Retention Bonuses: More Direction and Oversight Needed (Letter Report,
11/06/95, GAO/NSIAD-96-42).

GAO reviewed whether the Department of Defense (DOD) has effectively
managed the Selective Reenlistment Bonus (SRB) Program.

GAO found that: (1) the services have awarded some SRB to personnel in
high-skill categories where a high percentage of the required positions
are already filled; (2) in fiscal year (FY) 1994, 43 percent of the new
SRB contracts went to service members in skill categories where 90
percent or more of the required positions were filled and in which many
higher skill level service members were paid incentives to leave the
service; (3) each SRB program is targeted to different segments of the
military, including personnel in different grades and year groups; and
(4) the Office of the Secretary of Defense has not provided adequate
oversight of the SRB program, having performed only one skills review in
FY 1991.

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

 REPORTNUM:  NSIAD-96-42
     TITLE:  Retention Bonuses: More Direction and Oversight Needed
      DATE:  11/06/95
   SUBJECT:  Reductions in force
             Military downsizing
             Personnel management
             Military compensation
             Officer personnel
             Human resources utilization
             Military recruiting
             Combat readiness
             Retirement benefits
             Military reenlistment bonuses
IDENTIFIER:  DOD Selective Reenlistment Bonus Program
             Variable Reenlistment Bonus Program
             JCS Status of Resources and Training System
             Navy Enlisted Classification System
             DOD Special Separation Benefit
             DOD Voluntary Separation Incentive Program
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* 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.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


Cover
================================================================ COVER


Report to Congressional Committees

November 1995

RETENTION BONUSES - MORE DIRECTION
AND OVERSIGHT NEEDED

GAO/NSIAD-96-42

Retention Bonuses

(703074)


Abbreviations
=============================================================== ABBREV

  DOD - Department of Defense
  GSM - Gas Turbine Mechanics
  IDC - Independent Duty Corpsmen
  SRB - Selective Reenlistment Bonus
  OSD - Office of the Secretary of Defense
  SORTS - Status of Resources and Training System
  VSI - Voluntary Separation Incentive
  SSB - Special Separation Benefit
  NCO - Noncommissioned Officer

Letter
=============================================================== LETTER


B-259744.2

November 24, 1995

The Honorable Robert K.  Dornan
Chairman
The Honorable Owen B.  Pickett
Ranking Minority Member
Subcommittee on Military Personnel
Committee on National Security
House of Representatives

This review was conducted to determine whether the Department of
Defense (DOD) is effectively managing the Selective Reenlistment
Bonus (SRB) program.  As agreed with your office, this report is
addressed to you because of the Subcommittee's ongoing interest in
special pays and allowances.  The specific objectives were to (1)
ascertain whether reenlistment bonuses were being paid to enlisted
servicemembers in skill categories that were not experiencing
significant personnel shortages or skills that were also receiving
separation incentives and (2) assess the oversight of the program by
the Office of the Secretary of Defense (OSD).  The SRB program is one
means DOD is using to achieve its force manning objectives as the
services downsize.  We have previously reported that the services
have been successful in maintaining high aggregate personnel levels
throughout the drawdown.\1 This report looks below aggregate
personnel levels and focuses on personnel levels within occupational
specialties. 


--------------------
\1 Military Personnel:  High Aggregate Personnel Levels Maintained
Throughout Drawdown (GAO/NSIAD-95-97, June 2, 1995). 


   BACKGROUND
------------------------------------------------------------ Letter :1

The SRB program is authorized by 37 U.S.C.  308 to help maintain an
adequate level of experienced and qualified enlisted personnel.  The
program authorizes bonuses of up to $45,000 to personnel in critical
skills who have between 21 months and 14 years of active-duty service
and who reenlist or extend their reenlistments for at least 3 years. 
The intent of the program, according to DOD, is to focus reenlistment
incentives on critical skills that are in short supply and have high
training costs. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

We found that the services are awarding some SRBs to skills where a
high percentage of the required positions are already filled.  In
fiscal year 1994, for example, 43 percent of the new SRB contracts\2
awarded went to servicemembers in skills where 90 percent or more of
the required positions were filled (the level used by the readiness
reporting system to indicate a unit's capability to perform all
mission requirements) and in which many higher skill level
servicemembers were paid incentives to leave the service.  The value
of these SRB contracts was about $64 million. 

