Recruiting and Retention in the Active Component Military: Are There Problems?

CRS Report for Congress
Recruiting and Retention in the Active Component
Military: Are There Problems?
February 25, 2002
Lawrence Kapp
Analyst in National Defense
Foreign Affairs, Defense, and Trade


Congressional Research Service ˜ The Library of Congress

Recruiting and Retention in the Active Component
Military: Are There Problems?
Summary
This report provides information on recruiting and retention trends in the active
component of the armed forces since 1989, including recruiting and retention
shortfalls that occurred within the past few years; the congressional and executive
branch response to these shortfalls; and an assessment of the current situation.
Additionally, it contains an analysis of factors that may have an impact on recruiting
and retention in the future, and discusses policy options that could be considered to
minimize any negative effects that these factors might cause.
In recent years, the Military Services have experienced recruiting and retention
shortfalls for their active component forces. The Army, Navy, and Air Force had
trouble meeting their goals for new recruits (also known as “accessions”) in the late
1990s, and all of the Services experienced declines in the quality of their recruits from
the early 1990s to 2000. With respect to retention, the Army, Navy, and Air Force
had difficulty meeting their retention goals for enlisted personnel in the late 1990s,
and the Navy and Air Force are still having some problems today. Also, concerns
have been raised about low retention rates in certain critical specialties and a declining
retention rate among younger officers.
Cited causes of recruiting shortfalls in the late 1990s include the residual effects
of the post-Cold War drawdown, competition with a robust civilian economy,
competition with institutions of higher education, demographic and attitudinal changes
among younger Americans, and a need for more recruiting resources. Commonly
cited causes of retention shortfalls include competition with the civilian economy and
job dissatisfaction due to a variety of factors, including the nature and pace of current
military operations, a lack of critical supplies and equipment, “quality of life” issues,
and changes in the military culture. However, the data used to ascertain the causes
of these shortfalls have often been rather limited.
Congress and the executive branch have initiated or modified a number of
policies in response to these shortfalls. Most notably, Congress provided more money
for recruiters, advertising, enlistment bonuses and re-enlistment bonuses, as well as
increasing military pay and improving retirement benefits. The executive branch
refocused its advertising, provided its recruiters with more and better resources,
launched several programs designed to expand the pool of potential enlistees, and
began working on ways to better manage deployments. Additionally, other policy
changes have been made, most notably with respect to improving military housing
benefits and reimbursements for moves.
These policy changes and other factors appear to have had the desired effect, at
least in the short term. In fiscal years 2000 and 2001, all of the Services met their
quantity goals for new recruits. Recruit quality in fiscal year 2001 was slightly better
than it had been the previous year. With respect to retention, the data indicate
improvements in some areas. In spite of these positive signs, a number of longer term
factors – such as demographic, attitudinal, and lifestyle shifts – could have a negative
impact on recruiting and retention in the future.



Contents
Recruiting ................................................. 1
Background ............................................ 1
Perceived Causes........................................6
The Congressional Response..............................15
The Executive Branch Response............................16
Results ............................................... 17
Factors That May Cause Shortfalls in the Future...............17
Options for Congress....................................19
Retention ................................................. 21
Background ........................................... 21
Perceived Causes.......................................31
The Congressional Response..............................37
The Executive Branch Response............................38
Results ............................................... 40
Factors That Could Cause Shortfalls in the Future..............40
Options for Congress....................................41
List of Tables
Table 1: Enlisted Accessions, Department of Defense....................3
Table 2: Enlisted Accessions, Army..................................4
Table 3: Enlisted Accessions, Navy..................................4
Table 4: Enlisted Accessions, Marine Corps............................5
Table 5: Enlisted Accessions, Air Force...............................5
Table 6: Size of Youth Cohort Ages 18-24, 1980-2000...................8
Table 7: Positive Propensity To Serve on Active Duty Among Those
16-21 Years Old............................................9
Table 8: College Enrollment Rates of High School Graduates, 1989-1999....13
Table 9: Active Enlisted Advertising Expenditure......................14
Table 10: Active Enlisted Recruiters................................15
Table 11: Army Retention Rates For Enlisted Personnel, FY1989-2001.....23
Table 12: Navy Retention Rates For Enlisted Personnel, FY1989-2000.....25
Table 13: Navy Retention Rates For Enlisted Personnel, FY2001..........25
Table 14: Marine Corps Retention Rates For Enlisted Personnel,
FY1989-2001 ............................................. 26
Table 15: Air Force Retention Rates For Enlisted Personnel, FY1989-2001...27
Table 16: Active Duty O3 Officer Continuation Rates by Service,
FY1989-2000 ............................................. 30
Table 17: Active Duty O4 Officer Continuation Rates by Service,
FY1989-2000 ............................................31



Recruiting and Retention in the Active
Component Military: Are There Problems?
This report provides information on recruiting and retention trends in the active
component of the armed forces since 1989, including recruiting and retention
shortfalls which occurred within the past few years; the congressional and executive
branch response to these shortfalls; and an assessment of the current situation.
Additionally, the report analyze factors that may have an impact on recruiting and
retention in the future and explores possible policy options that could address negative
effects that these factors might cause.
This report is broken down into two sections: the first section deals with the
issue of recruiting; the second with retention. Both of these sections are structured
in the following manner:
!Background information on recent shortfalls in recruiting or retention.
!A discussion of perceived causes of the shortfalls.
!A summary of the major congressional and executive branch response to those
shortfalls.
!An assessment of the impact of the congressional and executive responses.
!An analysis of factors that could have a negative impact on recruiting or
retention trends in the future.
!A discussion of policy options that Congress might consider.
Recruiting
Background. Recruiting has been called the life blood of the Military Services.
Without a robust ability to bring new members into the military, the Services would
lack sufficient manpower to carry out mission essential tasks in the near term and
would lack a sufficient pool of entry-level personnel to develop into the mid-level and
upper-level leaders of the future. To protect against this, the individual military
Services set goals for “accessions” or new recruits each year. There are both1
“quantity” goals and “quality” goals. Officer and enlisted goals are set separately.


1This quantity goal is based primarily on the difference between the congressionally authorized
end strength of the Service for a given fiscal year and the projected number of personnel the
Service will actually have at the end of the year. Officer and enlisted accession goals are set
separately. For example, if a Service has an authorized end strength of 200,000 enlisted
personnel in a given year, and it projects that it will retain 175,000 of its current enlisted
members through the year, it will set a goal 25,000 new enlisted recruits for that year. The
actual number of new enlisted recruits it needs, however, may fluctuate throughout the year
as new projections are made about the retention of currently serving enlisted personnel.

As the Services have generally achieved their goals for officer accessions in recent
years, this section will focus on enlisted accessions only.
In FY1998, the Army and the Navy failed to meet their quantity goals for
enlisted accessions, and in fiscal year 1999, the Army and the Air Force failed to meet
their quantity goals as well. This was perceived by some as a serious problem not
only because of its potential impact on the force structure, but also because none of
the military Services had failed to meet their enlisted accession quantity goal since
1979. See Tables 1 through 5 for a breakdown of actual enlisted accessions versus
goals for DOD and for each of the individual Services.
Although the quantity of enlisted recruits is an important measure of recruiting
success, so too is recruit quality. DOD measures enlisted recruit “quality” based on
two criteria: graduation from high school and score on the Armed Forces
Qualification Test (AFQT). Since fiscal year 1993, DOD’s benchmarks for recruit
quality stipulate that at least 90% of new recruits must be high school graduates and
at least 60% must score above average on the AFQT.2 Those recruits who are both
high school graduates and score above average on the AFQT are considered “high
quality recruits.”
Every Service has met or exceeded these recruit quality standards in every year
since they were established; indeed, every Service met or exceeded these standards
in each of the seven years preceding the establishment of the standard. However,
recruit quality peaked in fiscal year 1992 and generally declined up through fiscal year
2000. For example, between 1992 and 2000, the proportion of new recruits with a
high school diploma dropped from 99% to 91%, the proportion who scored above
average on the AFQT dropped from 74% to 66%, and the proportion of recruits who
were deemed “high quality” dropped from 74% to 57%. This trend is even more
pronounced with respect to the Army, which historically has always been the Service
with the greatest recruiting problems. This downward trend line of recruit quality
caused some concern in recent years, especially when coupled with the quantity
shortfalls of fiscal years 1998 and 1999.


2These benchmarks were derived from a model called the Accession Quality Cost Performance
Trade-off Model (CPTM). The CPTM is an optimization model that solves for the “number
and quality mix of recruits predicted to generate a given level of performance and to fill
required positions over the first term of service at the lowest cost...The rationale underlying
the model is relatively straightforward. High-quality recruits – high school diploma graduates
scoring in the top half of the Armed Forces Qualification Test – are more capable in virtually
all military occupations. Their level of performance is higher, and they complete their
enlistment at higher rates than do other recruits. Hence, it takes fewer of them to generate a
given level of performance, and training costs are lower. However, the cost of each additional
high quality recruit increases at a greater rate than for other types of recruits...The CPTM
optimization model trades off the higher recruiting costs, higher expected performance, higher
retention, and lower training costs of the high-quality recruit with the lower recruiting cost,
lower expected performance, lower retention, and higher training costs of recruits from other
quality categories.” Office of the Assistant Secretary of Defense (Force Management Policy)
Report To Congress entitled Review of Minimum Active Enlisted Recruit Quality
Benchmarks: Do They Remain Valid?, March 2000, p. 5-6.

However, FY1992 was not a “normal” year: the resounding victory of the Gulf
War was fresh in people’s minds and the Services were taking in a comparatively low
number of new recruits, allowing them to be more selective. It is also important to
point out that no Service has fallen below DOD’s quality benchmarks. Thus, this
decline in recruit quality may simply be a return to “normalcy.” Additionally, the data
for fiscal year 2001 show slight improvements in the quality of Army, Marine Corps
and Air Force recruits. See Tables 1 through 5 for an overview of recruit quantity
and quality trends between 1989 and 2000. Additional data from 1975, 1980, and

1985 is included for historical perspective.


Table 1: Enlisted Accessions, Department of Defense
Recruit QuantityRecruit Quality
AFQT I-“High
FY Goal Actual Percent HSDG IIIA Quality”

1975 456,600 457,547 100% 65% 62% 45%


1980 388,785 389,861 100% 66% 38% 35%


1985 316,449 316,676 100% 93% 62% 56%


1989 292,021 293,896 101% 92% 64% 58%


1990 229,172 232,315 101% 95% 68% 64%


1991 206,573 206,617 100% 97% 75% 72%


1992 200,168 202,752 101% 99% 74% 74%


1993 206,290 206,927 100% 95% 71% 67%


1994 184,020 184,096 100% 96% 72% 68%


1995 174,806 175,783 101% 96% 71% 67%


1996 185,646 185,987 100% 96% 69% 65%


1997 196,957 197,081 100% 94% 69% 63%


1998 192,309 186,132 97% 94% 68% 63%


1999 195,092 187,180 96% 93% 66% 59%


2000 202,017 202,917 100% 91% 66% 57%


2001 195,324 196,355 101% 93% 66% 61%


Source: Department of Defense, Directorate for Accessions Policy.



Table 2: Enlisted Accessions, Army
Recruit QuantityRecruit Quality
“High
FYGoalActualPercentHSDGAFQT I-IIIAQuality”

1975 209,100 208,915 100% 57% 61% 38%


1980 172,800 173,228 100% 52% 22% 21%


1985 125,300 125,443 100% 91% 61% 52%


1989 119,875 120,535 101% 90% 62% 55%


1990 87,000 89,620 103% 95% 67% 62%


1991 78,241 78,241 100% 98% 80% 78%


1992 75,000 77,583 103% 100% 78% 78%


1993 76,900 77,563 101% 95% 70% 66%


1994 68,000 68,039 100% 95% 71% 66%


1995 62,929 62,929 100% 96% 69% 65%


1996 73,400 73,418 100% 95% 67% 63%


1997 82,000 82,088 100% 90% 68% 58%


1998 72,550 71,733 99% 90% 68% 58%


1999 74,500 68,209 92% 90% 63% 53%


2000 80,000 80,113 100% 90% 65% 52%


2001 75,800 75,855 100% 90% 63% 58%


Source: Department of Defense, Directorate for Accessions Policy.
Table 3: Enlisted Accessions, Navy
Recruit QuantityRecruit Quality
AFQT I-“High
FY Goal Actual Percent HSDG IIIA Quality”

1975 110,000 110,030 100% 74% 61% 49%


1980 97,627 97,678 100% 74% 51% 44%


1985 87,592 87,592 100% 89% 61% 51%


1989 94,286 95,186 101% 90% 57% 48%


1990 72,402 72,846 101% 92% 62% 55%


1991 68,311 68,311 100% 96% 66% 62%


1992 58,208 58,208 100% 98% 68% 66%


1993 63,073 63,073 100% 94% 70% 64%


1994 53,964 53,982 100% 95% 68% 63%


1995 48,637 48,637 100% 95% 66% 61%


1996 48,206 48,206 100% 95% 66% 61%


1997 50,135 50,135 100% 95% 66% 61%


1998 55,321 48,429 88% 95% 64% 60%


1999 52,524 52,595 100% 90% 65% 55%


2000 55,000 55,147 100% 90% 64% 54%


2001 53,520 53,690 100% 90% 63% 53%


Source: Department of Defense, Directorate for Accessions Policy.



