DOD Competitive Sourcing: Savings Are Occurring, but Actions Are Needed
to Improve Accuracy of Savings Estimates (Letter Report, 08/08/2000,
GAO/NSIAD-00-107).

Pursuant to a congressional request, GAO reviewed the Department of
Defense's (DOD) competitive sourcing activities, focusing on: (1) the
extent to which actual savings have been achieved or can be expected as
a result of competitions; and (2) DOD's efforts to improve processes for
identifying and tracking changes to cost and savings estimates.

GAO noted that: (1) available data indicated that DOD realized savings
from 7 of the 9 Office of Management and Budget's Circular A-76 cases
GAO reviewed, although less than the $290 million savings DOD components
initially projected; (2) neither GAO nor DOD could precisely quantify
the extent of savings from these 9 cases; (3) savings estimates were
imprecise for a number of reasons; (4) baseline cost estimates from
which savings were estimated were usually calculated using an average
cost of salary and benefits for the number of authorized positions
rather than using actual costs for the positions actually filled, which
would have been more precise; (5) while most baseline cost estimates
were based largely on personnel costs, up to 15 percent of the costs
associated with the government's most efficient organizations' plans or
the contractors' offers were not personnel costs; (6) because these
types of costs were not included in the baseline, a comparison of the
baseline to the government's most efficient organization or contractor
costs may have resulted in understanding the cost savings; (7) on the
other hand, savings estimates did not reflect the study and
implementation costs, which offset savings in the short term; (8) data
limitations made it impractical to identify precise amounts of savings;
(9) DOD has begun efforts to revise its information management systems
to better track the estimated and actual costs of activities studied,
though not to revise previous savings estimates; (10) DOD is also
emphasizing development of standardized baseline cost data for use in
determining initial savings estimates; (11) however, many of the cost
elements that are used in A-76 studies will continue to be estimated
because DOD lacks a cost accounting system to provide actual costs; (12)
reported savings from A-76 studies will continue to have some element of
uncertainty and imprecision and will be difficult to track in the out
years because workload requirements change, affecting program costs and
baseline from which savings are calculated; and (13) given that DOD is
reducing its operating budgets based on projected savings from A-76
studies, it is important that it have as much information as possible on
savings being realized, including adjustments for up-front investment
costs and other changes that may occur over time.

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

 REPORTNUM:  NSIAD-00-107
     TITLE:  DOD Competitive Sourcing: Savings Are Occurring, but
	     Actions Are Needed to Improve Accuracy of Savings
	     Estimates
      DATE:  08/08/2000
   SUBJECT:  Defense cost control
	     Defense procurement
	     Privatization
	     Comparative analysis
	     Department of Defense contractors
	     Performance measures
	     Management information systems
	     Future budget projections
	     Military downsizing
	     Contract costs
IDENTIFIER:  DOD Commercial Activities Management Information System
	     OMB Circular A-76 Program

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO/NSIAD-00-107

Appendix I: Summary Information for Nine A-76 Case Studies

20

Appendix II: Objectives, Scope, and Methodology

38

Appendix III: A-76 Study Process

40

Appendix IV: Comments From the Department of Defense

44

Appendix V: GAO Contacts and Staff Acknowledgments

46

47

Table 1: Competition Outcomes and Our Assessment of Whether
Savings Were Achieved in Nine Case Studies 8

Table 2: A-76 Cost Comparison Studies Reviewed 38

Figure 1: Overview of the A-76 Process 41

DOD Department of Defense

OMB Office of Management and Budget

OSD Office of the Secretary of Defense

National Security and
International Affairs Division

B-285302

August 8, 2000

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

Dear Mr. Chairman:

Since late 1995, the Department of Defense (DOD) has encouraged the services
and the Defense agencies to conduct cost comparison studies as provided for
in the Office of Management and Budget's Circular A-76. Under the A-76
process, otherwise known as competitive sourcing, Defense components study
commercial activities being performed by government personnel to determine
whether it would be more cost-effective to maintain the activities in-house
or to contract with the private sector for their performance. Between fiscal
year 1997 and 2005, DOD plans to study activities involving about 203,000
positions under competitive sourcing; studies of activities involving about
9,000 positions had been completed as of September 30, 1999.1

By studying activities involving about 203,000 positions, DOD expects to
achieve about $9.2 billion in savings during the fiscal year 1997-2005
period and $2.8 billion in annual recurring savings after fiscal year 2005,
which it plans to use to fund other priority needs such as modernization.2
We have reported that costs for completing A-76 studies and implementing the
results can reduce savings expected in the short term. After the study and
implementation are completed, changes to operating requirements can occur
that can affect the extent to which savings are realized and, in some cases,
make tracking cost and savings estimates increasingly less practical over
time. We have also questioned the reliability of historical data DOD uses to
record costs and savings,3 however, the data are currently being used to
reduce budgets in anticipation of expected savings from current
A-76 studies.

On the basis of your request, we (1) assessed the extent to which actual
savings have been achieved or can be expected as a result of competitions
and (2) identified DOD's efforts to improve processes for identifying and
tracking changes to cost and savings estimates. We judgmentally selected
nine A-76 studies that had been completed by the Army, the Navy, the Air
Force, and the Defense agencies between October 1995 and March 1998 for a
thorough analysis of projected savings. We judgmentally selected these
studies to provide a mix of A-76 studies won by the public and private
sectors as well as studies completed across the various Defense components.
While our sample is insufficient to identify trends, it gives some
indication of the challenges DOD faces in accurately determining savings
realized from its A-76 studies. Appendix I provides summary information on
each case study. Appendix II provides additional information regarding our
scope and methodology.

Available data indicated that DOD realized savings from seven of the nine
A-76 cases we reviewed, although less than the $290 million savings the
Defense components initially projected. Neither we nor DOD could precisely
quantify the extent of savings from these nine cases. Savings estimates were
imprecise for a number of reasons. Baseline cost estimates from which
savings were estimated were usually calculated using an average cost of
salary and benefits for the number of authorized positions rather than using
actual costs for the positions actually filled, which would have been more
precise. While most baseline cost estimates were based largely on personnel
costs, up to 15 percent of the costs associated with the government's most
efficient organizations' plans or the contractors' offers were not personnel
costs. Because these types of costs were not included in the baseline, a
comparison of the baseline to the government's most efficient organization
or contractor costs may have resulted in understating the cost savings. On
the other hand, savings estimates did not reflect the study and
implementation costs, which offset savings in the short term. Data
limitations made it impractical to identify precise amounts of savings.

DOD has begun efforts to revise its information management systems to better
track the estimated and actual costs of activities studied, though not to
revise previous savings estimates. DOD is also emphasizing development of
standardized baseline cost data for use in determining initial savings
estimates. As a practical matter, however, many of the cost elements that
are used in A-76 studies will continue to be estimated because DOD lacks a
cost accounting system to provide actual costs. Further, reported savings
from A-76 studies will continue to have some element of uncertainty and
imprecision and will be difficult to track in the out years because workload
requirements change, affecting program costs and the baseline from which
savings are calculated. Given that the Department is reducing its operating
budgets based on projected savings from A-76 studies, it is important that
it have as much information as possible on savings being realized, including
adjustments for up-front investment costs and other changes that may occur
over time.

