Public-Private Competitions: DOD's Additional Support for Combining Depot
Workloads Contains Weaknesses (Letter Report, 04/17/98,
GAO/NSIAD-98-143).

Pursuant to a congressional request, GAO reviewed the Department of
Defense's (DOD) supporting rationale for combining certain depot-level
maintenance and repair workloads.

GAO noted that: (1) the Air Force's support for DOD's determinations
that it is more logical and economical to combine the workloads being
competed at the closing depots is based on a wide variety of information
accumulated during the acquisition strategy development process started
in September 1995; (2) while GAO recognizes that the determinations
ultimately represent a management judgment based on various qualitative
and quantitative factors and that DOD's determinations may well be
appropriate, the rationale presented in the February 24, 1998,
Sacramento white paper and San Antonio report for combining the
workloads in single solicitations at each location is not well
supported; (3) GAO's assessment indicates that there are significant
weaknesses in logic, assumptions, and data; (4) DOD did not consider
other alternatives that appear to be logical and potentially
cost-effective, and its assumption that efficiencies from shared
personnel and facilities are best achieved with a single solicitation
for combined workloads at each location is questionable; (5) also, the
Air Force's claim that the effects of sequential personnel reductions
and transition delays can be problematic is questionable in view of
DOD's demonstrated success in the past handling multiple transitions and
sequential reductions; (6) in addition, the workload stability rationale
for Sacramento is questionable because the inherent inefficiencies of
the commodity workload are not likely to be improved by combination with
the more predictable and consistent aircraft workload; and (7) finally,
the Air Force's cost analysis, which concluded that workload combination
would save $22 million to $130 million at Sacramento and $92 million to
$259 million at San Antonio, is questionable because it did not consider
all cost factors, such as the cost benefits of increased competition
resulting from solicitations for individual workloads.

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

 REPORTNUM:  NSIAD-98-143
     TITLE:  Public-Private Competitions: DOD's Additional Support for 
             Combining Depot Workloads Contains Weaknesses
      DATE:  04/17/98
   SUBJECT:  Privatization
             Base closures
             Military downsizing
             Defense economic analysis
             Aircraft maintenance
             Military cost control
             Air Force procurement
             Cost effectiveness analysis

             
******************************************************************
** 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.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


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


Report to Congressional Requesters

April 1998

PUBLIC-PRIVATE COMPETITIONS -
DOD'S ADDITIONAL SUPPORT FOR
COMBINING DEPOT WORKLOADS CONTAINS
WEAKNESSES

GAO/NSIAD-98-143

Public-Private Competitions

(709335)


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

  BRAC - Base Closure and Realignment
  DOD - Department of Defense

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


B-279625

April 17, 1998

The Honorable James M.  Inhofe
Chairman, Subcommittee on Readiness
Committee on Armed Services
United States Senate

The Honorable Herbert H.  Bateman
Chairman, Subcommittee on Military Readiness
Committee on National Security
House of Representatives

The National Defense Authorization Act for Fiscal Year 1998 requires
that we review the Department of Defense's (DOD) December 19, 1997,
determinations and supporting rationale for combining certain
depot-level maintenance and repair workloads.  DOD determined that
workloads now being performed at the closing San Antonio, Texas, and
Sacramento, California, Air Force maintenance depots cannot be
performed as logically and economically without combination.  Our
required report, issued on January 20, 1998, concluded that DOD's
determinations to combine workloads were not adequately supported.\1
On February 24, 1998, DOD provided additional support for the earlier
determination to Congress and us.\2 As you requested, we reviewed
that additional material.  This report addresses the reasons the Air
Force believes it is more logical and economical to combine the
workloads and our overall views on the rationale for DOD's
determinations and the information provided on February 24, 1998 to
support this rationale. 


--------------------
\1 Public-Private Competitions:  DOD's Determination to Combine Depot
Workloads Is Not Adequately Supported (GAO/NSIAD-98-76, Jan.  20,
1998). 

\2 The Air Force developed the White Paper On Single Vs.  Multiple
Workload Competitions (Sacramento) and the Rationale for Combining
Multiple Depot-Level Maintenance and Repair Workloads (San Antonio)
to provide additional support for the Department's determinations to
combine multiple workloads into a single solicitation at each
location. 


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

As a result of a 1995 Base Closure and Realignment Commission (BRAC)
decision, Kelly Air Force Base, Texas, is to be realigned, and the
San Antonio Air Logistics Center, including its Air Force maintenance
depot, is to be closed by July 2001.  Similarly, McClellan Air Force
Base, California, and the Sacramento Air Logistics Center, including
its Air Force maintenance depot, is to be closed by July 2001.  To
mitigate the impact of the closures on the local communities and
center employees, the administration announced its decision to
maintain certain employment levels at these locations. 
Privatization-in-place\3

was one initiative for achieving these employment goals. 

Since that time, Congress and the administration have debated the
process and procedures for deciding where and by whom the depot
maintenance workloads at the closing depots should be performed.\4
Central to this debate are concerns about the excess facility
capacity at the Air Force's three remaining maintenance depots and
the legislative requirement in 10 U.S.C.  2469 that, for workloads
exceeding $3 million in value, a public-private competition must be
held before the workloads can be moved from a public depot to a
private sector company.\5 Because of congressional concerns raised in
1996, the Air Force revised its privatization-in-place plans to
provide for competitions between the public and private sectors as a
means to decide where the depot maintenance workloads would be
performed.  The first competition was for the C-5 aircraft depot
maintenance workload, which had been performed at the San Antonio
depot.  The Air Force awarded the workload to the Warner Robins depot
in Georgia on September 4, 1997.  During 1997, Congress continued to
oversee DOD's strategy for allocating workloads currently performed
at the closing depots. 