Table 4: Enlisted Accessions, Marine Corps
Recruit QuantityRecruit Quality
AFQT I-“High
FY Goal Actual Percent HSDG IIIA Quality”

1975 60,300 60,500 100% 47% 55% 32%


1980 43,684 44,281 101% 70% 39% 35%


1985 36,536 36,620 100% 97% 55% 53%


1989 34,130 34,424 101% 95% 67% 63%


1990 33,521 33,600 100% 95% 66% 62%


1991 30,015 30,059 100% 98% 69% 67%


1992 31,851 31,852 100% 99% 71% 70%


1993 34,802 34,776 100% 97% 68% 66%


1994 32,056 32,056 100% 95% 71% 68%


1995 32,346 33,217 103% 96% 66% 63%


1996 33,173 33,496 101% 96% 65% 63%


1997 34,512 34,548 100% 96% 65% 62%


1998 34,244 34,285 100% 96% 64% 62%


1999 33,668 33,703 100% 96% 64% 61%


2000 32,417 32,440 100% 95% 64% 60%


2001 31,404 31,429 100% 96% 64% 62%


Source: Department of Defense, Directorate for Accessions Policy.
Table 5: Enlisted Accessions, Air Force
Recruit QuantityRecruit Quality
AFQT I-“High
FY Goal Actual Percent HSDG IIIA Quality”

1975 77,200 78,102 101% 87% 69% 63%


1980 74,674 74,674 100% 84% 55% 56%


1985 67,021 67,021 100% 99% 69% 69%


1989 43,730 43,751 100% 99% 84% 82%


1990 36,249 36,249 100% 99% 85% 85%


1991 30,006 30,006 100% 99% 86% 85%


1992 35,109 35,109 100% 99% 86% 85%


1993 31,515 31,515 100% 99% 80% 79%


1994 30,000 30,019 100% 99% 81% 80%


1995 30,894 31,000 100% 99% 84% 83%


1996 30,867 30,867 100% 99% 83% 82%


1997 30,310 30,310 100% 99% 79% 78%


1998 30,194 31,685 105% 99% 78% 77%


1999 34,400 32,673 95% 99% 76% 75%


2000 34,600 35,217 102% 99% 73% 72%


2001 34,600 35,381 102% 99% 75% 74%


Source: Department of Defense, Directorate for Accessions Policy.



Perceived Causes. A number of factors have been offered as the “cause”
of the quantity shortfalls in the late 1990s and the quality levels that, though still
above DOD’s benchmarks and quite high from a historical perspective, were declining
throughout most of the 1990s. Commonly cited causes include the residual effects
of the post-Cold War drawdown, competition with a robust civilian economy,
competition with institutions of higher education, demographic and attitudinal changes
among younger Americans, and insufficient recruiting resources. Such factors may
not be mutually exclusive and are not necessarily competing explanations. Rather,
they may each be seen as partial answers to the question “What caused these
recruiting trends?” Each is listed below and discussed in some detail; however, the
reader should be aware that they are not necessarily listed in order of actual
importance.
Perceived Cause #1: Residual Effects of the Post-Cold War
Drawdown. The end of the Cold War forced the United States to rethink the
security threats it faced and the type of military it needed to respond to those threats.
As a result of this re-evaluation, Congress substantially reduced the size of U.S.
military forces, primarily between fiscal years 1991 and 1996. This “drawdown” was
accomplished in part by offering military personnel early retirement and voluntary
separation bonuses, as well as by involuntarily separating a small number of those who
did not take advantage of these voluntary separation incentives. However, a key tool
in managing the drawdown was a large reduction in the number of new recruits
brought into the Services. As one report noted, “During the drawdown,
accessions...were deliberately cut below the level needed to sustain the force. This
avoided having to induce even more people to leave the military who otherwise might3
want to stay.”
Because accessions were cut below the level needed to sustain the force, a
substantial increase in the number of accessions was inevitable in subsequent years.
As Table 1 indicates, goals for accessions for all the Services dropped from 292,021
in FY1989, the last pre-drawdown year, to a low of 174,806 in FY1995. In FY1996
and subsequent years, the goals for accessions began increasing, to 202,017 in
FY2000. To put it another way, between FY1995 and FY2000, the Services’ demand
for new recruits increased by 16%. For the Army, the change was even more
pronounced: between FY1995 and FY2000, its need for new recruits grew by 27%.4
This increase might not have been a problem for the Services if it had not been
compounded by a variety of other factors that decreased the size of its primary
recruiting pool and inhibited recruiting effectiveness. These factors are discussed
below.


3Military Recruiting Outlook: Recent Trends in Enlistment Propensity and Conversion of
Potential Enlisted Supply, 1996, RAND, p.1. Of all the services, the Air Force utilized this
strategy the most. See David Grissmer and Bernard D. Rostker, “Military Personnel in a
Changing World,” in 1991-1992 American Defense Annual, Lexington Books, 1992, 132-

135.


4Note also that, during the drawdown, the Marine Corps’ was the Service whose strength was
decreased the least. As a result, it was able to maintain fairly constant accession levels
throughout the drawdown and subsequently. It is also the only Service that consistently met
its recruiting goals throughout the 1990s.

Perceived Cause #2: Demographic and Attitudinal Changes
Among Young Americans. The vast majority of new recruits are young men,
usually those who have just graduated from high school and those who have spent a
few years in a civilian job or in college. Thus, the primary demographic group
targeted by recruiters is 18-24 year old males. Since 1981, the size of this age cohort
has been decreasing, the product of declining fertility rates in the 1960s and 1970s
(the “baby bust”). In the late 1990s, there were roughly 1 million fewer 18-24 year
old men than there were in 1989. The size of the 18-24 year old male cohort reached
its lowest point in 1996, and has been gradually increasing since then (See Table 6).
Population projections indicate that this cohort will continue to increase in size until5

2013, at that point it will begin to decline again.


Women make up a small but significant proportion of new recruits. In 1999,
they made up 18% of enlisted accessions. The female 18-24 year old cohort declined
from 1981 to 1996 and has been gradually increasing since then. It, too, is expected
to increase in size until 2013, at that point it will begin to decline again.
In addition to the smaller size of the 18-24 year old cohort during the 1990s,
attitudinal changes among younger Americans also made it more difficult for the
Services to recruit during that time-frame. Until 1999, the Department of Defense
conducted an annual survey known as the Youth Attitude Tracking Study (YATS),
which measured the attitudes of youth between the ages of 16 and 24.6 One of the
key types of questions asked in that survey concerned the individual’s “propensity to
serve” or willingness to consider joining the military. The correlation between
propensity to serve and the likelihood of eventually joining the military is quite strong.
One study compared the propensity to serve of high-quality youth with actual
enlistment rates and found that among those with a highly positive propensity to
serve, 34.4% actually enlisted; among those with a lower, but still positive, propensity
to serve, 19% actually enlisted; and among those with a negative propensity to serve,
only 5.1% actually enlisted.7 Thus, changes in a cohort’s propensity to serve can
make it easier or harder for the Services to recruit.
Positive propensity to serve among 16-21 year old males dropped significantly
between 1989 and 1999. In 1989, the last pre-drawdown year, 33.7% of young men
(16-21 years old) indicated a positive propensity to serve. That number generally
declined in subsequent years, although it jumped up in 1991, in the aftermath of the
Gulf War victory, and showed a small increase in 1995. Positive propensity to serve
also showed an increase in 1999, with 28.6% of young men indicating a positive


5U.S. Census Bureau, Population Division, Population Estimates Program, for years 1980-
2000. U.S. Census Bureau, Population Division, Population Projections Program for years

2001 and later.


6In 1999, YATS was replaced by a trimester quick-turn polling process.
7Military Recruiting Outlook, Recent Trends in Enlisted Propensity and Conversion of
Potential Enlisted Supply, RAND Corporation, 1996, 10-11. Note, however, that even
though those with a negative propensity to serve have a very low likelihood of joining the
military, they make up about half of each year’s accessions. This is due to the very large size
of the “negative propensity” group, which constitutes the great majority of the target
population.

propensity to serve. A similar but less pronounced trend in positive propensity to
serve occurred among young women, although their level of positive propensity to
serve has historically been much lower than that of young men. Additionally, in
recent years their level of positive propensity to serve has returned to its pre-
drawdown level. The trend in positive propensity to serve for 22-24 year old males
and females has generally tracked that of their younger peers, but at a lower level.
Table 6: Size of Youth Cohort Ages 18-24, 1980-2000
(In Thousands)
YearMales 18-24Females 18-24Total18-24
1980 15,101 15,002 30,103
1981 15,192 15,053 30,245
1982 15,165 14,994 30,159
1983 15,070 14,854 29,924
1984 14,860 14,601 29,461
1985 14,596 14,306 28,902
1986 14,283 13,944 28,227
1987 14,032 13,660 27,692
1988 13,888 13,467 27,355
1989 13,817 13,338 27,155
1990 13,679 13,155 26,834
1991 13,418 12,935 26,353
1992 13,240 12,736 25,976
1993 13,111 12,629 25,740
1994 12,931 12,466 25,397
1995 12,793 12,320 25,113
1996 12,662 12,180 24,842
1997 12,737 12,243 24,980
1998 12,995 12,482 25,477
1999 13,276 12,736 26,012
2000 13,658 13,090 26,748
Source: U.S. Census Bureau, Population Division.



Table 7: Positive Propensity To Serve on Active Duty Among
Those 16-21 Years Old
AfricanHispanicTotalWhiteAfricanHispanic Total
Year WhiteMales American Males Males Females American Females Females
Males Females
1989 28.6 54.2 46.1 33.7 8.9 32.2 24.6 14.2
1990 27.2 41.9 47.0 32.3 8.6 23.9 20.9 13.2
1991 28.5 49.2 45.9 34.3 9.3 28.2 28.4 15.2
1992 24.7 35.6 45.3 29.4 8.6 16.2 22.2 12.0
1993 24.7 36.8 42.1 29.4 7.5 23.4 20.7 12.0
1994 22.2 31.6 39.4 26.3 8.8 20.1 24.9 13.4
1995 22.5 32.4 43.6 27.5 7.4 24.3 25.1 13.5
1996 20.4 33.6 42.7 26.5 9.2 23.4 25.0 14.4
1997 21.2 34.2 36.8 26.2 7.0 19.0 20.5 11.6
1998 20.2 29.6 43.7 26.2 7.2 23.0 25.9 13.4
1999 22.0 35.7 46.1 28.6 9.1 28.9 22.4 15.0
Source: Department of Defense, Youth Attitude Tracking Study.
The reasons for the decline in propensity to serve during the 1990s are not
entirely clear. Some have attributed it to the low advertising budgets of the military
during the drawdown years; others have attributed it to the declining number of
veterans among youth “influencers” such as parents, teachers, and guidance
counselors. While plausible, these accounts do not adequately explain why the decline
has been heavily concentrated among one specific group: African American males.
African Americans have long had a higher propensity to serve in the military than
whites or Hispanics, so this trend may simply be a normalization of African American
attitudes towards the military. Nonetheless, if such a decline were to persist, it would
have troubling ramifications for the military’s ability to recruit, as it has long
benefitted from the relatively high propensity to serve of African American men to fill
its ranks. The propensity of this group to serve did rebound in 1999.
Perceived Cause #3: A Robust Civilian Economy. Traditionally, the
primary recruiting pool for enlisted personnel has been young men8 who have
graduated from high school within the past few years. Essentially, these people have
three options available to them: enlist in the military, work in a civilian job, or attend


8About 80% of recruits are male. The age breakdown of new recruits for the military in
FY1999 was as follows: age 17 (6.1%), age 18 (34.3%), age 19 (22.2%), age 20 (11.6%), age

21 (7.4%), age 22 (5.1%), age 23 (3.5%), age 24 (2.6%), older than 24 (7.2%). Source:


Office of the Assistant Secretary of Defense (Force Management Policy), Population
Representation in the Military Services, November, 2000, p. 2-10.

college.9 The decision among these options is based on a number of factors, one of
which is a rough comparison of the costs and benefits that could be derived from each
choice. When the civilian economy is “booming,” the anticipated costs of pursuing
a civilian job decrease while the anticipated benefits increase. One measure of the
relative attractiveness of civilian employment versus military employment is the
unemployment rates. The higher the unemployment rate, the lower the chance an
individual has of securing and keeping a civilian job, and this in turn makes military
employment more attractive. Historical data indicates that the unemployment rate in
the United States was generally falling from 1992 to 2000.10
Another measure of the attractiveness of civilian employment versus military
employment is relative levels of compensation. If military compensation is lower than
civilian compensation, military service becomes a less attractive option, and vice
versa. However, compensation levels can be difficult to quantify accurately. For
example, the valuation of employer provided benefits can differ markedly depending
on the methodology used. Even seemingly simple comparisons, like pay rates, can be
complicated due to the differing structure of military and civilian pay. For example,
military pay includes not only basic pay, but allowances for housing and subsistence,
and tax advantages as well. Some members of the military receive bonuses and
special pays too. As a result, estimations of average military pay levels can vary
substantially depending on how these different elements are accounted for.
An alternative to comparing average military and civilian pay is to compare rates
of pay growth over time. Typically, military basic pay is the variable used to calculate
changes in military pay levels and the Employment Cost Index (ECI)11 is the variable
used to calculate changes in civilian pay levels. Using this methodology, if civilian
pay grew by 25% over a given time period and military pay grew by 15% over the
same time-frame, one might conclude that military pay was becoming less competitive
relative to civilian pay during that period. This methodology has been widely used
to create a measure of pay competitiveness during the past two decades and, during
the 1990s, it indicated a “pay gap” of about eleven to 13% in favor of civilian1213


employment. It also indicated that the gap grew as the decade progressed.
9Kilburn, Rebecca and Jacob Alex Klerman, Enlistment Decisions in the 1990s: Evidence
from Individual Level Data, RAND Corporation, 1999.
10The unemployment rate in 1992 was 7.5%; from there it dropped steadily to 4.0% by 2000.
11The Employment Cost Index used measures changes in private sector wages and salaries
12Congressional Budget Office report entitled What Does The Military “Pay Gap” Mean,
June 1999, by Richard L. Fernandez, figure 2.
13A likely reason for this increase in the “pay gap” in the 1990s was the way in that military
pay raises were calculated for much of that decade. Since 1967, federal law has provided that
military pay be adjusted upward, at the same time and by the same percentage, as the average
increase in the pay rates for federal General Schedule (GS) employees. In 1990, Congress
enacted a statutory formula for federal civilian pay raises that provided for an annual increase,
beginning in fiscal year 1992, of 0.5% less than the annual increase in the employment
compensation index (ECI). Thus, by statute, military pay raises were supposed to be
determined by the an “ECI minus 0.5%” formula. As the “pay gap” formula is based on a
(continued...)