Service officials defended their management of the retention and
separation incentive programs, stating that each is targeted at
different segments of the force, that retention and separation
incentives went to personnel in different grades and year groups
(cohorts of personnel with the same number of years of military
experience), and that payment of separation incentives did not mean
they were satisfied with manning levels.  We believe that if a skill
is experiencing shortages that warrant paying retention incentives,
it is not prudent to pay incentives to others in that same skill to
leave the service. 

OSD is not providing adequate direction and oversight of the SRB
program.  Its guidance to the services for determining which skill
categories should receive SRBs is too general in nature.  As a
result, each service uses a different procedure for identifying which
skill categories are to receive SRBs.  Also, OSD's oversight of the
SRB program is lacking.  While OSD guidance requires detailed annual
reviews of the skill categories that the services plan to include in
their programs, these reviews are not being conducted.  OSD performed
only one such review--in fiscal year 1991--during which the need for
34 percent of the proposed skill categories was questioned.  However,
OSD did not require the services to respond to the report's findings,
did not take any action on the findings, and has not conducted any
subsequent reviews. 


--------------------
\2 Generally, half the amount payable for an SRB reenlistment is paid
at the beginning of the reenlistment period, with the remaining half
paid in equal annual installments over the remaining term of the
reenlistment.  In this report, the term "new contracts" is used to
indicate SRB reenlistment contracts initiated within the given fiscal
year.  When discussing the cost of new contracts, we refer to the
total value of the contracts (initial 50-percent payment and outyear
installments). 


   EVOLUTION OF CURRENT PROGRAM
------------------------------------------------------------ Letter :3

The current SRB program can be traced to 1965, when the services
began to experience increasing problems in first-term retention and
career manning in a number of technical, high training cost skills. 
In addressing the problem, DOD recommended the creation of a flexible
reenlistment bonus program that could be tailored to fit particular
skill-retention requirements and that could be changed as those
requirements changed.  As a result, Congress established the Variable
Reenlistment Bonus program in 1965.  In the ensuing years, this
became the SRB program and was modified and extended to address
concerns about retention and manning problems. 

The Secretary of Defense has established three eligibility zones for
the payment of SRBs.  Zones are defined in terms of years of
active-duty service.  Zone A includes reenlistments falling between
21 months and 6 years of active duty; zone B, between 6 and 10 years;
and zone C, between 10 and 14 years.  The service secretaries
designate which skills and which zones within those skills are
eligible to receive SRBs.  Servicemembers may receive only one SRB
within any one zone. 

The total cost of new SRB contracts awarded has declined over the
past
5 years (see fig.  1).  According to service officials and budget
justification documents submitted to Congress, the main reason for
the declines was the force downsizing occurring during this period,
which reduced the need for military personnel.  According to DOD, SRB
contracts declined by nearly 60 percent during the last 5 years while
the force declined about 30 percent. 

   Figure 1:  Cost of New SRB
   Contracts, Fiscal Year 1990-94

   (See figure in printed
   edition.)

In fiscal years 1993 and 1994, personnel in approximately 20 percent
of DOD's enlisted skills were awarded SRBs.  More than 30 percent of
the enlisted personnel were in those skills.  However, not all of
these servicemembers would be eligible for SRBs in a given year
because they would not be up for reenlistment in that year or would
not be in a zone that was eligible for SRBs.  According to DOD, 1.1
percent of all active-duty personnel received a new SRB contract in
1994, down from 2.4 percent in 1990. 

