Military Personnel: Services Need to Assess Efforts to Meet Recruiting
Goals and Cut Attrition (Letter Report, 06/23/2000, GAO/NSIAD-00-146).

Pursuant to a congressional request, GAO provided information on the
military services' efforts to meet their recruiting goals and cut
attrition, focusing on the services': (1) responses to recent recruiting
shortfalls; and (2) efforts to reduce their historically high attrition
rates for first-term enlistees.

GAO noted that: (1) to address problems in recruiting personnel, the
Army, the Navy, and the Air Force have increased their numbers of
recruiters and their advertising budgets and have offered larger
enlistment bonuses and more money for college; (2) these tools have been
shown by past research to help the services attract new recruits; (3)
the services have also sought innovative ways of expanding their
recruiting market without reducing the quality of recruits, for example,
by targeting persons attending community colleges and persons without
high school degrees who meet other quality standards; (4) because so
little time has passed since the services have begun to respond to their
recent recruiting problems, they cannot yet assess the long-term success
of their efforts; (5) also, the services do not yet know which of their
new recruiting initiatives work best; (6) while each of the services
might point to localized successes, the Department of Defense does not
know the extent to which the services might be competing with each other
for the same potential recruits; (7) until sufficient time has passed
and each of the services consistently meets its goal, DOD cannot be
assured that individual service strategies will collectively enable DOD
to meet its overall recruiting requirements; (8) in addition to
improving the screening of applicants before they enlist, the services
have also begun many efforts to reduce the attrition of first-term
enlistees while they are in training and after they have been assigned
to their first duty stations; (9) these efforts include providing extra
attention to recruits struggling during basic training and disciplining
and working with enlistees who have completed training and are
experiencing minor behavioral problems; (10) the latest attrition data
available indicate that first-term attrition has reached all-time highs
for DOD enlistees; (11) the attrition rate for enlistees entering the
services in the mid- to late 1980s hovered between 30 and 34 percent,
and this rate gradually rose in the 1990s from a low of 33 percent to a
peak of nearly 37 percent for enlistees entering the services in fiscal
year 1994 and 1995; (12) greater success in reducing attrition may not
yet be apparent because the services have just begun many of their
efforts; and (13) the services, however, are not developing tools needed
to measure the long-term success of their efforts, thus limiting their
ability to judge the effectiveness of those efforts in reducing
attrition.

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

 REPORTNUM:  NSIAD-00-146
     TITLE:  Military Personnel: Services Need to Assess Efforts to
	     Meet Recruiting Goals and Cut Attrition
      DATE:  06/23/2000
   SUBJECT:  Military recruiting
	     Attrition rates
	     Interagency relations
	     Military training
	     Cost effectiveness analysis
	     Enlisted personnel
	     Performance measures
IDENTIFIER:  Army General Educational Development Plus Program
	     Army College First Program
	     Army Partnership for Youth Success Program
	     Armed Forces Qualifications Test
	     Navy College Assistance Student Headstart program
	     Navy Technical Partnership Program
	     Navy College Program
	     Air Force Enlistment Bonus Program

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO/NSIAD-00-146

Appendix I: DOD and Service Actions to Implement Our Recommendations

32

Appendix II: Objectives, Scope, and Methodology

42

Appendix III: Comments From the Department of Defense

44

Appendix IV: GAO Contacts and Staff Acknowledgments

47

Table 1: Services' Success in Meeting Their Annual Recruiting
Goals for New Active-Duty Enlistees 5

Table 2: DOD Attrition Rates for Active-Duty Enlistees 19

Table 3: Remedial Programs Offered at Fort Jackson's Basic
Training Site 24

Table 4: Status of DOD and Service Implementation of Our
Previous Recommendations 34

Figure 1: The Army's Expenditures for Advertising and
Enlistment Bonuses (fiscal year 2000 constant
dollars in millions) 7

Figure 2: The Navy's Expenditures for Advertising and
Enlistment Bonuses (fiscal year 2000 constant
dollars in millions) 9

Figure 3: The Air Force's Expenditures for Advertising and
Enlistment Bonuses (fiscal year 2000 constant
dollars in millions) 11

Figure 4: The Marine Corps' Expenditures for Advertising and
Enlistment Bonuses (fiscal year 2000 constant dollars
in millions) 13

Figure 5: Percentage of Fiscal Year 1995 Enlistees Separating
During Particular Periods of Their First Terms 20

Figure 6: Trends in the Timing of First-Term Enlisted Attrition 21

Figure 7: Primary Reasons That Enlistees Who Entered the
Services in Fiscal Year 1998 Were Separated in Their
First 6 Months of Service 22

DOD Department of Defense

National Security and
International Affairs Division

B-285028

June 23, 2000

The Honorable Tim Hutchinson
Chairman
The Honorable Max Cleland
Ranking Minority Member
Subcommittee on Personnel
Committee on Armed Services
United States Senate

The Department of Defense (DOD) faces a significant challenge in recruiting
and retaining the hundreds of thousands of new recruits it enlists each
year. The last 2 years, in particular, have been difficult for the military
services as they have struggled to meet their recruiting goals. This
difficulty, which some believe represents a recruiting crisis, makes the
services' problems with first-term attrition rates even more critical. The
early separation of new recruits is costly in that the services' recruiting
and training investment in each enlistee averages almost $38,000. In
response to the request of the former Chairman and the current Ranking
Member, we assessed (1) the services' responses to recent recruiting
shortfalls and
(2) the services' efforts to reduce their historically high attrition rates
for first-term enlistees.

To address mounting problems in recruiting sufficient numbers of qualified
enlisted personnel, three services--the Army, the Navy, and the Air
Force--have increased their numbers of recruiters and their advertising
budgets and have offered larger enlistment bonuses and more money for
college. These tools have been shown by past research to help the services
attract new recruits. The services have also sought innovative ways of
expanding their recruiting market without reducing the quality of recruits,
for example, by targeting persons attending community colleges and persons
without high school degrees who meet other quality standards. Because so
little time has passed since the services have begun to respond to their
recent recruiting problems, they cannot yet assess the long-term success of
their efforts. Also, the services do not yet know which of their new
recruiting initiatives work best. For example, the Navy does not know the
extent to which each of the changes it has made to its recruiting
program--increasing its number of recruiters, its advertising budget, or its
enlistment bonuses--contributed toward meeting its goal in fiscal year 1999
and whether that strategy will work in the future. Finally, while each of
the services might point to localized successes, such as the Army's ability
to channel young people into hard-to-fill jobs by offering large enlistment
bonuses, DOD does not know the extent to which the services might be
competing with each other for the same potential recruits. Until sufficient
time has passed and each of the services consistently meets its goal, DOD
cannot be assured that individual service strategies will collectively
enable DOD to meet its overall recruiting requirements.

In addition to improving the screening of applicants before they enlist, the
services have also begun many efforts to reduce the attrition of first-term
enlistees while they are in training and after they have been assigned to
their first duty stations. These efforts include providing extra attention
to recruits struggling during basic training and disciplining and working
with enlistees who have completed training and are experiencing minor
behavioral problems. These actions appear promising as they target recruits
who might previously have been summarily discharged. Nonetheless, the latest
attrition data available indicate that first-term attrition has reached
all-time highs for DOD enlistees. The attrition rate for enlistees entering
the services in the mid- to late 1980s hovered between
30 and 34 percent, and this rate gradually rose in the 1990s from a low of
33 percent to a peak of nearly 37 percent for enlistees entering the
services in fiscal years 1994 and 1995. Greater success in reducing
attrition may not yet be apparent because the services have just begun many
of their efforts, or the continued high rate of attrition may indicate that
without these efforts, the services' losses would be even higher. The
services, however, are not developing tools needed to measure the long-term
success of their efforts, thus limiting their ability to judge the
effectiveness of those efforts in reducing attrition.