However, the methodology used to create this measure has serious shortfalls as well,14
including the previously mentioned problem of which elements of military pay to
count as “pay.” Indeed, some studies have disputed the existence of a “pay gap,” or
have found no link between the pay gap as it is usually measured and recruiting or
retention.15 However, whether this “pay gap” actually existed, or was as large as
claimed, was to a certain extent overshadowed by the wide media coverage of the
issue, which often reported the existence of a “pay gap” as though it were an
established fact.
Thus, although the data on relative pay levels is ambiguous, it seems fair to say
that military pay was widely perceived as being lower than civilian pay during the
1990s. Coupled with the previously mentioned data on unemployment rates, this
lends some support to the argument that civilian jobs became an increasingly
attractive option to recent high school graduates during the 1990s, thus increasing
competition for the pool of potential enlistees.
Perceived Cause #4: Increase in College Bound Youth. As
mentioned above, the military’s primary recruiting pool for enlisted personnel has
traditionally been young men who have graduated from high school within the past
few years. These individuals either enlist in the military, work in a civilian job, or
attend college. The latter choice has become increasingly popular in recent years. In
1980, just under half (46.7%) of male high school graduates went straight on to
college. That proportion rose to 57.6% in 1989 and peaked at 63.5% in 1997. Since
then, the proportion has tapered off slightly to 61.4% in 1999.16 Among women, a
similar trend is evident, although women have gone straight from high school to
college at a consistently higher rate than men in every year but one since 1988.
In part, this trend is due to the perceived benefits of a college education: better
employment prospects, higher wages, and higher social status. Nor is this perception
misguided. One study showed that, in 1998, the earnings of those who had completed
at least a bachelor’s degree were dramatically higher than those who had only a high
school education. For men, the “college premium” was 56% – that is, men with at
least a bachelor’s degree earned 56% more than those with just a high school diploma
– while for women, it was 100%. Additionally, this college premium has increased


13 (...continued)
comparison of the ECI and military basic pay, it is not surprising that the “pay gap” grew.
14For a more extensive discussion on the difficulty of comparing military and civilian
compensation generally, and pay specifically, see CRS Issue Brief IB10089, Military Pay and
Benefits: Key Questions and Answers, by Robert Goldich, p. 8-10. See also the CBO report
entitled What Does The Military “Pay Gap” Mean?, by Richard L. Fernandez.
15See Eugene J. Devine, The Myth Of The Military Pay Gap: Comparison of Military and
Civilian Earnings Levels, paper presented at the Conference of the Western Economic
Association International, July 1, 1996; and the CBO report entitled What Does The Military
“Pay Gap” Mean?, by Richard L. Fernandez. See also Andrew Compart, “Study Casts
Doubt On Military pay Gap,” Navy Times, October 6, 1997, 8.
16U.S. Department of Commerce, Bureau of the Census, Current Population Reports, Series
P-20.

fairly steadily over the past two decades. In 1980, for example, the college premium
for men was only 19%, while it was 52% for women.17
An additional reason for the increasing rates of college attendance may be found
in the decreasing financial barriers to attending college. Although it is true that the
costs of attending college have grown significantly in recent years, this has been
accompanied by an increasing number of ways to finance the costs. Students now
have access not only to scholarships, grants, and student loans, but are increasingly
able to get tuition reimbursement from their employers or to receive financial
assistance in exchange for some type of public service other than active military
service. Since World War II, the premier educational assistance package for public18
service has been the GI Bill, but the GI Bill now has competition from other sources.
As military sociologist Charles Moskos points out:
“...we now spend over $15 billion a year on Federal grants and loan subsidies to
college students. In other words, we’re paying people not to serve their country.
We have created...a GI Bill without the GI. In fact, if you have a GI Bill, if you
get the Montgomery GI Bill, colleges will deduct other student aid. They’ll count
the GI Bill as part of their income and therefore reduce student aid you might19
otherwise get.”
In addition to federal educational grants, the GI Bill also faces competition from
federal programs like AmeriCorps, which provides participants with educational
grants of up to $9,450 upon completion of service, and from various state National
Guards, which often offer free tuition at state schools as a benefit of membership in
addition to the scaled down version of the GI Bill available to all reservists.
Additionally, it is not uncommon today for private businesses to provide educational
benefits to their employees.


17U.S. Department of Education, National Center for Education Statistics, The Condition of
Education 2000, NCES 2000-062 (Washington, D.C.: U.S. Government Printing Office)
June, 2000, 34.
18“GI Bill” is a generic term used to describe the veteran’s educational benefits authorized by
various acts of Congress since 1944. The current GI Bill is formally known as the
Montgomery GI Bill and it provides up to 36 months of educational benefits to eligible
veterans for college or certain other educational programs. The monthly benefit for an
individual varies depending on a variety of factors, but the basic benefit for full time students
at eligible educational institutions was $672 per month in 2000. As a result of provisions in
the Veterans Education and Benefits Expansion Act of 2001 (P.L. 107-103, section 101), the
benefit was increased to $800 per month on January 1, 2002, and will increase to $900 per
month on October 1, 2002 and to $985 per month on October 1, 2003.
19Speech by Dr. Charles Moskos, Professor of Sociology at Northwestern University,
delivered to the 121st National Guard Association of the United States General Conference,
Sept. 5, 1999.

Table 8: College Enrollment Rates of High School Graduates,
1989-1999
(Percent of Total)
Year Total Male Female
1989 59.6% 57.6% 61.6%
1990 59.9% 57.8% 62.0%
1991 62.4% 57.6% 67.1%
1992 61.7% 59.6% 63.8%
1993 62.6% 59.7% 65.4%
1994 61.9% 60.6% 63.2%
1995 61.9% 62.6% 61.4%
1996 65.0% 60.1% 69.7%
1997 67.0% 63.5% 70.3%
1998 65.6% 62.4% 69.1%
1999 62.9% 61.4% 64.4%
Source: Department of Education, Digest of Educations Statistics: 2000
The data on college attendance certainly supports the argument that competition
from higher education decreased the supply of potential recruits in the 1990s. The
historically high proportion of people enrolling in college immediately after high
school posed a direct challenge to the military’s enlisted recruiting efforts, which has
traditionally focused on recruiting recent high school graduates. Put simply, with
more high school graduates choosing to go straight to college, fewer potential recruits
were in the military’s primary recruiting pool.
Perceived Cause #5: Insufficient Recruiting Resources. The
preceding paragraphs point to both an increase in the military’s demand for recruits
and a decrease in the available supply of potential enlistees as the fundamental causes
of the military’s recruiting shortfalls in the mid to late 1990s. However, another
factor certainly bears examination: the resources available to the Services to achieve
their recruiting goals. Most agree that these resources – such as advertising,
recruiters, enlistment bonuses, and educational benefits – have historically been
valuable tools for convincing potential recruits to sign an enlistment contract. While
it would be impossible to specify the “correct” level of these resources, it appears that
the levels were not adequate for the challenging recruiting environment that emerged
as the 1990s progressed.
As mentioned previously, a key tool the military used in managing the drawdown
was a large reduction in the number of new recruits brought into the Services. Based
on this policy of reduced accessions, the military spent less on recruiting related
expenditures during the drawdown period. For example, as Table 9 indicates, the
total advertising expenditures for all the Services fell by over 50% between 1989 and



1993. The number of recruiters employed by the Services also dropped between 1990
and 1994 by about 10%, as indicated in Table 10. This aggregate figure, however,
masks the fact that the Army reduced the number of recruiters it employed by 15%,
while the Navy cut its recruiting force by 25%. The Marine Corps, it should be
pointed out, actually increased the number of recruiters it employed between 1990
and 1994 by nearly 20%.
Table 9: Active Enlisted Advertising Expenditure
(Constant FY2001 Dollars, in Millions)
Fiscal Army Navy Air Marine TOTAL
Year Force Corps
1989 76.5 20.1 11.3 18.0 125.9
1990 73.0 26.0 7.6 17.2 123.9
1991 49.4 15.2 3.7 10.8 79.1
1992 41.8 11.6 4.6 11.4 69.4
1993 34.3 13.9 5.2 11.5 64.9
1994 45.2 33.6 6.3 11.3 96.4
1995 57.9 39.8 11.5 10.7 120.0
1996 72.8 42.7 10.5 15.2 141.2
1997 98.0 38.2 12.0 21.3 169.5
1998 97.8 63.8 12.3 27.2 201.1
1999 102.7 67.3 60.5 32.1 262.6
2000 106.3 64.7 49.7 33.0 253.7



Table 10: Active Enlisted Recruiters
Fiscal Army Navy Air Marine TOTAL
Year Force Corps
1989 N.A. N.A. N.A. N.A. N.A.
1990 6,228 4,184 1,175 2,136 13,723
1991 5,860 4,337 1,132 2,388 13,717
1992 5,938 3,842 1,178 2,325 13,283
1993 5,577 3,702 1,151 2,325 12,755
1994 5,307 3,132 1,153 2,550 12,142
1995 5,771 3,408 1,010 2,550 12,739
1996 5,774 3,450 1,010 2,550 12,784
1997 5,772 3,492 1,017 2,550 12,831
1998 5,691 3,561 1,068 2,650 12,897
1999 5,961 5,074 1,364 2,650 14,753
2000 6,257 4,931 1,426 2,650 15,202
Recruiting resources increased steadily between 1994 and 1998, and rose
dramatically after 1998, but some believe that the decline in recruiting resources in the
early part of the decade had a negative long term affect on recruiting. According to
the Department of Defense, “With the military drawdown, we cut advertising
budgets... and reduced the number of people that we enlisted.... We believe that this
created a perception that the military was no longer hiring. We also believe that by
not advertising during the middle of the 1990s we missed the opportunity to
communicate with people who are now enlistment age.”20
The Congressional Response. Congress has expressed concern about
recruiting trends since at least 1994. The committee report that accompanied H.R.

4301, the National Defense Authorization Act for Fiscal Year 1995, stated:


The committee has heard compelling testimony from [the] Department of Defense
and Service witnesses that the armed forces are experiencing difficulty in meeting
recruiting objectives. If left unchecked, these trends will ultimately require the
Services to access lower quality recruits than those that fueled the resurgence of21


the armed forces in the 1980s.
20Response provided by the Directorate for Accessions Policy, Office of the Assistant
Secretary of Defense (Force Management Policy), in response to questions asked by the
author of this report.
21H.Rept. 103-499, Report of the Committee on Armed Services on H.R. 4301, the National
Defense Authorization Act for Fiscal Year 1995, May 10, 1994, 184.

Similar concerns are found in the committee reports accompanying many subsequent
Defense Authorization bills through FY2001. The level of concern evidenced in these
reports increased substantially in FY1998 and FY1999, when several of the Services
failed to meet their accessions quantity goals.
Congress has responded to the military’s recruiting challenges in a number of
ways. The most direct way has been to increase the amount of money available for
activities related to recruiting, such as advertising (see Table 9), enlistment bonuses,
and educational benefits. It has also authorized substantial pay increases for the
military.22 Additionally, Congress has taken action to preserve recruiters’ access to
students in the following ways: by prohibiting almost all colleges that accept federal
funds (except student aid) from discriminating against recruiters on campus; by
prohibiting secondary schools that receive certain types of federal assistance from
discriminating against recruiters on campus; and by requiring those secondary schools
to provide military recruiters with access to the names, addresses and phone numbers
of their students, unless the student or parent of the student objects.
Less directly, Congress has also funded an increasing number of Junior Reserve
Officer Training Corps (JROTC) programs, which are located at high schools.
Although recruiting is not the purpose of JROTC, and its members are in no way
legally affiliated with the armed forces, it may instill participants with a positive
attitude towards the military. The Army Times, for example, reported that “more
than half of the students who participated [in JROTC] become affiliated with the23
military in some fashion.” Nonetheless, the extent to which JROTC instills positive
attitudes towards the military, as opposed to simply attracting people who already
view military service positively, is unclear.
The Executive Branch Response. The executive branch has taken a
number of actions designed to mitigate its recruiting shortfalls. The Department of
Defense has conducted more market research on both potential enlistees and their
“influencers” (e.g. parents, teachers, and guidance counselors), and has used this
information to refocus its advertising campaigns. Thus, most of the Services have
recently devised new advertising “messages” to appeal to the attitudes of young
Americans and have developed a stronger presence on the Internet, a popular medium
of communication for these potential recruits.
Significantly, the military has also begun to rethink its strategy of recruiting
enlistees primarily from the ranks of recent high school graduates. Increasingly, the
Services are implementing programs that are designed to appeal to college minded
students. For example, the Army’s “College First” program pays a stipend to people
while they are in college in exchange for enlisted service after graduation. Recruiters
are also working with colleges and universities to contact those students who “drop
out” (leave college with no intention of returning) or “stop out” (leave college with
the intention of coming back in the future). The Army has also launched a pilot
program known as “GED Plus,” which helps high school drop-outs with exceptionally