The 1998 Defense Authorization Act required that we and DOD analyze
various issues related to the competitions at the closing depots and
report to Congress regarding several areas, which are discussed in
appendix I.  One of these areas involves the combination into single
solicitations of aircraft and multi-commodity workloads at the
Sacramento depot and multi-engine workloads at the San Antonio depot. 
Appendix II provides additional information about the maintenance
workloads currently performed at these facilities.  As required by
the act, a solicitation may be issued for a single contract for the
performance of multiple depot-level maintenance or repair workloads. 
However, the Secretary of Defense must first (1) determine in writing
that the individual workloads cannot be performed as logically and
economically without combination by sources that are potentially
qualified to submit an offer and be awarded a contract to perform
those individual workloads and (2) submit a report to Congress
setting forth the reasons for the determination.  Further, the Air
Force cannot issue a solicitation for combined workloads until at
least 60 days after the Secretary submits the required report. 

Our January 20, 1998, report made two key points about DOD's
determinations.  First, we stated that there was no analysis of the
logic and economies associated with having the workload performed
individually by potentially qualified offerors.  Consequently, there
was no support for determining that the individual workloads cannot
as logically and economically be performed without combination. 
Second, we noted that the reports and available supporting data did
not adequately support DOD's determinations.  Appendix III contains a
summary of this report. 

We discussed our findings in a February 24, 1998, hearing conducted
by the Subcommittee on Military Readiness, House National Security
Committee, and a March 4, 1998, hearing conducted by the Subcommittee
on Readiness, Senate Armed Services Committee.\6 At those hearings,
Office of Secretary of Defense and Air Force officials provided
additional rationale supporting DOD's determinations to combine the
workloads.  Subcommittee members expressed their concerns regarding
whether the new data provided adequate support for the
determinations.  Both subcommittees requested that we analyze the
additional data and report to them on our findings.\7


--------------------
\3 Privatization-in-place is a term used to describe contracting with
the private sector for the performance of activities previously
accomplished by government employees at a government facility.  The
contractor would use the former government facility to perform the
work.  Privatization-in-place may involve transferring ownership of
government assets, such as facilities and equipment, to the private
sector. 

\4 The workloads at the Sacramento depot include KC-135, A-10, and
F-15 aircraft; ground communications and electronics equipment and
hydraulics; avionics and instruments; and electrical accessories. 
The workloads at the San Antonio depot include the F100, TF39, and
T56 engines; gas turbine engines; and fuel accessories. 

\5 We have issued several reports addressing these issues.  For more
details, see Related GAO Products at the end of this report. 

\6 Public-Private Competitions:  Access to Records is Inhibiting Work
on Congressional Mandates (GAO/T-NSIAD-98-101, Feb.  24, 1998, and
GAO/T-NSIAD-98-111, Mar.  4, 1998). 

\7 The rationale and support we assessed included the December 19,
1997, determinations and reports; the February 24, 1998, Sacramento
white paper and San Antonio report; and the supporting documentation
provided by the Air Force. 


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

The Air Force's support for DOD's determinations that it is more
logical and economical to combine the workloads being competed at the
closing depots is based on a wide variety of information accumulated
during the acquisition strategy development process started in
September 1995.  While we recognize that the determinations
ultimately represent a management judgment based on various
qualitative and quantitative factors and that DOD's determinations
may well be appropriate, the rationale presented in the February 24,
1998, Sacramento white paper and San Antonio report for combining the
workloads in single solicitations at each location is not well
supported. 

Our assessment indicates that there are significant weaknesses in
logic, assumptions, and data.  DOD did not consider other
alternatives that appear to be logical and potentially
cost-effective, and its assumption that efficiencies from shared
personnel and facilities are best achieved with a single solicitation
for combined workloads at each location is questionable.  Also, the
Air Force's claim that the effects of sequential personnel reductions
and transition delays can be problematic is questionable in view of
DOD's demonstrated success in the past handling multiple transitions
and sequential reductions.  In addition, the workload stability
rationale for Sacramento is questionable because the inherent
inefficiencies of the commodity workload are not likely to be
improved by combination with the more predictable and consistent
aircraft workload.  Finally, the Air Force's cost analysis, which
concluded that workload combination would save $22 million to $130
million at Sacramento and $92 million to $259 million at San Antonio,
is questionable because it did not consider all cost factors, such as
the cost benefits of increased competition resulting from
solicitations for individual workloads. 


   AIR FORCE RATIONALE FOR
   WORKLOAD COMBINATION
   DETERMINATIONS
------------------------------------------------------------ Letter :3

On February 24, 1998, the Air Force provided additional information
in support of DOD's December 19, 1997, determinations.  This
information included two documents:  a white paper containing the
rationale for combining the Sacramento depot's aircraft and commodity
workloads into a single solicitation and a report containing the
rationale for combining the San Antonio depot's engine workloads into
a single solicitation. 

Air Force officials stated that the decision to combine most of the
aircraft and commodity workloads at the Sacramento depot and the
engine workloads at the San Antonio depot was made before the mandate
in the 1998 National Defense Authorization Act.  The officials also
said that the process used to make the decision was valid and that a
reassessment of alternative acquisition strategies was not required
in response to the act. 

The Sacramento white paper described the rationale supporting the
workload combination determination as an iterative process that
evolved over a 2-1/2-year period beginning in September 1995.  This
process included conferences and discussions with potential offerors,
strategy panels with Air Force acquisition experts, repair base
analyses,\8 unsolicited input from industry representatives, and
reviews of recent DOD outsourcing efforts.  Sacramento officials
explained that the initial approach involved a privatization-in-place
strategy, including separate solicitations for seven individual
workloads and separate transition schedules for some of the
individual workloads.  In July 1996 the Air Force decided to conduct
a public-private competition combining the Sacramento KC-135 and A-10
aircraft and various commodity workloads including hydraulics,
instruments and avionics, and electrical accessories.  According to
Air Force officials, the Air Force has pursued workload combination
as its acquisition strategy since that time. 

The San Antonio report recognized that the Air Force had not
conducted an economic analysis regarding the potential savings of
issuing single versus multiple solicitations.  Instead, the Air Force
relied on reviews of engine workload data, repair processes, and
market surveys to identify the acquisition strategy for determining
how San Antonio's engine workloads will be performed in the future. 