The cost of the SRB program varies considerably by service.  Table 1
shows the number of people who received new SRB payments in fiscal
year 1994, the total cost, and the average cost per recipient of
those new SRB contracts.  Nearly 60 percent of the total cost for new
SRB contracts was incurred by the Navy.  Also, the average new SRB
contract cost per recipient was higher in the Marine Corps and the
Navy than in the other two services. 



                                Table 1
                
                 Number and Cost of New SRB Contracts,
                            Fiscal Year 1994

                                              Contract         Average
                                  Bonus           cost   contract cost
                             recipients  (in millions)  (in thousands)
------------------------  -------------  -------------  --------------
Air Force                         2,408          $20.1            $8.3
Army                              5,641           46.3             8.2
Marine Corps                        918           13.6            14.8
Navy                              9,170          113.5            12.4
======================================================================
Total                            18,137         $193.5           $10.7
----------------------------------------------------------------------

   SRBS AWARDED TO PERSONNEL IN
   HIGH-FILL SKILL CATEGORIES
------------------------------------------------------------ Letter :4

Many SRBs have gone to personnel who are not in skill categories
where extensive shortages exist.  To determine whether SRBs are
awarded only where needed to overcome shortages, we applied two
measures to each skill category that received SRBs in either fiscal
year 1994 or 1993:  (1) overall fill rate at the beginning of the
fiscal year (defined as the percent of required positions that were
filled) and (2) whether individuals in that same skill category had
been given financial incentives to leave the service. 


      A SUBSTANTIAL PROPORTION OF
      SRB PAYMENTS WENT TO
      PERSONNEL IN SKILLS WITH
      HIGH FILL RATES
---------------------------------------------------------- Letter :4.1

We used the proportion of required positions filled as an indicator
of whether a skill was experiencing a significant personnel shortage. 
The Status of Resources and Training System (SORTS), which is the
system used by the services for reporting unit readiness, has
established criteria that units with 90 percent of their assigned
personnel on hand are considered prepared to conduct all required
wartime missions.  Therefore, we used the 90-percent fill rate as an
indicator of high fill.  Also, according to representatives of the
Air Force and the Marine Corps, a fill rate of 90 percent or less in
a skill category flags that category for consideration for an SRB. 
Neither the Army nor the Navy had a specific fill rate threshold for
SRB consideration. 

Using service-provided fill rates and SRB information, we found that
81 percent of people awarded SRBs across DOD in fiscal year 1994 and
78 percent in fiscal year 1993 were in skill categories that were
filled at least at the 90-percent level.  The cost of these contracts
was about
$155 million in fiscal year 1994 and about $165 million in fiscal
year 1993. 

Figures 2 and 3 show the percentage of new SRB contracts given to
personnel in high-fill skill categories in fiscal years 1994 and
1993.  The figures show the results of analyses at three levels of
fill (90, 95, and
100 percent) by service.  As these figures show, a substantial
proportion of the SRB payments went to personnel in skill categories
that were not experiencing large shortfalls.  While the percentages
drop as the fill rate increases, each service paid a substantial
proportion of its new SRBs to personnel in skill categories that were
already filled 100 percent or higher.  Across DOD, 25 percent of
fiscal year 1994 and 30 percent of fiscal year 1993 SRB recipients
were in skill categories with fill rates of 100 percent or higher. 
The cost of these contracts was about $58 million in fiscal year 1994
and about $71 million in fiscal year 1993. 

   Figure 2:  Fiscal Year 1994
   Contracts to Personnel in
   High-Fill Skill Categories

   (See figure in printed
   edition.)

   Figure 3:  Fiscal Year 1993
   Contracts to Personnel in
   High-Fill Skill Categories

   (See figure in printed
   edition.)


