees; (2) how wage compensation benefits for
         retirement-age, long-term, full-time FECA beneficiaries compare with federal retirees’ annuities (not
         including USPS employees); and (3) the experiences of states that limit state workers’ compensation
         benefits for workers at retirement age. For the first question, we could include USPS employees
         because Labor provided us with data on FECA beneficiaries for all federal agencies, including USPS.
         However, we could not include USPS employees in answering the second question. Our analysis
         required determining the work histories for FECA beneficiaries and retired annuitants, using an Office
         of Personnel Management (OPM) database to obtain these work histories. USPS employees are not
         included in the OPM database. 3 We subsequently obtained data from USPS; however, the data were
         missing a significant amount of information necessary to determine the work histories of USPS




         1
             Codified at 5 U.S.C.§ 8101 et seq.
         2
          Cash benefits are paid for lost wages, loss of, or loss of use of, a body part or function (schedule awards), and death of an
         employee as a result of a workplace injury (survivor benefits). Schedule awards and survivor benefits can be in addition to
         any benefits payable for lost wages. The total for cash benefits includes amounts for long-term, full-time benefits; short-
         term benefits; part-time benefits; schedule awards; and survivor benefits.
         3
          USPS employees are not included in OPM’s Central Personnel Data File (CPDF) because USPS does not report
         personnel data to OPM.




                                                                                   GAO-12-309R Federal Employees’ Compensation Act
employees for the years covered by our analysis. 4 According to USPS officials, they changed data
systems in 1995 and some key data were not available.

To determine characteristics and compensation costs of long-term, full-time FECA beneficiaries, we
analyzed data from the Department of Labor’s FECA claimant database for chargeback year 2010. 5
(References to FECA data in 2010 from this point forward refer to the chargeback year.) We
conducted this analysis for these FECA beneficiaries, using their full retirement age as set forth in
the Social Security Act. 6 To compare FECA benefit levels with federal retirees’ annuities, we
conducted an analysis of retirees who were at least 55 years old and long-term, full-time FECA
beneficiaries of the same age who did not work for the USPS who were covered by the Civil Service
Retirement System (CSRS) in 2010. 7 We focused on this age group of FECA beneficiaries because
the minimum age of retirement eligibility under CSRS is 55 years old and using the lower age
increased the number of individuals we could include in our analysis. We compared the FECA
benefits to the annuities of federal retirees who were similar in terms of demographics, agencies they
worked for, occupations, retirement plans, incomes prior to injury or retirement, and years of federal
service at the point of the FECA beneficiaries’ date of injury. The analysis did not distinguish
between FECA beneficiaries who have no dependents, and received 66-2/3 percent of their preinjury
salaries and those beneficiaries with dependents who received 75 percent of their preinjury salaries.
Because FECA benefits are not taxed, and in order to make those benefits comparable to annuities
that are taxed, our analysis accounted for the effects of federal and state income taxes on retirees’
annuities. 8 Our comparison did not include individuals receiving part-time Loss of Wage Earning
Capacity (LWEC) benefits. These individuals do not receive full disability benefits from FECA,
because it has been determined that they are capable of working part-time or with restrictions. 9 For
our review of the state workers’ compensation programs, we conducted case studies in four states
that limit these benefits by retirement age—Kentucky, Minnesota, Montana, and Tennessee—out of
a total of eight states that have age-based workers’ compensation limits, as identified by the
Workers’ Compensation Research Institute. We interviewed workers’ compensation board officials
and private sector attorneys who specialize in workers’ compensation about their experiences in
managing these programs, specifically relating to retirement-age limitations.


4
 For example, we needed a service computation date to determine the length of service for annuitants. However, the
service computation data we received from USPS for the period 1988 to 2010 was missing these data for 16 to 90 percent
of employees, depending on the year.
5
 FECA benefits are paid out of the Employees’ Compensation Fund, and most are charged back to the employees’ agency.
Labor’s chargeback year for FECA agency billing purposes ends June 30.
6
 The age at which an individual can receive full retirement benefits under the Social Security Act (known as the “full
retirement age” or the “normal retirement age,” or NRA) ranges from 65 to 67, depending on the individual’s year of birth.
For those born in 1937 and earlier, NRA is 65; for those born in years 1943-1954, the NRA is 66; for those born in 1960 and
later, the NRA is 67; for years of birth in between those just cited, the NRA is interpolated by 2 months per year of birth.
See http://www.ssa.gov/oact/progdata/nra.html.
7
 We included only FECA recipients who were covered by CSRS because the majority of annuitants currently receiving
annuities are covered by CSRS. According to OPM, approximately 81 percent of fiscal year 2010 employee annuitants
were CSRS annuitants.
8
 We used the National Bureau of Economic Research’s (NBER) TAXSIM to determine federal and state income taxes for
annuitants. TAXSIM is NBER's FORTRAN program for calculating liabilities under U.S. federal and state income tax laws
from individual data. The TAXSIM Model (http://www.nber.org/taxsim) simulates the U.S. federal and state income tax
rules. See Daniel Richard Feenberg and Elizabeth Coutts, “An Introduction to the TAXSIM Model,” Journal of Policy
Analysis and Management, vol. 12, no. 1(winter, 1993), 189-194.
9
In general, part-time benefits are determined by calculating the difference between the employee’s preinjury and postinjury
monthly wages.




Page 2                                                                   GAO-12-309R Federal Employees’ Compensation Act
Enclosures I and II contain a detailed description of our scope and methodology.

We requested comments on a draft of this report from Labor, OPM, and USPS. The agencies’
comments are discussed at the end of this report.

We conducted this performance audit from February 2011 through February 2012 in accordance with
generally accepted government auditing standards. Those standards require that we plan and
perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe that the evidence obtained
provides a reasonable basis for our findings and conclusions based on our audit objectives.


Background
Labor’s Division of Federal Employees’ Compensation in the Office of Workers’ Compensation
Programs (OWCP) administers the FECA program. This program provides cash benefits and
medical benefits to federal employees who suffer temporary or permanent disabilities resulting from
work-related injuries or diseases. FECA cash benefits include payments for wages lost when
employees cannot work because of work-related disabilities due to traumatic injuries or occupational
diseases; schedule awards for loss of, or loss of use of, a body part or function; death benefits for
survivors; and burial allowances. Medical benefits include vocational rehabilitation and medical care
for injured workers. FECA cash benefits are not taxed.

OWCP charges agencies for whom injured employees worked for benefits provided. These agencies
subsequently reimburse Labor’s Employees’ Compensation Fund from their next annual
appropriation. USPS and the Department of Veterans Affairs have the largest number of FECA
beneficiaries, as shown in table 1. 10




10
   USPS receives no annual appropriations for purposes other than revenue forgone on free and reduced rate mail. USPS
generates revenue through the sale of postage and postal-related products and services and borrows money from the U.S.
Treasury via the Federal Financing Bank. It has relied increasingly on this debt to fund its operations. According to USPS,
it paid $1.3 billion for workers’ compensation costs in 2010.




Page 3                                                                   GAO-12-309R Federal Employees’ Compensation Act
Table 1: Number of FECA Beneficiaries per Agency in 2010

    Agency                                                                                    All FECA beneficiaries
                                                                                   Number                                  Percentage
    USPS                                                                            130,483                                         43
    Department of Veterans Affairs                                                   26,157                                           9
    Department of Homeland Security                                                  25,408                                           8
    Navy                                                                             19,919                                           6
    Army                                                                             19,852                                           6
    Air Force                                                                        12,728                                           4
    Department of Justice                                                            11,001                                           4
    Department of Agriculture                                                        10,691                                           3
    Department of the Interior                                                        9,205                                           3
    Defense agenciesa                                                                 6,101                                           2
                       b
    Other agencies                                                                   35,360                                         12
    Total                                                                           306,905                                         100
Source: GAO analysis of Labor data.
a
Defense Agencies covered include the Defense Contract Audit Agency, Defense Logistics Agency, and the Defense Contract
Management Agency, among others.
b
 The remaining agencies listed each have less than 2 percent of the total number of beneficiaries receiving workers’ compensation
benefits, and fewer than 5,275 beneficiaries each.


Eligible workers with temporary or permanent total disabilities who have no spouse or dependent
generally receive wage-loss compensation equal to 66-2/3 percent of their salary at injury. Those
with a spouse or dependent receive 75 percent. OWCP can reduce wage-loss compensation based
on employees’ wage-earning capacities when Labor determines they are capable of working again.
OWCP provides wage-loss compensation until claimants can return to work in either their original
positions or other suitable positions that meet medical work restrictions.

Claimants are not allowed to receive FECA benefits at the same time they receive certain other
federal disability or retirement benefits. 11 For example, claimants are not allowed to receive both
FECA wage-loss-compensation payments and disability payments from the Department of Veterans
Affairs for the same injury. Further, claimants cannot receive federal retirement benefits, such as
CSRS or Federal Employees’ Retirement System (FERS) benefits, paid through OPM concurrently
with FECA wage-loss benefits; they must elect to receive one or the other. However, the law
authorizing FECA does not require beneficiaries to “retire” at a certain age by, for example,
transitioning to a federal pension program like CSRS. They can continue receiving FECA wage-loss
compensation payments for as long as they remain unable to work due to a workplace injury.




11
  Employees eligible for FECA benefits could also be eligible for retirement disability benefits from OPM or Social Security
Disability Insurance benefits. Depending on which benefits employees are entitled to, employees might have to make an
election between them. In many cases in which individuals receive benefits from different programs simultaneously, one
benefit would likely be offset against the other to some extent.




Page 4                                                                           GAO-12-309R Federal Employees’ Compensation Act
Typically, federal workers participate in one of two retirement systems, which are administered by
OPM: CSRS or FERS. 12 CSRS covers most civilian federal employees who were hired before 1984.
Under CSRS, employees generally do not pay Social Security taxes or earn Social Security benefits.
Federal employees first hired in 1984 or later are covered by FERS. 13

Concerns that beneficiaries remain in the FECA program past traditional retirement age have led to
two types of proposals to change the program. As we have discussed in previous work, one type of
proposal would convert FECA benefits to regular federal employee retirement benefits at retirement
age. A second type of proposal would convert FECA benefits to a retirement-age FECA benefit. 14

States also administer individual state workers’ compensation programs for injured workers, but
FECA and the state programs differ in important ways. For example, the FECA program provides
coverage for injured workers from one fund, administered by Labor. In contrast, in states, multiple
private insurers generally pay most claims, and state programs typically provide coverage for both
private and public sector workers. In addition, state compensation awards may often be appealed to
state courts. In contrast, FECA is intended to be nonadversarial and remedial in nature. 15


Summary
In 2010, 31,880—or 10 percent—of all FECA beneficiaries were long-term, full-time beneficiaries and
10,873 of those—or 34 percent—were at full retirement age, as defined under the Social Security
Act. Of the $1.9 billion total in cash benefits paid to FECA beneficiaries, over half (58 percent) went
to long-term, full-time beneficiaries. Of that half, long-term, full-time beneficiaries at or above full
Social Security retirement age received 21 percent. This analysis covered all FECA beneficiaries,
including USPS and non-USPS employees.

Compared to their federal CSRS retired counterparts, non-USPS long-term, full-time FECA
beneficiaries typically received higher benefits in 2010. The median annual FECA benefit of $35,614
was about 26 percent higher than the median annual annuity received by retirees, which was
$28,289, after adjusting for the effects of taxes. The difference between FECA benefits and CSRS
annuities is typically larger when the FECA beneficiary was injured after fewer years of service. The
differences between annual FECA and CSRS benefits in our comparison are largely explained by the
benefit calculation formulas used for each set of benefits. The CSRS formula generally awards a
smaller percentage of salary than the FECA formula for most workers, except those with long tenure.