Both documents discuss the logic and economies supporting DOD's
determinations to combine workloads into a single solicitation at
each of the closing depots.  The key points in the Air Force's
rationale and support for DOD's determinations are summarized below. 


--------------------
\8 In late 1996, the Air Force accomplished repair base analyses for
six depot-level workloads currently performed by the Sacramento
depot.  The objective of each of the analyses was to identify
industry capabilities and capacity to repair and overhaul specific
workloads. 


      LOGIC FACTORS CITED TO
      SUPPORT WORKLOAD
      COMBINATIONS
---------------------------------------------------------- Letter :3.1

The Air Force stated that its decisions to combine the Sacramento and
San Antonio workloads into single solicitations at each location were
based on the following logic factors: 

  -- Workload commonality and overhead sharing.  The Air Force
     believes that shared personnel skills and backshops\9 provide an
     opportunity for achieving improved efficiencies and lower prices
     in peacetime while providing flexibility to better plan for
     wartime surge requirements.  Air Force officials noted that
     shared fixed overhead costs for such functions, such as
     planning, scheduling, and providing materiel support over a
     larger workload base, provide opportunities for improved
     economies and reduced costs at both the Sacramento and San
     Antonio depots.  Further, using the same backshops for multiple
     workloads should reduce the overall cost of the combined work at
     each location. 

  -- Avoidance of multiple transitions and personnel turbulence.  The
     Air Force believes that managing multiple transitions increases
     the readiness risks associated with closing complex, integrated
     industrial facilities.  Further, delaying the award of the
     contract by splitting the competition into multiple awards could
     subject the workforce to multiple reduction-in-force actions,
     which would disrupt the skill mix and result in productivity
     losses and production delays that adversely affect the readiness
     of the Air Force's operational units. 

  -- Workload stability.  This factor was also cited to support the
     rationale at the Sacramento depot.  The Air Force stated that,
     because the aircraft workload is stable, it can be competed
     using a guaranteed minimum quantity.  However, the Air Force
     noted that many of the commodity workloads have been erratic and
     therefore cannot be competed with a minimum guaranteed workload. 
     Consequently, the Air Force stated that combining the aircraft
     and commodity workloads into one solicitation would allow the
     winning offeror to smooth peaks in one workload segment and
     offset valleys in other workload segments, providing a more
     stable production capability.  Further, the Air Force stated
     that a more stable workload would increase efficiency and
     savings by providing potential offerors a more reliable basis
     for employment levels and cost planning. 

  -- Market surveys.  To support workload consolidation at the San
     Antonio depot, the Air Force said that the majority of
     respondents to its October 1995 market survey indicated a
     preference for a single contract for the C-5 aircraft and a
     single contract for the combined engine workloads.  Further, the
     Air Force concluded from survey results that more competitors
     would participate under the single solicitation for the multiple
     engine workloads. 


--------------------
\9 Depot industrial facilities have backshops to provide support for
common processes which may include cleaning, inspection, metal
plating, painting, and welding. 


      ECONOMIC FACTORS CITED TO
      SUPPORT WORKLOAD
      COMBINATIONS
---------------------------------------------------------- Letter :3.2

The Air Force cited the following factors supporting the economies of
workload combination at the closing depots: 

  -- Time delays.  The Sacramento white paper stated that separating
     the Sacramento workload into five segments would delay contract
     award and transition completion dates by 16 months, which would
     impact closure, increase costs, and reduce projected BRAC
     savings.  Similarly, the San Antonio report stated that
     separating the San Antonio engine workloads into three
     solicitations would extend the planned contract award from 225
     to 740 days, impacting closure and increasing costs. 

  -- Cost increases.  The Air Force stated that conducting multiple
     competitions at Sacramento could result in cost increases to the
     offerors and the government, which the Air Force estimates to be
     between $22 million and $130 million.  At San Antonio, the Air
     Force estimated the increased cost to be between $92 million and
     $259 million. 

  -- Increased risks.  The Air Force believes that changing the
     strategy from single to multiple awards would increase risks and
     translate into higher costs. 


   DOD'S DETERMINATIONS TO COMBINE
   WORKLOADS ARE NOT WELL
   SUPPORTED
------------------------------------------------------------ Letter :4

The additional rationale that the Air Force provided to further
justify DOD's December 19, 1997, determinations is not well
supported.  We identified significant weaknesses in both the logic
and economic rationale presented to support combining workloads at
the Sacramento and San Antonio depots into single solicitations at
each location. 


      SIGNIFICANT WEAKNESSES IN
      SUPPORTING LOGIC
---------------------------------------------------------- Letter :4.1

We identified significant weaknesses in the rationale presented by
the Air Force to support DOD's determinations to combine workloads at
the closing Sacramento and San Antonio depots into single
solicitations at each location.  First, the Air Force did not
adequately consider some other viable alternatives as a part of its
assessment.  Second, some assumptions are creditable only if the
combined workloads are performed in place.  Third, each of the
supporting points has specific weaknesses that create additional
questions regarding the adequacy of DOD's support for workload
combination determinations.  Our concerns regarding the economic
rationale are discussed in the following section. 


         LIMITED CONSIDERATION OF
         ALTERNATIVES
-------------------------------------------------------- Letter :4.1.1

Although the Air Force gave limited consideration to options other
than combining the workloads at the two locations, they did not
consider, or gave only limited consideration to, some other feasible
alternatives.  According to the 1998 Defense Authorization Act,
alternatives that appear logical and potentially cost-effective
should have been evaluated.  Options not considered include (1) using
solicitations that permit the competitors to offer on any combination
of workloads, from one to all and (2) having another contracting
activity conduct simultaneous competitions for segments of the
Sacramento or San Antonio workloads to avoid delays from sequential
competitions for individual segments of the competition. 