      SRBS AND SEPARATION
      INCENTIVES WERE GIVEN TO THE
      SAME SKILL CATEGORIES
---------------------------------------------------------- Letter :4.2

In recent years, retention needs have declined with reduced force
levels.  To facilitate military downsizing, Congress authorized two
types of special separation pay to personnel who voluntarily leave
the military by September 30, 1999, but are not eligible to retire: 
(1) the Voluntary Separation Incentive (VSI), which is a variable
annuity payment, and (2) the Special Separation Benefit (SSB), which
is a one-time, lump-sum payment.\3 We were initially told by OSD and
service representatives that retention and exit bonuses should not be
going to personnel in the same skill categories.  However, in fiscal
years 1994 and 1993, 48 percent of the personnel awarded new SRB
contracts were in skill categories in which other personnel in the
same skill categories received financial separation incentives.  In
fiscal year 1994, nearly 8,800 military personnel who received new
SRB contracts (at a cost of about $73 million) were in the same skill
categories as about 2,100 of the separation-incentive recipients (who
received about $82 million to leave the military).  In fiscal year
1993, nearly 10,300 military personnel who received new SRB contracts
(at a cost of about $75 million) were in the same skill categories as
about 2,100 of the separation incentive recipients (who received
about $82 million to leave).  Thus, either the services are paying
SRBs to people with skills that are not in short supply or they are
paying exit incentives to people with skills that are in short
supply. 

Table 2 shows the number of new fiscal year 1994 SRB recipients in
each service who were in skill categories where separation incentives
were paid, the percentage of total SRB recipients that this group
comprised, and the cost of those new SRB contracts.  Eighty-four
percent of the Army's new SRB recipients were in skill categories in
which separation incentives were also paid. 



                                Table 2
                
                    Selective Reenlistment Bonus and
                Separation Incentive Match (fiscal year
                                 1994)

                         (Dollars in millions)


                         Total SRB
                        recipients      Number     Percent        Cost
----------------------  ----------  ----------  ----------  ----------
Air Force                    2,408         670         28%        $5.6
Army                         5,641       4,720          84        38.7
Marine Corps                   918         636          69         9.1
Navy                         9,170       2,753          30        20.0
======================================================================
Total                       18,137       8,779         48%       $73.4
----------------------------------------------------------------------

--------------------
\3 10 U.S.C.  1174a and 1175. 


      ABOUT HALF OF SRBS AWARDED
      IN HIGH FILL SKILL
      CATEGORIES THAT ALSO
      RECEIVED SEPARATION
      INCENTIVES
---------------------------------------------------------- Letter :4.3

A more stringent test of whether SRBs were going to personnel who
were not in shortage categories involves the determination of how
many SRB recipients were in skill categories that had high fill rates
and where other personnel in the same skill categories received
incentive payments to leave the military.  In fiscal years 1994 and
1993, 43 percent of the new SRB contracts awarded (at a cost of about
$64 million in fiscal year 1994 and $65 million in fiscal year 1993)
were in skill categories that met both of the measures we
applied--fill rates of 90 percent or higher and payment of separation
incentives.  Furthermore, 9 percent of new SRB contracts awarded in
fiscal year 1994 and 17 percent in fiscal year 1993 were in skill
categories with fill rates of 100 percent or higher and to which exit
incentives were paid.  The cost of these contracts was about $14
million for fiscal year 1994 and about $29 million for fiscal year
1993. 

Figures 4 and 5 show the percentage of new SRB contracts by service
that went to personnel in skill categories having high fill rates and
where other personnel in the same skill categories received
separation incentives in fiscal years 1994 and 1993.  The number of
separation incentives given by the Air Force in fiscal year 1993
includes those given in fiscal year 1992.  Air Force officials told
us that they ran fiscal years 1992 and 1993 exit incentive programs
as one program and were unable to provide information on fiscal year
1993 by itself.  The reduction of the percentages from fiscal years
1993 to 1994 in the Air Force and the Marine Corps results primarily
from reductions in the number of separation incentives given. 

   Figure 4:  Fiscal Year 1994 SRB
   Contracts in High-Fill Skill
   Categories That Also Received
   Separation Bonuses

   (See figure in printed
   edition.)

   Figure 5:  Fiscal Year 1993 SRB
   Contracts in High-Fill Skill
   Categories That Also Received
   Separation Bonuses

   (See figure in printed
   edition.)