12
  Under both CSRS and FERS, the date of employees’ eligibility to retire with an annuity depends on their age and years of
service. The amount of the retirement annuity is determined by three factors: the number of years of service, the accrual
rate at which benefits are earned for each year of service, and the salary base to which the accrual rate is applied. In both
CSRS and FERS, the salary base is the average of the highest 3 consecutive years of basic pay. This is often called “high-
3” pay. Under both systems, a worker with at least 30 years of service can retire at the age of 55 (for FERS, this is true if
the worker was born before 1948).
13
  All federal employees who are enrolled in FERS pay Social Security taxes and earn Social Security benefits. Federal
employees enrolled in either CSRS or FERS also may contribute to the federal 401(k)-like program, the Thrift Savings Plan
(TSP); however, only employees enrolled in FERS are eligible for employer matching contributions to the TSP.
14
   GAO, Federal Workers’ Compensation: Questions to Consider in Changing Benefits for Older Beneficiaries,
GAO-11-854T (Washington, D.C.: July 26, 2011); and Federal Employees' Compensation Act: Issues Associated With
Changing Benefits for Older Beneficiaries, GAO/GGD-96-138BR (Washington, D.C.: Aug. 14, 1996).
15
 A federal employee or surviving dependent is not entitled to sue the United States or recover damages for such injury or
death under any other law for a work injury.




Page 5                                                                    GAO-12-309R Federal Employees’ Compensation Act
It is important to note that our finding regarding the difference in benefits levels does not allow us to
conclude whether such a difference exists for USPS employees or for current or future annuitants
under FERS, a population that is increasing given that FERS covers federal employees first hired in
1984 or later. USPS employees were not included in this analysis because USPS could not provide
sufficient data to reliably determine its employees’ work histories for the years covered by our
analysis.

We examined the experiences of four states that limit state workers’ compensation benefits based on
retirement age: Kentucky, Minnesota, Montana, and Tennessee. State officials and attorneys in
those states highlighted several aspects of their experiences with these provisions, including cost
savings, legal challenges, and financial hardships for some beneficiaries. For example, a workers’
compensation board official from Kentucky stated there has been a decrease in workers’
compensation costs since the retirement-age limitation went into effect; however, their office could
not attribute the savings to this provision because they lack statistical data since private insurance
carriers pay the benefits rather than the state.


In 2010, Long-Term, Full-Time FECA Beneficiaries, Many at Retirement Age,
Received More Than Half of All FECA Cash Benefits
A sizable portion of FECA beneficiaries who received cash benefits 16 are long-term, full-time
beneficiaries. 17 Many of these were at or above full Social Security retirement age. Of the 306,905
individuals who received FECA benefits in 2010, about 79,261 received cash benefits (see fig. 1). 18
Somewhat less than half (31,880) of these recipients were classified as long-term, full-time
beneficiaries. Among long-term, full-time beneficiaries, about 34 percent (10,873) were at retirement
age and most were injured more than 20 years ago. 19 This analysis covered all FECA beneficiaries,
including USPS and non-USPS employees.




16
 In 2010, of the 79,261 people who received cash benefits, 24,549 received short-term benefits, 10,594 received part-time
benefits, 7,596 received schedule awards, and 4,642 received survivor benefits.
17
 This group of long-term, full-time beneficiaries does not include those individuals receiving part-time benefits, schedule
awards, and survivor benefits.
18
  Because of the way Labor maintains its data, it is not possible to know whether an individual whose case closed in 2010
did or did not receive cash benefits. Cases may be closed for various reasons, including re-employment or because an
individual has recovered from their work-related injury.
19
   Our method of analysis ensured that only those who had reached full Social Security retirement age in 2010 were
included in this calculation. About 11 percent (32,563 of 306,905) of all FECA beneficiaries are eligible for full retirement
under the Social Security Act.




Page 6                                                                      GAO-12-309R Federal Employees’ Compensation Act
Figure 1: Number of FECA Beneficiaries Receiving FECA Cash Benefits in 2010




Note: Because of the way Labor maintains its data, it is not possible to know whether an individual whose case closed in 2010 did or did
not receive cash benefits. For additional information on FECA case status and compensation, see enclosure IV.


Long-term, full-time beneficiaries received 58 percent of total cash benefits in 2010. In addition, 21
percent of all cash benefits was paid to long-term, full-time, retirement-age beneficiaries, who were 3
percent of all FECA beneficiaries, as shown in figure 2.

Figure 2: FECA Beneficiaries and Cash Benefits Received in 2010




Note: Because of the way Labor maintains its data, it is not possible to know whether a case closed in 2010 did or did not receive cash
benefits. For additional information on FECA case status and compensation, see enclosure IV.


Although long-term, full-time beneficiaries received over half of all cash benefits paid in 2010, most of
these individuals (89 percent) received less than $50,000 per year in cash benefits (see enc. IV,
table 18). The median annual cash benefit was about $34,000 per year. A small number of FECA


Page 7                                                                            GAO-12-309R Federal Employees’ Compensation Act
beneficiaries received more than $100,000 in benefits in 2010. That year, 116 individuals (or less
than 1 percent of long-term, full-time beneficiaries) received more than $100,000. 20 USPS employees
made up about one-third of these individuals receiving over $100,000. While our analysis shows that
about one-third of long-term, full-time beneficiaries were retirement age, a number of beneficiaries
were between the ages of 50 and 59 and approaching retirement age (see enc. IV, table 15).

We calculated that one-third of long-term, full-time beneficiaries were injured more than 20 years
ago, as shown in figure 3. Labor’s data system is not designed in a way that would allow us to
determine the cumulative total amount of time a person has been receiving FECA benefits. Given
these constraints, we calculated how long long-term, full-time beneficiaries had been receiving
benefits from their date of injury. This calculation does not account for any possible breaks in
benefits during that time—e.g., if an employee returned to work. As a result, this proxy probably
overestimates the time they have been receiving FECA benefits of any kind.

Enclosure IV provides detailed information on the characteristics of three groups: (1) all FECA
beneficiaries, (2) long-term, full-time FECA beneficiaries, and (3) long-term, full-time retirement-age
FECA beneficiaries.

Figure 3: Years from the Date of Injury for Long-Term, Full-Time Beneficiaries in 2010




Note: Labor does not maintain this data in a way that would allow us to determine the cumulative total amount of time a FECA beneficiary
has received benefits since the date of injury.




20
   Of all FECA beneficiaries, 407 received $100,000 or more in cash benefits in 2010. According to Labor, this can occur if
a beneficiary received lump sum schedule award payments, retroactive claims payments, or payments provided for by
special legislation.




Page 8                                                                           GAO-12-309R Federal Employees’ Compensation Act
In 2010, Long-Term, Full-Time, Non-USPS, FECA Beneficiaries Typically
Received Cash Benefits That Were Higher Than CSRS Annuities
In this analysis, we compared non-USPS long-term, full-time FECA beneficiaries who were covered
by CSRS and at least 55 years old to an equivalent group of actual CSRS recipients of the same
age. 21 Enclosure II provides more detailed information on our comparison of the benefits and
annuities for the non-USPS FECA beneficiaries and CSRS retirees.

Annual Non-USPS FECA Cash Benefits Were Typically Higher
Than Federal CSRS Annuities
In 2010, the FECA benefits of long-term, full-time beneficiaries we analyzed were typically higher
than the retirement annuities of CSRS retirees (see table 2). 22 These FECA beneficiaries received a
median annual benefit of $35,614, while the comparison group of CSRS retirees received a median
annual annuity of $29,196, before adjusting for taxes. The median FECA benefit was $7,325 higher
than the median annual annuity received by retirees, which was $28,289, after adjusting for taxes. 23

The difference between FECA benefits and CSRS annuities is typically larger when the FECA
beneficiary was injured after fewer years of service. The median annual FECA benefit for
beneficiaries with 10 or fewer years of service when injured was $12,450 more than the median
CSRS annuity. In contrast, the median annual FECA benefit for beneficiaries with 30 or more years
of service when injured—and thus closer to retirement—was $3,692 more than the median CSRS
annuity. 24




21
   The available historical data on federal employees limited our analysis to those who (a) were non-USPS, and (b) had
electronic records in the OPM current employee database between 1988 and 2010.
22
   This analysis only included cash benefits for long-term, full-time beneficiaries, who are considered unable to work in any
capacity, for 2010. This analysis does not include closed cases, claims with medical payments only, FECA beneficiaries
who received survivor benefits, or schedule awards. The analysis also does not distinguish between FECA beneficiaries
who have no dependents, and received 66-2/3 percent of their preinjury salaries and those beneficiaries with dependents
who received 75 percent of their preinjury salaries. Further, this analysis does not include individuals receiving part- time
Loss of Wage Earning Capacity (LWEC) benefits. In general, part-time benefits are determined by calculating the
difference between the employee’s preinjury and postinjury monthly wage. These individuals do not receive full disability
benefits from FECA, because it has been determined that they are capable of working part-time or with restrictions.
23
   Unlike FECA benefits, a portion of each CSRS retiree’s annuity is taxed. Because OPM uses complex formulas to
determine the taxed portion of the annuity, we estimated federal and state income taxes using the National Bureau of
Economic Research’s (NBER) TAXSIM program. TAXSIM is NBER's program for calculating liabilities under U.S. federal
and state income tax laws from individual data. The TAXSIM Model (http://www.nber.org/taxsim) simulates the U.S. federal
and state income tax rules. See Daniel Richard Feenberg and Elizabeth Coutts, “An Introduction to the TAXSIM Model,”
Journal of Policy Analysis and Management, vol. 12, no. 1 (winter, 1993), 189-194. We could not estimate taxes for a
number of annuitants; therefore, in our tax analyses, we assumed these annuitants owed zero taxes.
24
  Our analysis showed that 71 percent of FECA beneficiaries had higher benefits than their retiree counterparts before
taxes; 77 percent had higher benefits after adjusting for taxes. Given the trend shown here, the remaining 29 percent of
FECA beneficiaries (23 percent after taxes) whose benefits were lower than their retiree counterparts’ annuities most likely
were injured later in their careers and after many years of service.




Page 9                                                                     GAO-12-309R Federal Employees’ Compensation Act
Table 2: Comparison of Median Non-USPS Annual FECA Cash Benefits and CSRS Annuities

                                                                                                                    Difference between
                                                                         Median annual        Median annual        median annual FECA
                                                  Median annual          CSRS annuity         CSRS annuity           benefit and CSRS
                                                   FECA benefit           before taxes           after taxes        annuity after taxes
    All long-term, full-time FECA                         $35,614a             $29,196                $28,289                        $7,324
    beneficiaries
    Years of serviceb
    Less than 10                                           $29,256             $16,806               $16,806c                      $12,450
    10 to 19                                               $32,911             $23,226                $22,956                        $9,956
    20 to 29                                               $38,303             $32,244                $31,007                        $7,296
    30 or more                                             $43,932             $43,020                $40,240                        $3,692
Source: GAO analysis of Labor and OPM data.