         QUESTIONABLE ASSUMPTIONS
-------------------------------------------------------- Letter :4.1.2

Our review indicates that several of the assumptions supporting the
Air Force's rationale are questionable unless the workload remains at
the existing locations.  For example, the Air Force states that
combining workloads will preclude multiple workload transitions,
thereby avoiding multiple reduction-in-force actions, limiting
personnel turbulence, and minimizing readiness impacts.  Further, the
Air Force states that, for the Sacramento workload, combining
aircraft and commodities into a single solicitation would provide the
winning offeror the ability to shift employees between workload
segments.  The advantages cited by the Air Force are not likely to
occur if the workloads are performed at a single location other than
Sacramento and San Antonio or at multiple locations. 


         OTHER WEAKNESSES
-------------------------------------------------------- Letter :4.1.3

We identified other weaknesses or deficiencies with each of the
factors cited by the Air Force including the following: 

  -- Workload commonality and sharing of overhead.  The Air Force's
     position that realizing efficiencies from shared personnel and
     facilities at Sacramento and San Antonio is best achieved with a
     single solicitation for combined workloads is questionable.  The
     efficiencies that are achievable from shared facilities and
     personnel may be greater if the workloads being combined are the
     same or more similar than the workloads being combined under the
     Sacramento and San Antonio solicitations.  For example, the Air
     Force may achieve greater efficiency by combining (1) the
     management of the Sacramento KC-135 workloads with other KC-135
     workloads to be competed and/or (2) the San Antonio Air Force
     T-56 engine workloads with other engine workloads also to be
     competed.  Both of these options provide opportunities for
     significant cost savings that were not considered by the Air
     Force. 

  -- Avoidance of multiple transitions and personnel turbulence.  We
     realize that risks can be associated with the transition of any
     depot maintenance workload.  However, we have reported that
     there is no inherent reason why these workloads cannot be
     transitioned without impacting equipment readiness if the
     transition is properly planned and effectively implemented.\10

Further, DOD has successfully closed 17 depots over the past 10 years
and has successfully managed multiple transitions and the resulting
sequential personnel reductions. 

  -- Workload stability for commodities and aircraft repair at
     Sacramento.  The Air Force data does not support the conclusion
     that the inherent inefficiencies of the commodity workload are
     improved by combining it with the more predictable and
     consistent aircraft workload.  For example, even though the Air
     Force states that stability will come from being able to
     transfer employees between the aircraft and commodity workloads,
     this transfer has rarely happened.  Although the Air Force has
     had the ability to shift workers among the aircraft and
     commodity workloads, Sacramento depot personnel data shows that
     on average, each year over the last 7 years, only 22 of the
     approximately 1,500 wage grade depot employees have been shifted
     between aircraft and commodities. 

  -- Results of market surveys.  We question whether the results of
     the 1995 market survey are applicable to the Air Force's current
     position that combining the San Antonio workloads is more
     logical and economical than issuing individual solicitations. 
     The survey was designed to collect potential offeror preferences
     under the then-current acquisition strategy of privatizing the
     San Antonio aircraft and engine workloads in place.  However, in
     1996 the Air Force revised this acquisition strategy and adopted
     a public-private competition strategy.  Further, in 1997 the Air
     Force conducted a market analysis of engine manufacturing
     companies to determine the availability and interest of public
     and private sector sources to perform the required repair of
     engines currently maintained in Air Force depots.  In this
     survey, engine manufacturers indicated a preference for
     repairing their own engines and were less interested in
     repairing other engines.  Additionally, our discussions with
     four potential offerors for the engine workload indicated that
     they are interested in participating regardless of whether or
     not the workloads are combined into a single solicitation. 


--------------------
\10 Depot Maintenance:  Lessons Learned From Transferring Alameda
Naval Aviation Depot Engine Workloads (GAO/NSIAD-98-10BR, Mar.  25,
1998). 


      SIGNIFICANT WEAKNESSES IN
      ECONOMIC ANALYSES
---------------------------------------------------------- Letter :4.2

We also identified two significant weaknesses in the Air Force's
economic analyses supporting the combination of workloads into single
solicitations at each site.  First, and most significantly, the
analyses were not comprehensive or consistent estimates of the
comparative costs associated with the alternatives examined.  Second,
the cost estimates are questionable for several key categories. 


         INCOMPREHENSIVE OR
         INCONSISTENT COMPARATIVE
         COST ANALYSES
-------------------------------------------------------- Letter :4.2.1

The Air Force analyses stated that workload combination would save
$22 million to $130 million at Sacramento and $92 million to $259
million at San Antonio.  These figures represent estimates of costs
associated with administering the additional contracts and delaying
contract award and transition.  However, the estimates contain two
significant weaknesses. 

First, all costs associated with performing the work are not
included.  For example, the analyses did not consider the cost of
performing maintenance operations, including the costs of labor,
parts, and overhead required to perform the repair under the two
alternatives considered, or the additional layer of cost associated
with subcontracting under the combined workload package scenario. 
Also, the possibility of the cost benefits of increased competition
resulting from solicitations for individual workloads was not
recognized.  Further, because the estimated value of the workload at
these locations is $2.4 billion at Sacramento and $8 billion at San
Antonio, the effect of not considering these costs could
significantly impact the outcome of the analyses.\11 To illustrate
the significance, a small difference of, for example, 5-percent
between cost estimates for single versus multiple solicitations would
represent $120 million and $400 million, for the Sacramento and San
Antonio workloads respectively.  These amounts would materially
affect the savings ranges projected by the Air Force. 

Second, the cost estimates for the two locations did not use
consistent cost elements.  For example, the San Antonio estimate
included a $40-million cost associated with delaying depot closure,
which would reduce the amount of estimated savings, whereas the
Sacramento estimate did not consider such costs.  We do believe costs
associated with delaying closure are relevant to both locations,
although we have some questions about the accuracy of the $40-million
cost estimate. 


--------------------
\11 The Air Force used an 8-year contract period for Sacramento
workloads and a 15-year period for San Antonio workloads in its cost
estimates.  For comparison purposes, we used the same periods. 