   SERVICES' RATIONALE FOR SRBS
   AND SEPARATION INCENTIVES IN
   SAME SKILLS
------------------------------------------------------------ Letter :5

Service officials told us that although they had paid some people to
stay and other people to leave in the same military skills, the
retention and separation incentive programs were directed at
different grade and year groups.  We agree.  However, we believe that
if a skill is critically short and warrants retention bonuses,
separation incentives should not be given to personnel in those
skills.  Air Force officials told us that they changed their policy
in fiscal year 1994 to not allow members in SRB skills to separate
using VSI/SSB except in cases of documented extreme hardships because
they did not think it was appropriate to pay some people to stay and
others to leave in the same skill. 

Service officials also said that their payment of VSI/SSB incentives
to personnel in an SRB skill did not mean that they were satisfied
with the fill rate in that skill.  Rather, they said the separation
incentives were given in an effort to comply with congressional
direction to use voluntary means to achieve force reductions.  While
Congress encouraged the services to use voluntary means wherever
possible to achieve needed reductions, we found no indication in the
legislative history that Congress intended that the services offer
voluntary separation incentives to personnel in critically short
skills to avoid involuntarily separating personnel in skills with
excesses. 

Service officials also took issue with our use of the 90-percent fill
level as an indication that a skill was not critically short.  We
agree that some skills might be considered critically short at
anything less than 100-percent fill.  That is why we also provided
data on the 95-percent and 100-percent fill levels.  However, the
services have not defined which skills require higher fill rates than
the 90-percent criterion used for readiness reporting. 


   OSD GUIDANCE AND OVERSIGHT OF
   THE SRB PROGRAM ARE LACKING
------------------------------------------------------------ Letter :6

OSD is not providing adequate direction and oversight of the SRB
program.  OSD guidance to the services for determining which skills
should receive SRBs is general and oversight review of the services'
programs is lacking. 

OSD guidance for determining those skills to receive SRBs instructs
the services to use a "balanced evaluation" that "should include, but
not be limited to, a full assessment of the following factors."

  Serious undermanning in three or more adjacent year-groups in the
     bonus zones. 

  Chronic and persistent shortages in total career manning. 

  High replacement cost. 

  Skills that are relatively arduous or otherwise unattractive
     compared to other military skills or civilian alternatives. 

  Skills that are essential to the accomplishment of defense
     missions. 

OSD has not defined many of the terms in its guidance, such as
"serious undermanning" and "chronic and persistent shortages," nor
has it established how much weight should be given to each of the
selection factors.  As a result, each service uses a different
procedure to identify and prioritize which skills will receive SRBs. 
Service officials said that, in deciding who will receive SRBs, they
consider factors similar to the OSD guidance, such as whether the
skill is currently receiving an SRB, reenlistment trends, fill rates,
the skill's criticality to accomplishing the defense mission, and the
cost, length, and availability of training.  They too have not
established criteria for determining how much weight to give these
various factors. 

OSD has proposed new guidelines for the SRB program, but these
guidelines do not clarify the selection criteria.  They state that
the purpose of the SRB program is "to encourage the reenlistment of
sufficient numbers of qualified enlisted uniformed services personnel
in critical military specialties with high training costs or
demonstrated retention shortfalls." The use of the connector "or"
appears to broaden the purpose of the program, which is stated in the
current guidelines as "intended to attract more reenlistments in
critical military specialties characterized by retention levels
insufficient to sustain the career force at an adequate level." While
we agree that training costs should be a consideration in deciding
whether to give retention bonuses, we do not believe that high
training cost alone justifies payment of retention incentives if the
personnel are not in specialties experiencing demonstrated retention
shortfalls. 

Although OSD guidance specifies that OSD conduct a detailed annual
review of the SRB program, examining each skill category programmed
for an SRB, such annual reviews have not been conducted.  A one-time
study conducted by OSD in 1991 of the skill categories that the
services were including in their programs, identified several areas
of concern and questioned the need to provide SRBs to 34 percent of
the proposed skill categories.  OSD did not require the services to
respond to the report's findings, took no action on the findings, and
has conducted no further reviews of the SRB program. 


