Military Retirement: Proposed Changes Warrant Careful Analysis
(Testimony, 02/25/99, GAO/T-NSIAD-99-94).

Pursuant to a congressional request, GAO discussed military retirement,
focusing on: (1) changes being proposed by the Department of Defense
(DOD); (2) areas where GAO believes more information is needed; and (3)
the opportunity to take a long-term strategic view of the military
compensation system.

GAO noted that: (1) there is no clear indication that the proposed
change to the retirement system, which would cost an estimated $13
billion in increased costs and unfunded liabilities, will address the
retention issue; (2) while the recently reported downturn in retention
rates is of concern, the nature of the retention problem is not clear;
(3) understanding the nature of the retention problem is critical in
choosing solutions--pockets of problems are best treated with targeted
rather than across-the-board solutions, and transitory problems are best
treated with actions that can be reversed or eliminated once the problem
has receded; (4) according to DOD, the 1986 Military Retirement Reform
Act (Redux) has become a symbol of eroding benefits to military members;
(5) although surveys of military personnel show an increasing level of
dissatisfaction with the retirement system, it is not clear what that
really means; (6) some of the surveys do not differentiate between
retirement pay and other retirement benefits; (7) also, many military
personnel appear to lack knowledge about their retirement system; (8)
the link between retirement pay and retention is unclear; (9) many
factors can influence a servicemember's decision; (10) the influence of
retirement in this decision has not been definitively determined; (11)
according to an analysis done by the Congressional Budget Office,
retention rates under Redux have not been markedly different than rates
under the prior system; (12) even if the retirement system is found to
be related to retention, it may not be the most cost-effective tool for
addressing any existing retention problems; (13) in addition, DOD's
proposal does not address other military retirement issues and their
impact on the structure of the force; (14) also, since the first
potential Redux retirees are still more than 7 years away from
retirement eligibility, DOD may be missing the opportunity for the kind
of comprehensive change to its compensation system suggested by the
Eighth Quadrennial Review of Military Compensation; and (15) the June
1997 Report of the Quadrennial Review called for DOD to take a broad
approach to align its policies with its strategy--rather than take a
piecemeal or one-size fits-all approach.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  T-NSIAD-99-94
     TITLE:  Military Retirement: Proposed Changes Warrant Careful 
             Analysis
      DATE:  02/25/99
   SUBJECT:  Military personnel
             Government retirement benefits
             Personnel management
             Military compensation
             Military cost control
             Service credit
IDENTIFIER:  DOD Quadrennial Defense Review
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  This text was extracted from a PDF file.        **
** Delineations within the text indicating chapter titles,      **
** headings, and bullets have not been preserved, and in some   **
** cases heading text has been incorrectly merged into          **
** body text in the adjacent column.  Graphic images have       **
** not been reproduced, but figure captions are included.       **
** Tables are included, but column deliniations have not been   **
** preserved.                                                   **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************
NS99094T.book GAO United States General Accounting Office

Testimony Before the Subcommittee on Military Personnel, Committee
on Armed Services, House of Representatives

For Release on Delivery Expected at 10: 00 a. m., EST Thursday,
February 25, 1999

MILITARY RETIREMENT Proposed Changes Warrant Careful Analysis

Statement of Mark E. Gebicke, Director, Military Operations and
Capabilities Issues, National Security and International Affairs
Division




GAO/T-NSIAD-99-94

  GAO/T-NSIAD-99-94

Page 1 GAO/T-NSIAD-99-94

Mr. Chairman and Members of the Subcommittee: I am here today to
discuss military retirement. Concerned about its ability to retain
personnel, the Department of Defense (DOD) has proposed pay and
retirement changes in its fiscal year 2000 budget. The pay
proposal includes a 4.4 percent, across- the- board raise and pay
table reform that will target increases to noncommissioned
officers and mid- grade

commissioned officers. The retirement change proposed by DOD is
essentially a partial repeal of the 1986 Military Retirement
Reform Act, which is commonly called Redux. After providing
background on the differences among the various military
retirement systems and how those differences came into being, I
will address (1) changes being proposed by DOD; (2) areas where we
believe

more information is needed; and (3) the opportunity to take a
long- term, strategic view of the military compensation system.
But first, I will summarize our observations regarding DOD's
proposed changes to the

military retirement system. Results in Brief Overall, we see no
clear indication that the proposed change to the

retirement system, which would cost an estimated $13 billion in
increased costs and unfunded liabilities, will address the
retention issue.