Notes: Statistics on FECA benefits apply to cash benefits received by beneficiaries who were at least 55 years old and covered by CSRS.
Statistics on annuities apply to CSRS retirees who were at least 55 years old. Due to the limited historical data available on federal
employees, both groups include only employees who had electronic records between 1988 and 2010 and were not USPS employees.
a
 This analysis included only wage-loss replacement payments for long-term, full-time beneficiaries. Schedule awards and survivor benefits
are not included.
b
    We matched the FECA beneficiary and the annuitant on equivalent years of service at the point of the FECA beneficiary’s date of injury.
c
    At this income level, many of the recipients had no tax liability.


The differences between annual FECA and CSRS benefits in our comparison are largely explained
by the benefit calculation formulas used for each set of benefits. The FECA formula awards 66 2/3
percent (for individuals) or 75 percent (for those with dependents) of the salary at time of injury. The
benefit also increases over time by the FECA cost of living adjustment (COLA). 25 The CSRS formula
generally awards a smaller percentage of salary than the FECA formula for most workers, except
those with long tenure. Also, FECA beneficiaries with a dependent have higher benefits, while the
provision of a joint and survivor annuity under CSRS generally provides a survivor benefit that lowers
the amount received by the beneficiary (thus increasing the differences between FECA benefits and
retirement annuities). Other factors may affect such comparisons, as well. By contrast, as seen in
table 2, differences due to taxes are relatively small.

See enclosure III for detailed information on the differences in FECA and retirement benefits and
annuities for non-USPS long-term, full-time beneficiaries and non-USPS retirees, with respect to
differences in each of the following factors: years of service at the time of injury, years since exit from
federal service, maximum prior income, and age in 2010.




25
     This is an annual adjustment of FECA benefits based on changes in the Consumer Price Index.




Page 10                                                                             GAO-12-309R Federal Employees’ Compensation Act
Selected States That Set Age Limits for Workers’ Compensation Programs
Reported Savings, Legal Challenges, and Financial Hardships for Some
Beneficiaries
We examined experiences in four states—Kentucky, Minnesota, Montana, and Tennessee—that limit
workers’ compensation benefits for wage-loss compensation based on retirement age. 26 These
states generally restrict wage-loss benefits for workers who reach retirement age if they have an
injury rated as a “permanent total disability”—i.e., a condition that prevents a return to work. 27
Kentucky, Montana, and Tennessee consider workers “retired” 28 when they are eligible for full
retirement benefits under the Social Security Act. 29 The remaining state—Minnesota—presumes
people are “retired from the labor market” upon reaching 67 years of age. However, that person may
present evidence to challenge this presumption, such as proof that they were already working past
age 67 when injured, according to a state official. (See table 3.)

Table 3: Retirement-Age Limits of Wage-loss Benefits for Permanent Total Disability in Selected State Workers’
Compensation Programs

    State                Retirement-age Limitations                                                                             Year enacted
    Kentucky             Wage-loss benefits end when claimant qualifies for full retirement benefits under the                             1996
                         Social Security Act or 2 years after the claimant’s last injury, whichever is later.
    Minnesota            Wage-loss benefits end when claimant reaches age 67.                                                              1996
    Montana              Wage-loss benefits end when claimant receives or is eligible to receive full retirement                           1981
                         benefits under the Social Security Act or alternative retirement system.
    Tennessee            • Wage-loss benefits end when claimant is, by age, eligible for full retirement benefits                          1994
                           under the Social Security Act.
                         • If a person is injured after age 60, wage-loss benefits continue for a maximum of
                           260 weeks (5 years), with the compensation rate reduced by half of the amount of
                           any Social Security benefits received during the period of the award.a
Source: State workers’ compensation officials.
a
 A person after age 60 may receive a maximum of 260 weeks for any work injury, whether it is permanent total or permanent partial
disability. Since permanent total disability benefits end with full retirement eligibility, many persons older than 60 elect permanent partial
disability even if they are unable to return to work with the preinjury employer, according to a state workers’ compensation official.



26
  According to the Workers’ Compensation Research Institute (WCRI) review of state workers’ compensation laws as of
January 2010, eight states limit workers’ compensation benefits by age. We initially selected five of those states—
Kentucky, Minnesota, Montana, North Dakota, and Tennessee—that limit workers’ compensation benefits for workers at
retirement age, according to WCRI. We did not receive information from officials in North Dakota on their experiences
within time frames that would have allowed us to include them in this report. There are other states that limit workers’
compensation benefits based on an age other than retirement age. For example, Florida limits workers’ compensation
benefits at age 75, according to the WCRI review. See Ramona Tanabe, Workers’ Compensation Laws as of January
2010, WCRI (Cambridge, MA: October 2010).
27
 Injured workers with a permanent total disability continue to receive medical care benefits after reaching retirement age
under the workers’ compensation program in all four states, according to workers’ compensation officials. Montana defines
permanent total disability as a condition that prevents a return to regular employment, which means work on a recurring
basis, according to a workers’ compensation official.
28
  However, Kentucky allows workers with a permanent total disability to receive wage-loss benefits from the state for up to
2 years after the employees’ last injury, so workers in Kentucky may in some instances continue receiving wage-loss
benefits after reaching retirement age.
29
   Although most employees are covered by Social Security, federal law generally allows states to enter into voluntary
agreements to provide Social Security coverage to certain state and local government employees. In the four states we
reviewed, the percentage of state and local government workers covered by Social Security ranged from about 75 percent to
94 percent in 2007, according to prior GAO work. See GAO, Social Security Administration: Management Oversight Needed to
Ensure Accurate Treatment of State and Local Government Employees, GAO-10-938 (Washington, D.C.: Oct. 4, 2010).




Page 11                                                                               GAO-12-309R Federal Employees’ Compensation Act
Stakeholders we interviewed from the four states highlighted several aspects of their experiences
regarding retirement-age limitations on workers’ compensation benefits, including cost savings, legal
challenges, and financial challenges for beneficiaries. Stakeholders included state workers’
compensation board officials and private attorneys who specialize in workers’ compensation cases.

Cost Savings
Workers’ compensation board officials from three of the states said that adding the retirement-age
limitation lowered costs, but officials from two states could not quantify the savings. For example, a
Minnesota workers’ compensation official said the state made many other changes simultaneously to
its workers’ compensation program, such as setting a new type of maximum weekly rate for all wage-
loss benefits and a new minimum weekly benefit rate for permanent total disability benefits;
consequently, the workers’ compensation office was unable to attribute savings to a single provision.
A workers’ compensation board official from Kentucky stated there has been a decrease in workers’
compensation costs since the retirement-age limitation went into effect; however, the official’s office
could not attribute the savings to this provision because it lacked statistical data since private
insurance carriers pay the benefits rather than the state. A workers’ compensation official from
Montana told us that the official’s office did not calculate the cost impact of the retirement-age
limitation; however, the official referred us to a 2005 report by the state’s audit organization that
estimated that repealing the retirement-age limitation would cost the Montana workers’ compensation
fund $228 million to $302 million for nonsettled permanent total disability claims that arose between
1981 and 2004. 30

Legal Challenges
Private attorneys from three of the states said that legal challenges to the retirement-age limitation
have been raised in their states, but the courts have generally upheld the retirement-age limitation.
For example, according to attorneys in Montana, in one case the state’s supreme court did not find a
rational basis to overturn the retirement-age limitation for workers receiving permanent total disability
benefits. In another case, however, the court established that workers receiving permanent partial
disability benefits can continue to receive them past retirement age. The state courts in Kentucky and
Tennessee have upheld the constitutionality of the retirement-age limitation of workers’
compensation benefits, according to private attorneys from those states.

Financial Hardships for Some Beneficiaries
Private attorneys from all four states told us that individuals whose benefits end when they reach
retirement age can face financial challenges. For example, attorneys in Kentucky said some people
work into their 80s because they need the income and cannot afford to retire. These attorneys said
that if such workers suffer a work-related injury at age 67, they potentially lose 15 to 20 years of
earnings, even though their wage-loss benefits would cover only 2 years. In addition, the Kentucky
attorneys noted that while wage-loss benefits continue for older beneficiaries who are ineligible for
full retirement Social Security benefits, wage-loss benefits cease for a claimant who qualifies for
even a small amount of Social Security retirement benefits.




30
 Legislative Audit Division, State of Montana, Financial Audit for the Fiscal Year Ended June 30, 2005, Montana State
Fund (Helena, MT: October 2005).




Page 12                                                                 GAO-12-309R Federal Employees’ Compensation Act
Agency Comments
We provided a draft of this report to Labor, OPM, and USPS for review and comment. In its
comments (see enclosure V), USPS said that under current statute, FECA provides benefits greater
than those provided under the traditional federal retirement systems for injured workers who are
retirement age. USPS added that it has become imperative that this matter be addressed through
legislative change. Labor provided technical comments, which we incorporated in the report as
appropriate. OPM indicated that it did not have any comments.


We are sending copies of this report to the Secretary of Labor, the U.S. Postmaster General, the
Director of OPM, interested committees, and others. In addition, the report is available at no charge
on the GAO website at http://www.gao.gov.

If you or your staff have questions about this report, please contact Andrew Sherrill at (202) 512-
7215 or sherrilla@gao.gov or Phillip Herr at (202) 512-2834 or herrp@gao.gov. Contact points for
our Offices of Congressional Relations and Public Affairs may be found on the last page of this
report. Key contributors to this report are listed in enclosure VI.




Andrew Sherrill
Director, Education, Workforce
  and Income Security Issues




Phillip Herr
Managing Director, Physical Infrastructure Issues

Enclosures




Page 13                                                     GAO-12-309R Federal Employees’ Compensation Act
List of Congressional Committees

The Honorable Susan M. Collins
Ranking Member
Committee on Homeland Security
and Government Affairs
United States Senate

The Honorable Thomas R. Carper
Chairman
Subcommittee on Federal Financial
Management, Government Information,
Federal Services, and International Security
Committee on Homeland Security and Governmental Affairs
United States Senate

The Honorable Claire McCaskill
Chair
Ad Hoc Subcommittee on Contracting Oversight
Committee on Homeland Security and Government Affairs
United States Senate

The Honorable Tom Coburn
Ranking Member
Permanent Subcommittee on Investigations
Committee on Homeland Security and Government Affairs
United States Senate

The Honorable Darrell Issa
Chairman
Committee on Oversight and Government Reform
House of Representatives




Page 14                                                 GAO-12-309R Federal Employees’ Compensation Act
Enclosure I: Details on Audit Scope and Methodology
To determine the characteristics and benefits for Federal Employees’ Compensation Act (FECA)
beneficiaries, we collected data from Department of Labor’s (Labor) Integrated Federal Employees’
Compensation System (iFECS), FECA’s claimant database for chargeback year 2010 31 to calculate
the number of beneficiaries by age and length of time receiving benefits from their injury date, including
those in the program past Social Security full retirement age. Because Labor’s iFECS system does not
track the length of time each beneficiary was receiving benefits, we used injury date as a proxy. We
also calculated and reported FECA benefits based on various recipient characteristics, such as
agency, benefit type, as well as total costs associated with cash benefits for FECA beneficiaries who
are past Social Security full retirement age. We assessed the reliability of the data by (1) electronically
testing required data elements, (2) reviewing existing information about the data and the system that
produced them, and (3) interviewing agency officials knowledgeable about the data. We determined
that the data we reviewed were reliable for the purposes of this report.