Because DOD does not have the tools at present to determine whether the
services' recruiting and retention efforts will be successful in the long
term, we are recommending that DOD and the services (1) assess the relative
success and cost-effectiveness of their recruiting strategies in meeting
DOD's overall needs by applying one service's best practices to the other
services whenever possible and by minimizing cross-service competition and
(2) put in place tools for measuring the long-term success of the services'
attempts to reduce attrition by confirming that the services' short-term
remedial efforts are not simply delaying attrition to later points in
enlistees' first terms. In written comments on a draft of this report, DOD
agreed with our recommendations and cited steps it plans to take to
implement them.

Until fiscal year 1998, the services had been successful in meeting their
recruiting goals for the all-volunteer force of enlistees. In fiscal year
1998, the Navy and the Army were the first services to miss their annual
recruiting goals for active-duty enlisted personnel. That year, the Navy
achieved 88 percent of its goal and the Army 99 percent. The following year,
the Army made only 92 percent of its goal and the Air Force made
95 percent of its objective. (See table 1.) For some Members of Congress,
the fact that the services were missing their recruiting goals indicated a
recruiting crisis. Added to the services' recent struggles to meet
recruiting goals is the fact that, historically, about one-third of their
enlistees do not complete their first terms of service.

          Fiscal year 1997          Fiscal year 1998          Fiscal year 1999

 Service  Goal   Actual Percentage  Goal   Actual Percentage  Goal   Actual Percentage
                        of goal                   of goal                   of goal
 Army     82,000 82,088 100         72,550 71,752 99          74,500 68,209 92
 Navy     50,135 50,135 100         55,321 48,429 88          52,524 52,595 100
 Marine
 Corps    34,512 34,548 100         34,267 34,284 100         33,668 33,703 100
 Air
 Force    30,310 30,310 100         30,194 31,685 105         34,400 32,673 95

Source: DOD.

In trying to reduce or eliminate recruiting shortfalls as quickly as
possible, the Army, the Navy, and the Air Force have increased their number
of recruiters, advertising budgets, and enlistment bonuses. Because the
Marine Corps has successfully met its recruiting goals, it does not plan to
initiate any major changes to its recruiting program in these areas. While
the individual services may be able to trace an increase in numbers
recruited to increases in recruiting resources, they have as yet been unable
to determine which of their separate efforts are most effective. For
example, they are not sure whether it is more effective, with limited
resources, to increase enlistment bonuses, numbers of recruiters, or
advertising. Without a history of what works with today's recruits, the
services must experiment with and document how well precise amounts of
benefits and bonuses result in the maximum attraction and retention of
enlistees. Finally, the services are at present unable to determine the
extent to which their individual efforts are resulting in greater
competition among themselves and thereby unnecessarily increasing the
overall cost of recruiting to DOD.

In response to missing recruiting goals for the past 2 years, the Army has
been putting additional resources into areas that historically have proven
to result in greater numbers of recruits. It has not yet, however, had time
to fully analyze whether what has worked in the past is currently working or
is likely to work in the future. From fiscal year 1993 through 1998, the
Army increased its number of recruiters from 4,368 to 6,331 and increased
its advertising expenditures from $34.3 million in fiscal year 1993 to
$112.9 million in fiscal year 1999 (in fiscal year 2000 constant dollars).
The Army has also offered an array of enlistment bonuses to qualified
personnel and increased the maximum amount offered from $12,000 to $20,000.
Enlistment bonus expenditures increased substantially in just the past year,
from $59.7 million in fiscal year 1998 to $105.2 million in fiscal year
1999. Figure 1 illustrates increases in the Army's advertising and
enlistment bonus expenditures for fiscal years 1993-99.

Source: DOD.

The Army has also recently announced new ways to expand its recruiting
market to persons it would not have sought out in the past. For example,
through its General Educational Development Plus program, the Army intends
to target youth who do not have high school diplomas but who have
higher-than-average aptitude scores and no histories of disciplinary
problems. It also plans to focus attention on youth who are college-bound.
The College First Program, for example, proposes to pay enlistees attending
college $150 per month for up to 2 years after they have signed a contract
to enlist but before they actually do so.

In an effort to more directly compete with private sector pay and benefits,
the Army is developing the Partnership for Youth Success Program. Under this
proposed program, the Army would obtain guarantees from private sector
companies that they would have jobs waiting for specially trained enlisted
personnel who successfully complete their tours. The companies would benefit
by obtaining highly skilled personnel, and the former Army personnel would
benefit by being assured a secure job, using their skills upon their
departure from the Army. While this program could have a very beneficial
effect on attracting quality recruits, it could also have an adverse effect
on retention if enlistees are provided incentives to separate from the
military.

Navy Recruiting Command officials noted several factors that they believe
enabled them to meet their recruiting goal in fiscal year 1999. They were
unable, however, to say precisely which of these factors worked or how well.
For example, the Navy increased its number of recruiters from 3,342 in
fiscal year 1998 to 4,725 in fiscal year 1999. Simultaneously, it opened
recruiting jobs to lower ranked, but "hard-charging," enlisted personnel.
Also, the Navy substantially increased its advertising expenditures--from
$37.8 million in fiscal year 1997 to $67.3 million in fiscal year 1999.
Finally, the Navy enhanced its bonus offerings to enlistees. For example, it
awarded special bonuses of $3,000 to enlistees who agreed to enter basic
training between February and May, a time when the Navy is traditionally
hard-pressed to fill their slots. Figure 2 illustrates the Navy's
expenditures for advertising and enlistment bonuses for fiscal years
1993-99.

Source: DOD.

While increasing the amounts of resources spent on recruiting, the Navy also
began to seek ways to expand its recruiting market. For example, it began to
target persons who have prior military service and persons without high
school diplomas who scored high on the Armed Forces Qualification Test, were
at least 19 years old, had a minimum of three character references, and had
proof of stable employment. The Navy is also targeting college-bound
recruits under three separate programs. Under the first program--the Navy
College Assistance Student Headstart Program--selected recruits in the
nuclear and submarine fields are placed on active duty while they are in
college, receive entry-level pay for up to 1 year, and then attend basic
training. In the second program, called Technical Preparation Partnerships,
the Navy coordinates with community colleges to allow recruits to earn their
associates' degrees while they are serving their first enlistment terms. A
third program, the Navy College Program, allows Navy servicemembers to
receive college credit for Navy-provided training.

Air Force recruiting officials told us that a false sense of security
contributed to their inability to meet fiscal year 1999 recruiting goals.
Because the Air Force had easily met its goals in the past, Air Force
officials did not believe that such recruiting investments were necessary.
Fiscal year 1999's recruiting results indicated that the pool of new
recruits the Air Force had historically counted on would not always be
readily available. To increase the chance of future success, the Air Force
has already begun to purchase television advertising, offer enlistment
bonuses to new recruits, and increase the number of Air Force recruiters.
Because the Air Force will be changing its recruiting methods in the future,
it has not yet had time to collect historical data on which recruiting
techniques work most successfully.