While the recently reported downturn in retention rates is of
concern, the nature of the retention problem is not clear. Is the
problem widespread or is it concentrated in certain military
occupations or year groups? Is it a transitory problem
attributable to such factors as reduced accessions during the
drawdown and the strong economy, or is it the beginning of a

long- term problem that will affect the military for the
foreseeable future? Understanding the nature of the retention
problem is critical in choosing solutions pockets of problems are
best treated with targeted rather than across- the- board
solutions, and transitory problems are best treated with actions
that can be reversed or eliminated once the problem has receded.

According to DOD, Redux has become a symbol of eroding benefits to
military members. Although surveys of military personnel show an
increasing level of dissatisfaction with the retirement system, it
is not clear what that really means. Some of the surveys do not
differentiate between retirement pay and other retirement
benefits. Also, many military personnel appear to lack knowledge
about their retirement system. For

Page 2 GAO/T-NSIAD-99-94

example, a 1998 Navy survey showed that over half the respondents
did not know which retirement system they were under. The link
between retirement pay and retention is unclear. The decision to
stay in or leave the military is a complex, highly individual
decision. Many factors (such as the availability of civilian jobs,
family considerations, and

satisfaction with military life) can influence a servicemember's
decision. The influence of retirement in this decision has not
been definitively determined. According to an analysis done by the
Congressional Budget Office, retention rates under Redux have not
been markedly different than rates under the prior system. Even if
the retirement system is found to be

related to retention, it may not be the most cost- effective tool
for addressing any existing retention problems. For example, when
given a choice, military members have tended to prefer up- front
compensation to deferred compensation.

In addition, DOD's proposal does not address other military
retirement issues and their impact on the structure of the force.
For example, DOD's proposal does not address a key purpose of the
Redux system to encourage personnel to remain on active duty after
reaching 20 years of service. Also, since the first potential
Redux retirees are still more than 7 years away from retirement
eligibility, DOD may be missing the opportunity for the kind of
comprehensive change to its compensation system suggested by the
Eighth Quadrennial Review of Military Compensation. The June 1997
Report of the Quadrennial Review called for

DOD to take a broad approach to align its policies with its
strategy rather than take a piecemeal or one- size- fits- all
approach.

Current Retirement Systems for Military Personnel

Military members are presently covered by three separate
retirement systems, depending on when they joined the military.
All three systems require no contribution from the servicemember
and allow retirement after

20 years of service. Servicemembers have no vesting before 20
years. Military members pay social security taxes and are eligible
for social security benefits, as well.

Depending on when servicemembers joined, the amount of their
military pension is calculated differently. Reforms in 1980
altered the pension calculations, changing the computation base
from 50 percent of the final monthly basic pay to 50 percent of an
average of the highest 3 years of basic pay (often called the
high- 3 system). Under these two systems, retirees are eligible
for 75 percent after 30 years.

Page 3 GAO/T-NSIAD-99-94

Major reforms followed in 1986. At that time, there were concerns
about the cost of the military retirement system, given the large
standing military force and the budget deficits at the time.
Further, some believed that the

retirement system was overly generous. These concerns ultimately
led to the passage of the Military Retirement Reform Act. Redux
reduced the retired pay calculation in such a way that retirees
with 20 years of service would receive 40 percent of their high- 3
basic pay rather than the 50 percent received under the previous
two systems. Under all three systems, however, servicemembers earn
75 percent after 30 years because Redux increased the amount
earned each year toward retirement after 20 years in such a way
that retirees receive 3.5 percent for each additional year of
service up to 30.