To determine if FECA beneficiaries’ benefits were higher or lower than federal retirees’ annuities, we
assembled data from Labor’s iFECS and the Office of Personnel Management’s (OPM) annuitant
database and Central Personnel Data File (CPDF), a central personnel database that has reliable
longitudinal data starting in fiscal year 1988. Our analysis focused on long-term, full-time FECA
beneficiaries who were covered by Civil Service Retirement System (CSRS). We excluded FECA
recipients who were part of the FERS retirement system, because there are more CSRS employee
annuitants currently receiving annuities. We excluded those FECA beneficiaries and retirees from
our analysis who were not working after 1988.

We excluded U.S. Postal Service (USPS) FECA beneficiaries from our analysis because USPS
could not provide reliable, comparable data on its employees. For example, we needed a service
computation date to determine the length of service for annuitants. 32 However, the service
computation data we received from USPS for the period 1988 to 2010 were missing these data for
16 percent to 90 percent of employees, depending on the year. Furthermore, data in 1988 and 1989
were missing a significant amount of information we needed.

Exclusion of USPS employees and our selection criteria left us with approximately 4,000 non-USPS
FECA beneficiaries and approximately 301,000 CSRS annuitants. We accounted for federal and
state taxes on retirees’ annuities using the NBER TAXSIM model, in order to make after-tax
comparisons with FECA benefits (which are not taxed). We could not calculate taxes for some
annuitants; therefore, we assumed that these annuitants owed zero taxes. Enclosure II describes in
more detail how we compared non-USPS FECA benefits to non-USPS retirement annuities.

To determine the experiences in states that limit workers’ compensation benefits for workers at
retirement age, we conducted case studies in four states that do so—Kentucky, Minnesota, Montana,
and Tennessee. To identify the states, we first reviewed reports and interviewed officials from
several organizations, including Workers Compensation Research Institute (WCRI); National
Academy of Social Insurance; Workers’ Injury Law & Advocacy Group; and National Council on
Compensation Insurance. We analyzed WCRI’s Workers’ Compensation Laws as of January 2010



31
 FECA benefits are paid out of the Employees’ Compensation Fund, and most are charged back to the employees’
agency. Labor’s chargeback year for FECA agency billing purposes ends June 30, 2010.
32
 A service computation date is a date that is used to determine benefits and is generally based on how long the person
has been in the federal service.




Page 15                                                                 GAO-12-309R Federal Employees’ Compensation Act
(Cambridge, MA: October 2010) to identify states that limit benefits by retirement age but did not
independently verify the information contained in this compilation. Of the eight states that, according
to the WCRI review, limit workers’ compensation benefits by age, we selected five states that limit
workers’ compensation benefits for workers at retirement age: Kentucky, Minnesota, Montana, North
Dakota and Tennessee. We spoke with stakeholders in all of the states, except for North Dakota. We
did not receive information from officials in North Dakota on their experiences within time frames that
would have allowed us to include them in this report.

In each of the four states, we interviewed workers’ compensation board officials who manage and
maintain state-level workers’ compensation programs to discuss their experiences in administering
workers’ compensation benefits, particularly for older workers; and private sector attorneys who
specialize in workers’ compensation within their states to discuss their experiences handling cases
involving older workers. We also reviewed specific elements of state law for the four selected states
and, as appropriate, verified our analysis of state laws with cognizant state officials.




Page 16                                                     GAO-12-309R Federal Employees’ Compensation Act
Enclosure II: Methodological Details on the Comparison of FECA and CSRS
Annuity Benefits

Overview of the Matched Comparison Group Method
Comparing FECA benefits and retirement annuities requires us to estimate something that cannot be
observed in the real world: the retirement benefits that FECA beneficiaries would have earned if they
were never injured. We can observe either annuity or FECA payments at retirement age, depending
on whether employees happened to be injured during their careers. In contrast, we cannot observe
the counterfactual payment that the FECA beneficiaries would have received if they had not become
sick or injured. The fact that employees become sick or injured prevents us from observing what they
would have earned had they not become sick or injured.

How do we estimate the counterfactual annuity incomes of FECA beneficiaries, given that we cannot
observe them directly? One simple method would be to calculate the difference between annuity and
FECA payments for retirement-age beneficiaries. If a typical beneficiary received more than a typical
annuitant, one might conclude that structural features of the FECA program made those benefits
more generous.

The problem with the simple comparison is that it reflects characteristics of employees who are
injured, in addition to characteristics of the FECA and CSRS annuity programs. An employee’s work
environment, long-term health, and attitudes toward risk could affect both income and the chance of
receiving FECA benefits. FECA beneficiaries and annuitants may vary with respect to these
characteristics when we observe them in 2010. For example, employees with physically demanding
jobs, such as letter carriers and border patrol agents, may have higher risks of being injured and
lower incomes than employees who work in less hazardous conditions and remain healthy enough to
retire. FECA benefits might appear less generous, when, in fact, the difference is due to lower prior
earnings among employees more prone to workplace injury.

In a matched comparison group analysis, we compare FECA beneficiaries to annuitants who have
similar incomes and potential to be injured on the job. By comparing FECA beneficiaries to
annuitants with similar demographics, work environments, periods of employment, and incomes prior
to injury or retirement, we can better isolate the difference in benefits due to program characteristics
alone. If we match the groups on all characteristics that affect both income and the chance of
becoming injured, statistical theory ensures that a simple comparison between groups can estimate
the counterfactual annuity income that FECA beneficiaries would have earned.


Theory of the Matched Comparison Group Method
For all employees, let Di equal 1 if the employees become eligible for FECA benefits and 0 otherwise.
Let Yid and Yir equal the employees’ potential FECA and retirement annuity benefits that they could
have earned if they were injured or worked to retirement, respectively. Because employees in the
population of interest either get injured or remain healthy and retire, their observed benefits at
retirement age equal

                                        Yi = Di Yid + (1 - Di)Yir .

This makes clear that the observed benefits depend on both (1) whether the employees are injured
(and the factors that lead to injury) and (2) the program features that determine Yid and Yir. We need




Page 17                                                         GAO-12-309R Federal Employees’ Compensation Act
to estimate the expected difference between what FECA beneficiaries earn and what they would
have earned in annuity benefits if they had not been injured. This is given by

                                 j = E(Yid – Yir | Di = 1) = E(Yid | Di = 1) – E(Yir | Di = 1) .

In statistics, this parameter is known as the Average Treatment Effect for the Treated. Here, the
treatment effect of interest is the mean difference in retirement-age earnings in the potential states of
injury and retirement, among those who were actually receiving FECA benefits in 2010.

The key assumption required to estimate j from the observed earnings and group membership data
is that Di is independent of (Yid, Yir), conditional on a vector of covariates Xi. In other words, FECA
and annuity benefits are unrelated to the propensity to be injured among employees who have
similar background characteristics. Using this assumption and the available data, causal inference
theory ensures that j = E(Yi | Xi , Di = 1) – E(Yi | Xi, Di = 0), which can be estimated from the observed
data by replacing the random variables with their sample values and consistent estimators of sample
means.

We can estimate j using various methods, including statistical models that specify a functional form
for E(Yi | Xi, Di). In contrast, the matched comparison group method selects a sample of annuitants
such that the empirical distributions of the covariates are as similar as possible to the covariate
distributions for the FECA beneficiaries. Achieving this covariate balance conditions the comparison
of sample means on Xi and estimates j without bias, assuming the treatment is unrelated to the
potential outcomes among employees with the same characteristics (or “conditionally ignorable”). 33


Matching Methods
To construct the matched samples, we used a computer algorithm that selected the single closest
annuitant for each FECA beneficiary, measured on the Mahalanobis measure of multivariate
distance in Xi, and then returned each annuitant to the pool of potential matches after matching each
FECA beneficiary. (This is known as one-to-one Mahalanobis matching with replacement. 34) With
about 75 annuitants available to match for each FECA beneficiary, we were able to combine exact
matching and matching in distribution. Because occupation and agency may strongly affect work
environment and income, we created one matched sample that was exactly matched on these
variables. Exact matching potentially reduces the sample available to match on other variables, so
we created a second sample that matched on the proportion of each group that had a “blue collar
occupation,” as defined by the Office of Personnel Management (OPM). 35 For all versions of the
analysis, we assessed covariate balance using the sample moments and quantiles before and after
matching.

Although our data are not a probability sample, we calculated the standard error of the difference in
mean benefits to assess the uncertainty of our estimates, using standard formulas for differences in



33
  Matching allows us to control for the propensity to get injured on the job without making assumptions about the functional
form of the relationships among Yi, Xi, and Di. We use this nonparametric method of adjusting for the covariates, instead of
parametric models, because we lack substantial prior knowledge on how federal employees become injured.
34
  We matched directly on Xi, rather than on an estimate of the propensity score, E(Di | Xi ). This avoided making potentially
inaccurate assumptions about the functional form relating employees’ characteristics to their chance of being injured.
35
     Examples of blue-collar occupations include aircraft electrician, roofer, and forklift operator.




Page 18                                                                         GAO-12-309R Federal Employees’ Compensation Act
means between independent samples. Our large sample sizes suggested that the estimated
variances would have been small, regardless of the particular method used.


Population of Interest and Data Sources
We analyzed two populations of federal employees: those who were long-term, full-time FECA
beneficiaries and at least 55 years old on June 30, 2010, and those who were CSRS annuitants on
approximately the same date who were in the same age group. These individuals did not include
those who received survivor benefits. 36 We assumed that the retirement age began at 55 years and
limited our analysis to these subpopulations of both groups. We focused on this age group because
96 percent of annuitants in 2010 were at least 55 years old and because a younger group provides
more data for analysis. In addition, 55 is the minimum age of retirement eligibility under CSRS.

The limited availability of historical data on federal employees also determined the population we
analyzed. OPM’s Central Personnel Data File (CPDF) was the primary source of data available on
employment histories. The database contained information measured on the last day of each fiscal
year since 1988. This period covered 54 percent of the people receiving long-term, full-time cash
benefits in chargeback year 2010 who were at least 55 years old and who did not work for USPS. As
a result, we could analyze only this subset of employees who worked after 1988.

The OPM personnel file provided relatively complete data on these employees. Data were available
for an average of 63 percent of each employee’s career, with 75 percent of employees having data
for at least 71 percent of their careers. 37 Data on the analysis variables were missing for only 0.4 to
1.8 percent of employees, depending on the variable. Although the limited scope of the OPM
personnel data limits our analysis to employees working since 1988, this subset of people likely
resembles current and future federal employees more closely than the employees we could not
analyze.

We constructed the final data for analysis using the OPM personnel data, an administrative database
on FECA beneficiaries from the Department of Labor (Labor), and an administrative database on
annuitants from OPM. First, we selected from the Labor database the subset of FECA beneficiaries
who worked for agencies other than USPS, were at least 55 years old in 2010, and received long-
term, full-time cash benefits in chargeback year 2010 (14,983 people). 38 We then selected all
employees who received annuities and were at least 55 years old in 2010 from the OPM annuitant
database (866,391 people). Finally, we matched the FECA beneficiaries and annuitants to their
historical personnel data from OPM and kept for analysis only those employees working after 1988
who were covered by the CSRS retirement system.

The final dataset included approximately 301,000 annuitants and 4,000 FECA beneficiaries,
depending on the variables required for analysis. These groups are smaller than the populations that
Labor and OPM originally provided, due to our screening by age and retirement system and the lack



36
  This analysis does not distinguish between FECA beneficiaries who have no dependents, and received 66-2/3 percent of
their preinjury salaries and those beneficiaries with dependents who received 75 percent of their preinjury salaries.
37
  We measured coverage rates using the ratio of the length of time an employee was observed in the CPDF to the time
elapsed between the minimum service computation year and the final year of employment observed in the CPDF.
38
 FECA benefits are paid out of the Employees’ Compensation Fund and most are charged back to the employees’ agency.
Labor’s chargeback year for FECA agency billing purposes ends June 30, 2010.