We made recommendations to the Secretary of Defense to use data obtained in
tracking the implementation of A-76 study results to modify initial savings
estimates and to study the potential to use a costing methodology to provide
a better calculation of the baseline and other costs used to estimate
savings. The Department generally agreed with the report and the
recommendations.

In 1966, the Office of Management and Budget (OMB) issued Circular A-76,
which established policy for acquiring commercial activities. In 1979, OMB
issued a supplemental handbook to the circular that included procedures for
A-76 cost comparison studies to determine whether commercial activities
should be performed by the government, by another federal agency, or by the
private sector. DOD currently refers to this program of
A-76 cost comparison studies as competitive sourcing.

In conducting an A-76 cost study, an agency develops a performance work
statement to identify the work to be done, prepares a government in-house
cost estimate based on the "most efficient organization" that can accomplish
the work, and compares this estimate with the lowest cost or best value
offer selected from the private sector. The most efficient organization
generally refers to a more streamlined, smaller version of the government
organization than is currently doing the work. The government activity will
be converted to performance by the private sector if the successful private
sector offer is either lower by an amount equal to
10 percent of direct personnel costs of the most efficient organization or
is $10 million less over the length of the specified performance period than
the in-house estimate. A more detailed description of the A-76 study process
is contained in appendix III.

DOD records the results of its A-76 competitions in the Commercial
Activities Management Information System. Each service and Defense agency
maintains its own version of this management information system, but DOD
requires each system to contain certain data elements for individual A-76
cost comparison studies, including numbers and length of individual studies,
numbers of in-house military and civilian positions to be affected,
comparisons of in-house and contractor-estimated costs, contract award
dates, and changes in costs for 3 years after a contract award. We have
previously reported some concerns about the accuracy and completeness of
data contained in components' Commercial Activities Management Information
Systems.4 Continuing concerns about the extent to which DOD was realizing
savings from its A-76 studies prompted Congress, in the National Defense
Authorization Act for Fiscal Year 1998,5 to require DOD to track cost
information on the performance of commercial activities for up to 5 years
during the term of a contract or an extension following implementation of
the A-76 study results.

DOD's savings from the nine A-76 studies we reviewed appeared significant,
although available data indicated that savings were less than the $290
million DOD initially projected. We found that savings occurred, regardless
of whether governmental organizations or private contractors won the
competitions. However, limitations in the way DOD components calculated
savings made the original savings estimates imprecise. Further, costs for
completing A-76 studies and implementing the results were not fully
recognized in savings estimates, which logically would reduce estimated
savings in the short run. After the study and implementation are completed,
changes to operating requirements can occur that, in some cases, can affect
the extent to which savings are realized and, in other cases, make tracking
cost and savings estimates increasingly less practical over time.

A-76 Competitions

Available data indicated that DOD realized savings as a result of A-76 cost
studies in at least seven of the nine cases, whether the in-house or private
sector organization prevailed in the cost studies. Of the remaining two
cases, data were insufficient for us to conclude whether savings were
realized in one instance, and savings for the other one appeared to be
negated by performance problems and contract termination after
20 months. Each of the cases examined presented circumstances that
complicated our review of savings likely to be realized from these studies.
In some cases, as discussed more fully in subsequent sections, we had to
make assumptions concerning how closely estimated baseline costs
approximated the actual costs. In other cases, we had to use the cost
difference between the contractor's offer and the in-house most efficient
organization's cost estimate as a starting point for estimated savings and
later adjusted the savings to account for study and implementation costs and
other factors. More complete observations on savings are included at the end
of each case study in appendix I. Table 1 identifies the winner of the
competitions and our assessment of the likelihood that savings occurred in
each case.

                                       Savings
                                       likely,
 Case study             Winner of the  although less Savings     Savings
                        competition                  uncertain   unlikely
                                       than original
                                       estimate
 1. Missile maintenance
 activity, Redstone     In-house       X
 Arsenal, Alabama
 2. Storage and
 warehouse activity,    Private sector                           X
 Fort Riley, Kansas
 3. Child care
 center--Naval Medical
 Center, San Diego,     Private sector X
 California
 4. Navy regional
 family services
 activity, San Diego,   Private sector X
 California
 5. Aircraft
 maintenance, Altus Air In-house       X
 Force Base, Oklahoma
 6. Base operating
 support, Laughlin Air  Private sector X
 Force Base, Texas
 7. Air Force regional
 engine repair,
 Laughlin Air Force     Private sector               X
 Base, Texas
 8. Base operating
 support,
 Wright-Patterson Air   Private sector X
 Force Base, Ohio
 9. Defense Finance and
 Accounting Service
 vendor payments,       In-house       X
 Columbus, Ohio

Note: Although the private sector won two-thirds of the A-76 studies we
analyzed, recent data indicated that the split between the public and
private sectors was about even.

Source: Our analysis of DOD data.

For the three cases where the government's most efficient organization
prevailed, savings were achieved because fewer persons were required to
perform the work than before the studies were conducted. The clearest
example of this was the Altus Air Force Base study involving aircraft
maintenance. In initiating the study, the Air Force planned to convert its
largely military workforce to civilian personnel, either government
employees or employees of the private sector contractor, depending on the
results of the A-76 study. Either way, civilian workers were expected to be
less costly. We have previously reported that historical data suggest the
potential for significant savings from such conversions because, on average,
individual civilian positions are less costly than military positions.6 At
Altus, the organization in place before the study had 1,444 authorized
positions, 1,401 of which were military.7 After the study, the government's
most efficient organization, which won, had 735 positions--all civilian.
While this case suggests significant reductions in program costs and
provides opportunities to reassign military personnel to other positions,
the amount of budgetary savings likely in the future is unclear because the
Air Force has recently indicated it does not expect to reduce military end
strengths as a result of future A-76 cost studies. In the remaining two
cases in which the government's most efficient organization prevailed,
personnel reductions resulted in savings, however, other factors made
determining the savings associated with these cases difficult.

In four of the six cases where contractors prevailed, savings were apparent
and continued even after some contract modifications were made. In the other
two cases, savings were not clear-cut.