   RECOMMENDATIONS
------------------------------------------------------------ Letter :7

We recommend that the Secretary of Defense establish guidance and
controls to ensure that the SRB program provides bonuses only for
reenlistments in skill categories that are in short supply. 
Specifically, we recommend that the Secretary (1) provide more
explicit guidance regarding the determination of shortage categories
and eligibility for SRBs and require the services to establish and
document more specific criteria for determining which skills will
receive SRBs and (2) monitor the services' adherence to this
guidance.  Because of the extent to which exit incentives have been
provided to personnel in skills which also received SRBs, we
recommend that the Secretary ensure that payment of exit and
retention incentives is coordinated so that they are not both
provided to personnel in the same skill categories. 


   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :8

DOD did not agree with our findings or recommendations, stating that
our methodology and analysis were flawed.  DOD's comments are
included in their entirety in appendix I. 

DOD stated that our analysis was flawed by an assumption that 90
percent manning is a satisfactory level of fill in all skill
categories.  DOD stated that certain skill categories are imminently
critical to the mission of each service and that, in many cases, 100
percent of authorized manning is not enough to do the job. 
Consequently, using 90 percent as the delineation of high fill for
critical skills is unacceptable to DOD. 

We did not assume that 90-percent fill is necessarily sufficient. 
Although 90 percent of authorized positions is the fill level used in
the official DOD unit readiness reporting system to indicate a
capability to perform all assigned missions, we agree with DOD that
what is considered to be an adequate fill level can vary by skill. 
For this reason, we provided data on the 90-percent, 95-percent, and
100-percent fill levels.  Our point, therefore, was not that 90
percent represents high fill, but that DOD has failed to adequately
define which skills require higher fill rates.  That is, when DOD
states that "certain skill categories are imminently critical to the
mission of each service," we expected to find some definition or
criteria that would identify which skills those were or how they
could be determined.  Furthermore, if it is true as DOD asserts that
"in many cases .  .  .  100 percent of authorized manning is not
enough to do the job," then manpower requirements need to be
reexamined.  In addition, if DOD believes that the 90-percent manning
figure used in readiness reporting does not represent a level that
enables a unit to perform all required missions, it needs to revise
its criteria so that an accurate picture of readiness can be conveyed
to military decisionmakers. 

DOD stated that our methodology was also flawed because we looked at
manning levels across entire skills rather than looking at manning
within SRB years of experience zones.  DOD stated that it is
essential to continue to administer the SRB program by zones "since
the services have requirements for minimum levels of manning within
each of these zones." However, enlisted force managers in each of the
services told us that they do not manage their enlisted force by SRB
zones nor do they routinely express their requirements by zone. 
Rather, they manage by grade level or years of service groups that
overlap the SRB zones. 

We originally attempted to analyze fill rates by SRB zone, but,
except for the Navy, the services could not readily provide us with
fill rates by zone.  In analyzing the Navy's data by zone, we found
that about 50 percent of the skill zones given SRBs in fiscal year
1994 were filled at rates of 90 percent or higher.  In fact, 35
percent were filled at rates of 100 percent or higher.  Consequently,
looking at fill rates by zone where the services were able to provide
the data did not change our conclusion that some SRBs were being paid
to people in skills that did not appear to have critical shortages. 

We found similar results when we looked at the Air Force.  Air Force
officials told us that they do not consider fill rates by zone when
making SRB decisions.  They stated that in most cases they provide
SRBs to personnel in zones A and B to ensure sufficient personnel at
the noncommissioned officer (NCO) level.  In examining this, however,
we found that most of the skills that were filled at or above the
90-percent level overall, also had NCO fill rates of at least 90
percent.  In fiscal year 1994, 64 percent of SRB skills with fill
rates of 90 percent or higher also had NCO fill rates of 90 percent
or higher and 21 percent had fill rates of 100 percent or higher.  In
fiscal year 1993, 78 percent of SRB skills with fill rates of 90
percent or higher also had NCO fill rates of 90 percent or higher and
44 percent had fill rates of 100 percent or higher.  Thus, looking at
Air Force NCO fill rates rather than overall fill rates does not
change the conclusion that some SRBs were being paid to people in
skills that did not appear to have critical shortages. 