22For more information on military pay raises, see CRS Issue Brief IB10089, Military Pay
and Benefits: Key Questions and Answers, by Robert L. Goldich.
23Army Times, “Congress Sets Own Goals For Recruiting,” June 7, 1999, p. 16.

high scores on the AFQT to complete their General Educational Development (GED)
credential and then enlist in the Army.24
Results. The response of Congress and the executive branch over the past few
years, together with other factors such as a slackening economy, has yielded some
success. In FY2000, for the first time in three years, all Services met their recruiting
goals for both quantity and quality, and this success was repeated in FY2001.
Additionally, recruit quality improved slightly for most of the Services in FY2001.
It seems fair to conclude that the increased recruiting budgets, improved pay and
benefits, and modified recruiting strategies have had a positive effect on recruiting.
Some of this improvement can also be attributed to factors largely unrelated to new
recruiting initiatives: for example, the growing youth cohort, an economic slowdown,
a slight drop off in college enrollment rates, and the hard work of individual military
recruiters.
Factors That May Cause Shortfalls in the Future. Despite the recent
improvements in meeting recruiting goals, a number of factors have the potential to
cause recruiting shortfalls in the future. These factors – changes in the military’s
recruit quality needs, demographic changes, and continuing competition with
institutions of higher education – have the potential to significantly reshape the
recruiting environment. Additionally, a resurgent economy and further declines in
“propensity to serve” could cause recruiting shortfalls as well.
Factor #1: Changing Recruit Quality Needs. Assuming that military
personnel requirements (i.e. end strengths) are not significantly increased in future
years, and assuming stable retention, the military’s demand for new recruits will not
increase significantly over the number of accessions required in FY2001. However,
the type of recruit needed may change dramatically over the next twenty years. Much
attention has been paid to the “revolution in military affairs” (RMA), a military
concept based on the premise that the integration of sophisticated information
technologies into military operations will transform the way in which the United
States fights its wars. Typically, discussions of RMA center on new types of
“networked” combat systems or on new tactics that can be employed while utilizing
these systems. Less frequently mentioned is the type of people who will be required
to man these systems and implement the tactics. They may need to have higher
cognitive ability than is presently required, and this could reduce the pool of people
qualified to enlist.25 Additionally, as these “high cognitive ability individuals” are
highly prized by corporations and institutions of higher education, competition for
them will be fierce.


24Allowing more recruits with GEDs into the Services is a controversial issue. See the section
entitled “Revised Recruit Criteria” later in this report.
25On the other hand, technology may simplify certain military tasks in such a way as to reduce
the need for high cognitive ability. For example, a test conducted in 1984 showed that the
performance differential between upper AFQT tank gunners and lower AQFT tank gunners
was lower on the new, more technologically sophisticated M-1 tank than on the older M-60
tank. Barry L. Scribner, D. Alton Smith, Robert H. Baldwin, and Robert W. Phillips, Are
Smart Tankers Better Tankers: AFQT and Military Productivity, Office of Economic and
Manpower Analysis, United States Military Academy.

Factor #2: Changing Demographics. Another important factor is the
changing demographics of the youth cohort. As indicated previously, the size of the
18-24 year old cohort is expected to expand in size through the year 2013. This is
good news for recruiters, as it means there is a larger pool of potential enlistees from
which to recruit. Additionally, as this cohort grows in size, it is also expected to have
an increasing proportion of Hispanics, who currently have the highest “propensity to
serve” of any major ethnic or racial group. This too could be good news for
recruiters, if it creates a primary recruiting pool with an increasingly positive attitude
towards military service. However, the increasing representation of Hispanics in the
youth cohort also has a drawback: one of DOD’s key measures of recruit quality is
possession of a high school diploma and Hispanics currently have a much higher high
school drop-out rate than other major groups. In 1999, for example, the drop-out
rate among Hispanics was 29%, compared to 7% for whites, and 13% for African
Americans.26 Thus, unless the high school graduation rate increases for Hispanics or
DOD changes its definition of recruit “quality,” the benefit from the increase in cohort
size may be less than proportional.
Factor #3: Competition With Institutions of Higher Education.
Competition with colleges appears to be one of the key recruiting challenges of the
future. As mentioned previously, the “wage premium” awarded to college graduates
is large and growing. Today, nearly two of every three high school students go
straight on to college after graduating from high school27 and that proportion may
increase in the future. This changing educational pattern will almost certainly require
DOD to shift its primary recruiting pool from recent high school graduates to those
with at least some college.28 Such a shift would pose significant challenges.
Recruiting incentives, for example, would need to be adapted from targeting those
with future plans for college toward targeting those with immediate plans for college,
those who are already in college, and those who have recently left college. Recruiters
would also need to be retrained to market the benefits of military service to people
whose career plans are substantially different than the typical high school graduate.
Additionally, colleges might be more reluctant to cooperate with recruiters than high
schools have been, as colleges might perceive recruiting efforts as attempts to “steal”
their undergraduates.
Factor #4: Competition with civilian employers. The United States is
currently in the midst of an economic slowdown, which started in early 2001. This
actually benefits military recruiting, as it reduces the competition it faces from civilian
employment opportunities. However, the economy tends to follow a cyclical trend,
so it is likely that the currently sluggish economic conditions will be replaced in the


26U.S. Department of Education, The Digest of Education Statistics 2000 (Washington, D.C.:
National Center For Education Statistics, 2001), Table 106.
27U.S. Department of Education, The Digest of Education Statistics 2000 (Washington, D.C.:
National Center For Education Statistics, 2001), Table 185. See Table 8 for historical data
on college enrollment rates of high school graduates.
28In FY1999, the percentage of new recruits with college experience was 6.6%. Office of the
Assistant Secretary of Defense (Force Management Policy), Population Representation in
the Military Services, Fiscal Year 1999, table 2.7.

not too distant future by a more robust economy. Such a revival could create
challenges for military recruiting like the economic boom of the 1990s did.
Factor #5: Changing Attitudes Towards Military Service. It appears
that the decline in “propensity to serve” that began in 1992 reached its low point in
1997 and 1998 (See Table 7). The data from 1999 show an increase in this key
predicator of willingness to serve in the military, and this upward trend may continue
in light of the growing proportion of Hispanics in the youth population. Such a trend
would be good news for the military Services. However, if a downward trend recurs
in future years, it would have a negative effect on recruiting.
Options for Congress. Should any of these factors manifest themselves in
the future, Congress may consider legislative remedies. Below is a list of policy
options that could address one or more of the factors mentioned above.
Increase Recruiting Resources. Historically, the first response to a
difficult recruiting environment has been to increase the amount of funds available for
recruiting operations; specifically, funds for advertising, enlistment bonuses, and
increasing the number of recruiters on duty. Experience has shown that this response
is usually an effective way of improving recruiting within a relatively short time
frame,29 and the amount of funding committed can be scaled to match the scope of the
recruiting shortfall. This approach likely would be effective if it appeared that
recruiting shortfalls were principally caused by increased recruit quality demands,
competition with higher education, or competition with civilian employers (factors

1, 3, and 4 above).


Revised Recruit Quality Criteria. This response might be considered if it
appeared that recruiting shortfalls were caused by changing demographics, specifically
the increased representation of a group that has a relatively low rate of high school
completion (factor 2 above). As mentioned earlier, completion of high school is one
of the two key indicators that DOD uses to assess recruit “quality.” It is important
to point out, however, that the primary reason that DOD uses graduation from high
school as an indicator of recruit quality – rather than a GED – rests on the strong
statistical link between completion of high school and completion of one’s enlistment
contract. In other words, those who graduate from high school are substantially more
likely to complete their enlistment than those who did not complete high school but
who have a GED.30 As a measure of quality then, high school graduation is less one
of intelligence or knowledge and more one of perseverance. Admittedly,
perseverance is a critical factor in recruit quality, for failure to complete an enlistment


29Author’s conversation with Bernard Rostker, former Under Secretary of Defense for
Personnel and Readiness.
30According to the Department of Defense, “About 80% of recruits with a high school diploma
will complete their first three years, whereas only about half of those who failed to complete
high school will make it. Those holding an alternative credential (e.g. General Educational
Development [GED] high school equivalency certificate) fall between those extremes.” Office
of the Assistant Secretary of Defense (Force Management Policy), Report to Congress,
Review of Minimum Active Enlisted Recruit Quality Benchmarks: Do They Remain Valid?,
March, 2000, p. 1

contract results in higher training costs for the military, undermanned units, and
personnel turbulence. However, there may be alternative ways by which an
individual’s “perseverance” could be measured. Currently, DOD is testing just such
an alternative measure in the form of a questionnaire known as the “Assessment of
Individual Motivation.” If it proves to be an accurate gauge of an individual’s
motivation, it could be coupled with a GED as a substitute for the high school
graduation requirement without lowering recruit quality. Congress, which has
generally been skeptical about suggestions to alter recruit quality standards, may
decide to closely review any such changes.
Expanding and Restructuring Military Educational Programs. This
response could be explored if it appeared that recruiting shortfalls were caused by
increased recruit quality demands or by increased competition with institutions of
higher education (factors 1 and 3 above). The military’s extensive educational
benefit programs are primarily structured to attract those who want to go to college
eventually, but not to attract those who want to go to college immediately after high
school, those who are already in college, or those who have completed at least some
college.
For example, the principal educational benefit offered to enlistees is the
Montgomery GI Bill. The Montgomery GI Bill provides a substantial educational
benefit to enlistees in return for their service, but until recently they could only use it
after they fulfilled their enlistment contract. (Even so, most service members do not
use their benefits until after they leave the military). The military also has an extensive
network of colleges that service members can attend while on active duty. The cost
of the courses is largely or entirely paid for by the military. Although it is possible to
complete an associate’s, bachelor’s, or even a graduate degree through these
“concurrent education” programs, it can be difficult for service members to take full
advantage of them because military duties can easily conflict with academic pursuits.
Recently, some of the Services have implemented student loan repayment programs
for those who have already attended college and are willing to enlist. There has also
been some experimentation with “college first” type programs, where the military
provides a stipend to individuals while they attend college in exchange for military
service after completion of college. At present, however, these programs are
relatively small in size and not offered by all Services.
If the Services want to expand their pool of recruits – to include those who are
motivated to go to college immediately after high school, those already in college, and
those who have some college education already – they could request Congress to
consider a restructured educational benefits program. Expansion of the student loan
repayment program and “college first” type programs would be options for doing this.
Another option might be to integrate the military’s “concurrent education” programs
more closely into an individual’s enlistment contract. For example, the Services might
guarantee an enlistee that in exchange for a 6-year enlistment, he or she would be
given enough “educational leave” during the course of the enlistment to complete an
associate’s degree. Incentives such as full military pay and benefits while attending
a college that accepted military training as partial credit towards an associate’s degree
could be considered. Tuition and fees might either be paid for out of the individual’s
Montgomery GI Bill benefit, or directly by the military.



More Generous Pay and Benefits. This approach could be effective if it
appeared that recruiting shortfalls were principally caused by increased recruit quality
demands, competition with higher education, or competition with civilian employers
(factors 1, 3, and 4 above). Potential recruits, especially the highest quality recruits,
have options other than military service, including civilian employment and higher
education. In confronting these options, individuals make a rough comparison of the
costs and benefits that could be derived from each choice. Increased pay and benefits
for military service would make the choice of enlisting more attractive. The
magnitude of the increase could be scaled to the magnitude of the recruiting shortfall.
Thus, modest recruiting shortfalls could be countered with modest increases in pay
and benefits, while major shortfalls would require more substantial increases.
However, this is usually a very expensive option, as increased pay and benefits usually
apply to the entire force rather than just new recruits. The costs of this approach also
continue even if competition for recruits wanes. Most analysts believe that enlistment
bonuses, which can be targeted solely towards new recruits, are a more efficient way
of improving recruiting than across-the-board improvements in pay or benefits.
Retention
Background. Retention in the military refers to the rate at which military
personnel voluntarily choose to stay in the military after their original obligated term
of service has ended.31 Imbalances in the retention rate can cause problems within the
military personnel system. If too few people stay in, the military will suffer from a
lack of experienced leaders, decreased military efficiency and lower job satisfaction.
If too many people stay in, promotion opportunities decrease and a higher percentage
of people must be involuntarily separated in order to prevent the organization from32
becoming “top heavy” with middle and upper level leaders.
In recent years, concerns have been raised that retention rates in the military
were too low in three areas: the retention rate of enlisted personnel generally, the
retention rate of enlisted and officer personnel in critical specialties, and the retention
rate of officers in certain paygrade. Retention data are presented and analyzed below
to explain the basis for these concerns.
Retention Among Enlisted Personnel Generally. Tables 11 through
14 provide a breakdown of retention data from each of the Services from fiscal year
1989, the last year before the end of the Cold War, to fiscal year 2001. As each of
the Services tracks its retention data in different ways,33 the tables do not lend


31For enlisted personnel, the obligated term of service is determined by their initial enlistment
contract (usually three to six years). For officers, the obligated term of service is determined
by rules governing the program through which they receive their commission (usually four or
five years).
32Each of these outcomes, in turn, can have a negative impact on recruiting by making the
military a less attractive career option.
33Between FY1989 and FY2000, retention computation methodologies varied among the
Services. Retention rates for each of the Services were computed by dividing the “number
(continued...)

themselves to direct comparison. Additionally, as each Service has a unique mission
and force structure, their retention requirements vary substantially. Thus, the data
from one Service should not be directly compared with the data from another
Service. Each data set should be viewed and analyzed individually.
Army. The retention rates for enlisted personnel in the Army are listed in Table
11. The Army tracks retention rates in three categories: initial term (serving in first
enlistment, regardless of length), mid-career (second or subsequent enlistment with
less than ten years of service), and career (second or subsequent enlistment with more
than ten but less than twenty years of service). It states its retention goals in terms
of raw numbers. A review of those numbers indicates that the Army did not meet
many of its retention goals in the first half of the 1990s: it failed to meet its initial term
retention goals in FY1991, 1992, 1993, and 1995; its mid-career retention goals in
FY1991 and 1995; and its career retention goals in FY1991, 1992, 1993, and 1994.
This trend is somewhat surprising in light of the generally declining retention goals
sought during that time-frame; but it is understandable if viewed as a response to the
post-Cold War drawdown. The drawdown was carried out principally between
FY1991 and FY1996 (highlighted in the table). Thus, the lower than expected
retention rates in the early 1990s might well have been the result of soldiers making
rational decisions to avoid the grim career prospects of the drawdown years in favor
of brighter prospects in the civilian world. Additionally, the imperative of reducing
the size of the Army due to drawdown pressures may have made meeting retention
goals a lower priority for Army officials in the early 1990s than it would otherwise
have been.