Defense components used a variety of approaches in calculating savings
estimates that limited the precision of their estimates. This occurred in
part because, at the time the nine studies began, there was no official DOD
guidance on calculating estimated costs of current operations for comparison
with projected future costs, as a way of estimating savings.8 However, a
calculation of estimated costs associated with the original
in-house activity is important in determining the extent of budgetary
savings resulting from the cost studies. Therefore, we sought to determine
whether and how DOD components established baseline cost estimates as a
basis for estimating savings from the cost comparison studies. For seven of
the nine cases we reviewed, the Defense components estimated savings by
subtracting the cost of the winning contractor's offer or most efficient
organization's cost estimate from an estimated original cost of operations,
which was largely derived from personnel costs. The approaches used in two
of the seven cases also overstated the savings estimates because savings
from other actions not related to the A-76 cost study were included. For
example, in the Redstone Arsenal case study, Army officials incorporated in
their A-76 savings estimated cost reductions resulting from workload
reductions that were unrelated to the A-76 study. Also, the remaining two
cases used different estimating methods.

In the seven cases where components used personnel reductions as a baseline
for estimating savings, several factors limited the precision of the
estimates. First, baseline costs were calculated using average salary and
benefit costs, which could be greater or less than actual costs. Second,
salary and benefit costs were computed on the number of authorized positions
versus the number of positions actually filled at the time of the
A-76 studies. Since the actual personnel levels were less than authorized in
four cases, savings estimates based on a comparison with a baseline
developed with authorized positions were overstated. (Actual personnel
levels were more than authorized in two cases and, in one case, the number
authorized and actual was not available.) Third, while most savings
estimates were based largely on personnel costs, up to 15 percent of the
costs associated with individual most efficient organization plans or
contractor offers were not personnel-related in some cases. Because no
amounts were included in the baseline cost original activity estimate for
nonpersonnel-related costs, the calculated savings could be underestimated
by the amount of those costs.

Finally, initial savings estimates for two of these seven cases were
inflated by savings from other actions unrelated to determining whether the
activity should be performed by the in-house organization or by a
contractor. In the first instance, which involved the missile maintenance
activity at Redstone Arsenal, changes were occurring concurrent with, but
independent of, the study that would reduce the workload by 40 percent as
well as decrease personnel requirements. Army officials included personnel
reductions from both actions in their savings estimates from the study and,
at the time of our review, they could not readily separate or estimate
reductions related solely to the A-76 study. In the other instance, which
involved the Defense Finance and Accounting Service's vendor payment
activity at Columbus, Ohio, reengineering savings as a result of
consolidating sites were captured as part of the A-76 study. However,
according to a Defense Finance and Accounting Service official, the
consolidation would likely have taken place regardless of whether an A-76
study was accomplished. Therefore, as with the Redstone case, officials
combined personnel reductions and savings from these actions with those
resulting from the A-76 study.

The remaining two cases at Fort Riley and San Diego used different
approaches to calculate savings. At Fort Riley, the Army subtracted the
costs associated with the winning contractor's offer from the costs
associated with the in-house most efficient organization. On the surface,
this approach would suggest savings were understated, since the most
efficient organization would generally be expected to be more
cost-effective than the existing organization. However, as noted earlier,
performance problems resulted in the contract being terminated 20 months
after award. In the second case, which involved the San Diego child care
center, the Navy calculated savings on a per child basis by comparing the
contractor's cost per child with the in-house cost per child. However, the
extent of actual savings over previous operations was made uncertain by the
fact that the winner of the cost study would be operating in a new facility
that was expected to be more efficient than the previous operation. Also,
the number of children actually served in the new facility was much greater
than the number used to project savings. Long-term savings from this A-76
study became even more difficult to assess since, at the end of the 3-year
contract period in September 1999, the activity was combined with other
child care activities in the region and studied under a regional A-76 study.
As a result of that A-76 study, the regionalized activity was returned
in-house.

Available data suggested that savings estimates were overstated because they
did not include (1) all costs of conducting and implementing the results of
the studies and (2) contract administration costs where applicable. In
addition, some changes to the scope of work after A-76 study results were
implemented which also reduced savings. Other changes that occur over time,
unrelated to the original A-76 studies, can increasingly make it impractical
to compare current costs of operation with those at the time of the A-76
studies.

Investment Costs Not Fully Considered in Developing Cost Estimates

The nine A-76 case studies we examined, like all such studies, incurred
up-front investment costs that must be offset before net savings begin to
accrue. They include costs of conducting the studies, personnel transition
costs, and contract administration, as applicable. Costs for the first two
are generally short term in nature, whereas, contract administration costs
extend over the life of a contract, typically up to 5 years. However,
components overstated savings estimates by not recognizing these costs in
estimating their savings.

Based on officials' estimates of the costs to conduct the nine A-76 studies,
we projected the cost per position studied ranged from an estimated
$364 for the Navy child care center, San Diego, California to an estimated
almost $9,000 for the Navy regional family services center, San Diego,
California. In eight of the nine studies we reviewed, in-house personnel
conducted the A-76 studies and the study costs represented installation
officials' estimates of the salaries of in-house staff. The Defense Finance
and Accounting Service vendor payment study was conducted with both
in-house personnel and contractor support.9

Personnel costs associated with transition to either the government's most
efficient organization or contractor performance of an activity were also
not included in savings estimates. Agencies may incur personnel separation
costs when reductions in force are required to implement the cost study
results. These costs can include those associated with voluntary separation
incentives or severance pay associated with involuntary reductions in force,
and can sometimes be reduced by placing affected employees in vacant
government positions. For the nine cases examined, personnel reduction cost
estimates ranged from none at the Redstone Arsenal, where personnel were
placed in other comparable jobs, to a high of $3 million at Wright-Patterson
Air Force Base, where about 500 employees were affected by the decision to
contract for the base operating support function. Further, for example, at
the Altus Air Force Base aircraft maintenance activity, the Laughlin Air
Force Base operating support activity, and the Wright-Patterson Air Force
Base operating support activity, reduction-in-force procedures were used and
affected personnel moved to lower graded positions. These employees retained
the pay of their previous pay grades (known as "save pay"). Each of these
costs served to reduce the savings expected in the short term and some, such
as the save pay, had the potential to carry over into succeeding years.

When cost studies resulted in work being contracted with the private sector,
the affected agencies also incurred costs to administer and oversee
individual A-76 contracts. For the cases reviewed, the cost estimates ranged
from an average of about $12,000 per year for the San Diego child care
contract to over $635,000 per year for the Wright-Patterson Air Force Base
operating support contract in fiscal year 1999. These costs were not
factored into the cost savings estimates.

Changes in Work Scope After the A-76 Competition Is Implemented Can Affect
Savings

Scope of work changes after A-76 study results have been implemented can
reduce the extent to which the original estimates of savings are realized
over the long term. In some cases, changes may include work that was
incorrectly or inadvertently omitted from the statement of work used for the
A-76 study. This type of change, which may increase operating costs, is
logical to consider as reducing original savings estimates. In other cases,
changes to work requirements, such as required by additional missions, could
not have been anticipated at the time of the study and would have arisen
regardless of who won the A-76 study. These changes may increase current
operating costs but are unrelated to any calculation of savings. As the
latter type occurs over time, it renders meaningless any attempt to compare
costs of operation with those that existed before the A-76 study.