Because Air Force recruiters have historically signed up about two times
more recruits than recruiters in other services while working fewer hours,
the Air Force has not needed to assign as many persons to recruiting duty as
the other services have. For example, in fiscal year 1999, the Marine Corps
employed 2,650 recruiters to recruit 33,685 new active-duty Marines, while
the Air Force employed only 950 recruiters to recruit 32,068 enlistees.
Banking on continued high productivity in fiscal year 1999, the Air Force
did not staff its field recruiting force to its authorized level of 1,209.
That is, even while the number of required annual enlistments increased and
the Air Force said it became more difficult to recruit, the Air Force not
only did not increase the number of its recruiters but allowed the number to
fall below authorized levels. For fiscal year 2000, the Air Force plans to
increase its number of actual recruiters to 1,446.

Prior to fiscal year 1999, the Air Force also maintained a minimal
advertising budget and had not requested money for paid advertising on
commercial television. In that year, for the first time, the Air Force
requested and received funding for television advertisements. The Air Force
increased expenditures for all advertising from approximately
$12.3 million in fiscal year 1998 to $56.8 million in fiscal year 1999.

In October 1998, the Air Force expanded its enlistment bonus program to
target persons willing to commit to 6- rather than 4-year contracts in
critical and highly technical skills, such as combat controllers, pararescue

personnel,1 linguists, and security forces. The Air Force believed that
offering such bonuses (1) positioned it for a better return on its
recruiting and training investment, (2) provided another tool to attract
youth into the Air Force, and (3) would result in improved retention over
time and ultimately in a reduction in future requirements for new recruits
without prior military service. Enlistees in approximately 100 occupations
are eligible for bonuses ranging from $2,000 to $12,000. Combat controllers
and pararescue personnel are eligible for the maximum bonus of $12,000.
Figure 3 illustrates the Air Force's increased expenditures for advertising
and enlistment bonuses for fiscal years 1993-99.

Source: DOD.

Finally, to address recruiting and retention issues, the Air Force formed a
Recruiting and Retention Task Force on March 1, 2000. Led by a brigadier
general, the task force reports to the Under Secretary of the Air Force and
the Vice Chief of Staff of the Air Force. The purpose of the task force is
to serve as the single point of contact for the integration, consolidation,
and coordination of over 120 recruiting and 89 retention initiatives.

The Marine Corps is the only service that has successfully met its
recruiting goals each year. Consequently, it does not plan to initiate major
changes in its recruiting strategy. Over the years, the Marine Corps has
slowly but steadily increased its recruiting workforce and advertising
budget. Recruiting officials believe they have been able to meet their goals
because the Marine Corps has a consistent and recognizable identity that
appeals to youth; because it has an institutional, financial, and resource
commitment to supporting recruiting services; and because its recruiters
work extremely hard. Data from the Defense Manpower Data Center's 1998
survey of recruiters indicate that the Marine Corps' recruiting success
comes at a price. Marine Corps recruiters work longer hours and take less
leave than recruiters in any other service.

The Marine Corps does not see itself in competition with the private sector
to the same extent that the other services do. As such, it is
institutionally against attempting to attract recruits with bonuses,
changing its entry-level requirements, or increasing the number of
recruiters. However, in order to maintain a "level playing field" with the
other services, the Corps did increase its advertising expenditures
substantially in fiscal year 1999 and nearly doubled its funding for
enlistment bonuses in the same year.
Figure 4 illustrates the Marine Corps' expenditures for advertising and
enlistment bonuses for fiscal years 1993-99.

Source: DOD.

Best and How to Minimize Cross-Service Competition

Despite years of research on how best to recruit into the military, the
Army, the Navy, and the Air Force are unsure of what recruiting strategies
will work best in today's environment. Their concerns are that private
sector competition, the economy, the attitudes and skills of youth, and the
views of their parents toward the military have so changed over time that
old ways of doing things may no longer be applicable. DOD and the services
cannot yet determine whether they are taking the appropriate steps to
increase the number of young people they enlist without reducing the chances
that these persons will perform acceptably and complete their enlistment
tours. A DOD study released in March 2000 suggests that the Army and the
Navy should consider recruiting between 85 and 90 percent high school
diploma graduates, as opposed to the previous 90-percent benchmark. DOD
believes that recruiting in this range should allow the services to adjust
better to extraordinarily tight recruiting markets. However, DOD is not yet
able to assess the long-term effect of the services' attempts to recruit
persons outside of their traditional markets or to bring in a somewhat lower
percentage of high school diploma graduates. Lastly, DOD and the services
have been unable to determine how the services' individual methods of
increasing the number of recruits they enlist affect competition among
themselves.

DOD and the services are undertaking many efforts to improve their knowledge
of the recruiting market and of recruiting methods. For example, a recently
completed study for DOD by a contractor pointed out how DOD could reduce
redundancies in the ways the services purchase marketing information and how
it could improve the way the services market themselves to those who
influence youth. Two other studies are in final stages: a RAND study that
will provide a foundation for better determining advertising strategies
using national and local media and a Navy College Fund study that will
analyze the effects of differing enlistment incentives on the enlistment of
high quality recruits. DOD has also begun many initiatives to reengineer
recruiting. For example, it plans to create an on-line recruiting station,
increase military internet traffic, take advantage of state-of-the-art
civilian telemarketing business practices, and test the feasibility of
processing applicants in areas closer to an applicant's home.

These initiatives and studies appear promising in opening up new avenues for
military recruiting. Two important areas, however, need further work:
determining the most cost-effective mix of recruiting resources and
analyzing the extent to which the services are competing with each other
when they increase their numbers of recruiters or the amounts spent on
enlistment bonuses and college incentives. In regard to the first area, DOD
believes that such a study would be of value and plans to update a model
that it hopes will form the basis for making wise decisions about how to
best use recruiting resources in today's economy. The existing model relates
the quantity of recruits, recruit quality, and total recruiting costs and
identifies potential tradeoffs among them. It does not, however, address the
relative effectiveness of alternative uses of recruiting dollars for such
things as additional recruiters, advertising, and enlistment bonuses. In the
second area, DOD states that it has not completed an in-depth analysis of
whether the services are competing with each other when they increase the
number of recruiters or the amounts spent on incentives. Currently, DOD has
no evidence to confirm or disprove whether increases in recruiters and
enlistment incentives have resulted in increased numbers of recruits for DOD
as a whole.

Such analyses of the effectiveness of the services' actions in meeting DOD's
overall recruiting needs would enable DOD and the services to develop
fact-based policies on how best to spend their recruiting resources and to
minimize interservice competition. These types of analyses are also in
keeping with the Government Performance and Results Act of 1993
(P.L. 103-62, Aug. 3, 1993), which was designed to create a new
results-oriented federal management and decision-making approach that
requires agencies to set goals, measure performance, and report on their
accomplishments. DOD's Performance Plan for Fiscal Year 2000 includes a
performance goal to "recruit, retain, and develop personnel to maintain a
highly skilled and motivated force capable of meeting tomorrow's
challenges." The plan includes two related performance measures: number of
recruits enlisted and enlisted retention rates. The plan also includes the
percentage of "quality" recruits as a related performance indicator.2

One example of a new program needing evaluation is the Army's experiment
with bringing on board a small number of motivated and intelligent high
school dropouts in an attempt to enlarge the recruiting market. Before this
time, the Army was reluctant to enlist persons without high school diplomas
because they have historically had a higher attrition rate. Army officials
noted that it will take about 3 years to determine whether these recruits
will perform at acceptable levels so that decisions about enlarging and
maintaining the program can be made. The Army has planned for such an
evaluation to be conducted by a contractor.