Another feature of Redux is a reduced cost of living adjustment
(COLA). High- 3 retirees receive an annual adjustment equal to the
consumer price index (CPI). However, Redux retirees' annual
adjustment is 1 percentage point less than the CPI. When retirees
reach age 62, retirement pay is recalculated at 2.5 percent

per year of service at retirement, and COLA is recalculated to
restore the purchasing power lost due to the annual COLA
reductions. Thus, at 62, Redux retirees with 20 years of service
would receive 50 percent, rather than 40 percent, of their high- 3
basic pay and all COLA decrements would

be made up. After age 62, retirees' pay remains at 50 percent of
their average high- 3 basic pay, but the 1- percent reduction in
COLA begins again. Table 1 describes the features of all three
retirement systems.

Page 4 GAO/T-NSIAD-99-94

Table 1: Current Retirement Systems for Military Personnel

a At age 62, the retiree's pay will be recalculated based upon the
number of years of service multiplied by 2.5 percent up to a
maximum of 75 percent. b A one- time recomputation of the
retiree's annuity to make up for lost purchasing power caused by
the holding of COLAs to the inflation rate minus 1 percent. #R- 11
COLAs for Military Retirees: Summary of

Congressional and Executive Branch Action, 1982- 1998.pop Source:
DOD.

None of the retirement changes were retroactive. Redux applied
only to servicemembers who joined on or after its effective date
of August 1, 1986. This change was made prospectively so that it
would not affect servicemembers who entered the service under the
previous systems.

Redux was adopted for two main reasons. The first was cost.
Congress wanted to reduce the overall cost of the military
retirement system but still provide fair and equitable retirement
benefits. Redux is a less costly system than its predecessors.
According to an estimate prepared in 1998, the DOD actuary
estimated that the military would have needed to put aside an
additional $7.5 billion between 1986 and 1999 had Redux not been

passed. The second reason for Redux was to provide an incentive
for personnel to stay past 20 years of service and thus to
increase the size of the senior career force. For each year of
service beyond 20, Redux provided

servicemembers with 3.5 percent toward their retirement pensions.
According to the DOD actuary, about 47 percent of retirees leave
within 1 year after reaching 20 years of service; by 3 years,
almost 70 percent have

left. Since the first Redux- covered servicemembers will not reach
20 years of service until 2006, it is unknown whether this
incentive will actually Final pay High- 3 Redux

Applies to servicemembers joining:

Before September 8, 1980 Between September 8, 1980, and July 31,
1986

On or after August 1, 1986

Basis of computation: Final basic pay Average basic pay for
highest 3 years Average basic pay for highest 3 years Multiplier
2. 5 percent per year

of service 2.5 percent per year of service 2 percent per year at
20 years; 3. 5 percent

per year thereafter up to 30 years of service (2. 5 percent
restored at age 62) a

Cost- of- living adjustments CPI CPI CPI minus 1 percent (one-
time catch- up at age 62) b

Page 5 GAO/T-NSIAD-99-94

encourage servicemembers to remain in the military. Table 2
compares the percent of base pay that retirees would receive under
the high- 3 and Redux systems depending upon the years of service
they had at the time of retirement.

Table 2: Percentages of Base Pay Used to Calculate Retirement Pay
Under the High- 3 and Redux Retirement Systems by Years of
Service.

Source: GAO.

Figures 1 and 2 compare the projected retirement pay for
hypothetical retirees under the high- 3 and Redux systems over a
30- year period. 1 Figure 1 shows what an officer at the O- 5 pay
level would earn after retiring at 42 years of age with 20 years
of service. Figure 2 shows what a noncommissioned officer at the
E- 7 pay level would earn after retiring at 38 years of age with
20 years of service. The figures illustrate the adjustment that
would be made under Redux at age 62 to equalize the retirement pay
with the high- 3 system. After age 62, the figures illustrate the
impact of the reduced COLA provision for the Redux retiree.

Years of service 20 21 22 23 24 25 26 27 28 29 30

High- 3 50 52. 5 55 57. 5 60 62.5 65 67.5 70 72. 5 75 Redux 40 43.
5 47 50. 5 54 57.5 61 64.5 68 71. 5 75

1 The charts use the over 20 column of the January 1998 pay tables
as the high- 3 average. In addition, the charts assume a 3.5-
percent annual change in the CPI.