Page 19                                                               GAO-12-309R Federal Employees’ Compensation Act
of OPM personnel data for employees working prior to 1988. Consequently, our results generalize
only to the population of non-USPS, long-term, full-time FECA beneficiaries working after 1988 who
were at least 55 years old in 2010.


Measurement of Characteristics for Matching Analysis
In this section, we describe the variables we used to conduct the analysis and our measurement of
important concepts. We measured most of the variables prior to the FECA beneficiary’s last date of
employment. 39 This created a comparison group of annuitants with similar work histories prior to
when the matched FECA beneficiary left federal employment.

     •    Benefits

The benefit for FECA beneficiaries was their annual gross cash payments for chargeback year 2010,
and the benefit for annuitants was their annual annuity income in fiscal year 2010. Using the TaxSim
program developed by the National Bureau of Economic Research, we estimated the amount of
taxes each annuitant might have paid on the gross benefit to create an after-tax measure. 40 Because
FECA benefits are not subject to income tax, the adjustment helped ensure that we compared after-
tax benefits for both groups. Due to missing and inconsistent data from OPM, we could not estimate
taxes for 414 of the approximately 301,000 annuitants in our population of interest. We assumed that
these annuitants owed zero taxes, which overestimated after-tax benefits. In addition, we assumed
that each annuitant had a spouse older than the age of 65—the scenario for which taxes are
lowest—because data on the annuitants’ families were not available. These assumptions reduce any
differences between after-tax annuities and FECA benefits for other otherwise equivalent employees.

     •    First year of federal employment

We measured the first year of federal employment for both groups using the earliest service
computation date for leave purposes in OPM’s personnel data. For employees who started after
1988, the earliest date likely equals their starting date because the personnel data would have
covered their first year working and would not have reflected later adjustments due to lapses in
service. For employees who started before 1988, the minimum date would equal the starting date for
those workers who were employed continuously (at least 75 percent of workers in the OPM
personnel data).

     •    Last year of federal employment

We measured the last year of federal employment for both groups as the latest year in which the
OPM personnel data recorded the employee as being on board.

     •    Length of service




39
  To do this, we created subsets of the FECA beneficiaries according to their final year of federal employment and then
found matched annuitants with similar covariates measured up to this year. We combined the year-specific samples to
create an overall matched sample.
40
   Daniel Richard Feenberg and Elizabeth Coutts, “An Introduction to the TAXSIM Model,” Journal of Policy Analysis and
Management, vol. 12, no 1 (winter 1993), 189-194. Accessed via the Internet on Nov. 30, 2011, at
http://www.nber.org/~taxsim/taxsim-calc9/index.html.




Page 20                                                                  GAO-12-309R Federal Employees’ Compensation Act
Our data allowed us to measure the length of service for FECA beneficiaries on (1) the date of injury,
(2) the date of last employment, and (3) the date of entering the long-term, full-time benefit rolls. We
chose to measure length of service as the time between the last year of employment and the
maximum service computation date in OPM’s personnel data. The maximum service computation
date reflects adjustments for multiple spells of employment. Consequently, the time elapsed between
the last year of employment and the service computation date should measure length of service at
the time of separation.

We calculated the same measure of length of service for annuitants, even though OPM provided the
measure it used for calculating annuity benefits. For most employees, the OPM measure differed
from the measure we calculated from OPM’s personnel data by no more than 1.4 years, but we used
our calculated measure to ensure that we measured length of service the same way for FECA
beneficiaries and annuitants.

    •     Spells of employment

We used OPM’s personnel data to calculate the number of spells of employment from 1988 through
2010. Our measure underestimates the number of spells for employees who were working prior to
1988 and had more than one spell of employment. This error is likely to be slight, however, given that
at least 75 percent of employees in OPM’s personnel data after 1988 worked continuously. We
controlled for number of spells to construct comparable work histories and to approximate an
employee’s risk preferences, assuming that having more spells is correlated with accepting more
occupational risk.

    •     Prior income

We measured income during each employee’s career using salary data from OPM’s personnel data.
We measured the minimum, median, and maximum incomes for the employees’ career for the
portion of their career prior to the date of disability. (For annuitants, we measured income prior to
their matched FECA beneficiaries’ last years of employment.)

    •     Occupation, agency, education, and gender

We used “occupation groups” defined by OPM to measure the type of work each employee
performed. These groups included occupations that involved a similar type of work, such as the
“engineering and architecture” group that included occupations involving “civil engineering” and
“architecture.” We used two-digit agency codes to identify the employing agency, such as the
Departments of Commerce and Defense. For the FECA beneficiaries, we measured occupation and
agency as close to the injury date as possible. For the annuitant group, we measured occupation and
agency in the annuitants’ last year of employment. Since occupation and agency do not change
frequently over time, the exact timing of the measurements should not be consequential.

We used OPM’s personnel data file to determine gender and education for both groups of employees
on the same dates as the measures of occupation and agency.




Page 21                                                      GAO-12-309R Federal Employees’ Compensation Act
Balance of Characteristics before and after Matching
Table 4 and figure 4 describe the characteristics of the FECA beneficiaries and annuitants, prior to
matching. As compared to the annuitants, the FECA beneficiaries stopped working at an earlier age,
had fewer years of service, had lower prior incomes, were less educated, and were more likely to
have an occupation defined by OPM as “blue-collar.” This suggests that the injured employees were
more likely to have jobs that involved lower-paid physical labor. Despite these differences, the
median FECA beneficiary earned 12.2 percent ($3,886) more in gross benefits than the median
annuitant prior to matching.

Table 4: Characteristics of FECA Beneficiaries and Annuitants before Matching

                                                           FECA beneficiaries                               Annuitants
                                                   25th                                75th         25th                     75th
                                              percentile     Median       Mean    percentile   percentile Median   Mean percentile
    Outcome
    FECA cash benefits or                       $28,779     $35,638 $39,922         $46,634      $20,868 $31,752 $36,296        $47,472
    CSRS annuity income in
    2010 (nominal dollars)
    FECA cash benefits or                       $28,779     $35,638 $39,922         $46,634      $20,856 $30,570 $33,917        $43,932
    estimated after-tax CSRS
    annuity income in 2010
    (nominal dollars)
    Covariates
    First year of employment                       1970        1974       1974         1978         1966    1970        1970      1975
    Last year of employment                        1990        1994       1995         1999         1993    1997        1998      2003
    Age in 2010                                       60         65        66.4           72         63.1    68.7        69.5      75.3
    Age in first year of                            23.8         29        30.2         35.3         22.3    26.8        28.9      34.3
    employment
    Age in last year of                            46.1         51.3      51.5         56.8         53.7     56.5       56.9       60.6
    employment
    Years of servicea                              15.6          21    21.1              26         23.6    28.7    27.9           32.5
    Spells of employment                              1           1       1               1            1       1       1              1
    Minimum prior income                        $38,409     $47,230 $50,655         $58,107      $39,331 $50,573 $55,398        $66,676
    (2011 dollars)
    Median prior income                         $41,248     $49,829 $54,191         $62,734      $42,772 $55,772 $60,617        $73,469
    (2011 dollars)
    Maximum prior income                        $43,380     $52,738 $57,217         $66,555      $45,861 $61,047 $66,128        $81,356
    (2011 dollars)
    No high school degree                                                 20.6                                          13.3
    (percent)
    High school degree                                                    35.4                                            32
    (percent)
    Some college (percent)                                                  28                                          28.1
    College degree (percent)                                               8.9                                          13.8
    Graduate degree (percent)                                              6.9                                          12.8
    OPM blue-collar occupation                                            38.4                                          14.3
    (percent)
    Occupation group                                            Various                                       Various
    Agency                                                      Various                                       Various
Source: GAO analysis of Labor and OPM data.

Note: Populations include 4,006 FECA beneficiaries and 301,433 retired workers who were at least 55 years old in 2010, working after
1988, and covered by the CSRS retirement system. Some percentages do not total 100 percent, due to rounding.
a
    Length of service is the time between the last year of employment and the maximum service computation date in OPM’s personnel data.




Page 22                                                                                 GAO-12-309R Federal Employees’ Compensation Act
Figure 4: Covariate Density Estimates for FECA Beneficiaries and Annuitants before Matching




Our matching algorithm effectively corrected for these imbalances, as shown in table 5 and figure 5.
The balance measures reflect the results of our analysis that used exact matching on occupation and
agency. All covariates are well-balanced in both the means and various quantiles. Other measures of
the sample distribution of the covariates, such as the ratio of variances and mean differences
between all empirical quantiles, showed similarly strong balance. We achieved similar balance when
we substituted an indicator for blue-collar occupation instead of exactly matching on occupation and
agency.




Page 23                                                          GAO-12-309R Federal Employees’ Compensation Act
Table 5: Characteristics of FECA Beneficiaries and Annuitants after Matching

                                                     FECA beneficiaries                                 Retired
                                         25th                               75th           25th                                  75th
                                    percentile        Median      Mean percentile     percentile   Median         Mean      percentile
    Outcome
    Before tax annuity or              $28,853        $35,614   $39,946   $46,504        $20,409   $29,196     $33,813        $41,967
    FECA cash benefits in
    2010 (nominal dollars)
    After-tax annuity or               $28,853        $35,614   $39,946   $46,504        $20,400   $28,289     $31,775        $39,335
    FECA cash benefits in
    2010 (nominal dollars)
    Covariates
    First year of                             1970      1974      1974      1978            1970      1974         1973           1978
    employment
    Age in 2010                                 61         65      66.5       72            61.1      65.3          66.6           71.7
    Age in first year of                      23.8         29      30.3      35.4           23.3      28.2          29.6           34.6
    employment
    Age in last year of                       45.9         51      51.2      56.3           45.6      50.8          50.8           55.8
    employment
    Years of servicea                         15.6       20.8      20.9      25.6           16.3      21.2          21.2           26.1
    Spells of employment                         1          1        1         1               1          1            1                 1
    Minimum prior income               $38,466        $47,342   $50,784   $58,174        $39,235   $47,682     $50,924        $58,073
    (2011 dollars)
    Median prior income                $41,211        $49,803   $54,178   $62,710        $41,798   $50,026     $54,113        $62,319
    (2011 dollars)
    Maximum prior income               $43,277        $52,511   $57,016   $66,384        $43,873   $52,513     $57,275        $66,944
    (2011 dollars)
    No high school degree                                          20.7                                             20.2
    (percent)
    High school degree                                             35.5                                             36.5
    (percent)
    Some college (percent)                                         27.9                                             27.9
    College degree                                                   9                                               8.6
    (percent)
    Graduate degree                                                 6.9                                              6.9
    (percent)
    Occupation group                                             Same                                             Same
    Agency                                                       Same                                             Same
Source: GAO analysis of Labor and OPM data.

Note: Populations include 3,864 disabled and 3,864 matched retired workers who were at least 55 years old in 2010, working after 1988,
and covered by the CSRS retirement system. The number of disabled workers excludes 142 who could not be matched. After-tax
estimates assume that taxpayers had a spouse older than 65. Some percentages do not total 100 percent, due to rounding.
a
 Years of service were based on the amount of time between the employees’ earliest date of service and the FECA beneficiaries’ date of
injury.