         QUESTIONABLE COST
         ESTIMATES
-------------------------------------------------------- Letter :4.2.2

Notwithstanding our concerns about the comprehensiveness and
consistency of the cost analysis, our review of the cost data
provided indicates that the estimates are overstated or questionable
in several areas, including the following: 

  -- Contract administration costs at Sacramento.  The Sacramento
     estimate included a 1.9 percent estimate for contract
     administration costs resulting from having more than one
     contract for the Sacramento workload.  This estimate was based
     on a contractor industrial performance metrics study.\12

This estimate may be overstated because participants in the original
study found the cost impacts projected by the contractor were
significantly overstated.\13 For example, five participants prepared
estimates of the top 10 cost drivers identified in the contractor
study and found that the study estimates were overstated from 14 to
70 percent. 

  -- Closure savings costs.  As mentioned above, the San Antonio cost
     estimate included a $40-million cost associated with delaying
     depot closure.\14

However, this estimate is overstated.  The $40-million estimate is
based on the closure of all logistics operations, some of which will
not close until 2001.\15 According to the BRAC estimates, savings
from closing the depot maintenance operations provided only 21
percent of the estimated annual savings from closure.  At this rate,
the cost of delay should be no higher than $8.4 million rather than
the $40 million estimated in the San Antonio report. 

  -- Transition costs.  Sacramento included a cost estimate for
     extending the transition period.\16 Under the multi-contract
     approach, Sacramento assumed workload segments would be
     transitioned incrementally over a 20- to 24-month period. 
     Although the Air Force may incur additional transition costs
     under a multiple contract strategy, we found transition costs
     were overstated.  The Air Force's transition cost methodology
     assumed that each individual winning offeror would require the
     full 20 to 24 months to complete the transition.  However,
     Sacramento officials recognized that the contractors'
     transitions for the individual workload segments will not
     require the entire 20 to 24-month period.  The officials stated
     that they were unable to separately identify a more precise cost
     estimate. 


--------------------
\12 DOD Regulatory Cost Premium, Coopers and Lybrand, December 1994. 

\13 Acquisition Reform:  DOD Faces Challenges in Reducing Oversight
Costs (GAO/NSIAD-97-48, Jan.  29,1997). 

\14 The Air Force based its $40-million estimate on our report, Air
Force Depot Maintenance:  Privatization-In-Place Plans Are Costly
While Excess Capacity Exists (GAO/NSIAD-97-13, Dec.  31, 1996).  This
report estimated that a 1-year delay in the closure of McClellan Air
Force Base and parts of Kelly Air Force Base would result in a
$90-million reduction of the BRAC Commission's estimated savings. 

\15 Under the Air Force's current closure plans, certain functions
and activities such as program and item management and various tenant
organizations will remain at the closing Sacramento and San Antonio
facilities until 2001. 

\16 The Air Force used an estimate of average transition costs per
day for the Aerospace Guidance and Metrology Center during its
privatization-in-place transition year.  We have testified and issued
two reports on the Center's closure and privatization:  Air Force
Privatization-In-Place:  Analysis of Aircraft and Missile Guidance
System Depot Repair Costs (GAO/NSIAD-98-35, Dec.  22, 1997) and
Aerospace Guidance and Metrology Center:  Cost Growth and other
Factors Affect Closure and Privatization (GAO/NSIAD-95-60, Dec.  9,
1994). 


   CONCLUSIONS
------------------------------------------------------------ Letter :5

The Air Force's support for DOD's determinations that it is more
logical and economical to combine the workloads being competed at the
closing depots is based on a wide variety of information accumulated
during the acquisition strategy development process started in
September 1995.  We recognize that this substantial body of data
includes certain information relevant to the determinations required
by the National Defense Authorization Act of 1998.  We also recognize
that the determinations ultimately represent a management judgment
based on various qualitative and quantitative factors. 

However, our assessment of these factors, as presented by the Air
Force in its February 24, 1998, Sacramento white paper and San
Antonio report shows significant weaknesses in logic, assumptions,
and data.  Consequently, DOD's determinations may well be
appropriate, but its rationale is not well supported. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :6

On April 10, 1998, we provided a draft of this report for comment. 
DOD informed us that, given the short amount of time available, it
chose not to comment on the report at this time. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :7

To determine the reasons the Air Force believes it is more logical
and economical to combine the workloads at the Sacramento and San
Antonio depots, we reviewed the December 19, 1997, reports DOD
provided to Congress, as required by 10 U.S.C.  2469a; the Sacramento
white paper and San Antonio report provided to Congress on February
24, 1998, which expanded on DOD's rationale for combining workloads
into single solicitations; and other information relevant to the
preparation of these reports. 

To analyze the rationale for DOD's determination, we reviewed (1)
information contained in the reports; (2) documentation and other
data supporting the reports; (3) discussions with Air Force officials
responsible for preparing the reports and managing depot maintenance
workloads; (4) discussions with contractor officials who are planning
to participate in the competitions for workloads currently performed
at the Sacramento and San Antonio depots; (5) discussions with Air
Force Audit Agency officials who provided advice on the preparation
of the Sacramento white paper and San Antonio report; (6) a review of
related Air Force studies, reports, and data; (7) our prior work
regarding related depot maintenance issues; and (8) a review of
applicable laws and regulations. 

We conducted our review between February and April 1998 in accordance
with generally accepted government auditing standards. 


---------------------------------------------------------- Letter :7.1

We are sending copies of this report to the Secretaries of Defense
and the Air Force; the Director, Office of Management and Budget; and
interested congressional committees.  Copies will also be made
available to others on request. 

If you or your staff have any questions about this report, please
contact me at (202) 512-8412.  Major contributors to this report are
listed in appendix IV. 

David R.  Warren, Director
Defense Management Issues


REQUIREMENTS FOR REPORTS ON
DEPOT-LEVEL MAINTENANCE
=========================================================== Appendix I

The National Defense Authorization Act for Fiscal Year 1998 contains
several depot-related reporting requirements. 