Among the six cases where work was awarded to contractors, two contracts had
modifications to correct work statement omissions that, according to
government officials, should have been included in the original statement of
work. One modification to the Wright-Patterson Air Force Base statement
added about $92,000, or less than 1 percent, to the total contract and the
other to the Fort Riley statement added $187,000, or
6 percent. Changes in the scope of work or mission after the studies were
completed increased the work statement for one most efficient organization
at Altus Air Force Base by $3.4 million, or 2 percent, and one for the base
operating support contractor at Laughlin Air Force Base by $830,000, or 3
percent. In three other cases, contracts were subsequently modified to
reflect increases in Department of Labor wage rates--$116,000, $334,000, and
$440,000, or about 1 percent of the total cost of each contract.

Unrecognized Costs

Once weaknesses in the components' savings estimates were taken into account
and adjustments were made for unrecognized costs, available data indicated
that savings remained in most instances. At the same time, each of the cases
examined presented unique circumstances that complicated our review of
savings likely to be realized from these studies. Neither we nor DOD could
precisely quantify the extent of savings from these nine cases.

In some instances, we had to make assumptions concerning how closely
estimated baseline costs approximated the actual costs. In other instances,
we had to use the difference between a contractor's offer and an in-house
most efficient organization cost estimate as a starting point for estimated
savings and make subsequent adjustments to account for study and
implementation costs and other factors. However, to the extent we could
account for these cost factors, it appeared that savings, however imprecise,
were realized in seven of the cases (as shown in table 1).

Savings From Competitive Sourcing

DOD officials are in the process of improving the Commercial Activities
Management Information Systems to address previously identified weaknesses,
as well as developing mechanisms intended to better track costs in response
to legislative requirements.10 However, DOD does not have plans for updating
projected savings data initially recorded in components' Commercial
Activities Management Information Systems. Having precise savings estimates
is important because of the Department's efforts to reduce operating budgets
in anticipation of savings.

The Office of the Secretary of Defense (OSD) is working to update its
overall commercial activities program guidance dated September 1985.
Department officials told us that they expect to issue the guidance later
this year. DOD issued procedural guidance in September 1999 to standardize
the database used for capturing the results of A-76 cost comparison studies.
The guidance emphasized that the original cost of an activity should be
determined by the estimated total cost (expressed in work years) before the
development of the government's most efficient organization. Guidance issued
on April 3, 2000, expanded the guidelines to state that estimates of annual
work years should be determined by the number of actual civilian work years
required to perform the work being studied during the 12 months prior to the
cost comparison study. According to an OSD official, savings estimates will
be based on the difference between this baseline estimate of personnel costs
and the costs associated with either the winning contractor's offer or the
government's most efficient organization cost estimate over the life of the
contract or specified term.

An official responsible for the DOD's A-76 program told us that they expect
to issue additional guidance that focuses on tracking operating costs after
A-76 cost studies are completed and the results are implemented. They
project that mechanisms for tracking this cost information will be in place
by July 2000. The official also told us that under the new guidance, when
changes cause costs to change, the reason for the changes will be reflected
in the components' Commercial Activities Management Information Systems. The
involved installation will maintain supporting documentation for the
changes. However, even when a change affects the initial savings estimate,
the Department currently does not plan to adjust the estimates recorded in
its Commercial Activities Management Information System. However, absent
adjusting the savings estimates in this system or elsewhere, the Department
lacks an effective basis for systematically assessing the extent to which
projected savings are being realized. Since the Department is reducing
future operating budgets in anticipation of savings, we believe having more
accurate savings estimates to review during the annual budget process would
be useful in reassessing whether or not the initial savings projections were
accurate.

Despite the planned information system improvements, many of the database
cost elements will continue to be estimated because, as we have previously
reported, DOD lacks reliable financial and cost data to establish actual
baseline costs and determine the effectiveness of its reforms.11 In 1995,
the Federal Accounting Standards Advisory Board12 developed a set of
managerial cost accounting standards that require federal agencies to
provide reliable and timely information on the full cost of federal
programs, their activities, and outputs, and to report costs continuously,
routinely, and consistently.13 The standards did not require the use of a
particular type of costing system or methodology. While acknowledging that
several costing methodologies have been successful in the private sector and
in some government entities, federal entities were encouraged to evaluate
the potential of activity-based costing as an effective managerial tool.
Activity-based costing focuses on the activities of a production cycle and
uses cost drivers (such as length of time an activity is performed) to
assign costs through activities to outputs. In 1999, we reported that
Defense components had begun to develop some activity-based costing models,
primarily to promote better business practices.14 We found little effort,
however, had been devoted to building costing models for the purpose of
providing a more accurate indication of baseline costs, which should provide
better savings estimates for A-76 cost studies. Further, no other costing
methodology is being promoted for use in A-76 studies, an OSD official said.

Most of the A-76 cases we examined showed that DOD attained savings although
weaknesses in baseline cost data and the omission of implementation costs
meant the savings were not as great as initially projected. Data
limitations, such as imprecise baseline costs, and mitigating factors that
affected individual cases made it impractical for us to attempt to determine
revised savings estimates that would be considered precise. The Department
likewise will be unable to develop more reliable estimates of savings from
its A-76 studies until it improves its initial cost estimates and includes
study and implementation costs, and other factors that may reduce expected
savings over time. Until DOD implements a financial management system
capable of producing accurate and reliable cost data, the use of a costing
methodology, such as activity-based costing, as an interim measure may help
to provide more accurate baseline and other costs and provide better savings
estimates for A-76 studies. Improved estimates are important because of the
Department's practice of reducing operating budgets that are based on
expected savings. The guidance recently provided or planned, if fully
implemented, can be an important step in the right direction toward
addressing this problem. However, we believe additional actions are needed
to ensure that changes in costs and savings are tracked on an ongoing basis
and that efforts are made to systematically track changes in savings
estimates and to use the data to periodically reassess planned reductions in
operating budgets based on projected savings.

To improve the precision of savings estimates resulting from A-76 studies,
we recommend that the Secretary of Defense (1) develop an appropriate
mechanism for assessing changes in A-76 savings in a systematic manner and
require components to use data obtained in tracking the implementation of
A-76 study results to update initial savings and projected savings estimates
during the annual budget process and (2) study the potential to use a
costing methodology as an interim measure to provide a better estimation of
baseline and other costs used to estimate savings from A-76 studies.

The Deputy Under Secretary of Defense (Installations) provided written
comments on a draft of this report. The Department generally agreed with the
report and its recommendations. The Department also provided some technical
comments, which were incorporated in the report as appropriate. The
Department's comments are included in appendix IV.

The Department agreed with our recommendation in our draft report that
Defense components use data obtained in tracking the implementation of A-76
study results to modify initial savings estimates in their Commercial
Activities Management Information Systems, although it emphasized that these
information systems capture actual costs, not savings. It plans, however, to
improve savings estimates based on the actual data captured in these systems
for use during the budget process. While the Department's response in our
discussions with Defense officials suggested that these systems may not
provide the optimum mechanism to track changes in savings, we believe it is
important to have some mechanism to capture savings on an ongoing basis.
Scope of work changes, necessitated due to omissions, after A-76 study
results have been implemented can reduce the extent to which original
estimates of savings are realized over the long term. We modified our
recommendation to emphasize the importance of generating reliable savings
data to use in the budget decision process.