Similarly, the impact of higher enlistment bonuses remains in question.
While higher enlistment bonuses may result in short-term successes in
enlisting persons into hard-to-fill jobs, neither the services nor DOD
planned to conduct studies to determine what the optimal amounts of such
monetary incentives should be or whether such bonuses primarily resulted in
competition among the services for the same limited number of recruits. At
present, no research exists to help the services determine how much these
bonuses should be. Recent decisions on how much to offer in bonuses have in
some cases been based on one service's attempt to remain competitive with
another. For example, Navy officials told us that they felt compelled to
"play catch-up" with the Army after the Army increased its bonuses to
$20,000 for some occupations. On the other hand, an Army official believed
that the Army had lost its traditional "market share" of recruits to the
Navy this year. Specifically, he believed that the Navy's increased number
of recruiters were taking away persons who would otherwise have joined the
Army. In other words, these services see evidence that success for one
service often comes at the expense of another. Service officials believe
that at least to some extent, they are simply shifting their "market shares"
of a fixed number of new recruits. While Army officials believe that they
can determine how much the services are "expanding the market" and how much
they are simply competing for a fixed number of recruits, we found no
evidence that the services or DOD are systematically tracking the extent to
which such shifting of market shares is resulting in unnecessarily increased
recruiting costs.

Another enlistment program--offering bonuses to recruits who enter the
service in certain hard-to-recruit months--could have an undesired negative
side effect. Both the Army and the Navy are offering incentives for recruits
to sign up and immediately begin training, even though the Navy believes
that recruits who spend time in the Delayed Entry Program are better
motivated and have better chances to succeed than those who immediately
enter active duty. For this reason, while these new programs may help the
Navy to meet immediate enlistment goals or the Army to fill costly training
seats, they may not be cost-effective in the long term if those who would
have dropped out during the Delayed Entry Program are now dropping out in
basic training. In fact, Navy basic training officials cited shorter time
spent in the Delayed Entry Program as a major reason that basic training
attrition has risen. Army officials disagreed, arguing that there is little
evidence that time spent in the Delayed Entry Program has any relation to
attrition rates. Army officials believe that it is more cost-effective to
fill costly training slots and to take the chance that recruits will drop
out during training than it is to allow the training slots to remain
unfilled.

Recently, efforts have intensified to bring together all services and the
private sector to discuss what needs to be done to address recruiting needs
in the 21st century. In a September 1999 symposium, top DOD military and
civilian officials concluded that the lack of a strategic, long-term
perspective hampers the development of effective recruiting programs and
that more accurate information is needed to better allocate resources.
Officials noted that better research was needed to help determine whether
their current approaches were cost-effective and whether current enlistment
quality standards remain valid.

The services are taking steps to reduce first-term attrition. Initiatives
are being taken to improve the screening of applicants before they enlist,
to provide extra attention to new recruits struggling during basic training,
and to discipline and retain enlistees who complete training and experience
minor behavioral problems. Many of the services' attrition-reduction actions
are being taken in response to recommendations we have made in previous
reports on this issue. These recommendations have included ways DOD and the
services could tie their recruiter incentive systems more closely to their
recruits' graduation rates from basic training, ways they could more
thoroughly screen out military applicants who have disqualifying medical or
criminal histories, and ways they could target and retain some first-term
enlistees who simply need remedial attention.3 Many of the services' efforts
to reduce first-term attrition appear promising. However, the most recently
available data for all the services indicate that attrition rates for
persons entering the services in fiscal years 1994 and 1995 and being
separated by the end of fiscal years 1998 and 1999 were at historically high
levels.4 Since many of the services' initiatives were introduced after these
groups of enlistees had completed training, they could not be expected to
have much impact on the attrition rate for these groups. For initiatives
that were introduced before fiscal year 1994, attrition rates might have
risen even higher without them.

Attrition Rates

Our analysis of service-wide attrition data from the Defense Manpower Data
Center on enlistees who entered the services in fiscal years 1994 and 1995,
the two most recent year groups for whom a full 4 years of data is
available, indicates that service efforts initiated before these years to
reduce training attrition had not yet been successful. Service actions
initiated to reduce training attrition after those years would not be
reflected in these groups' rates. In fact, the rate of attrition at the
48-month

point for these enlistees has risen.5 As shown in table 2, over the last
11 years, the DOD-wide attrition rate has varied from a low of 29.6 percent
for enlistees who entered the services in fiscal year 1985 to an all-time
high of 36.9 percent for those who entered in fiscal year 1994. For
enlistees entering the services in fiscal year 1995 and separating early
before the end of fiscal year 1999, DOD's overall rate remained high, at
36.8 percent. The rates for the Marine Corps and the Air Force remained
virtually the same in fiscal years 1994 and 1995; the rate declined by
around 1 percentage point, and the Navy's rate increased by 1.7 percentage
points. The Marine Corps' rate peaked for enlistees entering the service in
fiscal year 1990.

During the past 6 years, only the Marine Corps has begun to lower its
attrition rate. Its rate for enlistees entering in fiscal year 1995 was
almost the same as it was for enlistees entering in fiscal year 1985, while
the rates of the other services remain 7-9 percentage points above their
1985 level. (See table 2.) Marine Corps officials attributed their success
in reducing attrition to a change in philosophy from weeding recruits out
quickly to taking more time to help them meet Marine Corps standards.

 Numbers in percentages
 Fiscal year
 of enlistment Army  Navy  Marine Corps  Air Force DOD-wide
 1985          30.5  30.6  33.7          24.6      29.6
 1986          31.3  33.8  35.4          26.3      31.4
 1987          31.4  31.5  33.0          25.3      30.5
 1988          33.7  31.6  30.4          25.5      31.3
 1989          35.3  34.0  32.7          30.1      33.7
 1990          36.4  32.6  36.2          30.4      34.2
 1991          36.8  30.5  34.2          31.7      33.6
 1992          35.9  32.2  32.2          30.0      33.2
 1993          39.3  35.8  31.5          32.5      35.8
 1994          40.1  37.8  33.2          32.7      36.9
 1995          39.0  39.6  33.2          31.9      36.8

Source: Our analysis of Defense Manpower Data Center data.

Training

A significant portion of first-term attrition occurs during enlistees' first
6 months of service, when most enlistees are in basic or follow-on training.
After the first 6 months of service, early separations decline. This pattern
holds true for all four services. Figure 5 shows the timing of premature
separations for enlistees who entered the services in fiscal year 1995.

Source: Our analysis of Defense Manpower Data Center data.

From fiscal year 1985 through fiscal year 1995, as overall enlisted
attrition has risen, the percentage of attrition occurring in the first 6
months of service has risen more quickly than it has during later intervals
of service (see fig. 6). For this reason, the services have focused many of
their initiatives to reduce attrition on activities that occur in an
enlistee's first
6 months.

Source: Our analysis of Defense Manpower Data Center data.