33 (...continued)
retained” by the total “loss population.” For the Army, Air Force, and the Marine Corps the
number retained was simply the number who re-enlisted. For the Navy, the number retained
was the number who re-enlisted plus the number who signed a long-term extension to their
current enlistment. With respect to the “loss population,” the Army, Air Force, and Marine
Corps calculated it in essentially the same way: it was simply the total number of people
eligible to re-enlist (whether they actually choose to re-enlist or not). Until 2001, however,
the Navy’s loss population included the total number of people eligible to re-enlist plus those
who were not eligible to re-enlist and who left the service. This difference in computation
methodology made the Navy’s retention rate appear lower than it would have been if
calculated in a manner similar to the other Services. In April 2001, the Navy adopted a
computation methodology that brought it in line with the other Services.

Table 11: Army Retention Rates For Enlisted Personnel,
FY1989-2001
Initial Term Mid-Career Career
Goal Achieved Goal Achieved Goal Achieved
Percent PercentPercentPercentPercentPercent
Year Number of of Number of of Number of of
Goal Eligibles Goal Eligibles Goal Eligibles
FY1989 34400 115% 48.6% 25400 127% 81.6% 28200 108% 74.3%
FY1990 27000 106% 40.7% 20000 158% 81.6% 20000 158% 70.3%
FY1991 21000 97% 41.4% 27000 91% 72.0% 29000 96% 71.8%
FY1992 25321 89% 30.7% 27052 102% 62.5% 29497 92% 61.6%
FY1993 30000 93% 46.1% 25500 100% 75.6% 22000 93% 58.7%
FY1994 22600 109% 49.2% 23000 105% 73.8% 21400 98% 54.4%
FY1995 20197 99% 45.6% 24551 97% 72.8% 27862 102% 58.2%
FY1996 21433 100% 46.5% 22671 102% 70.0% 28966 98% 65.7%
FY1997 23935 102% 54.4% 29699 102% 75.4% 26325 95% 65.5%
FY1998 20542 106% 51.2% 22912 102% 74.1% 18671 96% 60.1%
FY1999 20200 103% 50.3% 23000 105% 75.9% 21800 120% 66.0%
FY2000 20000 107% 49.7% 23200 104% 82.4% 24800 104% 70.6%
FY2001 19750 101% 53.0% 23350 102% 76.3% 20900 102% 66.5%
Source: Department of Defense, Directorate for Officer & Enlisted Personnel
Management
Since FY1996, the last year of the drawdown, the Army has met or exceeded its
retention goals in both its initial term and mid-career categories. However, it
continued to have trouble meeting its retention goals for career personnel until
FY1999. From FY1996 to FY1998, the Army missed its retention goals for career
personnel by between 2 and 5%. Since FY1999, however, the Army has been able
to meet its career retention goal by substantial margins.
Navy. The retention rates for enlisted personnel in the Navy from FY1989-2000
are listed in Table 12. During this time-frame the Navy tracked retention rates in
three categories: first term (serving in first enlistment, regardless of length), second
term (second or subsequent enlistment with less than ten years of service), and career
(second or subsequent enlistment with more than ten years of service). It is important
to reiterate, however, that during this time period the Navy calculated its retention
rate in a way that made its rates appear lower than those of the other Services; thus,
Navy retention rates cannot be directly compared with the retention rates of other
Services.
A review of Navy retention data does indicate an overall decline in retention
rates between fiscal years 1989 and 1999 in all three retention categories. Although
it is not surprising that retention rates for the Navy dropped during the drawdown
years, it is surprising that retention rates continued to decline after FY1996, which is
generally considered to be the last drawdown year. However, the Navy started its



drawdown later than the other Services (its peak drawdown years in terms of declines
in endstrength were FY1994 and 95), and the Navy continued to experience
significant cuts in endstrength even after the principal drawdown years had passed
(Navy endstrength declined by 5% in FY1997 and by 3% in FY1998). Thus, it is
reasonable to conclude that the Navy’s retention rates were affected by the drawdown
for a longer period of time than the other Services.
Additionally, caution must also be used in interpreting these data because the
Navy did not have any formal retention goals during this time-frame. Therefore, it is
difficult to assess the meaning of these retention rates. Was the decline an indicator
of retention problems in the Navy, or was it simply a reflection of force structure
readjustments in the post-drawdown Navy? In all likelihood, it was some combination
of these two explanations, but it is unclear from the data alone. Still, testimony from
senior Navy leaders indicates that the Navy had substantial retention problems with
its enlisted force between FY1996 and FY2000. For example, despite retention levels
substantially higher than those of 1999, the Chief of Naval Operations indicated that
as of June 2001, retention rates were not at a “steady state” level.34
Recently the Navy changed the way it tracks retention. Beginning in FY2001,
it has modified the definitions of its retention categories, set goals for each of them,
and significantly changed the formula by which its retention rates are calculated,
bringing them in line with the formula used by the other Services. The Navy’s new
retention categories are Zone A (17 months to 6 years of service), Zone B (6 to 10
years of service) and Zone C (10-14 years of service). The Navy’s goals and
retention rates for FY2001 are listed in Table 13 below and indicate that the Navy
came very close to meeting its retention goals in FY2001. Nonetheless, the retention
pattern exhibited in previous years, coupled with the comments by Admiral Clark,
suggest that enlisted retention in the Navy may still be a cause for concern.


34Statement of Admiral Vernon E. Clark, Chief of Naval Operations, before the Senate
Appropriations Subcommittee on Defense, June 20, 2001.

Table 12: Navy Retention Rates For Enlisted Personnel,
FY1989-2000
(No official goals were set by the Navy during this time)
1ST Term 2ND TermCareer
Achieved Achieved Achieved
Percent ofPercent ofPercent of
Year Eligibles Eligibles Eligibles
FY1989 37.4% 56.0% 66.9%
FY1990 38.3% 55.7% 70.6%
FY1991 41.4% 58.0% 71.4%
FY1992 38.9% 56.8% 65.4%
FY1993 34.8% 51.9% 61.7%
FY1994 35.7% 47.4% 57.7%
FY1995 36.4% 49.2% 56.5%
FY1996 32.9% 48.9% 55.7%
FY1997 30.8% 48.4% 57.5%
FY1998 30.5% 46.3% 54.6%
FY1999 28.2% 43.8% 53.3%
FY2000 29.6% 46.5% 56.6%
Source: Department of Defense, Directorate for Officer & Enlisted Personnel
Management.
Table 13: Navy Retention Rates For Enlisted Personnel, FY2001
Zone AZone BZone C
GoalAchievedGoalAchievedGoalAchieved
YearPercent ofEligiblesPercent ofEligiblesPercent ofEligiblesPercent ofEligiblesPercent ofEligiblesPercent ofEligibles
FY2001 57% 56.9% 69% 68.2% 89% 84.6%
Source: Department of Defense, Directorate for Officer & Enlisted Personnel
Management.



Marine Corps. The retention rates for enlisted personnel in the Marine Corps
are listed in Table 14. The Marine Corps also tracks retention rates in three
categories: first term (serving in first enlistment, regardless of length), second term
(second or subsequent enlistment with less than ten years of service), and career
(second or subsequent enlistment with more than ten but less than twenty years of
service). Since FY1992, the Marine Corps has set retention goals, which are stated
as a percentage of those eligible for retention, for its first term personnel. However,
it has not set goals for second term or career personnel, which makes evaluating the
retention data in these categories difficult. A review of the data for first term
personnel indicates that the Marine Corps had some difficulty meeting its goals during
the post-Cold War drawdown, but recovered in subsequent years. In fiscal years
1998 through 2001, the Marine Corps essentially met its retention goals for first term
personnel. It appears that a similar trend occurred in the second term and career
categories: retention rates dropped during the drawdown, followed by a gradual
upswing that eventually produced retention rates higher than those immediately before
the drawdown. Indeed, second term retention rates in FY2001 are about 13
percentage points higher than those in FY1989, and career retention rates for FY2001
are about 19 percentage points higher than those of FY1989.
Table 14: Marine Corps Retention Rates For Enlisted Personnel,
FY1989-2001
(percent of eligibles)
1ST TERM2ND TERMCAREER
Year Goal Achieved Achieved Achieved
FY1989 26.3% 47.7% 77.2%
FY1990 29.9% 54.7% 78.7%
FY1991 21.3% 47.5% 72.4%
FY1992 22-26% 19.3% 43.8% 71.2%
FY1993 22-26% 17.2% 46.6% 74.0%
FY1994 22-26% 19.4% 42.5% 73.7%
FY1995 22-26% 20.5% 43.8% 73.1%
FY1996 22-26% 22.0% 46.5% 77.5%
FY1997 22-26% 21.4% 45.8% 80.2%
FY1998 22-26% 21.8% 47.1% 80.4%
FY1999 22-26% 23.8% 47.5% 80.4%
FY2000 26.3% 26.6% 63.4% 95.6%
FY2001 26.8% 26.7% 60.4% 96.1%
Source: Department of Defense, Directorate for Officer & Enlisted Personnel
Management.



Air Force. The retention rates for enlisted personnel in the Air Force are listed
in Table 15. The Air Force tracks retention rates in three categories: first term
(serving in first enlistment, regardless of length), second term (second or subsequent
enlistment with less than ten years of service), and career (second or subsequent
enlistment with more than ten years of service and up through the point where
retirement papers are approved). The Air Force sets retention goals stated in
percentage terms. A review of the retention data for the Air Force indicates that it
has had some difficulty meeting its retention goals in recent years.
Throughout most of the 1990s, the Air Force rarely had difficulties meeting or
exceeding its retention goals. However, that pattern changed toward the end of the
decade. The Air Force did not meet its retention goals for first term personnel in
FY1998 - 2000; for second term personnel in FY1997 - 2001; and for career
personnel in FY1998 - 2001. The shortfall was most pronounced with respect to
second term personnel where the Air Force fell short of its goal by about six
percentage points per year.
Table 15: Air Force Retention Rates For Enlisted Personnel,
FY1989-2001
1ST TERM2ND TERMCAREER
Goal Achieve Goal Achieve Goal Achieve
FY1989 55% 59.0% 75% 76.0% 95% 96.0%
FY1990 55% 52.0% 75% 69.0% 95% 93.0%
FY1991 55% 59.0% 75% 77.0% 95% 95.0%
FY1992 55% 58.0% 75% 76.0% 95% 96.0%
FY1993 55% 61.0% 75% 82.0% 95% 97.0%
FY1994 55% 59.0% 75% 81.0% 95% 96.0%
FY1995 55% 63.0% 75% 77.0% 95% 96.0%
FY1996 55% 59.0% 75% 76.0% 95% 95.0%
FY1997 55% 56.0% 75% 71.0% 95% 95.0%
FY1998 55% 53.9% 75% 69.0% 95% 93.0%
FY1999 55% 49.1% 75% 69.4% 95% 92.0%
FY2000 55% 52.0% 75% 68.8% 95% 90.6%
FY2001 55% 55.4% 75% 68.8% 95% 90.6%
Source: Department of Defense, Directorate for Officer & Enlisted Personnel
Management
Retention In Critical Specialties. In addition to concerns about overall
retention patterns among enlisted personnel, there has been ongoing concern about
the military’s ability to retain people – both officers and enlisted personnel – in certain
key occupational specialties. Retention of pilots has been a highly publicized area of



concern for years,35 but significant concerns have also been raised about retention in
fields such as communications, intelligence, maintenance, electronics, information
technology, and the health professions.
This shortfall appears to have gotten worse, at least in some specialties, in the
post drawdown era. A report issued by the General Accounting Office (GAO) in
March of 2000 compared retention rates at the occupational level in the 1988-1990
and 1996-1998 time-frames. The study found that substantial drops in retention had
occurred among enlisted personnel in a number of critical specialties and that smaller
drops had occurred among officers as well. The more severe declines for enlisted
personnel occurred in the intelligence field (retention rate down 16 points in the Army
and 11 points in the Air Force), data processing (down 10 points in the Army, 11
points in the Navy, and 15 points in the Air Force), and air crews (down 20 points in
the Navy and 14 points in the Air Force).36
According to this same report, reductions in retention rates among officers
occurred as well. However, these reductions occurred in a lower proportion of
occupational fields and the declines were, on average, substantially smaller than those
among enlisted personnel. Additionally, the fields where the declines occurred varied
substantially among the different Services. In the Army, retention rate declines of
between 3 and 5% occurred among officers in the fields of nursing, dentistry, health
service administration, automotive maintenance, data processing, and fiscal
administration. In the Navy, declines of between 3 and 4% occurred in the fields of
ordnance and electronics, and in some supply related fields. In the Air Force, declines
of 3 to 5% occurred in the fields of counterintelligence, electronics, missile
maintenance, fiscal administration, psychology, and rotary wing aviation (helicopter
pilots). In the Marine Corps, declines of 3 to 4% occurred in the fields of intelligence,37
electronics, supply, and aircraft crew operations.
Retention In Certain Grades Of Officers. A final area of concern in
recent years has been the retention of officers in the O-3 and O-4 paygrades (These
paygrades correspond to the ranks of Captain and Major in the Army, Air Force, and
Marine Corps; Lieutenant and Lieutenant Commander in the Navy). These officers
are important in several respects: their large numbers make them a “center of gravity”
of the officer corps; they often function in key front line leadership roles; and they are
the pool from which the senior military leaders of the future will be drawn. In recent
years, there have been reports that retention among these officers has been declining,
perhaps precipitously. For example, one recent news article stated that “Anecdotal
reports to DOD officials suggest Army, Air Force, and Marine Corps captains and


35Pilot retention is a very specialized issue that is largely dependent on the hiring status of
commercial airlines. For more information on this topic, see CRS Report 98-764, Military
Pilot Retention: Issues and Options, by Mike Ryan.
36GAO Report to Congressional Requestors, Military Personnel: Systematic Analyses Needed
to Measure Retention in Key Careers and Occupations, March 2000, page 19.
37GAO Report to Congressional Requestors, Military Personnel: Systematic Analyses Needed
to Measure Retention in Key Careers and Occupations, March 2000, pages 62-63.