The Department also agreed with our recommendation to study the potential to
use a costing methodology as an interim measure to provide better baseline
and other cost data to estimate savings from A-76 studies. According to the
Department, the services have efforts underway to develop activity-based
cost accounting systems it believes will improve management oversight.

We are sending copies of this report to Representative Solomon Ortiz,
Ranking Minority Member, Subcommittee on Military Readiness, House Committee
on Armed Services; and Senator James M. Inhofe, Chairman, and Senator
Charles Robb, Ranking Minority Member, Subcommittee on Readiness and
Management Support, Senate Committee on Armed Services. We are also sending
copies of this report to the Honorable William S. Cohen, Secretary of
Defense; the Honorable Louis Caldera, Secretary of the Army; the Honorable
Richard Danzig, Secretary of the Navy; the Honorable F.W. Peters, Secretary
of the Air Force; General James L. Jones, Commandant of the Marine Corps;
and the Honorable Jacob J. Lew,

Director, Office of Management and Budget. Copies will be made available to
others upon request. GAO contacts and other key contributors are listed in
appendix V.

Sincerely yours,
David R. Warren, Director
Defense Management Issues

Summary Information for Nine A-76 Case Studies

The following provides key information surrounding the nine A-76 case
studies included in this report. In general, the results of the A-76 studies
indicated there are savings. In all but one of the nine cases, the winner of
the competition was determined by selecting the lowest cost option between
the selected private sector offer and the government's most efficient
organization. In the one case, the higher cost government's most efficient
organization won the competition because the private sector's offer was not
either lower by an amount equal to 10 percent of direct personnel costs of
the most efficient organization or at least $10 million less over the length
of the performance period than the in-house estimate. In all cases, the most
efficient organization contained fewer personnel than the previous in-house
organization. This would suggest that regardless of the winner, cost studies
resulted in a more streamlined, cost-efficient operation and savings may be
assumed. However, in one case, while the lowest cost option was selected,
serious performance problems ensued, and the contract was terminated.

In an effort to make a more precise determination of savings, we focused on
determining whether and how baseline costs were established as a basis for
projecting savings in the Department of of Defense (DOD) components'
Commercial Activities Management Information Systems. When baseline data
were not established, we focused on the differences between costs projected
for in-house most efficient organizations and contractor proposals. We also
examined what costs were associated with conducting the studies and
implementing the results, what changes that may have occurred following
implementation of the results that could affect costs and savings, and
collectively what these data indicated about the level of projected savings.
In selected instances, we identified unanticipated factors that adversely
affected the ability to determine savings.

Overview: The Army performs maintenance operations on various missile
systems and related equipment at its Redstone Arsenal. In fiscal year 1998,
after an A-76 cost study in which the in-house organization prevailed, the
Army converted its missile maintenance activity to the government most
efficient organization. According to the Army's Commercial Activities
Management Information System, estimated savings were $2 million per year,
representing a 65-percent savings over the original cost of operations, for
a total of about $6 million over a 3-year period.15 Of the costs identified
in the government's most efficient organization estimate, about 89 percent
were personnel-related and an Army installation official stated that about
90 percent of the costs identified in the competing contractor's offer were
personnel-related.

Basis for Initial Savings Projection: In completing this A-76 study, Army
officials did not calculate a baseline cost for their existing maintenance
activity. Army officials predicated their savings on the elimination of all
13 military positions and 21 of the 39 civilian positions actually
performing this activity--a 65-percent reduction in the 52 authorized
positions. Savings were calculated using an average cost per position.
However, as noted later, some unspecified portion of the reduction was tied
to workload reductions unrelated to the A-76 study.

Study and Implementation Costs: Redstone Arsenal officials estimated that
they incurred in-house staff costs of almost $63,000 in fiscal year 1998 to
conduct the A-76 study. No contractor support was used to conduct the study.
Officials identified no civilian transition assistance costs because all
affected personnel were placed in other positions at the arsenal.

Changes in Work Requirements Following Implementation of Study Results: We
identified no significant changes to work requirements after completion of
the cost study that might have affected the level of savings initially
projected.

Other Mitigating Factors: At about the same time that Redstone Arsenal
officials were studying their missile maintenance activity, the Army was
removing missile systems from its inventory so the systems no longer needed
maintenance support. This change was expected to reduce the activity's
workload hours by about 40 percent. Accordingly, a reduction in personnel
was planned before the A-76 study began and reductions began during the
study.

The revised workload provided the basis for developing the government's most
efficient organization estimate and contractor proposals and ensured a
common basis for comparison. However, in calculating the savings resulting
from the A-76 study, Army officials included personnel reductions associated
with both the workload reduction and the study because DOD did not require
that calculations be adjusted to reflect A-76 savings only. Officials said
they could not separate out or estimate reductions related solely to the
A-76 study.

Although the government's most efficient organization estimate of almost
$3.2 million exceeded the contractor's nearly $3 million offer, the most
efficient organization was selected because, as provided in Office of
Management and Budget (OMB) Circular A-76 guidelines, the contractor's offer
was not lower by an amount equal to 10 percent of the direct personnel costs
of the most efficient organization or at least $10 million less over the
length of the performance period than the in-house estimate.

Summary Observations Regarding A-76 Savings: Available data indicated
savings were achieved, but the precise amount could not be determined since
baseline operating costs were not established and many of the personnel
reductions were not related to the A-76 study. Redstone Arsenal officials
said their accounting systems did not provide the level of detail necessary
to determine activity costs, including the cost to maintain various missile
systems before or after conversion to the most efficient organization.
Therefore, we could not determine actual costs, and we had to use an
estimated cost to adjust the government's projected net savings to reflect
the effect of up-front A-76 study costs. However, assuming an approximate
correlation between the percentage change in workload (missiles to be
maintained) of 40 percent and the 65 percent in personnel reductions, a
majority of the projected savings could have resulted from the study. Thus,
study and implementation costs of about $63,000 appeared to play a limited
role in reducing the remaining projected savings.

Overview: In fiscal year 1996, after completing an A-76 study, Fort Riley
awarded a firm fixed price contract for its storage and warehousing
activity. The Army's commercial activities database showed $1.2 million in
annual savings were estimated, representing a 66-percent savings over the
most efficient organization estimate, for a total of almost $6 million over
a 5-year period.16 The contract was terminated after 20 months because of
performance problems. Costs identified in the government's most efficient
organization were 98 percent personnel-related and, according to a Fort
Riley official, costs identified in the contractor's offer were 98 to
100 percent personnel-related.