Service-wide attrition data from the Defense Manpower Data Center indicate
that enlistees who leave the military in their first 6 months are separated
for three general reasons: (1) medical/physical problems, that is, medical
conditions that existed prior to service or physical problems that developed
while enlistees were in training; (2) erroneous or fraudulent enlistment,
indicating either that the services did not detect military applicants'
disqualifying conditions prior to their enlistments or that the applicants
deliberately withheld disqualifying information from the services; and (3)
performance problems, such as failure to pass the physical training test,
loss of motivation, or inability to adapt to military life. The remaining
separations at the 6-month point are for various other reasons, none of
which is particularly predominant.

These reasons include misconduct, excessive weight or body fat, character
and behavior disorders, alcoholism, drug use, and homosexuality. (See
fig. 7.)

Notes:

Performance problems include such behaviors as failure to pass physical
training tests, losing motivation, or inability to adapt to military life.

Medical and physical problems include conditions that were not detected when
enlistees were physically examined before they entered the service,
conditions that developed while enlistees were in the service, or injuries
that enlistees suffer while in the military.

Fraudulent enlistment is the entry into service of someone who knowingly
concealed information that would have disqualified him or her from military
service. Erroneous enlistment is the entry into service of someone who had a
disqualifying condition that was not discovered beforehand and was not
deliberately concealed at the time of enlistment. Separations for fraudulent
or erroneous enlistment apply to discoveries of disqualifying conditions
covering all military service requirements, for example, maximum number of
allowable dependents, drug use, medical problems, psychological problems, or
criminal histories.

Source: Our analysis of Defense Manpower Data Center data.

Service-wide data from the Defense Manpower Data Center indicate that about
34 percent of all separations during training are for performance problems.
Officials at the Army, the Navy, and the Marine Corps training bases we
visited believe that attrition because of performance problems can be
reduced and have instituted remedial programs to provide extra attention to
recruits struggling with behavioral, physical training, or academic
problems. While the Army and the Marine Corps report that they have recently
been able to reduce attrition during training, the most recently available
data from the Defense Manpower Data Center does not yet confirm this.
Defense Manpower Data Center data indicate that attrition rates remained
high for all the services' enlistees during their first 6 months of service.
For enlistees who entered the services in fiscal year 1998, 15.8 percent of
all Army enlistees were separated before they had served 6 months; 16.2
percent of all Navy enlistees; 17.5 percent of all Marine Corps enlistees;
and 12.1 percent of all Air Force enlistees.

Table 3 describes the remedial training programs that target groups of
enlistees for retention at the Army's basic training site at Fort Jackson,
South Carolina. Training officials reported that in fiscal year 1999, they
placed 7,612 enlistees in these programs, or 22 percent of the 34,265
recruits who were trained at Fort Jackson that year. Of those who entered
remedial programs, 68 percent successfully completed their programs and were
returned to regular basic training units.

 Program                            Description
                                    Remedial attention paid to recruits who
 Think It Over                      have second thoughts about basic
                                    training in the first week after they
                                    have arrived
                                    A 4-day language skills program
 English as a Second Language       designed to help recruits whose primary
                                    language is not English
                                    A required remedial physical fitness
 Fitness Training Unit              training course for all recruits who
                                    fail the physical fitness test given on
                                    the second day of basic training
                                    A remedial academic program for those
                                    whose scores on the DOD entrance
 Prepare to Train                   aptitude test suggest that they might
                                    have difficulty in the academic portion
                                    of training
                                    A program for recruits who have
 Retraining/Holding Unit, "See It   performance and behavioral problems
 Through"                           during their first through final weeks
                                    of basic training
                                    A mandatory program to help injured
 Physical Training and              enlistees remain physically fit, heal
 Rehabilitation Program             properly, and quickly return to
                                    training
                                    A 3-week remedial physical training
 Physical Training Course           course for enlistees who fail the Army
                                    Physical Fitness Test required for
                                    graduation

Three of these programs are offered at all Army basic training sites: the
Fitness Training Unit, the Retraining/Holding Unit, and the Physical
Training and Rehabilitation Program. Army officials believe that such
programs have resulted in reduced attrition in the first 6 months of
service. Army data indicate that 6-month attrition peaked in fiscal year
1998 at
17.8 percent and then dropped to 14.6 percent in fiscal year 1999 and
14.3 percent in fiscal year 2000.

When Navy basic training attrition rates grew steadily from fiscal years
1996 through 1999--from 13.2 to 18.3 percent--the Navy became concerned
about how to retain more recruits. Navy officials believe that attrition
rates would have gone even higher in fiscal year 1999 if they did not have
three remedial programs: the Personal Applied Skills Streaming Program,
which helped 861 recruits to address anger, motivational, and behavioral
problems; the Fundamental Applied Skills Training Program, which provided
instruction for 1,526 recruits struggling with academic problems; and PT-O,
a program that helped 1,555 recruits address physical training problems.
Before these programs were implemented, recruits struggling for these
reasons might have been separated. While these programs appear promising,
Navy officials have not collected data on how many enlistees who were
enrolled in these remedial programs ultimately completed their first tours
of duty. Therefore, the Navy cannot determine whether such programs were
only delaying rather than preventing early separations.

Data maintained by the Marine Corps indicate that it has had some success in
reducing attrition during basic training. For example, the Marine Corps
reports that attrition at Parris Island, South Carolina, dropped from
17.8 percent in fiscal year 1998 to around 12.1 percent in fiscal year 1999.
Training officials attributed this drop almost entirely to a change in
training philosophy that involves working with troubled recruits rather than
discharging them. Training officials stressed that such a change does not
reduce quality or graduation requirements.

In keeping with the Marine Corps' change in philosophy, training officials
instituted new or reinforced ongoing programs to reduce attrition. These
programs include a week-long remedial program to help recruits pass their
marksmanship qualification and a physical conditioning program for enlistees
who fail their initial physical fitness test. Marine Corps training
officials said that 81 percent of all male recruits and 90 percent of all
female recruits who were sent to the physical conditioning program were
returned to training. Officials did not have information on the attrition
rates of persons who completed their first tours of duty after they
successfully graduated from Parris Island's remedial programs.

Recruits can be separated for fraudulent enlistment if they knowingly
conceal information that would disqualify them from military service. Such a
discharge can result from concealing the number of one's dependents, from
lying about prior drug use, or from failing to report a prior medically or
criminally disqualifying condition. Recruits can be separated for erroneous
enlistment if their disqualifying conditions are discovered after they enter
the military, though this type of separation has also been used to discharge
Navy enlistees who test positive on the drug tests at basic training.

To reduce the numbers of fraudulent and erroneous enlistments involving
medical and psychiatric histories, the services have attempted to improve
the gathering of prior medical histories of military applicants. To reduce
the number of enlistments involving the concealment of criminal backgrounds,
DOD and the services have taken actions to improve the criminal screening
process. For example, the Military Entrance Processing Command, which is the
organization responsible for operating the 65 stations located throughout
the country where recruits receive their entry screening, has obtained
funding from the services to purchase fingerprint machines that will allow
the services to more efficiently conduct background searches before military
applicants enter the service.