Page 6 GAO/T-NSIAD-99-94

Figure 1: Projected Annual Retirement Pay (42 year old O- 5 after
20 years of service)

)

Figure 2: Projected Annual Retirement Pay (38 year old E- 7 after
20 years of service)

Without having had to contribute toward retirement, retirees
become eligible to receive an income for life at a relatively
young age in their late 30s to early 40s allowing them to engage
in other, full- time employment. Military retirees do not usually
completely retire from the workforce; instead, they have a second
career. The catch- up provisions in Redux provide for restored
benefits at age 62, when military retirees are more likely to
actually be fully retired and are eligible for social security
benefits.

More than two- thirds of the current military force came in after
the effective date of the 1986 Reform Act and would be covered by
Redux,

$0 $10,000

$20,000 $30,000

$40,000 $50,000

$60,000 $70,000

$80,000 $90,000

High- 3 Redux

$0 $5, 000

$10,000 $15,000

$20,000 $25,000

$30,000 $35,000

$40,000 $45,000

High- 3 Redux

Page 7 GAO/T-NSIAD-99-94

should they retire. This includes large numbers of junior enlisted
personnel in the force. At the end of fiscal year 1997, the DOD
actuary estimated that about 16 percent of the enlisted personnel
who entered the service stay to earn a military retirement. A
higher proportion of officers, about 46 percent of those who
entered the service over that same period,

are likely to stay for 20 years or more. It is important to note
that no one has yet taken a nondisability retirement under Redux.
The first Redux retirees will reach retirement eligibility in
2006, 20 years after the effective date of the legislation. The
most senior military members covered by Redux those who entered
the military just after the effective date of August 1, 1986 now
have approximately

12- years of military service. Thus, the first potential Redux
retirees are still more than 7 years from retirement. DOD Proposal
to Change Military Retirement

DOD has proposed pay and retirement changes in its fiscal year
2000 budget as a means of addressing declines in retention. The
pay portion of the proposal contains two parts: a 4.4- percent,
across- the- board pay raise

and pay table reform that will provide targeted pay increases to
noncommissioned officers and mid- grade commissioned officers. The
retirement change proposed by DOD would essentially be a partial
repeal of Redux. DOD's proposal would restore the multiplier
provision to allow servicemembers to earn 50 percent of their
high- 3 salary at 20 years of service. DOD proposes to alter the
way the COLA is calculated, as well. When the CPI is more than 3
percent, the formula would still be CPI minus 1 percent. However,
the proposal provides some low- inflation protection

when the CPI drops below 3 percent. If the CPI is between 2 and 3
percent, the cost- of- living adjustment would be 2 percent. If
the CPI is below 2 percent, retirees would receive a full cost-
of- living adjustment. DOD estimates that the retirement changes
it proposes will cost about $6 billion over the next 6 fiscal
years (2000- 2005). This is about 24 percent of the $24.7 billion
total cost of the DOD military pay and retirement proposal. 2 This
$6 billion represents the additional amount that DOD

projects it will need to pay into the retirement trust fund to
provide increased benefits. The year- by- year accrual projections
are shown in table 3.

2 The total estimate does not include the cost of giving DOD
civilians a commensurate 4. 4- percent pay raise. According to
DOD, the cost of the DOD civilian pay raise would be about $10
billion.

Page 8 GAO/T-NSIAD-99-94

Table 3: DOD Projections of the Additional Accrual Costs to Fund
Its Proposed Retirement Changes in the Fiscal Year 2000
President's Budget

Source: DOD.

DOD's cost figures do not include an estimated $7 billion 3
increase in the unfunded liability for the military retirement
trust fund if Redux were repealed. This liability would increase
because DOD has been setting aside money for the less- costly
Redux retirement plan since its adoption in 1986. At the end of
fiscal year 1997, the total existing unfunded liability in the
military retirement trust fund was almost $500 billion. The trust
fund has a large unfunded liability because, prior to 1984, DOD
did not set aside funds

to pay for retirement liabilities. The $500 billion is what
remains of the unfunded liability identified when DOD switched to
accrual accounting practices. The unfunded liability is assigned
to the Treasury to pay off over a very long period of time; the
DOD actuary estimates that the initial unfunded liability will be
fully amortized in 2033.