Navy lieutenants are leaving the military in numbers not seen since 1973....”38 An
April 2000, article stated that “Dissatisfied younger Army officers are leaving the
service in droves, worrying the service’s leadership and provoking intense debates39
about the problem at military bases across the nation.” An analysis of retention data
indicates that these assertions are overstated for officers in the O-3 paygrade, but do
not appear accurate for those in the O-4 paygrade.
The continuation rates40 of officers in paygrade O-3 are displayed in Table 16
and broken out by Service and fiscal year. The pattern is quite similar for each
Service: all show a drop in continuation rates early in the drawdown, a recovery
several years later, followed by another drop off in more recent years. For all
Services, the continuation rate for 0-3 officers was lower in FY2000 than in FY1989.
However, this decline was more pronounced in the Army than in the other Services.
The Army’s continuation rate in FY2000 was 3.4 percentage points lower than it was
in FY1989, compared with a 2.6 percentage point decline in the Air Force, a 2.1
percentage point decline in the Navy, and a 0.7 percentage point decline in the Marine
Corps. While these declines appear to be small, their impact on the officer corps can
be larger due to the numbers of officers in the 0-3 paygrade. For example, if the
Army’s continuation rate for captains were the same in FY2000 as it had been in
FY1989, it would have kept about 725 more of them on active duty then actually
occurred. If the Air Force’s continuation rate for captains were the same in FY2000
as it had been in FY1989, it would have kept about 670 more of them on active duty.
Although the change in the Navy’s continuation rate for 0-3 officers between
FY1989 and FY2000 has been less than that of the Army and Air Force, the Navy’s
continuation rate for these officers is troubling for another reason: its showed less of
a recovery after the “drawdown drop” then the other Services did. Although all the
Services showed a significant drop in continuation rates in FY1992 and 1993, most
recovered strongly in subsequent years. The Army and Air Force saw their
continuation rates rebound in FY1994, and the Marine Corps continuation rate began
to improve in FY1995. The Navy, however, did not see a turnaround in its 0-3
continuation rate until FY1996, and the recovery proved short-lived: continuation
dropped in FY1997, 1998, and 1999. Thus, the Navy’s decline in 0-3 continuation
rates, while not as large as the Army’s, may be a greater problem as it did not recover
as robustly from the lows of the drawdown era. The Navy’s 0-3 continuation rate
in FY1999 was actually lower than in any year of the drawdown era.


38Jim Garamone, “DOD Examines Captain/Lieutenant Retention,” American Forces Press
Service, May 17, 2001.
39Thomas E. Ricks, “Younger Officers Quit Army At Fast Clip,” The Washington Post, April

17, 2000, p 1.


40Continuation rates are the measure of retention commonly used with commissioned officers.
Unlike enlisted personnel, officers do not sign enlistment and re-enlistment contracts for
specified periods of time; they may resign at any time after the end of their service obligation
(usually four or five years after commissioning). As such, continuation rates – which
measure the percentage of officers who continue in the service from one year to the next – are
commonly used to measure their retention.

Thus, with the exception of the Marine Corps – whose continuation rates have
recovered to nearly the same level as before the drawdown – the data indicate that
continuation rates for officers in the 0-3 paygrade are an area of concern. This
concern is magnified by the possibility that the overall retention data presented here
may mask troubling shifts in voluntary separation of officers in the O-3 pay grade
versus involuntary separation. One study by the Army indicates that voluntary
separation has climbed significantly for Army captains in a recent five year period,41
while the proportion of involuntary separations has decreased. This has led some to
infer that fewer officers are being released for substandard performance, which might
indicate a decline in the quality of the officers being retained. Nonetheless, in light of
the currently available data, the assertion that there has been a massive exodus of
officers in the O-3 paygrade does not appear to be supportable.
Table 16: Active Duty O3 Officer Continuation Rates by Service,
FY1989-2000
%
Marine% Air
% Army% NavyCorpsForce% DOD
FY1989 92.0 88.2 89.2 91.5 90.8
FY1990 91.4 88.5 89.2 90.5 90.3
FY1991 94.0 89.2 90.4 91.5 91.7
FY1992 86.7 87.3 88.5 87.3 87.2
FY1993 84.3 87.2 88.0 84.6 85.3
FY1994 89.3 86.2 85.4 93.7 90.0
FY1995 90.3 86.0 90.8 90.6 89.3
FY1996 90.2 88.6 89.2 91.4 90.2
FY1997 90.8 87.7 87.1 91.1 89.9
FY1998 90.7 86.4 89.2 90.0 89.2
FY1999 88.8 85.5 88.5 89.2 88.1
FY2000 88.6 86.1 88.5 88.9 88.0
Source: Department of Defense, Directorate for Officer & Enlisted Personnel
Management.
The retention rates of officers in paygrade 0-4 between FY1989 and FY2000 are
displayed in Table 17. The pattern of retention rates for these officers is slightly
different from that exhibited by officers in the 0-3 paygrade: retention rates changed
little between FY1989 and FY1991, dropped from about FY1992 to FY1994 and
recovered from about FY1995 to FY1998. Since FY1998, the continuation rates
have shifted somewhat from year to year, but have remained at or slightly below the
level they were at during the pre-drawdown time-frame. This pattern is consistent
with the imperatives of the drawdown: voluntary and involuntary separations induced
a decline in retention rates as the drawdown era began, followed by a recovery of


41Taken from briefing slides of “Commander’s Conference,” presented by Lieutenant General
Timothy Maude, Army Deputy Chief of Staff for Personnel, October 19, 2000.

retention rates as the drawdown era came to a close. While continuation rates for
officers in the O-4 paygrade deserve continued observation, the data do not indicate
a significant shortfall in aggregate O-4 retention rates as of FY2000.
Table 17: Active Duty O4 Officer Continuation Rates by Service,
FY1989-2000
% % Marine% Air
% ArmyNavyCorpsForce.% DOD
FY1989 94.4 92.1 93.1 92.0 92.8
FY1990 94.2 91.1 92.4 92.0 92.5
FY1991 94.2 92.0 92.7 92.3 92.9
FY1992 87.9 91.2 91.6 89.4 89.5
FY1993 88.3 90.7 91.5 89.4 89.5
FY1994 88.5 81.6 90.0 84.5 85.4
FY1995 89.0 89.0 93.2 86.8 88.5
FY1996 87.5 90.3 91.4 89.4 89.2
FY1997 90.2 90.7 91.6 89.8 90.3
FY1998 94.5 92.8 91.4 91.0 92.6
FY1999 94.2 89.1 91.6 91.3 91.7
FY2000 93.8 91.0 92.1 90.6 91.9
Source: Department of Defense, Directorate for Officer & Enlisted Personnel
Management.
Perceived Causes. Aggregate retention rates in the military are the product
of thousands of individual decisions on whether or not to stay in the military. These
individual decisions, analysts believe, are based on the individual’s answer to one basic
question: “Would I be better off if I stayed in or left the military?” The question,
while simple to state, can be an extraordinarily complex one to answer, as the
definition of “better off” entails a broad array of factors including pay and benefits,
job satisfaction, and quality of life. Some of these factors, like pay, are tangible and
can be quantified and compared with civilian jobs.42 Others, like quality of life, are
less tangible and can be very difficult or impossible to quantify and compare with
civilian life. Additionally, different individuals attach different levels of importance
to these various factors. Given the exact same environment, the answer to the
question “Would I be better off if I stayed in or left the military?” will be answered
differently by different people depending on their personal priorities. Thus, it is
impossible to identify some sort of objective “tipping point” that leads military
personnel to decide for or against continued service.
However, based on survey research, focus groups, and individual interviews it
is possible to get a general idea of the most important factors affecting retention
decisions at a given point in time. The factors most commonly cited as the “causes”
of recent military retention shortfalls are identified and discussed below. These


42Although, as discussed previously with respect to the “pay gap,” attempts to do this can be
a very controversial.

factors are not mutually exclusive and should not be seen as competing explanations,
but instead as partial answers to the question “Why have the military Services
experienced retention shortfalls?” Additionally, the factors are not necessarily listed
in order of importance, for it is not entirely clear what the relative importance of
these factors is. Finally, it should be pointed out that Congress and the executive
branch have already taken a number of steps to address these issues. They are listed
in the next section.
Perceived Cause # 1: Relative Value of Pay and Benefits. In a 1999
survey of active duty personnel administered by the Department of Defense,
respondents were asked to choose the “most important factor for leaving or
considering leaving active duty.” The most commonly cited response was basic pay
(28.2%). Indeed, many commentators have argued that the major reason for the
military’s recent retention shortfalls is the gap between military pay levels and civilian
pay levels. As pointed out in the previous section on recruiting, there is robust debate
among compensation analysts over whether such a “pay gap” actually exists,43
however, it is clear that the vast majority of military personnel believe that it does.
In the 1999 survey just mentioned, two-thirds of the respondents indicated that they
would receive a better total compensation package (pay, benefits, bonuses, and
allowances) in the civilian world than they were receiving in the military.
Contributing to this belief may be the well publicized manner in which military
pay raises were calculated throughout most of the 1990s. Since 1967, federal law has
provided that military pay be adjusted upward, at the same time and by the same
percentage, as the average increase in the pay rates for federal General Schedule (GS)
employees.44 In 1990, Congress enacted a statutory formula for federal civilian pay
raises that, beginning in fiscal year 1992, provided for an annual increase of 0.5% less
than the annual increase in the private sector wages and salaries component of the45
employment compensation index (ECI). Thus, by statute, military pay raises were
supposed to be determined by the an “ECI minus 0.5%” formula (although this was46
not always the case in fact). As the ECI is considered a standard measure of the rate
at which civilian compensation rates are increasing, military personnel likely saw the
“ECI minus 0.5%” formula as evidence that their wages were not keeping up with
those of private sector employees.
Another factor cited by some as contributing to military retention shortfalls was
changes in military benefits, most notably retirement pay. A major change was made


43See CRS Issue Brief IB10089, Military Pay and Benefits: Key Questions and Answers by
Robert Goldich.
44The Uniformed Services Pay Act of 1967, P.L. 90-207, December 16, 1967. Codified at

37 USC 1009.


45The Federal Employees’ Pay Comparability Act of 1990, P.L. 101-509, November 5, 1990.
Codified at 5 USC 5303.
46Despite this statutory formula, which could operate each year without any intervention,
Congress passed laws that mandated specific pay raises for the military each year. In general,
these pay raises were identical to that determined by the “ECI minus 0.5%” formula, although
Congress specified pay raises higher than the statutory formula in FY1997 and FY1999.

to the military retirement system in 1986. Under the Military Retirement Reform Act
of 1986, military personnel who first entered military service on or after August 1,
1986 would receive a less generous retirement plan than those who first entered
military service before that date. The new retirement plan, known as “Redux,” was
unpopular among the servicemembers who fell under its provisions.47 Although no
one actually received reduced benefits under this plan – compulsory Redux was
repealed in 1999,48 seven years before the first Redux payments would have been
made – some have argued that the existence of Redux had a negative effect on
retention rates in the later half of the 1990s. From this perspective, the diminished
retirement benefit available to Redux eligible servicemembers did not provide a
sufficient incentive for them to stay in the military for a full career. It does not appear
that researchers have been able to document an empirical link between Redux and
lower retention rates.
Retirement benefits are also cited as a special concern for personnel in critical49
occupational specialties. In a survey conducted in 1998 and 1999, officers in critical
skill specialties who indicated an intention to leave the military picked retirement pay
as one of the top reasons for leaving the military, along with frequency of
deployments. For enlisted personnel in critical specialties, basic pay was the most
important reason for leaving the military, but retirement pay was tied with frequency
of deployments for second place.50 These concerns about retirement pay can probably
be attributed – at least in part – to the same concerns about Redux that other
servicemembers may have had at that time. However, another cause of this concern
about retired pay may be the way in which retired pay is calculated, and the impact
this has on people in critical skill specialties.
Depending on when a person first entered military service, retired pay is based
on either the level of basic pay that the servicemember earned just prior to retirement
or on the average of the highest 36 months of basic pay that the servicemember has
received during his or her career. Regardless of which formula is used, however,
basic pay is the only aspect of total compensation that is included in the calculation:
special pays and bonuses are omitted. This is significant for those in critical
specialties because they often receive very large bonuses or special pays as an
incentive to keep them in the service. Although these bonuses and special pays may
substantially increase their average monthly pay while on active duty, it does not
enhance their retirement pay. Thus, the value of their retirement pay is relatively low