To reduce premature separations for medical and physical reasons, DOD and
the services have taken actions to improve the medical screening of
enlistees before they enlist. For example, DOD has adopted a recommendation
we made in our report on attrition during training to revise the medical
form used to gather historical medical information on applicants for
military service in order to gather more precise medical information on
conditions that often result in early separations.6 Also in response to a
recommendation we made in that report, DOD now requires military applicants
to list their medical providers and insurers so that applicants will be more
forthcoming in reporting their past medical histories. Ultimately, the
Military Entrance Processing Command hopes to add medical screening tests,
such as tests for asthma, at its processing stations to help screen out
enlistees with medical conditions known to be high attrition risks.

Officials at training bases for all four services cited attempts to reduce
early separations for enlistees who suffer physical injuries.

ï¿½ At Fort Jackson, Army training officials determined that providing
athletic shoes designed to properly fit each recruit would reduce the risk
of injury for all trainees. While Fort Jackson officials believe that better
athletic shoes have resulted in fewer trainee injuries, these officials have
not collected data to verify this assumption.

ï¿½ Navy basic training officials have also taken several actions to reduce
injury-related separations by redesigning their training program. This
redesign has included rewriting the physical conditioning schedule,
including how often recruits are required to run and march, and retraining
drill instructors to emphasize the importance of keeping to a regimented
routine to reduce stress fractures and other injuries. Data maintained by
the Navy indicate that these efforts resulted in a reduction in the number
of severe stress fractures. Specifically, Navy data indicate that the number
of severe stress fractures peaked at almost 3.5 per thousand recruits in
fiscal year 1998 but was reduced in fiscal year 1999 to about 0.5 per
thousand.

ï¿½ After a spike in attrition in the first quarter of 1998, Parris Island's
training officials introduced initiatives to reduce training-related
injuries, such as a Sports Medicine and Rehabilitation Therapy Clinic to
rehabilitate rather than separate recruits who suffer injuries and to
expedite their return to training. Marine Corps training officials noted
that attrition due to lower extremity problems fell by 10 percent during the
first 6 months after the Sports Medicine Clinic was opened.

ï¿½ The Air Force began to change its basic training regimen after a spike in
the attrition of persons suffering injuries in 1995. In that year, training
officials began to alternate enlistees' wearing of boots and athletic shoes,
and in 1997, they added a central physical therapy clinic closer to where
trainees were located. In January 1999, a new sports medicine specialist
began to emphasize the importance of rehabilitating rather than separating
enlistees who suffered injuries. Because of this new approach to treating
injured enlistees, as well as other initiatives, the Air Force reports that
the number of separations for medical reasons, including injuries, was
reduced by half from fiscal year 1998 to fiscal year 1999.

Separations for enlistees who have completed training, that is, between
their 7th and 48th months of service, have less to do with screening and
more to do with performance and discipline. For example, 32.4 percent of
enlistees who entered the services in fiscal year 1995 and separated after
serving at least 6 months were separated for misconduct; 15 percent for
medical or physical problems; 9.7 percent for drug use; 8.3 percent for
performance problems; and 7.1 percent for character or behavior disorders.

The services have initiated a number of programs to reduce the attrition of
enlistees who have completed training. Like their programs to reduce
attrition from basic training, however, most of these efforts are too new to
demonstrate their long-term effects. Examples include the following:

ï¿½ The Air Force has created First Term Airman Centers at all bases. The
centers bring together all first-term airmen immediately upon arrival at
their new bases for indoctrination and mandatory training. These classes
last from 10 days to nearly 1 month.

ï¿½ The Navy is developing a retraining program that is nonpunitive in nature
and targeted at sailors lacking good life skills, such as sound
decision-making, effective goal-setting, and good financial management.

ï¿½ The Army has begun to rehabilitate and retain or punish enlistees
previously separated as deserters.

ï¿½ For a 1-year test period beginning in November 1999, the Army changed the
probationary status of first-term enlistees who had completed training.
Enlistees whose training is completed before their first
180 days of service are no longer treated as on probation; commanders may no
longer separate these persons using an expedited discharge process. The
intent of this program is to compel commanders to consider rehabilitation
rather than separation for enlistees in this category.

The only available DOD-wide data on enlisted separations are not very useful
in specifying why enlistees are leaving early.7 As we reported in 1997, the
codes used to categorize separations are vague, more than one code can be
chosen to classify the same separation, and the services use these codes
differently. As a result, we recommended that the Secretary of Defense issue
implementing guidance for DOD's separation codes to provide a reliable
database for DOD to manage attrition and for the services to set appropriate
targets for reducing it. While DOD formed a working group in April 1998 to
improve the separation codes, it has not yet issued its final set of codes
or its accompanying guidance. It was therefore too soon for us to assess
whether the new codes and guidance would correct problems we reported
earlier. DOD has also established long-term efforts to collect more precise
information on the medical reasons enlistees are being separated early.
Collecting such data will enable DOD and the services--at some point in the
future--to make fact-based decisions on which medical conditions result in
greater or lesser attrition risks.

DOD and the services also plan to collect survey information from enlistees
on why they are separating from the military. As required by the National
Defense Authorization Act for Fiscal Year 2000,8 DOD plans to administer a
survey on the attitudes toward service of individuals separating from the
military. This survey will be given to all personnel leaving the service or
transferring to the reserves during the 6-month period between April and
September 2000. The Defense Manpower Data Center has been tasked with
developing and analyzing the survey results.

Also, all four services have begun or will shortly begin to collect
information on their enlistees' perceptions of quality of life through entry
and exit surveys.

Attrition-Reducing Efforts

While all the services appear to have begun initiatives that will help them
either to reduce first-term attrition or to prevent it from rising further,
the effect of these initiatives will not be known until enlistees have had a
chance to serve their entire contract terms after these initiatives have
been in place. For example, to measure the effectiveness of initiatives
begun in fiscal year 1999, it will be necessary to track enlistees who
entered that year until 48 months later, in fiscal year 2003. While, in some
cases, the services have mechanisms in place to track the success of
remediation programs in boot camp, we did not see that the services had the
tools necessary to track the long-term success of individuals who had
successfully completed these programs. In other words, the services did not
know what happened to these persons after they completed their training.

DOD is experiencing a recruiting challenge that has called for an
extraordinary increase in the attention and the resources focused on this
area. As the services have reacted to this challenge, however, they have not
had the time to carefully determine which of their improvement efforts are
most effective and how their efforts affect the other services'
recruitments. As a result, DOD has been unable to determine the extent to
which the services are simply competing against each other and unnecessarily
raising the cost of recruiting. Nor has it been able to suggest to the
services the most cost-effective mix of their recruiting resources in
today's market.

In the area of first-term attrition, DOD and the services also need more
complete information on the root causes of attrition and on what initiatives
work in the long term. At present, the services have no way to demonstrate
whether or not the enlistees they are retaining in the short term by
spending more time and resources are good long-term investments. That is,
the services cannot yet demonstrate that these programs are not simply
delaying the attrition of enlistees whom they help to graduate from basic
training.

To maximize the effectiveness of DOD's recruiting efforts, we recommend that
the Secretary of Defense and the service secretaries assess the relative
success of their various recruiting strategies in meeting DOD's future
needs. This assessment should address how one service's best practices might
be applied to the other services, how the services can minimize competition
with each other for the same limited number of enlistees, and how the
services can create the most cost-effective mix of numbers of recruiters,
amounts of enlistment bonuses, college incentives, advertising, and other
recruiting tools.