1.  Report on DOD's Compliance with 50-Percent Limitation (section
358) The act amends 10 U.S.C.  ï¿½ 2466(a) by increasing the amount of
depot-level maintenance and repair workload funds that the Department
of Defense (DOD) can use for contractors from 40 to 50 percent and
revises 10 U.S.C.  ï¿½ 2466(e) by requiring the Secretary of Defense to
submit a report to Congress identifying the percentage of funds
expended for contractors' performance by February 1 of each year. 

Within 90 days of DOD's submission of its annual report to Congress,
we must review the DOD report and report to Congress whether DOD has
complied with the 50-percent limitation. 

2.  Reports Concerning Public-Private Competitions for the Depot
Maintenance Workloads at the Closing San Antonio and Sacramento
Depots (section 359)

The act adds to 10 U.S.C.  a new section, 2469a, which provides for
special procedures for public-private competitions for the workloads
of these two closing depots.  It also requires that we report in the
following areas: 

First, the Secretary of Defense is required to submit a determination
to Congress if DOD finds it necessary to combine any of the workloads
into a single solicitation.  We must report our views on the DOD
determination within 30 days. 

Second, we are required to review all DOD solicitations for the
workloads at San Antonio and Sacramento and to report to Congress
within 45 days of the solicitations' issuance whether the
solicitations provide "substantially equal" opportunity to compete
without regard to performance location and otherwise comply with
applicable laws and regulations. 

Third, we must review all DOD awards for the workloads at the two
closing Air Logistics Centers and report to Congress within 45 days
of the contract awards whether the procedures used complied with
applicable laws and regulations and provided a "substantially equal"
opportunity to compete without regard to performance location,
determine whether "appropriate consideration was given to factors
other than cost" in the selection, and ascertain whether the
selection resulted in the lowest total cost to DOD for performance of
the workload. 

Fourth, within 60 days of its enactment, the 1998 Defense
Authorization Act requires us to review the C-5 aircraft workload
competition and subsequent award to the Warner-Robins Air Logistics
Center and report to Congress on whether the procedures used provided
an equal opportunity for offerors to compete without regard to
performance location, whether the procedures complied with applicable
laws and the Federal Acquisition Regulation, and whether the award
resulted in the lowest total cost to DOD. 

3.  Report on Navy's Practice of Using Temporary Duty Assignments for
Ship Maintenance and Repair (section 366)

The act requires us to report by May 1, 1998, on the Navy's use of
temporary duty workers to perform ship maintenance and repairs at
homeports not having shipyards. 


SACRAMENTO AND SAN ANTONIO DEPOT
MAINTENANCE WORKLOADS
========================================================== Appendix II


   SACRAMENTO
-------------------------------------------------------- Appendix II:1

At the time it was identified for closure during the 1995 base
realignment and closure (BRAC) process, the Air Force's Sacramento
depot had responsibility for the repair of four aircraft and four
commodity groups.  The depot also had a significant body of
manufacturing or repair work it performed in small quantities for
various non-Air Force customers.  Additionally, it had a
microelectronics facility that performed reverse engineering on parts
to provide technical data for manufacturing support parts or for
developing repair procedures. 

Two of the four aircraft repaired at the Sacramento depot will not be
included in the competition package--the F-15 and EF-111.  F-15
repairs are being consolidated at the Warner Robins depot, which is
the F-15 center of excellence and already performs most of the F-15
work.  The EF-111 repair requirement is expected to end as the
aircraft is phased out of operations.  KC-135 and A-10 aircraft
requirements are expected to be included in the Sacramento
competition package.  The KC-135 aircraft is currently repaired at
the Oklahoma City depot and at a contractor facility in Birmingham,
Alabama.  Table II.1 shows the production hours for 1995, 1996, and
1997 for the KC-135 and A-10 aircraft.  The KC-135 workload may be
increased in the competition package, but the A-10 workload is
expected to decrease and to be erratic as the aircraft is phased out
of the inventory. 



                               Table II.1
                
                   Sacramento Depot Aircraft Workload
                         (fiscal years 1995-97)

                                            Fiscal Year
                              ----------------------------------------
Aircraft                              1995          1996          1997
----------------------------  ------------  ------------  ------------
KC-135                             823,755     1,045,027       696,760
A-10                                77,090       102,819        87,939
======================================================================
Total                              900,845     1,147,846       784,699
----------------------------------------------------------------------
Note:  The direct production actual hours are based on customer
orders. 

In accordance with a 1995 BRAC Commission decision, the Sacramento
depot's largest commodity grouping--ground communications and
electronics--which has a projected workload of about 825,000 hours,
is being transitioned to the Tobyhanna Army Depot between 1998 and
2001.  The Sacramento depot's software maintenance workload has
declined significantly, and the remaining software work is expected
to be transferred outside the competition process to the Ogden depot. 
The remaining commodity groups currently repaired at Sacramento
include hydraulics, instruments and avionics, and electrical
accessories. 

Table II.2 provides an overview of the actual direct labor hours used
during fiscal years 1995-97 for the commodity groupings that are
currently repaired at the Sacramento depot and are expected to be a
part of the competitive package. 



                               Table II.2
                
                  Sacramento Depot Commodity Workload
                         (fiscal years 1995-97)

                                            Fiscal Year
                              ----------------------------------------
Commodity Group                       1995          1996          1997
----------------------------  ------------  ------------  ------------
Hydraulics                         449,803       479,702       436,659
Electrical accessories             377,765       350,979       291,449
Instruments and avionics           325,626       289,300       312,226
======================================================================
Total                            1,153,194     1,119,981     1,040,334
----------------------------------------------------------------------
Note:  The direct production actual hours are based on customer
orders. 

The Air Force assessed Sacramento's core capabilities and analyzed
the private sector's repair base.\1 Through this process, which was
approved by the Defense Depot Maintenance Council, none of the
Sacramento workload was determined to be core. 


--------------------
\1 Core capabilities consist of the minimum facilities, equipment,
and skilled personnel necessary to ensure a high level of technical
expertise and combat readiness by maintaining weapon systems
determined to be necessary to support the nation's strategic or
contingency plans.  The objective of the repair base analysis was to
identify industry capabilities and capacity to repair and overhaul
specific workloads. 