Page 24                                                                             GAO-12-309R Federal Employees’ Compensation Act
Figure 5: Covariate Density Estimates for FECA Beneficiaries and Annuitants after Matching




We found results similar to those in table 6 of enclosure III using samples matched on the proportion
of each group having a blue-collar occupation instead of exactly matched on occupation and agency.
Using this alternative matched sample, the median before-tax estimates of the difference in benefits
for all FECA beneficiaries varied by no more than 1 percent from the estimates we produced using
the matched sample described above.




Page 25                                                           GAO-12-309R Federal Employees’ Compensation Act
Enclosure III: Difference in Benefits between non-USPS FECA Beneficiaries and
Matched Retired Employees
Entries are the difference in FECA beneficiaries versus matched annuitant benefits, using the
retirement comparison and exact matching on occupation and agency. After-tax estimates assume
that the taxpayer had a spouse of the same age.

For additional technical information on methodology in comparing FECA benefits to retirement
annuities, refer to enclosure II.

Table 6: Differences in Benefits between Non-USPS FECA Beneficiaries and Matched Retired Employees

                                                                                           Proportional difference between FECA
                                                    Dollars (2010)                             benefit and annuity (percent)
                                        Before taxes            After taxes                 Before taxes            After taxes
    Group                            Median         Mean       Median         Mean          Median     Mean       Median          Mean
    Overall                             6,418        6,133       7,324        8,172              22     18.1         25.9          25.7
                         a
    Years of service
    Less than 10                       12,450        9,352      12,450    10,071                74.1    44.6         74.1          49.8
    10 to 19                            9,685        8,219       9,956        9,494             41.7    29.9         43.4          36.2
    20 to 29                            6,059        5,603       7,296        8,013             18.8       15        23.5          22.9
    30 or more                                912      740       3,692        4,619              2.1       1.6         9.2         10.6
    Years since exitb
    Less than 4                         5,675        7,297       7,781    10,525                14.8       17        21.5          26.5
    4 to 7                              6,237        5,705       7,941        8,351             18.1    14.6         24.3          22.9
    8 to 11                            12,280       11,003      13,010    12,971                44.8    32.7         48.7          40.9
    12 or more                         13,030       12,652      13,145    13,702                60.9    49.5         61.8          55.9
    Maximum prior income
    1st quartile                        6,295        6,006       6,325        6,358             33.6    30.4         33.9          32.8
    2nd quartile                        7,315        7,060       7,772        7,841             29.5    27.4         31.9          31.3
    3rd quartile                        5,990        6,195       7,412        7,930              18     18.3         23.3          24.6
    4th quartile                        4,031        6,157       8,686    11,377                 7.6    11.1         17.9          22.7
    Age in 2010
    55 to 59                            2,015        1,823       3,632        4,702              5.8       4.7         11            13
    60 to 64                            4,409        4,211       5,781        6,623             13.6    11.5         18.7          19.3
    65 to 69                            6,693        6,286       7,682        8,238             22.7    18.3           27          25.4
    70 and older                       11,159       10,834      11,337    12,041                49.2    39.8         50.4          46.2
Source: GAO analysis of Labor and OPM data.
a
 Years of service were based on the amount of time between the employees’ earliest date of service and the FECA beneficiaries’ date of
injury.
b
    Years since exit measures the time elapsed between the employees’ last year of federal employment and 2010.




Page 26                                                                               GAO-12-309R Federal Employees’ Compensation Act
Enclosure IV: Characteristics of (1) All FECA Beneficiaries, (2) Long-Term, Full-
Time FECA Beneficiaries, and (3) Long-Term, Full-Time Retirement-Age FECA
Beneficiaries for 2010 41

All FECA Beneficiaries
Table 7: Distribution of Benefit Type for All FECA Beneficiaries (2010)

                                        All                              USPS                                  Non-USPS
                               Number of FECA                  Number of FECA                         Number of FECA
                                  beneficiaries Percent           beneficiaries Percent                  beneficiaries         Percent
 Medical                               109,913       35.8%                 52,891       40.5%                     57,022         32.3%
 benefits only
 Long-term, full-                       31,880       10.4%                 12,349        9.5%                     19,531         11.1%
 time
 Part-time                              10,594        3.5%                   2,501       1.9%                       8,093         4.6%
 Short-term                             24,549        8.0%                 15,567       11.9%                       8,982         5.1%
 Schedule                                7,596        2.5%                   3,816       2.9%                       3,780         2.1%
 award
 Survivor                                4,642        1.5%                     625       0.5%                       4,017         2.3%
 benefits
 Closed case                           108,495       35.4%                 39,398       30.2%                     69,097         39.2%
 Other                                   9,236        3.0%                   3,336       2.6%                       5,900         3.3%
 Total                                 306,905       100.0%               130,483     100.0%                     176,422       100.0%
Source: GAO analysis of Labor data.

Note: Other includes administrative reviewed, reopened, case retired or awaiting retirement, claim under development, claim unreviewed,
claim destroyed, and claims with overpayments.



Table 8: Distribution of Age for All FECA Beneficiaries (2010)

                                         All                              USPS                                 Non-USPS
                                Number of FECA                 Number of FECA                         Number of FECA
                                   beneficiaries Percent          beneficiaries Percent                  beneficiaries         Percent
 Less than 20                             1,404        0.5%                      29      0.0%                       1,375         0.8%
 20-29                                   20,548        6.7%                  3,273       2.5%                     17,275          9.8%
 30-39                                   39,194       12.8%                 13,933      10.7%                     25,261         14.3%
 40-49                                   70,829       23.1%                 35,883      27.5%                     34,946         19.8%
 50-59                                  102,336       33.3%                 53,143      40.7%                     49,193         27.9%
 60-64                                   36,076       11.8%                 15,556      11.9%                     20,520         11.6%
 65 and older                            36,333       11.8%                  8,655       6.6%                     27,678         15.7%
 Missing, not                                  185     0.1%                      11      0.0%                         174         0.1%
 applicable
 Total                                  306,905      100.0%               130,483      100.0%                    176,422       100.0%
 Median Age                            52 years                           52 years                              52 years
Source: GAO analysis of Labor data.




41
  All data presented in this enclosure are from chargeback year 2010. See enclosure I for more information on the source of
the data.




Page 27                                                                         GAO-12-309R Federal Employees’ Compensation Act
Table 9: Distribution of Gender for All FECA Beneficiaries (2010)

                                 All                                  USPS                                Non-USPS
                        Number of FECA                      Number of FECA                          Number of FECA
                           beneficiaries       Percent         beneficiaries         Percent           beneficiaries            Percent
 Male                           171,927         56.1%                62,745           48.1%                 109,182              62.1%
 Female                         134,311         43.9%                67,696           51.9%                  66,615              37.9%
 Total                          306,238        100.0%               130,441          100.0%                 175,797             100.0%
Source: GAO analysis of Labor data.

Note: Total excludes 667 cases where gender was missing.



Table 10: Years from Date of Injury to 2010 for All FECA Beneficiaries

                                                 All                         USPS                              Non-USPS
                                        Number of                     Number of                            Number of
                                             FECA                          FECA                                 FECA
                                      beneficiaries      Percent    beneficiaries        Percent         beneficiaries  Percent
 Less than 1 year ago                      114,321        37.2%             42,880         32.9%                71,441           40.5%
 1 to 5 years ago                          104,829        34.2%            51,174         39.2%                 53,655           30.4%
 6 to 10 years ago                          30,779        10.0%            17,985         13.8%                 12,794            7.3%
 11 to 20 years ago                         29,182         9.5%            12,599          9.7%                 16,583            9.4%
 21 to 30 years ago                         16,012         5.2%             3,514          2.7%                 12,498            7.1%
 31 to 40 years ago                          9,244         3.0%             1,910          1.5%                  7,334            4.2%
 41 or more years ago                        2,538         0.8%               421          0.3%                  2,117            1.2%
 Total                                     306,905       100.0%           130,483        100.0%                176,422          100.0%
Source: GAO analysis of Labor data.




Table 11: Distribution of Annual Cash Benefits for All FECA Beneficiaries (2010)

                                                All                          USPS                             Non-USPS
                                        Number of                     Number of                           Number of
                                             FECA                          FECA                                FECA
                                      beneficiaries      Percent    beneficiaries        Percent        beneficiaries  Percent
 $0                                        221,522        73.4%           93,555          73.4%             127,967     73.4%
 $1 - $9,999                                23,368         7.7%           12,268           9.6%               11,100     6.4%
 $10,000 - $19,999                          14,519         4.8%            6,206           4.9%                8,313     4.8%
 $20,000 - $29,999                          15,394         5.1%            4,741           3.7%               10,653     6.1%
 $30,000 - $39,999                          16,160         5.4%             8,386          6.6%                 7,774             4.5%
 $40,000 - $49,999                           5,558         1.8%             1,654          1.3%                 3,904             2.2%
 $50,000 - $59,999                           2,172         0.7%               265          0.2%                 1,907             1.1%
 $60,000 - $69,999                           1,252         0.4%               138          0.1%                 1,114             0.6%
 $70,000 - $79,999                             681         0.2%                62          0.0%                   619             0.4%
 $80,000 - $89,999                             386         0.1%                35          0.0%                   351             0.2%
 $90,000 - $99,999                             407         0.1%                25          0.0%                   382             0.2%
 $100,000 or more                              407         0.1%                88          0.1%                   319             0.2%
 Total                                     301,826       100.0%           127,423        100.0%               174,403           100.0%
 Median benefit                            $21,504                        $17,371                             $23,574
 (excluding $0)
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.




Page 28                                                                         GAO-12-309R Federal Employees’ Compensation Act
Table 12: Distribution of Annual Medical Benefits for All FECA Beneficiaries (2010)

                                                 All                           USPS                             Non-USPS
                                        Number of                     Number of                           Number of
                                             FECA                          FECA                                FECA
                                      beneficiaries    Percent      beneficiaries        Percent        beneficiaries           Percent
 $0                                          79,305     26.0%               23,071         17.8%                56,234           32.1%
 $1 - $9,999                                203,493     66.7%               95,254         73.4%               108,239           61.7%
 $10,000 - $19,999                           13,715      4.5%                6,998          5.4%                 6,717            3.8%
 $20,000 - $29,999                            4,228      1.4%                2,136          1.6%                 2,092            1.2%
 $30,000 - $39,999                            1,793      0.6%                  929          0.7%                   864            0.5%
 $40,000 - $49,999                              907      0.3%                  457          0.4%                   450            0.3%
 $50,000 - $59,999                              532      0.2%                  273          0.2%                   259            0.1%
 $60,000 - $69,999                              349      0.1%                  175          0.1%                   174            0.1%
 $70,000 - $79,999                              189      0.1%                    90         0.1%                     99           0.1%
 $80,000 - $89,999                              158      0.1%                    79         0.1%                     79           0.0%
 $90,000 - $99,999                               99      0.0%                    53         0.0%                     46           0.0%
 $100,000 or more                               386      0.1%                  185          0.1%                   201            0.1%
 Total                                      305,154    100.0%             129,700        100.0%                175,454          100.0%
 Median benefit                                $888                           $975                                $823
 (excluding $0)
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.