Basis for Initial Savings Projection: Army officials estimated savings based
on the difference between the cost projected for its most efficient
organization and the competing contractor's proposal. Assuming the same
level of work as previously performed, using this method could result in
understating the expected savings because an A-76 study usually results in
performing the work with less personnel.

Study and Implementation Costs: Costs associated with the A-76 study and its
implementation totaled $524,000. Fort Riley officials estimated it cost
$147,000 to conduct the A-76 study in fiscal years 1995 and 1996 using
in-house personnel. The study resulted in eliminating all 29 government
positions, including some temporary workers. While the study was ongoing,
most of the employees found other jobs; only eight permanent employees
remained by the time the study was completed. During a reduction in force
held to implement the results of this study, four of the eight employees
were placed in comparable government jobs and four were placed in lower
graded positions, although they retained their previous pay levels under
"save pay" provisions. Fort Riley officials, however, could not provide the
specific costs associated with the reduction in force of these eight
employees because this action was included in a basewide reduction in force
unrelated to the A-76 study. Officials also estimated contract
administration costs were $126,000 and $251,000 in fiscal years 1996 and
1997, respectively.

Changes to Work Requirements Following Implementation of Study Results:
During the 20-month contract period, modifications to the contract were made
that increased the contract costs by about $453,000 for that period.
Contract payments substantially exceeded the offer price by about $238,000
during the 8 months the contract was in effect in fiscal year 1996 and by
$215,000 in fiscal year 1997. Roughly three-quarters of the 1996 cost
increase was due to a requirement added to the work statement that added
$187,000 (or 6 percent) to the total cost of the contract. The remaining
contract cost increases for both fiscal years 1996 and 1997 were due to the
addition of work because of a change in mission and to resolve workload
backlogs and disputes.

Other Mitigating Factors: Contracting officials stated that there were
numerous performance problems with contractor operations since the beginning
of the contract. Officials stated that the contractor significantly underbid
the requirement and that there was no way to quantify the cost of the lack
of service and support provided the customer. This contract was terminated
at the conclusion of fiscal year 1997, and the contractor was paid about
$828,000 to settle disputed costs.

Summary Observations Regarding A-76 Savings: Given the difficulties that
occurred under this contract and the early termination, we did not attempt
to calculate any savings for the period the contract was in effect, given
that savings were likely not achieved.

Overview: In fiscal year 1997, as the result of its simplified A-76 cost
study,17 the Navy contracted for its child care activity at the Naval
Medical Center in San Diego using a firm fixed price contract. The contract
price was on a per child basis tied to age and the number of children
actually enrolled. The Navy estimated that it would realize savings of about
$3,400 per child annually, a savings of 55 percent from costs per child in
the government's most efficient organization developed for the cost study
(which was the same cost per child of the previous activity). Installation
officials projected savings of about $1 million over a 3-year period based
on serving an average of 100 children over that time period.18 Officials
estimated the costs identified in the most efficient organization and the
original organization were 100-percent personnel−related. The
competing contractor's offer did not identify what percentage was
personnel-related costs because it was based on a per child fixed price.

Basis for Initial Savings Projection: In determining the expected level of
savings, Navy officials used a baseline cost of existing operations as the
basis for the proposed most efficient organization, even though the most
efficient organization was using a low capacity, obsolete, and more costly
facility than the new facility the winning organization would occupy. They
projected savings of about 55 percent based on the difference between the
cost of service per child included in the government's most efficient
organization ($6,200) estimate and the winning contractor's proposal
($2,800) multiplied against an expected caseload of 100 children in the new
facility. The cost comparison was based on serving 79 children; however, the
initial savings estimate was projected for an average of 100 children per
year.

Study and Implementation Costs: Navy officials estimated study costs of
$12,000 using in-house personnel. The study resulted in eliminating 33
government positions; however, all personnel, except one, obtained
comparable positions in the area. That one individual, according to Navy
officials, was separated subject to the reduction in force at a cost of
$2,600. The officials reported costs of $11,000 to administer the contract
in fiscal year 1997. In the following 2 years, the government incurred costs
of about $12,000 each year for contract administration.

Changes in Work Requirements Following Implementation of Study Results:
Government costs for the contract increased from initial estimates because
more children (about 170 per year) were enrolled each year of the contract
than the number used to calculate estimates (100). The increased population
served would also suggest a higher cost avoidance between the contractor's
proposal and the government's most efficient organization estimate. Costs to
run the child care center were based on a combination of costs per child
(variable based on the child's age) and a parent co-payment. The contract
called for the contractor to subtract the co-payment from the fixed rate for
the child and bill the government for the remaining amount. The average cost
per child actually averaged about $3,400 based on varying age groups and
parent co-payments.

Other Mitigating Factors: An original baseline cost per child was used, even
though the government had built a new, larger child care facility to replace
the original center, which was obsolete. The Navy estimate did not make any
adjustments to estimate the cost efficiencies to be gained from occupying a
modern and higher capacity facility. Not adjusting for possible efficiencies
may have overstated the government's estimated cost per child and thus the
savings per child compared with the contractor's cost per child. Both the
in-house organization's estimate and the contractor's price were based on
serving 79 children in the new facility. Navy officials based their initial
savings estimate on an enrollment of 100 children. However, as noted, the
number of children actually enrolled was greater than anticipated, averaging
about 170 over the 3-year contract. The Navy may have understated the total
savings because it underestimated the children to be served.

By the end of the 3-year contract in September 1999, however, the activity
had been combined with other child care activities in the region, studied
under another A-76 cost comparison, and implemented under a regional
government most efficient organization. Therefore, no additional savings or
cost avoidance was attributable to this single study after September 1999.
Any subsequent savings would be reported as part of the new A-76 regional
child care study.

Summary Observations Regarding A-76 Savings: Determining a precise level of
savings was not possible because of the factors that clouded the savings
calculation. Since baseline operations changed, it was not possible to
determine how savings would have been affected if adjustments would have
been made during the A-76 study for cost efficiencies based on the new,
higher capacity facility and a more accurate count of children served.
Nevertheless, available data indicated contractor operations were
significantly less costly than the proposed in-house operation and, the
difference was not significantly reduced by the cost of conducting and
implementing the study. A subsequent A-76 study of providing child care
services on a regional basis identified the potential for additional savings
and returned the function in-house.

Overview: In fiscal year 1998, the Navy consolidated its family services
activity at three locations in San Diego, California, and contracted for it
using a firm fixed price plus award fee contract. The family service
activity provides personal support services, such as marital and financial
counseling. Installation officials said that about $1.8 million in annual
savings were estimated, or savings of 35 percent over existing operations
(based on authorized positions), for a total of $9 million over a 5-year
contract period.19 Officials estimated that, historically, about 84 percent
of in-house costs were personnel-related. During fiscal years 1998 and 1999,
roughly 96 percent of contract costs were personnel-related.