DOD noted that a person with 2 years' experience cannot be
substituted for a person with 10 to 14 years of experience.  We
agree.  Therefore, when we found the services paying separation
bonuses to personnel with 10 to
14 years of experience in a skill area, we viewed it as an indication
that the service personnel managers did not consider that skill area
to be experiencing a critical shortfall.  If service personnel
managers believed that a skill area was critically undermanned, it
would make no sense to provide incentives to the higher experienced
personnel in that skill to leave the service and thus exacerbate the
undermanning. 

DOD also noted that it is not cost-efficient for a senior person to
perform a function for which he is overqualified.  Again, we agree. 
However, if there is really a critical shortage of lower skilled
personnel and an excess of higher skilled personnel in that same
occupation, we would expect the service to backfill with the higher
skilled personnel rather than paying them bonuses to leave. 

DOD also took issue with our finding that additional OSD oversight is
required.  DOD stated that the services and the Department spend a
great deal of time and effort on the SRB program and it already goes
through several lengthy review processes, including an annual budget
justification.  DOD also argues that the 1991 study declared the SRB
program to be well-run and, therefore, no additional OSD oversight is
required.  The 1991 study, however, does not really support that
conclusion.  While making the general comment that the services' SRB
programs were in compliance with DOD policy and were well-managed,
the study identified 84 skills out of 250 (about 34 percent) that
should be considered for further review.\4

We could find no indication that those 84 skills were reexamined. 
Also, the study noted that the OSD policy guidance is very general
and that there are numerous ways it can be interpreted, each
interpretation leading to a very different analytic criteria.  The
study proposed an automated approach that would apply a set of
objective criteria to each skill, resulting in two groups of
skills--those that were acceptable and those that needed further
consideration.  The services would then be asked to comment on any
skills that were identified by the criteria as needing further
review. 

OSD also stated that it does not want to add more complications to an
already cumbersome system by layering additional restrictions on the
services.  DOD also noted that the services need flexibility to be
able to respond to rapidly changing requirements for readiness.  We
do not see the exercise of adequate oversight as necessarily
decreasing flexibility.  The 1991 study stated that the approach it
proposed would allow each of the services to develop, execute, and
justify its SRB plans based on its unique requirements and objectives
as long as they fit within the overall policy guidance. 

Rather than use the approach suggested by the 1991 study or develop a
similarly streamlined method of maintaining adequate oversight, OSD
has opted for reducing its oversight of the SRB program.  DOD
Instruction 1304.22 stated that OSD "shall conduct a detailed annual
review of the enlistment bonus, selective reenlistment bonus, and
special duty assignment pay programs" in conjunction with Program
Objectives Memorandum cycle.  It further stated that each military
specialty programmed for a bonus in the next 2 fiscal years shall be
examined.  However, OSD has not performed such a review since the
1991 study and it has drafted new guidance that eliminates the
detailed review requirement. 


--------------------
\4 The 1991 study also acknowledged that it could only evaluate a
portion of the Navy's SRB program because the Navy did not have any
realistic requirements for skills under the more detailed Navy
Enlisted Classification system. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :9

We examined the legislative history of the SRB program and OSD and
service regulations for the program.  We also interviewed OSD and
service representatives to determine their policies on designating
SRB skills, awarding SRBs, and paying of VSI and SSB to
servicemembers in SRB skills. 