Table 13: Distribution of Total Annual Benefits (Cash and Medical Benefits Combined) for All FECA
Beneficiaries (2010)

                                                 All                         USPS                           Non-USPS
                                        Number of                     Number of                           Number of
                                             FECA                          FECA                                FECA
                                      beneficiaries    Percent      beneficiaries        Percent        beneficiaries Percent
 $0                                         56,688      18.8%             17,064          13.4%               39,624   22.8%
 $1 - $9,999                               173,538      57.6%             81,301          63.8%               92,237   53.1%
 $10,000 - $19,999                          19,117       6.3%              8,547           6.7%               10,570    6.1%
 $20,000 - $29,999                          16,079       5.3%              5,685           4.5%               10,394    6.0%
 $30,000 - $39,999                          14,921       5.0%              6,661           5.2%                8,260    4.8%
 $40,000 - $49,999                           9,044       3.0%              4,098           3.2%                4,946    2.8%
 $50,000 - $59,999                           4,615       1.5%              1,766           1.4%                2,849    1.6%
 $60,000 - $69,999                           2,637       0.9%                854           0.7%                1,783    1.0%
 $70,000 - $79,999                           1,511       0.5%                437           0.3%                1,074    0.6%
 $80,000 - $89,999                             926       0.3%                290           0.2%                  636    0.4%
 $90,000 - $99,999                             691       0.2%                182           0.1%                  509    0.3%
 $100,000 or more                            1,411       0.5%                475           0.4%                  936    0.5%
 Total                                     301,178     100.0%            127,360         100.0%              173,818 100.0%
 Median benefit                             $1,563                        $1,463                              $1,707
 (excluding $0)
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.




Page 29                                                                         GAO-12-309R Federal Employees’ Compensation Act
Long-Term, Full-Time FECA Beneficiaries42
Table 14: Distribution of Long-Term, Full-Time Beneficiaries (2010)

                                      All                                   USPS                               Non-USPS
                    Number of long-term,                         Number of long-term,                 Number of long-term,
                    full-time beneficiaries Percent              full-time beneficiaries Percent      full-time beneficiaries Percent
 Long term,                            31,880 100.0%                            12,349       38.7%                   19,531   61.3%
 full-time
Source: GAO analysis of Labor data.




Table 15: Distribution of Age of Long-Term, Full-Time Beneficiaries (2010)

                                            All                                USPS                             Non-USPS
                       Number of long-term,                       Number of long-term,                Number of long-term,
                       full-time beneficiaries Percent            full-time beneficiaries Percent     full-time beneficiaries Percent
 Less than 20                                       7    0.0%                            0     0.0%                       7      0.0%
 20-29                                            187    0.6%                         64       0.5%                     123      0.6%
 30-39                                        1,090      3.4%                       591        4.8%                     499      2.6%
 40-49                                        4,299     13.5%                      2,474      20.0%                   1,825      9.3%
 50-59                                        9,544     29.9%                      4,871      39.4%                   4,673    23.9%
 60-64                                        5,147     16.1%                      1,876      15.2%                   3,271    16.7%
 65 and older                                11,606     36.4%                      2,473      20.0%                   9,133    46.8%
 Total                                       31,880     100.0%                   12,349      100.0%                  19,531   100.0%
 Median age                                 60 years                            56 years                           63 years
Source: GAO analysis of Labor data.




Table 16: Distribution of Gender of Long-Term, Full-Time Beneficiaries (2010)

                                        All                                    USPS                             Non-USPS
                       Number of long-term,                       Number of long-term,                Number of long-term,
                       full-time beneficiaries Percent            full-time beneficiaries Percent     full-time beneficiaries Percent
 Male                                        17,705     55.5%                      5,554      45.0%                  12,151    62.2%
 Female                                      14,174     44.5%                      6,795      55.0%                   7,379    37.8%
 Total                                       31,879     100.0%                   12,349      100.0%                  19,530   100.0%
Source: GAO analysis of Labor data.

Note: Table excludes cases where gender was missing.




42
 This group of long-term, full-time beneficiaries does not include those individuals receiving part-time benefits, schedule
awards, and survivor benefits.




Page 30                                                                               GAO-12-309R Federal Employees’ Compensation Act
Table 17: Years from Date of Injury to 2010 for Long-Term, Full-Time Beneficiaries

                                               All                         USPS                         Non-USPS
                                  Number of long-               Number of long-                 Number of long-
                                    term, full-time               term, full-time                 term, full-time
                                     beneficiaries    Percent      beneficiaries    Percent        beneficiaries    Percent
 Less than 1 year ago                        1,110     3.48%                 674     5.46%                   436      2.23%
 1 year ago                                  1,437     4.51%                 923     7.47%                   514      2.63%
 2 years ago                                 1,171     3.67%                 711     5.76%                   460      2.36%
 3 years ago                                 1,131     3.55%                 613     4.96%                   518      2.65%
 4 years ago                                 1,053     3.30%                 605     4.90%                   448      2.29%
 5 years ago                                 1,107     3.47%                 575     4.66%                   532      2.72%
 6 years ago                                 1,106     3.47%                 580     4.70%                   526      2.69%
 7 years ago                                 1,144     3.59%                 585     4.74%                   559      2.86%
 8 years ago                                 1,019     3.20%                 564     4.57%                   455      2.33%
 9 years ago                                 1,040     3.26%                 569     4.61%                   471      2.41%
 10 years ago                                1,042     3.27%                 540     4.37%                   502      2.57%
 11-15 years ago                             4,072    12.77%               1,716    13.90%                 2,536    12.06%
 16-20 years ago                             4,126    12.94%               1,212     9.81%                 2,914    14.92%
 21-25 years ago                             3,870    12.14%                 867     7.02%                 3,003    15.38%
 26-30 years ago                             2,731     8.57%                 486     3.94%                 2,245    11.49%
 31-35 years ago                             2,497     7.83%                 571     4.62%                 1,926      9.86%
 36-40 years ago                             1,396     4.38%                 351     2.84%                 1,045      5.35%
 41 or more years ago                          828     2.60%                 207     1.68%                   621      3.18%
 Total                                      31,880                        12,349                          19,531
Source: GAO analysis of Labor data.

Note: Some percentages do not total 100 percent, due to rounding.




Page 31                                                                      GAO-12-309R Federal Employees’ Compensation Act
Table 18: Distribution of Annual Cash Benefits for Long-Term, Full-Time Beneficiaries (2010)

                                              All                            USPS                               Non-USPS
                                  Number of long-               Number of long-                       Number of long-
                                    term, full-time               term, full-time                       term, full-time
                                     beneficiaries    Percent      beneficiaries        Percent          beneficiaries          Percent
 $0                                            163      0.5%                     70        0.6%                       93          0.5%
 $1 - $9,999                                 1,096      3.4%                   691         5.6%                      405          2.1%
 $10,000 - $19,999                           3,161      9.9%                 1,170         9.5%                    1,991         10.2%
 $20,000 - $29,999                           7,558     23.7%                 1,812        14.7%                    5,746         29.5%
 $30,000 - $39,999                          12,391     38.9%                 6,899        56.0%                    5,492         28.2%
 $40,000 - $49,999                           4,029     12.7%                 1,339        10.9%                    2,690         13.8%
 $50,000 - $59,999                           1,411      4.4%                   159         1.3%                    1,252          6.4%
 $60,000 - $69,999                             848      2.7%                     83        0.7%                      765          3.9%
 $70,000 - $79,999                             501      1.6%                     36        0.3%                      465          2.4%
 $80,000 - $89,999                             261      0.8%                     16        0.1%                      245          1.3%
 $90,000 - $99,999                             298      0.9%                     10        0.1%                      288          1.5%
 $100,000 or more                              116      0.4%                     39        0.3%                       77          0.4%
 Total                                      31,833    100.0%                12,324      100.0%                   19,509         100.0%
 Median benefit                            $33,554                        $34,459                               $32,492
 (excluding $0)
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.



Table 19: Distribution of Annual Medical Benefits for Long-Term, Full-Time Beneficiaries (2010)

                                              All                            USPS                               Non-USPS
                                  Number of long-               Number of long-                       Number of long-
                                    term, full-time               term, full-time                       term, full-time
                                     beneficiaries    Percent      beneficiaries        Percent          beneficiaries          Percent
 $0                                          6,946     21.8%                 1,555        12.6%                    5,391         27.6%
 $1 - $9,999                                17,026     53.5%                 6,931        56.2%                  10,095          51.8%
 $10,000 - $19,999                           4,177     13.1%                 2,053        16.6%                    2,124         10.9%
 $20,000 - $29,999                           1,551      4.9%                   747         6.1%                      804          4.1%
 $30,000 - $39,999                             754      2.4%                   377         3.1%                      377          1.9%
 $40,000 - $49,999                             452      1.4%                   225         1.8%                      227          1.2%
 $50,000 - $59,999                             272      0.9%                   133         1.1%                      139          0.7%
 $60,000 - $69,999                             188      0.6%                     95        0.8%                       93          0.5%
 $70,000 - $79,999                             105      0.3%                     49        0.4%                       56          0.3%
 $80,000 - $89,999                              83      0.3%                     37        0.3%                       46          0.2%
 $90,000 - $99,999                              57      0.2%                     29        0.2%                       28          0.1%
 $100,000 or more                              228      0.7%                   105         0.9%                      123          0.6%
 Total                                      31,839    100.0%                12,336      100.0%                   19,503         100.0%
 Median benefit                             $4,931                          $6,056                               $4,016
 (excluding $0)
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.




Page 32                                                                         GAO-12-309R Federal Employees’ Compensation Act
Table 20: Distribution of Total Annual Benefits (Cash and Medical Benefits Combined) for Long-Term, Full-Time
Beneficiaries (2010)

                                                 All                        USPS                                Non-USPS
                                  Number of long-                 Number of long-                      Number of long-
                                    term, full-time                 term, full-time                      term, full-time
                                     beneficiaries      Percent      beneficiaries     Percent            beneficiaries         Percent
 $0                                               11      0.0%                    2       0.0%                           9        0.0%
 $1 - $9,999                                     639      2.0%                 352        2.9%                        287         1.5%
 $10,000 - $19,999                             2,067      6.5%                 749        6.1%                      1,318         6.8%
 $20,000 - $29,999                             5,507     17.3%               1,237       10.0%                      4,270        21.9%
 $30,000 - $39,999                             9,366     29.4%               4,241       34.4%                      5,125        26.3%
 $40,000 - $49,999                             6,207     19.5%               2,993       24.3%                      3,214        16.5%
 $50,000 - $59,999                             3,124      9.8%               1,248       10.1%                      1,876         9.6%
 $60,000 - $69,999                             1,794      5.6%                 582        4.7%                      1,212         6.2%
 $70,000 - $79,999                             1,136      3.6%                 323        2.6%                        813         4.2%
 $80,000 - $89,999                               647      2.0%                 189        1.5%                        458         2.3%
 $90,000 - $99,999                               475      1.5%                 116        0.9%                        359         1.8%
 $100,000 or more                                870      2.7%                 300        2.4%                        570         2.9%
 Total                                        31,843    100.0%              12,332      100.0%                    19,511        100.0%
 Median benefit                              $38,282                       $39,177                               $37,289
 (excluding $0)
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.



Retirement-Age Beneficiaries on Long-Term, Full-Time Rolls 43
Table 21: Distribution of Retirement-Age Beneficiaries on Long-Term, Full-Time Rolls (2010)

                                                 All                          USPS                              Non-USPS
                                           Number of                    Number of                             Number of
                                      retirement-age               retirement-age                        retirement-age
                                        beneficiaries   Percent      beneficiaries      Percent            beneficiaries        Percent
 Long term, full-time                         10,873    100.0%               2,286        21.0%                     8,587        79.0%
Source: GAO analysis of Labor data.