47Nonetheless, even under Redux, military retirement would compare quite favorably with
most civilian retirement systems, as it continued to contain a distinctive feature: eligibility for
retired pay after 20 years of service, regardless of age.
48P.L. 106-65, sections 641-644. The law repealed compulsory Redux and gave military
personnel the option of retiring under the old, more generous, pre-Redux retirement pay
computation formula, or retiring under Redux with an immediate cash bonus.
49General Accounting Office Report (GAO/NSIAD-99-197BR), Perspectives of Surveyed
Service Members in Retention Critical Specialties, August, 1999.
50General Accounting Office Report (GAO/NSIAD-99-197BR), Perspectives of Surveyed
Service Members in Retention Critical Specialties, August, 1999, p 25

in comparison to their active duty pay, and this may decrease the perceived value of
staying in the military long enough to receive retirement benefits.
Perceived Cause #2: Lower Quality of Life. Many servicemembers and
outside observers believe that there has been a decline in the quality of life associated
with military service in the post-Cold War era. During the Cold War era, the United
States was focused primarily on containing the spread of communism, with a
particular emphasis on protecting Europe from a potential Soviet invasion and
protecting South Korea from a potential North Korean invasion. To facilitate this
military posture, major bases were maintained in Western Europe and Asia. Military
personnel stationed at these overseas bases had demanding jobs, but great efforts
were made to provide servicemembers with a “normal” life. For example, overseas
bases typically had most of the amenities of stateside bases and servicemembers’
families were usually allowed to accompany them.
With the end of the Cold War, however, the situation changed dramatically.
Although many overseas bases were closed and a large number of servicemembers
were transferred to stateside bases, U.S. military operations around the globe grew.
In the absence of the Soviet threat, the U.S. began to focus more on peacekeeping
and humanitarian missions in underdeveloped parts of the world. Deployments on
these missions typically took servicemembers away from their families for six to nine
months and sent them to places with comparatively primitive infrastructures.
Additionally, as the number of these military commitments increased during the 1990s
and as the size of the military decreased during the drawdown, the pace of
deployments for individuals increased.51 This put new pressures on servicemembers,
as the time spent on mission preparation, deployment, and post-deployment
reconstitution activities decreased the amount of leisure and family time available.
Nonetheless, military deployments do not always correlate with lowered
retention rates. For example, non-commissioned officers (NCOs) in the Army’s V
Corps reportedly have a higher retention rate than NCOs in the other three Army
Corps, despite V Corps higher rate of deployment.52 Most studies indicate that,
provided they are well managed and not excessive, deployments can enhance retention
by providing participants with a sense of accomplishment.53 However, these studies
also indicate that after a certain threshold level, deployments can have a negative
effect on retention.
Another quality of life issue that has been raised in recent years is the quality of
housing used by military personnel and their families. In the military, servicemembers
either live free of charge in military housing or receive an allowance to pay for civilian


51See CRS Report 98-41, Military Readiness, Operations Tempo (OPTEMPO) and
Personnel Tempo (PERSTEMPO): Are U.S. Forces Doing To Much?, by Mike Ryan, p. 29.
52Eric B. Pilgrim, “V Corps Tops On NCO Retention,” Stars and Stripes, August 28, 2001.
53See for example James Hosek and Mark Totten, Does Perstempo Hurt Reenlistment? The
Effect of Long or Hostile Perstempo on Reenlistment, RAND, 1998; Paul Sticha, Paul Hogan
and Maris Diane, Personnel Tempo: Definition, Measurement, and Effects on Retention,
Readiness and Quality of Life, Army Research Institute, 1999; and Peter Francis, OPTEMPO
and Readiness, Center for Naval Analysis, 1999.

housing. Military housing, although free, has often been criticized for being old and
poorly maintained. Recently, for example, a senior defense official testified before
Congress that, “over two-thirds of our [DOD] housing inventory in the United States54
is substandard” and Air Force General Joseph Ralston, commander in chief of the
U.S. European Command, recently stated “although we continue to make progress
as a whole, family housing throughout Europe remains old, is in need of extensive
repair and modernization and is well below contemporary standards in the U.S.”55
Those who receive a housing allowance (known as Basic Allowance for Housing or
BAH) in lieu of military housing usually have greater flexibility in selecting well
maintained housing with modern amenities. However, as BAH currently covers

92.5% of the average rental cost of civilian housing that meets DOD standards,


servicemembers receiving BAH are essentially forced to choose between accepting
lower-cost, below average housing or paying money “out of pocket” to live in better
accommodations. Recently, the senior enlisted members from each of the military
Services testified that quality of housing was a major quality of life concern for their56
members. The Department of Defense is in the midst of a multi-year plan to raise
BAH to 100% of the average rental cost, but it will not be completed until 2005.
Perceived Cause #3: Decreased Job Satisfaction. Retention shortfalls
may also be attributed to a decrease in the level of “job satisfaction” reported by
military personnel. Some of the common reasons cited for this decline in job
satisfaction include unhappiness with the actions of military superiors, a lack of
enthusiasm for the new missions undertaken by the military in the post Cold War era,
difficulties in accomplishing assigned tasks due to lack of resources, and disapproval
of changes in the military culture. Each of these topics is discussed in more detail
below.
Dissatisfaction with Leadership. A commonly repeated complaint in recent
years, especially among lower level officers, has been that their supervisors
“micromanage” them, do not properly mentor them for more challenging assignments,
and punish them with career ending evaluations for comparatively minor mistakes.
Though this type of complaint is not new to the military Services, some factors may
have exacerbated the issue in recent years, including the more competitive promotion
environment that emerged during the drawdown, which may have promoted excessive
risk avoidance behaviors on the part of some leaders; more extensive and critical
media coverage of the behavior of members of the military, which may prompt leaders
to monitor and control their subordinates more closely; and advances in
communications technologies, which allow superiors at all levels to keep closer track
of subordinates.


54Undersecretary of Defense Randall Yim, Testimony before the House Appropriations
Subcomittee on Military Construction, March 15, 2001.
55Rick Maze, “Ralston Decries Conditions At European Bases,” Army Times, February 14,

2002.


56Testimony of before the House Appropriations Subcommittee on Military Construction,
March 8, 2001.

Changing Mission. During the Cold War era, the mission of the armed forces
was clearly defined – protect the free world from communism – and unmistakably
oriented towards fighting a massive war. For many, the knowledge that they were
defending the nation and its allies from a major threat was psychologically rewarding
in itself. In the post-Cold War era, the military’s mission shifted toward peacekeeping
and humanitarian relief operations. While these types of missions certainly have their
own psychological rewards and were appealing to many military personnel, some may
have found them less rewarding and not important enough to justify the sacrifices of
military life. This factor is probably irrelevant now, however, as the recent terrorist
attacks on the United States and the ongoing operations in Afghanistan have
dramatically reemphasized the military’s mission of defending the nation.
Lack of Resources. Another complaint revolves around shortages of
resources necessary to accomplish assigned missions. Some of the resource shortages
cited as interfering with mission accomplishment include: insufficient personnel
assigned to the unit, a lack of required equipment, shortages of spare parts to repair
needed equipment, and a lack of fuel, ammunition, and training areas to perform
realistic training. Most of these shortages can be attributed to budgetary decisions,
although personnel issues are also linked more broadly with recruiting, retention,
training and assignment policies, and training area issues are linked with base closure
and environmental protection issues as well.
Changing Military Culture. A final cluster of issues cited in relation to job
dissatisfaction revolve around the ongoing changes in the military’s internal culture
and its relationship to the culture of the civilian world. The culture of the military is
unique and, in many ways, stands in sharp contrast to the broader American culture:
the military is hierarchical, not egalitarian; it is regimented in areas where civilians
expect liberty; decision making is typically “top down,” rather than democratic.
Although many Americans would find this type of environment oppressive, others find
it quite rewarding, for it can instill a strong sense of purpose, promote deep
interpersonal bonds and create a powerful sense of community. Those who do enjoy
the unique culture of military life naturally want to see it preserved and express
dissatisfaction when they think that it is being changed for the worse.
Complaints that the military way of life is changing for the worse are nothing
new, but they may have been more highly publicized in the 1990s due to related
policy disputes at the federal level. Some servicemembers resented attempts by
policymakers to allow homosexuals to serve in the military, to allow women to serve
in combat, or to change military rules prohibiting adultery and fraternization. Others
perceived a de-emphasis on traditional military values like courage, fidelity, and self-
sacrifice and an increased focus on values like tolerance and compassion.57


57For examples of these perspectives, see opinions contained in Joseph C. Myers,
Legitimization and Tolerance: Not the Same, Army, July, 1993, 37-39; James Webb, The
War on Military Culture, The Weekly Standard, January 20, 1997, 17-22; Adam G.
Mersereau, Diversity May Prove Deadly on the Battlefield, The Wall Street Journal,
November 14, 1996, A22; Tranette Ledford, Many Soldiers Unhappy With Fraternization
Proposals, Army Times, January 18, 1999, 7; Stephanie Gutmann, Today’s Army: Warriors
(continued...)

The Congressional Response. In recent years, Congress has passed a
number of laws aimed at improving military retention. Perhaps the most visible of
these initiatives has been a string of military pay raises and the repeal of the “Redux”
retirement plan. Other legislative initiatives include pay table reform, increased
funding for reenlistment bonuses, and increased ceilings on reenlistment bonuses and
incentive pays.
With respect to pay, Congress passed a law in 1999 that suspended the “ECI
minus 0.5%” formula (discussed previously) and replaced it with an “ECI plus 0.5%”
formula for the period FY2001-2006. Under this new formula, military personnel will
receive pay raises one half of a percent higher than the ECI and a full one percent
higher than they would have under the old formula. Assuming that it is fully
implemented, this policy will increase the basic pay of military personnel relative to
civilian occupations; additionally, it will probably help dispel any negative perceptions
generated by the “ECI minus 0.5%” formula. Congress also approved a “pay table
reform” initiative in 1999 that increased the pay of certain enlisted personnel and
officers. The change was most beneficial for mid-career officers, providing extra pay
to those officers at precisely the time in their careers when they make their most
important retention decisions. It was also beneficial to some junior and mid career
enlisted personnel, although they received smaller increases than the officers.
Another major retention initiative passed by Congress in 1999 was the repeal of
the Redux military retirement system. As mentioned previously, the Military
Retirement Reform Act of 1986 made major compulsory cuts in the future retired pay
of those people who first entered the military on or after August 1, 1986. The
National Defense Authorization Act for FY2000 repealed the compulsory nature of
those cuts by providing affected military personnel with the option, upon reaching 15
years of service, of electing retirement under the old, pre-Redux system, or retirement
under the Redux system with a substantial cash bonus.58


57 (...continued)
Vs. Schoolmarms, The Wall Street Journal, May 9, 2000.
58Another major change to the military retirement system recently approved by Congress
allows members of the military to participate in the federal Thrift Savings Plan (TSP). The
TSP, which is similar to the 401(K) plans offered by private corporations, allows participants
to invest pre-tax dollars in a special account where the invested money and any earnings are
not taxed until they are withdrawn. Beginning in late 2001, military personnel were allowed
to contribute up to 5% of their base pay to their TSP. In addition, they can contribute all or
part of any special or incentive pay they earn, provided their total contibutions do not exceed
the limits set by the Internal Revenue Service. This latter provision is considered to be
especially attractive to those military personnel serving in retention critical specialities. These
people often receive large retention bonuses or incentive pays to encourage them to continue
military service; however, these bonuses and special pays are not considered to be part of the
salary base for calculating retired pay. By allowing these people to deposit all or part of their
bonuses and special pays in a TSP, they can enhance their retirement income. At this point,
however, the effect of the TSP on military retention is unclear. Some have argued that it will
have a negative effect on retention by providing military personnel with a sufficient retirement
“nest egg” to leave the service after ten or fifteen years.

Congress has also responded to retention shortfalls in the military by substantially
increasing the amount of money available for re-enlistment bonuses, officer retention
bonuses, and various special and incentive pays. For example, Congress provided
more money than the Services requested for Selective Reenlistment Bonuses in fiscal
years 1999-2001. The 106th Congress increased the maximum amount of the active
duty re-enlistment bonus from $45,000 to $60,000 and established a special retention
bonus for members of armed forces qualified in critical military skills.59 The 106th
Congress also established, extended and/or increased bonuses, special pays, and
incentive pays designed to retain personnel in critical military occupations such as
aviation, health care, special warfare, and naval surface warfare.60 The 107th Congress
authorized an accession bonus of up to $100,000 for new officers in certain critical
skill areas and provided authority to the Services to allow the transfer of GI Bill
benefits from an enlisted person in a critical skill areas to certain family members as
a reenlistment incentive.61
The Executive Branch Response. The executive branch has been very
involved in developing policies aimed at improving military retention. In addition to
recommending and/or supporting the legislative policy changes mentioned above, the
various military Services have also enacted administrative policies designed to
improve retention rates. The most notable of these efforts have been those directed
toward better managing personnel tempo62 (PERSTEMPO), reducing manpower
shortages in units, and improving military housing.
Perhaps the most well known PERSTEMPO management initiative has been the
implementation of the Expeditionary Aerospace Force (EAF) concept by the Air
Force. The EAF, established in 1999, established ten “packages” of Air Force units,
known as Aerospace Expeditionary Forces (AEFs). At any given time, two of these
AEFs are “on call” for deployment to handle ongoing military operations; the other
eight are conducting normal training and operations. Every three months, two new
AEFs rotate into “on call” status. This system provides Air Force personnel with
greater stability, provides more planning time for deployments, and better distributes
the burden of deployment throughout the Air Force.