Personnel Surveys Provide Mixed Results

In support of proposed retirement changes, the military leadership
has cited growing dissatisfaction among military members about
their retirement benefits. Reports of dissatisfaction have often
been anecdotal arising when military leaders visit the troops. In
addition, various DOD and service surveys show that military
members have increasingly cited dissatisfaction with military
retirement. For example,

Dollars in millions Fiscal year Multiplier repeal COLA reform
Total accrual

projection

2000 $601 $195 $796 2001 656 232 888 2002 713 270 983 2003 773 276
1, 049 2004 797 285 1, 082 2005 862 291 1, 153

Total $4, 402 $1, 549 $5, 951

3 The difference between the estimated $7 billion increase in the
unfunded liability and the estimated $7.5 billion in reduced
accrual payments under Redux is due to the fact that the DOD
proposal represents only a partial repeal of Redux. As discussed,
DOD's proposal does not call for totally repealing the reduced
COLA.

Page 9 GAO/T-NSIAD-99-94

the Army conducts a semi- annual personnel survey that addresses
satisfaction with 56 aspects of military life, one of which is
retirement benefits. Since 1992, satisfaction with retirement
benefits has declined

steadily. However, the results may not be directly attributable to
retirement pay because the survey lumps all retirement benefits
together, including pay, health care, commissary privileges,
access to recreation facilities, and others.

Another example is a late- 1998 Air Force survey of 633 departing
personnel. Retirement pay was cited as an important reason for
separating from the Air Force by 58 percent of the enlisted
respondents and 42 percent of the officers. Other reasons cited
were the availability of civilian jobs, pay, frequent moves, and
compatibility of the military with family life. Approximately 86
percent indicated that retirement was not a good incentive to
serve 20 years or more. However, when asked if there was one

single thing that the Air Force could do to keep them in the
service, only 35 percent of enlisted personnel and 48 percent of
officers said that there was. Of those who indicated that there
was one single change that could be made, the most frequently
cited changes were more choice in assignments and a decrease in
the amount of time spent away from their home station. Retirement
was not among the single changes that were cited as being
sufficient to keep them in the service. Officers wanted more say/
choice in assignments and a decrease in OPSTEMPO; Enlisted
Personnel wanted

more pay, more control over assignments and to decrease OPSTEMPO.
These and other survey results also reveal that many military
members do not understand their retirement systems. The Army's
fall 1996 Sample Survey of Military Personnel showed that, of
those who planned to stay in the Army until retirement, more than
one- half of the officer respondents

and two- thirds of the enlisted respondents did not know which
retirement plan they were under. Even with the recent publicity
surrounding retirement issues, approximately 58 percent of the
enlisted personnel responding to a Navy survey in late 1998
indicated that they did not know what retirement system they were
under. Such lack of knowledge makes the basis for their
dissatisfaction unclear. Issues Warranting

Further Analysis Before proceeding with changes to the military
retirement system,

Congress may want to seek the answers to several questions: (1)
What is the nature of the retention problem? Is it widespread or
limited primarily to relatively small pockets? Is it a transitory
problem or is it expected to continue for the foreseeable future?
(2) What is the link between

Page 10 GAO/T-NSIAD-99-94

retirement pay and retention? (3) Should other aspects of
retirement or the military compensation system be addressed in a
more comprehensive way? Only with further examination of these
questions can Congress be confident that changing the retirement
system will solve DOD's reported retention problems and be cost-
effective. I do not have the answers to these questions today but
believe that they are important enough to warrant further
analysis.

Extent of Retention Problems Not Yet Known

According to DOD officials, aggregate historical retention
patterns are relatively stable, but recent downturns in retention
are indicators of an impending problem. The Air Force and Navy
have recently fallen short of their retention goals. Some of the
most critical shortfalls are in the Air Force, where overall
retention rates are below target and second- term

retention rates are at 68 percent, below the target rate of 75
percent. Retention problems have reportedly been severe in certain
jobs, such as pilots and computer specialists, among others. We
are now reviewing the retention of pilots and other specialties
where retention has been problematic. The results are expected to
be available this summer.