Basis for Initial Savings Projection: Navy officials calculated the
projected savings by comparing an estimate of baseline costs for fiscal year
1996 with the contractor's offer. The estimated baseline costs were
calculated by multiplying the number of authorized positions for the
existing activity by the salary and fringe benefit costs projected on the
average mid-point in each affected grade. Using this approach, officials
identified a baseline cost of about $5 million for the 79 civilian and 10
military positions, representing authorized positions at the three
consolidated activities. However, according to a Navy official, the actual
number of personnel performing the activity totaled 117, plus about 40
volunteers. Navy officials pointed out that the additional personnel were
predominately military that were temporarily placed at the center awaiting a
change in duty or a release from service and there was no specific
requirement that military personnel operate the center. The difference
between the authorized positions and the actual suggested the baseline costs
were understated. At the same time, another perspective on savings is found
in the difference between the total costs associated with the most efficient
organization used in the cost study and the winning contractor's price,
which was about $3 million less over a 5-year contract period.

Study and Implementation Costs: Costs to conduct the A-76 study using
in-house personnel were estimated at about $800,000 and $51,000 in costs
associated with implementing a reduction in force at one of the three
locations. We could not determine if similar costs were incurred at the
other two locations because a Navy official said data from the locations
were not tracked and could not be reconstructed. Navy officials said
contract administration cost about $148,000 annually.

Changes in Work Requirements Following Implementation of Study Results: We
identified no changes in statement of work nor any mandated wage increases
during the period covered by our review.

Other Mitigating Factors: Navy officials estimated that during the
8 months of fiscal year 1998 that the contract was in place and for all of
fiscal year 1999, contract payments were 83 and 95 percent of projected
costs, respectively. The decrease occurred because the contractor did not
earn all award fees. Award fee criteria included the results of customer
satisfaction surveys.

Summary Observations Regarding A-76 Savings: This cost study appears to have
produced sizeable savings over previous operating costs, yet the precise
amount of those savings remains unclear given limitations in baseline costs
and uncertainties over the number of personnel previously required to
perform the work. Additionally, the government's most efficient organization
indicated about 14 percent of its costs were nonpersonnel-related. Precisely
how much this will affect the initial projection of about $1.8 million in
annual savings is unclear. However, data suggested a majority of the savings
would remain, or increase, if those costs were fully accounted for. The
savings would then be reduced by about $1.6 million in study,
implementation, and contract administration costs, but those reductions
represent less than a year's projected savings.

Overview: In fiscal year 1996, following an A-76 study, Altus Air Force Base
converted its aircraft maintenance activity performed predominately by
military personnel to a government most efficient organization performed by
civilian personnel only. According to the Air Force's Commercial Activities
Management Information System, annual savings of almost $100 million over a
5-year period were estimated: a savings of
38 percent over existing operations. This case represents the largest
estimated savings amount of the nine A-76 studies reviewed. Of the costs
identified in the most efficient organization, about 97 percent were
personnel-related. An Air Force official estimated that about 92 percent of
the competing contractor's offer consisted of personnel-related costs.

Basis for Initial Savings Projection: The Air Force developed its savings
projection by taking the difference between its estimated 5-year baseline
cost for the existing operation of $265 million and the in-house most
efficient organization of $166 million. The baseline cost was developed
using an estimated cost of 1,401 authorized military and 43 authorized
civilian positions. The actual number of total personnel assigned was 1,248
(1,206 military and 42 civilians), according to Air Force officials, or 14
percent less than the 1,444 authorized. The most efficient organization was
based on an estimated average cost for 735 authorized civilian positions,
about 97 percent of the total cost reflected personnel costs. Air Force
officials said they could not determine whether
non-personnel costs were included in the original operating cost figure.

Study and Implementation Costs: Air Force officials reported that they
incurred estimated costs of about $948,000 to perform the study and about
$2.2 million of that was related to civilian personnel reductions for "save
pay" through the end of fiscal year 1999. Estimated costs for save pay for
fiscal years 1996 to 2001 were about $3.7 million or almost $706,000 per
year. We can expect that some of these costs can continue for some time
unless the affected employees leave government service or are promoted to
their previous pay grades.

Changes in Work Requirements Following Implementation of Study Results:
Costs for the period we reviewed increased by about $3.4 million after
implementation of the most efficient organization due to a change in
mission. The mission change was unanticipated and was not an omission from
the original performance work statement.

Other Mitigating Factors: The government's most efficient organization was
scheduled to be fully implemented by December 1996; however, hiring problems
delayed implementation until April 1998. During conversion to the government
most efficient organization, Altus Air Force Base officials experienced
problems recruiting fully qualified aircraft mechanics. Lower wage scales at
Altus Air Force Base compared to bases located in areas with higher wage
scales caused retention problems. In addition, during the time Altus Air
Force Base officials were trying to hire qualified personnel, the Air Force
centralized its hiring office at Randolph Air Force Base, further slowing
the hiring process. Also during this time, some aircraft were maintained at
other installations, military personnel were brought to or retained at Altus
Air Force Base to maintain aircraft, and the Air Force began an intensive
aircraft mechanic training program. As a result, Air Force officials told us
they encountered unanticipated transition costs of about $2.4 million. By
April 1998, according to officials, the most efficient organization was
fully operational and accomplishing its mission. According to available
data, the activity spent, in total, about $5 million less than budgeted
amounts during the last 3 months of fiscal year 1996 and fiscal years 1997
to 1999.

Summary Observations Regarding A-76 Savings: The initial savings estimate
was overstated because it was based on the cost of authorized personnel
(includes salary and fringe benefits), whereas the actual number of
personnel on board was about 14 percent less. Also, officials said they did
not have documentation that other nonpersonnel costs were not accounted for.
Even so, the magnitude of savings likely to be realized from going from an
organization of 1,248 military and civilian positions to 793 authorized
civilian positions (a revised authorization figure reflecting the workload
increases) still represents a sizeable savings, even after accounting for
study and implementation costs.

Overview: In fiscal year 1997, Laughlin Air Force Base contracted for its
base operating support activity on a firm fixed price basis. Base operating
support encompasses the civil engineering, supply, and transportation
functions. According to the Air Force's Commercial Activities Management
Information System, almost $6 million in annual savings were estimated, a
savings of about 50 percent over existing operations, for a total of almost
$30 million over a 5-year contract period. According to a Laughlin official,
about 92 percent of the winning contractor's offer represented
personnel-related costs. Available data indicated estimated costs under the
most efficient organization were nearly all personnel-related.

Basis for Initial Savings Projection: The Air Force developed its savings
projection by taking the difference between its estimated baseline costs for
the existing operation compared with the contract price. The baseline costs
were developed by calculating an average cost of 278 authorized positions
(including salary and fringe benefits). The actual number of personnel
assigned as of the date the A-76 study was completed was 121. (The number of
actual personnel assigned as of the date the A-76 study was announced was
unavailable.)