43
 Includes only those FECA beneficiaries on the long-term, full-time rolls who were of full retirement age as set forth in the
Social Security Act.




Page 33                                                                         GAO-12-309R Federal Employees’ Compensation Act
Table 22: Distribution of Gender of Retirement-Age Beneficiaries on Long-Term, Full-Time Rolls (2010)

                                                 All                         USPS                         Non-USPS
                                           Number of                     Number of                      Number of
                                      retirement-age                retirement-age                 retirement-age
                                        beneficiaries   Percent       beneficiaries   Percent        beneficiaries   Percent
 Male                                          7,200     66.2%               1,472      64.4%               5,728      66.7%
 Female                                        3,673     33.8%                 814      35.6%               2,859      33.3%
 Total                                        10,873    100.0%               2,286    100.0%                8,587     100.0%
Source: GAO analysis of Labor data.

Note: Table excludes case where gender was missing.



Table 23: Years from Date of Injury to 2010 for Retirement-age Beneficiaries on Long-Term, Full-Time Rolls (2010)

                                                  All                        USPS                         Non-USPS
                                           Number of                     Number of                      Number of
                                      retirement-age                retirement-age                 retirement-age
                                        beneficiaries   Percent       beneficiaries   Percent        beneficiaries    Percent
 Less than 1 year ago                             28     0.26%                  11      0.48%                   17      0.20%
 1 year ago                                       30     0.28%                  13      0.57%                   17      0.20%
 2 years ago                                      37     0.34%                  16      0.70%                   21      0.24%
 3 years ago                                      39     0.36%                  19      0.83%                   20      0.23%
 4 years ago                                      44     0.40%                  13      0.57%                   31      0.36%
 5 years ago                                      64     0.59%                  20      0.87%                   44      0.51%
 6 years ago                                      72     0.66%                  32      1.40%                   40      0.47%
 7 years ago                                      88     0.81%                  27      1.18%                   61      0.71%
 8 years ago                                      82     0.75%                  20      0.87%                   62      0.72%
 9 years ago                                     109     1.00%                  43      1.88%                   66      0.77%
 10 years ago                                    118     1.09%                  36      1.57%                   82      0.95%
 11-15 years ago                                 703     6.47%                 193      8.44%                 510       5.94%
 16-20 years ago                               1,424    13.10%                 240     10.50%                1,184    13.79%
 21-25 years ago                               1,973    18.15%                 306     13.39%                1,667    19.41%
 26-30 years ago                               1,843    16.95%                 298     13.04%                1,545    17.99%
 31-35 years ago                               2,126    19.55%                 475     20.78%                1,651    19.23%
 36-40 years ago                               1,289    11.86%                 320     14.00%                 969     11.28%
 41 or more years ago                            804     7.39%                 204      8.92%                 600       6.99%
 Total                                        10,873                         2,286                           8,587
Source: GAO analysis of Labor data.

Note: Some percentages do not total 100 percent, due to rounding.




Page 34                                                                        GAO-12-309R Federal Employees’ Compensation Act
Table 24: Distribution of Annual Cash Benefits for Retirement-Age Beneficiaries on Long-Term, Full-Time Rolls (2010)

                                                 All                               USPS                         Non-USPS
                                           Number of                           Number of                      Number of
                                      retirement-age                      retirement-age                 retirement-age
                                        beneficiaries      Percent          beneficiaries   Percent        beneficiaries        Percent
 $0                                               12         0.1%                      0       0.0%                    12         0.1%
 $1 - $9,999                                      84         0.8%                     20       0.9%                    64         0.7%
 $10,000 - $19,999                               900         8.3%                     97       4.2%                   803         9.4%
 $20,000 - $29,999                             2,889        26.6%                    340      14.9%                2,549         29.7%
 $30,000 - $39,999                             4,119        37.9%                  1,551      67.8%                2,568         29.9%
 $40,000 - $49,999                             1,382        12.7%                    213       9.3%                1,169         13.6%
 $50,000 - $59,999                               597         5.5%                     29       1.3%                   568         6.6%
 $60,000 - $69,999                               399         3.7%                     19       0.8%                   380         4.4%
 $70,000 - $79,999                               259         2.4%                      8       0.3%                   251         2.9%
 $80,000 - $89,999                               120         1.1%                      1       0.0%                   119         1.4%
 $90,000 - $99,999                                88         0.8%                      3       0.1%                    85         1.0%
 $100,000 or more                                 20         0.2%                      5       0.2%                    15         0.2%
 Total                                        10,869       100.0%                  2,286     100.0%                8,583        100.0%
 Median benefit                              $33,620                             $34,459                         $33,019
 (excluding $0)
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.



Table 25: Distribution of Annual Medical Benefits for Retirement-Age Beneficiaries on Long-Term, Full-Time Rolls (2010)

                                                        All                            USPS                      Non-USPS
                                                  Number of                        Number of                  Number of
                                             retirement-age                   retirement-age             retirement-age
                                               beneficiaries Percent            beneficiaries Percent      beneficiaries  Percent
 $0                                                   4,492   41.4%                      829 36.3%                3,663    42.7%
 $1 - $9,999                                          5,506   50.7%                    1,252 54.8%                4,254    49.6%
 $10,000 - $19,999                                      540    5.0%                      124    5.4%                416     4.9%
 $20,000 - $29,999                                         134     1.2%                   30   1.3%                  104          1.2%
 $30,000 - $39,999                                          63     0.6%                   16   0.7%                   47          0.5%
 $40,000 - $49,999                                          32     0.3%                   11   0.5%                   21          0.2%
 $50,000 - $59,999                                          17     0.2%                    2   0.1%                   15          0.2%
 $60,000 - $69,999                                          23     0.2%                    9   0.4%                   14          0.2%
 $70,000 - $79,999                                          16     0.1%                    3   0.1%                   13          0.2%
 $80,000 - $89,999                                          11     0.1%                    0   0.0%                   11          0.1%
 $90,000 - $99,999                                           8     0.1%                    3   0.1%                    5          0.1%
 $100,000 or more                                           19     0.2%                    6   0.3%                   13          0.2%
 Total                                                  10,861   100.0%                2,285 100.0%                8,576        100.0%
 Median benefit (excluding $0)                          $1,620                        $1,937                      $1,500
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.




Page 35                                                                              GAO-12-309R Federal Employees’ Compensation Act
Table 26: Distribution of Total Annual Benefits (Cash and Medical Benefits Combined) for Retirement-Age
Beneficiaries on Long-Term, Full-Time Rolls (2010)

                                                  All                            USPS                            Non-USPS
                                           Number of                        Number of                        Number of
                                      retirement-age                   retirement-age                   retirement-age
                                        beneficiaries Percent            beneficiaries Percent            beneficiaries         Percent
 $0                                                     0    0.0%                      0      0.0%                      0         0.0%
 $1 - $9,999                                        74       0.7%                    11       0.5%                     63         0.7%
 $10,000 - $19,999                                711        6.5%                    78       3.4%                   633          7.4%
 $20,000 - $29,999                              2,510       23.1%                   260      11.4%                 2,250         26.2%
 $30,000 - $39,999                              3,848       35.4%                 1,330      58.2%                 2,518         29.3%
 $40,000 - $49,999                              1,731       15.9%                   414      18.1%                 1,317         15.3%
 $50,000 - $59,999                                789        7.3%                    89       3.9%                   700          8.2%
 $60,000 - $69,999                                457        4.2%                    39       1.7%                   418          4.9%
 $70,000 - $79,999                                358        3.3%                    24       1.0%                   334          3.9%
 $80,000 - $89,999                                164        1.5%                    12       0.5%                   152          1.8%
 $90,000 - $99,999                                115        1.1%                      5      0.2%                   110          1.3%
 $100,000 or more                                 112        1.0%                    24       1.0%                     88         1.0%
 Total                                         10,869       100.0%                2,286    100.0%                  8,583        100.0%
 Median benefit (excluding $0)                $35,457                          $36,019                          $35,142
Source: GAO analysis of Labor data.

Note: Table excludes cases where benefits were less than zero. Amounts less than zero are due to accounting for, for example,
chargebacks to the claim for overpayments.




Page 36                                                                         GAO-12-309R Federal Employees’ Compensation Act
Enclosure V: Comments from the U.S. Postal Service




Page 37                                     GAO-12-309R Federal Employees’ Compensation Act
           Enclosure VI: GAO Contacts and Staff Acknowledgments

           GAO Contacts
           Andrew Sherrill at (202) 512-7215 or sherrilla@gao.gov, and Phillip Herr at (202) 512-2834 or
           herrp@gao.gov


           Staff Acknowledgments
           In addition to the contacts named above, key contributors to this report were Patrick Dibattista, H.
           Brandon Haller, James Bennett, Michelle Bracy, Tonnye Conner-White, Melinda Cordero, Carla
           Craddock, Robert Heilman, Gene Kuehneman, Christopher Morehouse, James Rebbe, Jeff Tessin,
           Walter Vance, Kathleen van Gelder, and James Whitcomb.




(131064)


           Page 38                                                     GAO-12-309R Federal Employees’ Compensation Act
This is a work of the U.S. government and is not subject to copyright protection in the
United States. The published product may be reproduced and distributed in its entirety
without further permission from GAO. However, because this work may contain
copyrighted images or other material, permission from the copyright holder may be
necessary if you wish to reproduce this material separately.
                      The Government Accountability Office, the audit, evaluation, and
GAO’s Mission         investigative arm of Congress, exists to support Congress in meeting its
                      constitutional responsibilities and to help improve the performance and
                      accountability of the federal government for the American people. GAO
                      examines the use of public funds; evaluates federal programs and
                      policies; and provides analyses, recommendations, and other assistance
                      to help Congress make informed oversight, policy, and funding decisions.
                      GAO’s commitment to good government is reflected in its core values of
                      accountability, integrity, and reliability.

Obtaining Copies of   The fastest and easiest way to obtain copies of GAO documents at no
                      cost is through GAO’s website (www.gao.gov). Each weekday afternoon,
GAO Reports and       GAO posts on its website newly released reports, testimony, and
Testimony             correspondence. To have GAO e-mail you a list of newly posted products,
                      go to www.gao.gov and select “E-mail Updates.”

Order by Phone        The price of each GAO publication reflects GAO’s actual cost of
                      production and distribution and depends on the number of pages in the
                      publication and whether the publication is printed in color or black and
                      white. Pricing and ordering information is posted on GAO’s website,
                      http://www.gao.gov/ordering.htm.
                      Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
                      TDD (202) 512-2537.
                      Orders may be paid for using American Express, Discover Card,
                      MasterCard, Visa, check, or money order. Call for additional information.

Connect with GAO      Connect with GAO on Facebook, Flickr, Twitter, and YouTube.
                      Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts.
                      Visit GAO on the web at www.gao.gov.

To Report Fraud,      Contact:

Waste, and Abuse in   Website: www.gao.gov/fraudnet/fraudnet.htm
                      E-mail: fraudnet@gao.gov
Federal Programs      Automated answering system: (800) 424-5454 or (202) 512-7470

Congressional         Katherine Siggerud, Managing Director, siggerudk@gao.gov, (202) 512-
                      4400, U.S. Government Accountability Office, 441 G Street NW, Room
Relations             7125, Washington, DC 20548

Public Affairs        Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
                      U.S. Government Accountability Office, 441 G Street NW, Room 7149
                      Washington, DC 20548




                        Please Print on Recycled Paper.