We analyzed information provided by the services from a number of
databases to determine the following for fiscal years 1993 and 1994: 

  the number and cost of new SRBs awarded by skill,

  fill rates at the beginning of the year for skills receiving SRBs,

  the number and cost of SRBs awarded to skills with high fill rates,
     and

  the number of VSIs and SSBs given to personnel in skills eligible
     for SRBs. 

We did not perform a reliability assessment of the databases from
which the services provided us data.  However, we compared the
information provided us to that contained in service reports and
discussed the information with service officials to ensure it
provided a reasonable and accurate profile of individuals receiving
SRBs, the fill rates for SRB skills, and VSI and SSB recipients.  Our
review was conducted from June 1994 to October 1995 in accordance
with generally accepted government auditing standards. 


---------------------------------------------------------- Letter :9.1

Please contact me at (202) 512-5140 if you or your staff have any
questions concerning this report.  Major contributors to this report
are listed in appendix II. 

Mark E.  Gebicke
Director, Military Operations
 and Capabilities Issues




(See figure in printed edition.)Appendix I
COMMENTS FROM THE DEPARTMENT OF
DEFENSE
============================================================== Letter 



(See figure in printed edition.)



(See figure in printed edition.)

Now on p.  1. 

Now on pp.  2-4. 

See comment 1. 



(See figure in printed edition.)

Now on pp.  4-7. 

Now on pp.  7-8. 



(See figure in printed edition.)

Now on pp.  8-10. 



(See figure in printed edition.)

Now on pp.  11-12. 



(See figure in printed edition.)

Now on p.  12. 


The following are GAO's comments on the Department of Defense's
letter dated September 28, 1995. 

GAO COMMENTS

1.  While there may have been other contributors, the drawdown was
the main reason for the reductions cited by service Selective
Reenlistment Bonus (SRB) program managers and stated in service
budget justification documents submitted to Congress.  In its budget
documents over this period, the Navy stated that "the number of new
payments declined .  .  .  due to force structure reductions." In
discussing the declining total cost of new SRB contracts, the Air
Force reported in its budget documents to Congress that ".  .  .  the
overall drawdown of the force is a contributing factor to the lower
totals .  .  ." In addition, SRB program managers in all four
services told us that, generally speaking, the declining total cost
of new SRB contracts resulted from the drawdown. 

2.  As of September 1995, the Navy had 989 Independent Duty Corpsmen
(IDC) against an authorization of 981 billets, a fill rate of over
100 percent.  The Navy has requirements for an IDC functioning as the
sole medical care provider on 231 of its approximately 372 ships. 
Even in the unlikely scenario that all 231 ships were deployed at one
time, the Navy should have no trouble providing the 231 IDCs from its
inventory of nearly
1,000 IDCs. 

3.  While Navy data shows that the first-term retention rate for Gas
Turbine Mechanics (GSM) declined from 63.3 percent to 37.8 percent
during fiscal year 1994, there is no apparent relationship between
the decline and the SRB program.  When the Navy reduced the
first-term SRB award payment for GSMs by nearly two-thirds over the
course of fiscal year 1991, the first-term retention rate actually
increased to 57.6 percent from 53.9 percent during the year.  With
the SRB award to first-term GSMs maintained at the reduced payment
level, retention rates were 68.2 percent and 63.3 percent at the end
of fiscal years 1992 and 1993, respectively.  According to Navy
officials, the drop in reenlistment of first-term GSMs that occurred
in fiscal year 1994 was the result of a perception among personnel
within that skill area that, because of reduced ship construction and
possible ship decommissionings, there was no future in the GSM
rating.  Despite the drop in the first-term reenlistment rate, as of
September 1995, the Navy had an inventory of 2,974 GSMs against an
authorization of 2,871 billets, a fill level of over 100 percent. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II

NATIONAL SECURITY AND
INTERNATIONAL AFFAIRS DIVISION,
WASHINGTON, D.C. 

Sharon A.  Cekala, Associate Director
William E.  Beusse, Assistant Director

NORFOLK REGIONAL OFFICE

Janet Keller, Evaluator-in-Charge
Sharon Reid, Evaluator

*** End of document. ***