Second, we recommend that the Secretary of Defense direct the service
secretaries to (1) continue to identify groups of enlistees whom they
believe could be rehabilitated and retained, both during and after training,
and (2) put in place tools for measuring the long-term success of these
rehabilitative efforts. These measurement tools should allow the services to
verify that attrition that is being reduced is not simply being delayed
until later in enlistees' first terms. These measurement tools should also
allow the services to determine what effect their recruitment of persons
outside their traditional market has on attrition.

In written comments on a draft of this report, DOD agreed with our
recommendations. Regarding our recommendation that DOD assess the relative
success of various recruiting strategies, DOD believes that some competition
among the services is healthy and productive but will consider using an
outside contractor to measure the impact of this competition and explore
ways to reduce duplicative efforts. DOD also indicated that it was looking
at a variety of ways to expand recruiting efforts and plans to develop a new
model that will allow tradeoff analyses to determine the relative
cost-effectiveness of the various recruiting resources. We believe that it
is important that this new model address the relative effectiveness of
alternative uses of recruiting dollars for such things as additional
recruiters, advertising, and enlistment bonuses. Regarding our
recommendation to take specific steps to reduce attrition rates, DOD agreed
to target enlistees for rehabilitation or remedial training and to develop
methods to measure accurately the success of such efforts in reducing
first-term attrition.

Appendix I presents a detailed assessment of DOD's and the services'
implementation of recommendations we have made in previous reports to
improve recruiting and reduce attrition. Appendix II presents a more
detailed description of our scope and methodology, and appendix III presents
DOD's comments on our draft in their entirety.

We are sending copies of this report to the Honorable William S. Cohen,
Secretary of Defense; the Honorable Louis Caldera, Secretary of the Army;
the Honorable Richard J. Danzig, Secretary of the Navy; the Honorable F.
Whitten Peters, Secretary of the Air Force; General James L. Jones,
Commandant of the Marine Corps; and the Honorable Jacob J. Lew, Director of
the Office of Management and Budget. We will also make copies available to
others upon request.

Please contact me at (202) 512-5140 if you or your staff have any questions
concerning this report. Key contacts and staff acknowledgments are listed in
appendix IV.

Norman J. Rabkin
Director, National Security Preparedness Issues

DOD and Service Actions to Implement Our Recommendations

At the request of the Senate Committee on Armed Services' Subcommittee on
Personnel, we have conducted a series of jobs to determine why the attrition
of enlisted personnel during their first terms of duty has remained
relatively constant despite the increased quality of new recruits and to
analyze the management and selection of recruiters. Our work has included
(1) a report outlining the reasons for attrition during the first 6 months
of an enlistee's term,9 (2) a report recommending how recruiter selection
and incentive systems could be improved to increase recruiter performance
and the likelihood that enlistees will complete their first terms,10 (3) a
study of reasons for enlisted attrition after basic training,11 and (4) a
study of the process of screening incoming recruits to detect criminal
backgrounds.12 In these four reports, we have made 20 recommendations on
ways the Department of Defense (DOD) could improve its management of
recruiter incentive systems, its process of screening incoming recruits, and
its retention of first-term enlistees.13

In this follow-on work, the Subcommittee asked us to evaluate the progress
DOD and the services have made in implementing our previous recommendations.
In summary, we found that DOD and the services had fully, substantially, or
partially implemented 17 of our 20 recommendations. One had not yet been
implemented, but some action had been taken. And in two cases, DOD had taken
no action. We judged the actions taken by DOD and the services using the
following categories:

ï¿½ Fully implemented. We used this category when we believed that the entire
wording of the recommendation had been fully implemented.

ï¿½ Substantially implemented. We used this category when we believed that DOD
and the services had either taken actions that met the intent of

      Recommendation        GAO report and number       DOD response
                                                  Revised the codes and
                                                  plans to issue
                                                  implementing guidance
                                                  telling the services how
 1. Revise separation codes                       and when to apply these
 and issue implementing     NSIAD-97-39, Jan. 6,  codes. Because the new
 guidance to collect data   1997                  codes and guidance have
 on reasons for attrition.                        not yet been issued, it
                                                  is too early to determine
                                                  whether identified
                                                  problems have been
                                                  corrected.

 2. Army, Marine Corps, and
 Air Force revise recruiter NSIAD-97-39, Jan. 6,
 incentive systems to       1997, and
 reward for recruits'       NSIAD-98-58, Jan. 30,
 graduation from basic      1998
 training.

 3. Require military
 applicants to provide the                        Fielded a new medical
 names of their medical                           history form to collect
 insurers and providers to  NSIAD-97-39, Jan. 6,  information on
 encourage more accurate    1997                  applicants' past medical
 reporting of past                                providers and insurers.
 problems.
 4. Revise medical history                        Fielded a new medical
 form to eliminate vague                          history form with more
 questions about            NSIAD-97-39, Jan. 6,  precisely worded
 applicants' past medical   1997                  questions on past
 problems.                                        problems.
                                                  Issued guidance to the
                                                  services on how to
 5. Use DOD's medical                             capture medical
 database to determine      NSIAD-97-39, Jan. 6,  information on
 whether new screening      1997                  applicants' specific
 tests should be added.                           diagnoses. The database
                                                  is now being developed
                                                  for future use.
 6. Place responsibility                          Does not believe such an
 for reviewing medical                            outside review is
 files outside MEPCOMa to   NSIAD-97-39, Jan. 6,  necessary. Believes that
 remove possible conflict   1997                  existing review by MEPCOM
 of interest.                                     is sufficient.

 7. Test all applicants for
 drugs at the MEPSb to
 reduce the numbers of      NSIAD-97-39, Jan. 6,
 enlistees separated for    1997
 drugs during basic
 training.

 8. Encourage recruits to
 undergo physical training  NSIAD-97-39, Jan. 6,
 while awaiting active      1997
 duty.

 9. Review policies
 allowing first-term
 enlistees to separate
 voluntarily more than 90   NSIAD-98-213, Sept.
 days before the end of     15, 1998
 their contractual
 obligations.

                                                  In late 1999,
                                                  administered a survey of
                                                  active-duty personnel
                                                  that included many items
                                                  aimed at assessing
                                                  quality of life. There is
                                                  also a legislative
                                                  mandate that the services
                                                  administer an exit
                                                  survey.
 10. Use quality-of-life
 surveys to target
 improvements that will     NSIAD-98-213, Sept.
 result in lower first-term 15, 1998
 attrition.

                                                  In testimonies and other
                                                  forums, DOD has made
                                                  reducing attrition an
                                                  important goal.

 11. Leadership should
 continually emphasize the  NSIAD-98-213, Sept.
 cost of attrition.         15, 1998

                            NSIAD-98-213, Sept.
                            15, 1998

 12. Target first-term
 enlistees for retention
 and remedial action.

 13. Reassess the
 appropriateness of
 providing favorable types
 of discharge to enlistees  NSIAD-98-213, Sept.   No action taken.
 whose behavior or          15, 1998
 performance led to their
 early separation.

 14. Use recruiters to      NSIAD-98-58, Jan. 30,
 select recruiters.         1998

 15. Explore the
 feasibility of developing
 or procuring assessment    NSIAD-98-58, Jan. 30,
 tests that can aid in the  1998
 selection of recruiters.