Some shortfalls may be a relatively short- run problem. For
example, the shortfalls may be related to how the services
approached reducing the size of the force in the early 1990s. To
meet reduced personnel ceilings and yet maintain readiness, the
services generally opted to reduce accessions

rather than separate experienced personnel. Those lowered
accessions during the drawdown may be contributing to shortfalls,
as the pipeline now has fewer junior personnel. With fewer junior
personnel in the pipeline, DOD needs to retain a higher percentage
to fill its now relatively constant personnel slots. This task is
made more difficult because of the low

unemployment rate and strong economy. Unclear Link Between
Retention and Retirement

The empirical link between retirement and retention has not been
convincingly drawn. When asked about the effect of Redux on
retention, the Marine Corps was blunt, saying that it is difficult
to quantify the impact of the Redux retirement system and that it
cannot definitively

substantiate direct impacts on retention. Army leadership has said
that retirement is less effective as a retention tool than it used
to be. The service chiefs are united in their belief that Redux is
hurting retention. This appears to be based on a combination of
anecdotal reports, survey data, and professional military
judgment.

Page 11 GAO/T-NSIAD-99-94

This lack of hard evidence concerns us and concerns some in
Congress. In October 1998, two members of the Senate Armed
Services Committee sent a letter to the Secretary of Defense
asking for an analysis of the likely impact of DOD's pay and
retirement proposals. During January 1999 testimony, the Chairman
of the Joint Chiefs of Staff was asked about this analysis. He
cited the concerns evidenced in the surveys and said that taking
time to perform detailed studies might lead to the demise of the

force. Similarly, the House National Security Committee has twice
requested that DOD report on the effects of Redux on retention,
among other items. The first of these reports, requested in the
House subcommittee report on the fiscal year 1998 Defense
authorization act, was never received. The second report,
requested in the House report on the fiscal year 1999 Defense
authorization act, is due June 30, 1999.

Several factors blur the link between retention and retirement. It
has not been shown that retention among personnel that entered the
service under Redux is lower than it is for those that entered
under the previous systems. Also, the decision to stay in or leave
the military is complex and based on a number of factors. Finally,
deferred compensation such as retirement pay is typically less of
an incentive than up- front compensation for retaining personnel.
I will expand on each of these points.

Historical data shows no conclusive evidence of a lower retention
rate for Redux personnel. The Congressional Budget Office (CBO)
analyzed retention decisions of personnel who entered the military
around the time

Redux was implemented. CBO has stated that its analysis generally
shows no significant difference in retention under Redux. Had
Redux been a major influence in the retention decision, fewer
Redux personnel would be staying in the military.

The decision of whether to stay in or leave the military is
complicated. DOD and the services contend that Redux is simply not
attractive enough to keep people in like the previous systems did.
In addition to retirement, several other factors civilian
employment opportunities, quality of life, and general
satisfaction with military life ranging from military housing to
rotation policies influence retention. These influences are
difficult to isolate. Finally, increasing retirement pay may not
be the most cost- effective way to increase retention. Rand
believes that targeted pay increases of the type that DOD is
proposing are the most effective way to address pockets of
retention problems. Increasing retirement pay is generally seen as
a less

Page 12 GAO/T-NSIAD-99-94

effective incentive compared to increasing current pay because of
the tendency of military personnel to deeply discount the value of
future dollars. For example, personnel who were offered early
separations during the drawdown showed a strong preference for the
cash incentives rather

than an annuity even though the annuity was worth more from a
present value perspective. According to an analysis performed by
DOD, approximately 50 percent of officers and around 90 percent of
enlisted people took the up- front payment, even though the
present value of the annuity was, in many cases, worth more than
twice as much. In addition, since retirement pay is computed on
basic pay, an increase in pay also results in an increase in
retirement pay.