   SAN ANTONIO
-------------------------------------------------------- Appendix II:2

At the time of its closure, the San Antonio depot largely did
modifications and repairs of aircraft, turbine engines, and support
equipment, and did a smaller amount of work on nuclear ordnance and
engine software.  The source of repairs for the C-5 aircraft was
determined through a separate public-private competition.  That
workload was won by the Warner Robins depot, which assumed
responsibility for the C-5 in November 1997; work-in-process will
continue at San Antonio until the summer of 1998.  The Warner Robins
depot inducted its first C-5 aircraft in January 1998.  The nuclear
ordnance commodity management workload is being transferred outside
the competition to the Ogden and Oklahoma City depots and Kirkland
Air Force Base, with the bulk of the work going to Ogden. 

Table II.3 shows a breakout of the San Antonio engine workload based
on direct production actual hours for fiscal years 1995 through 1997. 



                               Table II.3
                
                   San Antonio Depot Engine Workload
                         (fiscal years 1995-97)

                                            Fiscal Year
                              ----------------------------------------
Engine                                1995          1996          1997
----------------------------  ------------  ------------  ------------
F100                             1,693,031     1,688,945     1,414,954
T56                                627,199       917,017       981,068
TF39                               462,704       676,837       654,632
======================================================================
Total                            2,782,934     3,282,799     3,050,654
----------------------------------------------------------------------
For various reasons, the competition for engine workloads will not
include all of the workload at the San Antonio depot.  For example,
the Navy is making independent source-of-repair decisions for its T56
engine workloads.  Further, core engine workload will be moved
outside the competition process to the Oklahoma City depot.  The Air
Force assessed the core engine capabilities at the San Antonio and
Oklahoma City depots and analyzed private industry's repair base.  As
a result of this process, the Air Force determined that it should
retain the capability to repair about 24 percent of the annual F100
engine module workload and 50 percent of the workload required to
maintain the capability to repair and check out whole engines--or
about nine whole engines.  Accordingly, the Air Force is moving the
F100 core workload to the Oklahoma City depot outside the engine
competition.  Finally, it is uncertain whether the Air Force could
outsource all the engine workload in the competitive package given
the statutory limits on the percentage of depot maintenance work that
can be performed by the private sector. 


   AIR FORCE MANAGEMENT STRUCTURE
   FOR THE SACRAMENTO AND SAN
   ANTONIO COMPETITIONS
-------------------------------------------------------- Appendix II:3

The Air Force is using a management structure for administering and
managing the Sacramento and San Antonio competitions similar to the
one it used for the C-5 competition.  The structure includes a
program office and evaluation team at each center as well as an
advisory council and source selection official at Air Force
headquarters.  The program office has general responsibility for
preparing and managing the request for proposals.  The evaluation
team will report its assessments to a council made up of
representatives from the Office of the Secretary of Defense, Air
Force headquarters, and Air Force Materiel Command staff.  The
council will review the team's assessment and advise the source
selection official. 


SUMMARY OF JANUARY 20, 1998 GAO
REPORT ON DOD'S DETERMINATIONS TO
COMBINE DEPOT MAINTENANCE
WORKLOADS
========================================================= Appendix III

It may be that the individual workloads at the closing San Antonio,
Texas, and Sacramento, California, Air Force depots cannot as
logically and economically be performed without combination by
sources that are potentially qualified to submit an offer and be
awarded a contract for individual workloads.  However, the DOD
reports and data do not provide adequate information to support DOD's
determinations. 

First, DOD has not analyzed the logic and economies associated with
having the workload performed individually by potentially qualified
offerors.  Consequently, it has no support for determining that the
individual workloads cannot as logically and economically be
performed without combination by sources that would do them
individually.  Air Force officials stated that they were uncertain as
to how they would analyze the performance of workloads on an
individual basis.  However, Air Force studies indicate that the
information to make such an analysis is available.  For example, in
1996 the Air Force performed analyses for six depot-level workloads
performed by the Sacramento depot to identify industry capabilities
and capacity.  Individual analyses were accomplished for hydraulics,
software, electrical accessories, flight instruments, A-10 aircraft,
and KC-135 aircraft depot-level workloads.  As a part of these
analyses, the Air Force identified sufficient numbers of qualified
contractors interested in various segments of the Sacramento workload
to support a conclusion that it could rely on the private sector to
support the workloads. 

Second, reports and available data did not adequately support DOD's
determinations "that the individual workloads cannot as logically and
economically be performed without combination by sources that are
potentially qualified to submit an offer and to be awarded a contract
to perform those individual workloads." For example, DOD's
determination report relating to the Sacramento Air Logistics Center,
McClellan Air Force Base, California, states that all competitors
indicated throughout their Sacramento workload studies that
consolidating workloads offered the most logical and economical
performance possibilities.  This statement was based on studies
performed by the offerors as part of the competition process.\1
However, one offeror's study states that the present competition
format is not in the best interest of the government and recommends
that the workload be separated into two competitive packages.  We
were unable to determine whether the other two contractor studies
support the statement in the DOD report that all competitors favored
consolidating the workloads because the Air Force did not provide us
adequate or timely access to the studies cited in the report. 


--------------------
\1 \2 Prior to the planned competition, the Air Force engaged three
offerors to identify work processes at Sacramento and determine how
those processes could be performed more efficiently. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix IV

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

Julia Denman, Assistant Director

OFFICE OF GENERAL COUNSEL

John Brosnan, Assistant General Counsel

DALLAS FIELD OFFICE

Penney Harwell, Site Senior
Larry Junek, Project Manager
Kim Seay, Senior Evaluator
John Strong, Deputy Project Manager
Pam Valentine, Evaluator




RELATED GAO PRODUCTS
============================================================ Chapter 0

Defense Depot Maintenance:  DOD Shifting More Workload for New Weapon
Systems to the Private Sector (GAO/NSIAD-98-8, Mar.  31, 1998). 

Depot Maintenance:  Lessons Learned From Transferring Alameda Naval
Aviation Depot Engine Workloads (GAO/NSIAD-98-10BR, Mar.  25, 1998). 