 16. Have drill instructors NSIAD-98-58, Jan. 30,
 and recruiters interact.   1998

 17. Give recruits a
 physical fitness test      NSIAD-98-58, Jan. 30,
 before they go to basic    1998
 training.

                                                  Has developed a timeline
                                                  for the services to
                                                  electronically transmit
                                                  background check forms
                                                  and fingerprints.

 18. Develop a DOD plan to
 use initiatives to
 electronically transmit    NSIAD-99-53, Feb. 23,
 the background check forms 1999
 and fingerprints.

                                                  Plans to purchase
                                                  automatic fingerprint
                                                  machines for all 65 MEPS
                                                  by the end of 2000. These
 19. Require that all                             machines will enable DOD
 national agency checks for NSIAD-99-53, Feb. 23, to do full fingerprint
 enlistment be based on     1999                  searches quickly.
 full fingerprint searches.

 20. Do not assign
 enlistees to their
 first-duty stations until  NSIAD-99-53, Feb. 23,
 their background checks    1999
 are complete.

Note: Gray shading in the table indicates areas in which either DOD or the
services have no responsibility for implementing the recommendation.

a MEPCOM = Military Entrance Processing Command.

b MEPS = Military Entrance Processing Station.

c DSS = Defense Security Service.

Objectives, Scope, and Methodology

DOD faces a significant challenge in recruiting and retaining the hundreds
of thousands of new recruits it enlists each year. In response to the
request of the former Chairman and the current Ranking Member, we assessed
(1) the services' responses to recent recruiting shortfalls and their
long-term effects and (2) the services' efforts to reduce their historically
high first-term enlisted attrition rates.

To determine what the services were doing to meet their recruiting needs, we
interviewed officials at each of the services' recruiting commands: the Army
Recruiting Command, Fort Knox, Kentucky; the Marine Corps Recruiting
Command, Quantico, Virginia; the Navy Recruiting Command, Millington,
Tennessee; and the Air Force Recruiting Service, Randolph Air Force Base,
San Antonio, Texas. We also reviewed past research on the effectiveness of
various approaches to recruiting.

To determine what the services have been doing to reduce attrition during
basic training, we visited one of each service's basic training sites: Army
Basic Training, Fort Jackson, South Carolina; Navy Recruit Training Command,
Great Lakes, Illinois; Marine Corps Basic Training, Parris Island, South
Carolina; and Air Force Basic Training, Lackland Air Force Base, San
Antonio, Texas. We interviewed officials responsible for implementing
attrition-reduction initiatives and reviewed documentation on the
effectiveness of these efforts.

To determine whether service efforts to reduce first-term enlisted attrition
have been successful, we updated our previously reported attrition data by
adding information on the most recent group of enlistees for whom the
Defense Manpower Data Center had data 4 years after entry into the service.
This group involved all enlistees who entered the services with no prior
service in fiscal year 1995 and were separated before the end of 1999. We
compared the attrition rates of this group to attrition rates for enlistees
who entered the services in earlier years. Although we did not extensively
test the reliability of the Center's database, we did compare attrition
rates using this data to the attrition rates the services calculate
themselves. While the services collect and maintain attrition data
differently, their rates of attrition revealed the same trends and patterns.

To determine what DOD is doing to improve the data on reasons for early
attrition, we interviewed officials from the Office of the Under Secretary
of Defense, Force Management Policy, Washington, D.C.; from the U.S.
Military Entrance Processing Command, in North Chicago, Illinois; and from
the Walter Reed Institute of Research, Washington, D.C. We also reviewed
revisions made to the separations codes that are used to classify enlistees'
discharges.

We conducted our review between August 1999 and June 2000 in accordance with
generally accepted government auditing standards.

Comments From the Department of Defense

GAO Contacts and Staff Acknowledgments

William Beusse (202) 512-3517

In addition to the name above, Beverly Schladt, Joan Slowitsky, Donna
Rogers, Julia Kennon, and Nancy Ragsdale made key contributions to this
report.

(702023)

Table 1: Services' Success in Meeting Their Annual Recruiting
Goals for New Active-Duty Enlistees 5

Table 2: DOD Attrition Rates for Active-Duty Enlistees 19

Table 3: Remedial Programs Offered at Fort Jackson's Basic
Training Site 24

Table 4: Status of DOD and Service Implementation of Our
Previous Recommendations 34

Figure 1: The Army's Expenditures for Advertising and
Enlistment Bonuses (fiscal year 2000 constant
dollars in millions) 7

Figure 2: The Navy's Expenditures for Advertising and
Enlistment Bonuses (fiscal year 2000 constant
dollars in millions) 9

Figure 3: The Air Force's Expenditures for Advertising and
Enlistment Bonuses (fiscal year 2000 constant
dollars in millions) 11

Figure 4: The Marine Corps' Expenditures for Advertising and
Enlistment Bonuses (fiscal year 2000 constant dollars
in millions) 13

Figure 5: Percentage of Fiscal Year 1995 Enlistees Separating
During Particular Periods of Their First Terms 20

Figure 6: Trends in the Timing of First-Term Enlisted Attrition 21

Figure 7: Primary Reasons That Enlistees Who Entered the
Services in Fiscal Year 1998 Were Separated in Their
First 6 Months of Service 22
  

1. Pararescue personnel are responsible for search, rescue, and recovery
operations. They provide rapid response in adverse geographic and
environmental conditions on land or water, including emergency and field
medical care.

2. See GAO/GGD/AIMD-99-215 and GAO/NSIAD-99-178R for our assessment of DOD's
performance plan.

3. See appendix I for a detailed listing of our prior recommendations and
DOD's response to them.

4. The most recent group of enlistees we could track DOD-wide 48 months
after they had enlisted was the group who entered in fiscal year 1995. For
example, to track enlistees who entered the services in fiscal year 1996, we
would have to track them until 48 months later, or the end of fiscal year
2000. At the time we did our data analysis, complete data on fiscal year
2000 was unavailable.

5. We measure the attrition rate 48 months after enlistees enter the service
because this captures the early separation of the majority of enlistees, who
have 2-, 3-, and 4-year contract terms. Measuring the rate at 48 months also
captures the early separation of persons with 5- or 6-year contracts during
their first 4 years of service.

6. See Military Attrition: DOD Could Save Millions by Better Screening
Enlisted Personnel (GAO/NSIAD-97-39 , Jan. 6, 1997).

7. See Military Attrition: DOD Could Save Millions by Better Screening
Enlisted Personnel (GAO/NSIAD-97-39 , Jan. 6, 1997).

8. Section 581 of P.L. 106-65 (Oct. 5, 1999).

9. Military Attrition: DOD Could Save Millions by Better Screening Enlisted
Personnel (GAO/NSIAD-97-39, Jan. 6, 1997).

10. Military Recruiting: DOD Could Improve Its Recruiter Selection and
Incentive Systems (GAO/NSIAD-98-58, Jan. 30, 1998).

11. Military Attrition: Better Data, Coupled With Policy Changes, Could Help
the Services Reduce Early Separations (GAO/NSIAD-98-213, Sept. 15, 1998).

12. Military Recruiting: New Initiatives Could Improve Criminal History
Screening (GAO/NSIAD-99-53, Feb. 23, 1999).

13. We withdrew a 21st recommendation, that the Secretary of Defense
encourage the use of quarterly floating recruiting goals, because the
services made compelling arguments against such goals.
*** End of document. ***