59P.L 106-65, section 618, National Defense Authorization Act For FY2000, October 5, 1999.
P.L. 106-398, section 633, National Defense Authorization Act For FY2001, October 30,

2000.


60P.L 106-65, section 612-629, National Defense Authorization Act For FY2000, October 5,

1999. P.L. 106-398, sections 622-634, National Defense Authorization Act For FY2001,


October 30, 2000.
61P.L. 107-107, section 621 and 654, National Defense Authorization Act For FY2002,
December 28, 2001.
62Personnel Tempo, or PERSTEMPO, is the rate at which military personnel are deployed
away from their home station. Until recently, this was usually measured on a unit basis,
rather than an individual basis. However, due to a 1999 law passed by Congress (P.L.106-
65), each of the Services now track the personnel tempo of each individual. This rate of
individual personnel tempo is commonly referred to as ITEMPO. ITEMPO is simply a
specific way of measuring PERSTEMPO.

Another important PERSTEMPO management tool has been the increased use
of Reserve Component forces, especially by the Air Force and the Army. In the Cold
War era (1946-1989), reservists were involuntarily activated for federal service in63
military operations four times, an average of about once every decade. In the post
Cold War era (1990-2001), reservists have been involuntarily activated for federal
service in military operations six times, an average of about once every two years
Additionally, the Services are also making increased use of reserve volunteers to help
them conduct operational missions. Moreover, reserve units are increasingly being
tasked with missions formerly reserved for active duty units. For example, the Army
made a historic decision when it selected the 49th Armored Division from the Texas
Army National Guard to serve as the headquarters element for the Bosnian
peacekeeping mission in 2000. The Army also selected six other Army National
Guard units to serve as the headquarters element in Bosnia during later rotations.
More recently, a National Guard infantry battalion was mobilized to take part in the
Multinational Force and Observers mission in the Sinai, a mission normally performed
by active component forces.
Another area where the Services have taken action is in attempting to reduce
manpower shortages in units. All of the Services have become more aggressive in
bringing former military personnel and reservists onto active duty. Different ways of
using existing manpower are also being tried. For example, the Army launched a
“man the force” initiative designed to fill combat divisions by transferring individuals
from non-combat units, such as the Training and Doctrine Command, to billets in the
Army’s divisions, armored cavalry regiments, and certain “early deploying” units.
Improving military housing has been designated as a top priority by the
Department of Defense and the various Services. As mentioned previously, the
Department of Defense is in the midst of a multi-year plan to raise the Basic
Allowance for Housing to 100% of the average rental cost. DOD has also requested
substantial increases in the amount of money appropriated for military housing in
FY2002. In comparison to the funds appropriated in FY2001, DOD requested a 60%
increase in military construction funds for barracks, dormitories, and similar facilities,
and requested a 13% increase in funds for family housing. Congress approved these
requests and added funds for additional housing projects.64 Finally, the Services are
also making increased use of their authority under the Military Housing Privatization
Initiative, a congressionally authorized program, to produce commercially owned and
operated housing for military personnel.65


63This category excludes those reservists who served on active duty under voluntary orders
or annual training orders and excludes members of the National Guard serving in a state
status. Additionally, with the exception of those mobilized in response to the terrorist attacks
of September 11th, it excludes involuntary activations of reservists for domestic reasons, such
as responding to civic disorders. For more information on this topic, see CRS Report
RL30637, Involuntary Reserve Activations For U.S. Military Operations Since World War
II, by Lawrence Kapp.
64P.L. 107-64, Military Construction FY2002 Appropriations Act, November 5, 2001.
65For more information on the Military Housing Privatization Initiative, see CRS Report
RL31039, Military Housing Privatization Initiative: Background and Issues, by Daniel H.
(continued...)

Results. The retention initiatives taken by both Congress and the executive
branch appear to be bearing some positive results, but there are still areas of concern.
With respect to retention of enlisted personnel generally, the Army and the Marine
Corps appear to be doing fairly well; however, the Air Force continues to experience
retention rates that fall short of its 2nd term and career goals, and Navy retention rates,
while improving in FY2001, still bear watching. With respect to retention in critical
specialties, the available data indicate that this continues to be a cause of concern for
the military. Recent increases in reenlistment bonuses, officer retention bonuses, and
various special and incentive pays, along with the general slowing of the economy and
the surge in patriotism following the September 11th terrorist attacks, have probably
mitigated this problem somewhat, but the proof (i.e. the data) is not yet in. With
respect to officers in the O-3 paygrade, there has been no massive exodus from the
services in recent years, but they are exhibiting a somewhat lower than normal
retention rate. Retention rates for officers in the 0-4 paygrade are little different than
they were before the drawdown began.
Factors That Could Cause Shortfalls in the Future. Despite the recent
improvements in military retention, a number of factors could change these trends.
Perhaps the most significant is the increasingly demanding retention environment
created by the small cohorts of new accessions brought in during the drawdown era
of the 1990s. Other variables include future PERSTEMPO levels and the state of the
economy. Each of these is discussed below.
Factor #1: Comparatively Small Size of Certain Retention Cohorts.
As mentioned earlier in this report, during the post-Cold War drawdown of the early
to mid-1990s, the Army, Navy and Air Force reduced their accessions below the level
needed to sustain the force as a way of meeting their endstrength goals. As a result,
the cohorts that entered the services during these years are comparatively small.
These smaller cohorts will create a more challenging retention environment in future
years when the comparatively low number of recruits in these cohorts finish their
enlistments and begin to make retention decisions. In order to meet their needs for
mid-level enlisted personnel, the affected Services will need to achieve a higher than
normal retention rate. This factor is even more of an issue in certain retention critical
specialties. According to Vice Admiral Patricia Tracey, deputy assistant secretary of
defense for military personnel policy, the Services will need “virtually a 100%
continuation rate”66 to sustain normal manning in certain specialized skill areas.
Factor #2: PERSTEMPO Level. Another factor that could negatively
impact retention rates would be a high PERSTEMPO level. Although deployments
can have a positive effect on retention, studies indicate that after a certain threshold
level, high PERSTEMPO levels have a negative effect on retention.67 While it is


65 (...continued)
Else.
66Cited by Jim Garamone,, “DOD Working On Retention Challenges,”American Forces Press
Service, May 9, 2001.
67See for example James Hosek and Mark Totten, Does Perstempo Hurt Reenlistment? The
(continued...)

difficult to assess just where this threshold level is, it is possible that future increases
in PERSTEMPO levels could have a negative effect on retention.
Factor #3: Competition With Civilian Employers. The United States is
currently in the midst of an economic slowdown. This benefits military retention, as
it reduces the attractiveness of leaving the military. However, as the economy tends
to follow a cyclical trend, it is likely that the currently sluggish economy will be
replaced in the not too distant future by a more robust one. Such a revival would
create more civilian employment opportunities for military personnel and could create
a more challenging retention environment. On the other hand, the vicissitudes of the
economy may lead some military members to view the security of a military career
with greater appreciation.
Options for Congress. There are a number of options that Congress could
consider to address retention shortfalls. Below is a list of policies that could be used
to respond to one or more of the factors mentioned above, grouped under four broad
headings.
More Competitive Pay and Benefits. This response could be effective if
it appeared that serious retention shortfalls were being caused primarily by increased
competition with civilian employers. As mentioned previously, aggregate retention
rates in the military are the product of thousands of individual decisions on whether
or not to stay in the military. These individual decisions, it is believed, are based on
the individual’s answer to one basic question: “Would I be better off if I stayed in
or left the military?” A major factor in this calculation is military compensation – pay
and benefits – or, more specifically, the perceived value of military compensation
relative to civilian compensation.
Pay. Across-the-board pay raises would be an option to improve the perceived
competitiveness of military compensation. However, they are expensive and some
argue that they are inefficient because they give raises to people in pay grades or
occupational specialties that do not have retention shortfalls. Another option would
be to target pay raises towards retention critical pay grades. This can also be
problematic, however, as it can distort the relationship between pay grades and
minimize the incentive to get promoted. For example, targeted pay raises might
enable lower ranking members to make nearly as much higher ranking members, a
phenomenon known as “pay compression,” or more than higher ranking members, a
phenomenon known as “pay inversion.” Another option, should Congress view this
as a problem, would be to increase total pay in a targeted manner by increasing re-
enlistment or retention bonuses for those in retention critical occupational skills, pay
grades, or retention cohorts. It might be considered if the projected shortfalls occur
due to the comparatively small size of certain retention cohorts mentioned above.


67 (...continued)
Effect of Long or Hostile Perstempo on Reenlistment, RAND, 1998; Paul Sticha, Paul Hogan
and Maris Diane, Personnel Tempo: Definition, Measurement, and Effects on Retention,
Readiness and Quality of Life, Army Research Institute, 1999; and Peter Francis, OPTEMPO
and Readiness, Center for Naval Analysis, 1999.

Benefits. With respect to benefits, the military currently offers a number of
valuable benefits to servicemembers, including health care, a generous non-
contributory retirement plan, educational benefits, subsidized grocery stores, and free
or inexpensive access to various recreational facilities. Congress might consider
enhancing these benefits or expanding certain benefits to family members.
Retention officers and career counselors will often repeat the adage that
“individuals enlist, but families reenlist.” This summarizes the belief that
servicemembers’ decisions are profoundly affected by the attitudes of their spouses
and children. In light of this, an option to encourage retention would be to provide
more benefits to family members, particularly spouses. For example, Congress
recently gave the Services the authority to allow certain military personnel to transfer
their GI Bill benefits to their spouse or children as a reenlistment incentive. While this
will likely increase the cost of the GI Bill program – as the benefit likely will be
utilized more often – it arguably provides family members with a significant incentive
to endure the hardships of military family life. If this program is successful, it may be
worth further consideration. The Department of Defense has opposed efforts to
transfer GI Bill benefits to family members in the past on the grounds that it is an
untested concept, could be very expensive, and could hurt recruiting efforts in the
future because children of servicemembers, who are an excellent source of recruits,68
would not need to join the service to acquire GI Bill benefits. However, Deputy
Secretary of Defense Paul Wolfowitz indicated at his Senate confirmation hearing in69

2001 that he favored expanded family educational benefits.


Increased Funding For Operations And Maintenance. One of the
issues commonly cited as a cause of retention shortfalls is decreased job satisfaction.
There are several components to this sense of dissatisfaction, not all of which are easy
to resolve through congressional action. Dissatisfaction over the level of resources
provided could be addressed by Congress if it were deemed a major problem.
Increasing the amount equipment, spare parts, ammunition, fuel, and related supplies
might mitigate a significant source of job dissatisfaction among some military
personnel. However, increasing funding of operations and maintenance accounts
might have a negative impact on other budgetary priorities.
More Flexible Career Paths. Another strategy for improving job
satisfaction, some analysts suggest, would be to make military career paths more
flexible, thereby giving military personnel more control over their professional
development. At present, the military personnel system usually requires
servicemembers to serve in a fairly rigid progression of assignments, involving
frequent moves and constant rotation into new jobs. Some have argued that this
system should be modified to allow individual’s greater involvement in their duty
assignments; for example, by allowing them to stay longer at a given assignment or
in a given pay grade if they so desire. On the other hand, longer tours would mean
that military personnel would receive a narrower array of job experiences during the


68Rick Maze, “Pentagon Opposes GI Bill Transfer To Family Members,” Army Times,
September 11, 2001, 10.
69Vince Crawley, “Expanding GI Bill Gets DOD Support,” Marine Corps Times, March 12,

2001, 20.



course of their careers, and staying longer in a given pay grade could reduce
promotion rates for more junior personnel.
Another option might be to expand current programs that allow military
personnel, usually officers, to earn an advanced degree at a civilian university or to
learn the best business practices of a given industry by working for a corporation
while remaining on active duty. In exchange for this opportunity, the servicemember
must agree to serve a specified number of years in the military at the completion of
the program (generally the service obligation is three years for every year in the
program). This type of program might be extended to more officers and enlisted
personnel as well. The military could benefit from this both by obtaining substantial
service obligations from participating servicemembers and by obtaining better
educated or more highly trained employees. The downside, of course, is that
servicemembers are unavailable for military duties while enrolled in these programs.
A less expensive variation of this would be to allow unpaid sabbaticals for a year or
eighteen months to allow military personnel to take a break from military life without
damaging their careers. This option might be an especially useful tool for retaining
those who basically enjoy military life, but who have need a temporary break in
service to attend to family problems or to recover from “burn out.”
Reduced PERSTEMPO Levels. This response would likely be effective if
retention shortfalls were caused primarily by dissatisfaction with high rates of
deployment away from home station. PERSTEMPO management is generally
considered to be an executive branch function, but Congress certainly has the ability
to influence PERSTEMPO levels in a number of ways. Two of the most obvious are
providing funding for fewer military commitments or, alternatively, increasing the
number of military personnel available to handle existing or future commitments.
Other options frequently mentioned include requiring greater use of Reserve
Component forces or civilian contractors. Another option, often mentioned with
respect to the various Services “transformation initiatives,” would be to promote the
procurements of various systems that reduce the demands on military manpower. For
instance, plans for a new Navy destroyer call for a crew of about 95 people, as
compared to a crew of more than 300 for current Navy destroyers.