Public-Private Competitions:  Access to Records Is Inhibiting Work on
Congressional Mandates (GAO/T-NSIAD-98-111, Mar.  4, 1998). 

Public-Private Competitions:  Access to Records Is Inhibiting Work on
Congressional Mandates (GAO/T-NSIAD-98-101, Feb.  24, 1998). 

Public-Private Competitions:  DOD's Determination to Combine Depot
Workloads Is Not Adequately Supported (GAO/NSIAD-98-76, Jan.  20,
1998).  Public-Private Competition:  Processes Used for C-5 Aircraft
Award Appear Reasonable (GAO/NSIAD-98-72, Jan.  20, 1998). 

DOD Depot Maintenance:  Information on Public and Private Sector
Workload Allocations (GAO/NSIAD-98-41, Jan.  20, 1998). 

Air Force Privatization-in-Place:  Analysis of Aircraft and Missile
System Depot Repair Costs (GAO/NSIAD-98-35, Dec.  22, 1997). 

Outsourcing DOD Logistics:  Savings Achievable but Defense Science
Board's Projections Are Overstated (GAO/NSIAD-98-48, Dec.  8, 1997). 

Air Force Depot Maintenance:  Information on the Cost Effectiveness
of B-1B and B-52 Support Options (GAO/NSIAD-97-210BR, Sept.  12,
1997). 

Navy Depot Maintenance:  Privatizing the Louisville Operations in
Place Is Not Cost-Effective (GAO/NSIAD-97-52, July 31, 1997). 

Defense Depot Maintenance:  Challenges Facing DOD in Managing Working
Capital Funds (GAO/T-NSIAD/AIMD-97-152, May 7, 1997). 

Depot Maintenance:  Uncertainties and Challenges DOD Faces in
Restructuring Its Depot Maintenance Program (GAO/T-NSIAD-97-111, Mar. 
18, 1997 and GAO/T-NSIAD-112, Apr.  10, 1997). 

Navy Ordnance:  Analysis of Business Area Price Increases and
Financial Losses (GAO/AIMD/NSIAD-97-74, Mar.  14, 1997). 

Defense Outsourcing:  Challenges Facing DOD as It Attempts to Save
Billions in Infrastructure Costs (GAO/T-NSIAD-97-110, Mar.  12,
1997). 

High-Risk Series:  Defense Infrastructure (GAO/HR-97-7, Feb.  1997). 

Air Force Depot Maintenance:  Privatization-in-Place Plans Are Costly
While Excess Capacity Exists (GAO/NSIAD-97-13, Dec.  31, 1996). 

Army Depot Maintenance:  Privatization Without Further Downsizing
Increases Costly Excess Capacity (GAO/NSIAD-96-201, Sept.  18, 1996). 

Navy Depot Maintenance:  Cost and Savings Issues Related to
Privatizing-in-Place the Louisville, Kentucky, Depot
(GAO/NSIAD-96-202, Sept.  18, 1996). 

Defense Depot Maintenance:  Commission on Roles and Mission's
Privatization Assumptions Are Questionable (GAO/NSIAD-96-161, July
15, 1996). 

Defense Depot Maintenance:  DOD's Policy Report Leaves Future Role of
Depot System Uncertain (GAO/NSIAD-96-165, May 21, 1996). 

Defense Depot Maintenance:  More Comprehensive and Consistent
Workload Data Needed for Decisionmakers (GAO/NSIAD-96-166, May 21,
1996). 

Defense Depot Maintenance:  Privatization and the Debate Over the
Public-Private Mix (GAO/T-NSIAD-96-146, Apr.  16, 1996, and
GAO/T-NSIAD-96-148, Apr.  17, 1996). 

Military Bases:  Closure and Realignment Savings Are Significant, but
Not Easily Quantified (GAO/NSIAD-96-67, Apr.  8, 1996). 

Depot Maintenance:  Opportunities to Privatize Repair of Military
Engines (GAO/NSIAD-96-33, Mar.  5, 1996). 

Closing Maintenance Depots:  Savings, Personnel, and Workload
Redistribution Issues (GAO/NSIAD-96-29, Mar.  4, 1996). 

Navy Maintenance:  Assessment of the Public-Private Competition
Program for Aviation Maintenance (GAO/NSIAD-96-30, Jan.  22, 1996). 

Depot Maintenance:  The Navy's Decision to Stop F/A-18 Repairs at
Ogden Air Logistics Center (GAO/NSIAD-96-31, Dec.  15, 1995). 

Military Bases:  Case Studies on Selected Bases Closed in 1988 and
1991 (GAO/NSIAD-95-139, Aug.  15, 1995). 

Military Base Closure:  Analysis of DOD's Process and Recommendations
for 1995 (GAO/T-NSIAD-95-132, Apr.  17, 1995). 

Military Bases:  Analysis of DOD's 1995 Process and Recommendations
for Closure and Realignment (GAO/NSIAD-95-133, Apr.  14, 1995). 

Aerospace Guidance and Metrology Center:  Cost Growth and Other
Factors Affect Closure and Privatization (GAO/NSIAD-95-60, Dec.  9,
1994). 

Navy Maintenance:  Assessment of the Public and Private Shipyard
Competition Program (GAO/NSIAD-94-184, May 25, 1994). 

Depot Maintenance:  Issues in Allocating Workload Between the Public
and Private Sectors (GAO/T-NSIAD-94-161, Apr.  12, 1994). 

Depot Maintenance (GAO/NSIAD-93-292R, Sept.  30, 1993). 

Depot Maintenance:  Issues in Management and Restructuring to Support
a Downsized Military (GAO/T-NSIAD-93-13, May 6, 1993). 

Air Logistics Center Indicators (GAO/NSIAD-93-146R, Feb.  25, 1993). 

Defense Force Management:  Challenges Facing DOD as It Continues to
Downsize Its Civilian Workforce (GAO/NSIAD-93-123, Feb.  12, 1993). 


*** End of document. ***