Other Issues to Consider DOD may be missing an opportunity to take
a more comprehensive approach to its compensation system. The
purpose of the military

compensation system is to attract, retain, and motivate the number
of quality personnel needed to maintain the desired level of
national security. The military compensation system is a complex
accretion of over 40 different pays and allowances, many of which
came into being to address a specific problem at the time. Over
the past 50 years, at least 15 presidentially or congressionally
commissioned study groups have

recommended improvements to the system. While the level of
compensation has changed dramatically over that period, the
features of the military compensation system have changed
relatively little. The most recent broad review of military
compensation, the Eighth Quadrennial Review of Military
Compensation, sought to design a compensation system that would
attract, retain, and motivate the diverse workforce of the 21 st
century. Its June 1997 executive report encouraged a broad,
strategic approach to human resource management systems through
which DOD would align its personnel practices with the strategic
direction of the organization. The report says that a
comprehensive review

of the compensation system requires the military to include
everything that servicemembers value, and it concludes that one
size does not fit all. In 1996, GAO hosted a roundtable of
military experts to discuss possible changes to the military
retirement system. 4 The consensus of the panel members was that
the military retirement system while exerting a strong

pull toward retirement after members reach 10 to 12 years of
service can 4 Military Retirement: Possible Changes Merit Further
Evaluation (GAO/NSIAD-97-17, Nov. 15, 1996).

Page 13 GAO/T-NSIAD-99-94

actually impede effective force management. Many of the
participants indicated that the 20- year vesting provision of the
retirement system is a hindrance to effectively managing the
force. Because military personnel are not entitled to any
retirement benefits unless they have served 20 years, the services
have been reluctant to involuntarily separate personnel as they
approached 20 years of service. And although some military combat
specialties (like infantry) require youth and vigor, experience
may be of greater value in occupations such as intelligence
analysis or systems acquisition.

Once a member reaches 20 years, the previous retirement systems
tend to exert a considerable push effect. As I noted earlier,
about 47 percent of retirees leave within one year after reaching
20 years of service; by the end of 22 years of service, almost 70
percent have left. Part of the rationale for the Redux reform was
to increase the size of the career force by reducing this push
effect. If Redux is repealed, we will not know whether it would
have achieved the objective of stemming the loss of senior career
personnel and repeal may have an unintended consequence of
reducing the incentive for people to stay in the military after
they become eligible for

retirement. In summary, I believe that changes to the military
retirement system must be carefully analyzed. Such changes would
apply across the board and would likely have a long- term effect.
Overall, DOD has not demonstrated that the proposed change to the
retirement system, estimated to cost $13 billion in higher accrual
payments to the retirement trust fund and an

increase in Treasury's unfunded liabilities, is a cost- effective
way to improve retention. A number of questions should be explored
before making changes to a system that has been changed only twice
in over 50 years. First, we need to understand what kind of a
retention problem exists. Are retention

problems focused in certain subpopulations or do they apply across
the force? Are current retention problems a transitory result of
such factors as restricted accessions during the recent military
downsizing and a good

economy, or do they signal a long- term decline in the
attractiveness of military careers? Second, we need to understand
the relationship between the retirement system and retention,
particularly its relative importance to servicemembers compared to
other changes that might be made in compensation or quality of
life in the military. Would spending scarce resources to increase
retirement benefits have a greater impact on DOD's

current and long- term manning than spending those resources in
other

Page 14 GAO/T-NSIAD-99-94

areas? Finally, given that the first retirees under Redux will not
reach retirement eligibility until 2006, an opportunity exists to
take a longer term, strategic perspective. Is the current military
compensation system, which is the result of an historical
accretion of elements, best suited for meeting our national
defense manning needs into the 21 st century?

Mr. Chairman, this concludes my prepared statement. We are
prepared to respond to any questions that you or other Members of
the Subcommittee may have.

(703271) Lett er

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

Orders for 100 or more copies to be mailed to a single address are
discounted 25 percent.

Orders by mail: U. S. General Accounting Office P. O. Box 37050
Washington, DC 20013

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

Orders may also be placed by calling (202) 512- 6000 or by using
fax number (202) 512- 6061, or TDD (202) 512- 2537.

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

For information on how to access GAO reports on the INTERNET, send
an e- mail message with info in the body to: info@ www. gao. gov
or visit GAO's World Wide Web Home Page at: http:// www. gao. gov

United States General Accounting Office Washington, D. C. 20548-
0001

Official Business Penalty for Private Use $300

Address Correction Requested Bulk Rate

Postage & Fees Paid GAO Permit No. GI00

*** End of document. ***