Study and Implementation Costs: In fiscal year 1997, the Air Force incurred
about $1.2 million in A-76 study costs, about $1.2 million for transition
costs, and almost $203,000 in save pay costs. Save pay costs of almost
$195,000 and $198,000 were also incurred in fiscal years 1998 and 1999, and
some unspecified portion of the save pay costs could continue into
subsequent years of the contract unless the affected workers leave
government service or again attain their previous pay grades. Additionally,
the Air Force identified annual recurring costs for contract
administration--more than $320,000 for fiscal year 1997, $323,000 for fiscal
year 1998, and $378,000 for fiscal year 1999.

Changes in Work Requirements Following Implementation of Study Results: We
identified no changes in work requirements that should have been known at
the time the performance work statement was developed for the A-76 study.
However, contract payments increased by about $830,000--about $207,000 in
fiscal year 1997, $583,000 in fiscal year 1998, and $40,000 in fiscal year
1999--due to increased workload.

Other Mitigating Factors: Contract costs increased by $334,000, or $71,000
in fiscal year 1998 and $263,000 in fiscal year 1999, reflecting Department
of Labor mandated wage increases. Actual contract payments were 4 percent,
11 percent, and 3 percent greater than original estimates for fiscal years
1997, 1998, and 1999, respectively, because of the mandated wage increases
and workload changes.

Summary Observations Regarding A-76 Savings: Available data indicated
savings were realized from this A-76 study, but less than initially
projected. If one assumes identified costs provide a rough approximation of
previous costs, the contractor's proposal represents a significant savings
over prior operations. However, there were some limitations in the accuracy
of the estimated baseline costs from which savings projections were
developed because personnel costs were developed on an average basis. Also,
the $6 million in projected annual savings would have to be offset by study
and implementation costs.

Overview: In fiscal year 1997, Laughlin Air Force Base contracted for its
engine regional repair center under a fixed price with award fee contract.
The center repairs aircraft engines that used to be repaired in-house and
under contracts at Randolph, Sheppard, Vance, and Laughlin Air Force Bases.
According to the Air Force's Commercial Activities Management Information
System, savings of $25.5 million over a 51-month period were estimated, a
savings of about 45 percent over its existing operations. The contractor's
offer was 92-percent personnel-related costs, and the government's most
efficient organization estimate was 89-percent personnel-related costs.

Basis for Initial Savings Projection: The Air Force projected
$25.5 million in savings by taking the difference between $56.9 million, the
estimated original cost of operations, and $31.4 million, the contractor
price. The savings calculation was based on consolidating and competing
workload at six Air Force Bases--Randolph, Sheppard, Vance, Laughlin,
Columbus, and Reese--however, workload at just four bases was, in fact,
consolidated. The Air Force did not have data readily available breaking out
projected savings for these four bases.

Study and Implementation Costs: The Air Force estimated almost
$1.4 million in A-76 study, implementation, and contract administration
costs for fiscal years 1997 to 1999. Specifically, Air Force officials
reported estimated costs of (1) $377,000 to conduct the A-76 study and (2)
$159,000 in fiscal year 1997 to pay for personnel retained during the
transition from the in-house activity to the contractor. They also reported
$70,000 for separation incentive pay in fiscal year 1997. Officials said
they incurred a total of $81,000 for save pay between fiscal year 1997 and
1999 due to reduction-in-force actions. While the amounts declined each
year, some amount of save pay costs are likely to continue for unspecified
subsequent years. The officials reported contract administration costs of
about $72,000 in fiscal year 1997, $282,000 in fiscal year 1998, and
$340,000 in fiscal year 1999.

Changes in Work Requirements Following Implementation of Study Results: We
identified no changes in work requirements that should have been known at
the time the performance work statement was developed for the A-76 study.

Other Mitigating Factors: According to Air Force officials, savings
estimates were based on an original cost of operations reflecting
authorizations at one location only, although the activity was also being
performed at the other locations by contractor personnel. Officials said the
cost of contract personnel was included in the estimates, but yet they could
not provide documentation supporting the calculation. Additionally, the Air
Force originally planned to consolidate the engine repair function at six
bases into one. However, it never consolidated the engine work at one base
because it closed. Engine work from a second base was also not consolidated
because of transition problems and the Air Force decided to retain a
separate repair contract at the second base. Yet, the savings projection in
the commercial activities database reflected the expected savings from
consolidating workload at all six bases.

Actual contract payments exceeded estimates by about 11 percent in fiscal
year 1997, 25 percent in fiscal year 1998, and 17 percent in fiscal year
1999, largely because the contractor had to hire additional personnel to
address work backlogs. Backlogs occurred because the contractor was unable
to hire skilled personnel quickly due to the installation's remote location.
Further, personnel that worked in the activity when it was in-house
transferred to other installation activities when the activity converted to
contract and, as a result, were unavailable to work for the contractor. The
contractor hired workers from the original in-house activity on a temporary
basis to meet repair requirements. Actual costs also exceeded estimates due
to mandated wage increases. There was one contract cost increase in 1998 of
almost $116,000 to pay for Department of Labor mandated wage increases.

Summary of Observations Regarding A-76 Savings: We could not determine
whether savings were achieved from this cost comparison because the baseline
data were inaccurate, overstating the savings estimate. The baseline
consisted of the costs of performing the function at six bases, but two of
the six bases were never consolidated into Laughlin.

Ohio

Overview: At the end of fiscal year 1998, Wright-Patterson Air Force Base
contracted for its base operating support activity under a firm fixed price
plus award fee contract. Wright-Patterson's base operating support
multifunction activity includes base supply, transportation, maintenance, a
laboratory, and a laboratory supply function. According to the Air Force's
Commercial Activities Management Information System, $14 million in annual
savings was estimated, for a total of almost $58 million over
49-month contract period, or a savings of about 59 percent over its
previously existing operating costs. About 86 percent of the costs
identified in the government's most efficient organization estimate were for
personnel. Officials could not estimate the contractor's personnel
percentage because the offer did not break down the contractor's costs.

Basis for Initial Savings Projection: The Air Force developed its savings
projection by taking the difference between its estimated baseline costs for
the existing operation and the contract price. The baseline costs included
personnel costs for the 499 authorized positions based on an estimated
average cost per position, including salary and fringe benefits. The actual
number of personnel assigned when the study was announced was 503--the
additional 4 positions being filled by military personnel temporarily
assigned to the organization.

Study, Implementation, and Contract Administration Costs: Study, personnel
reduction, and contract administration costs totaled about
$6 million. Of this amount, according to officials, about $2.2 million was
to conduct the A-76 study. During fiscal years 1998 and 1999, the officials
indicated they had incurred a total personnel reduction cost of about
$3 million (about $691,000 for reduction-in-force actions, $641,000 for
priority placement, $1.3 million for separation incentive payments, $10,000
for career transition assistance, and $381,000 for save pay associated with
reduction-in-force actions). Contract administration costs were $163,539 and
$635,399 for fiscal years 1998 and 1999.

Changes in the Contract Following Implementation of Study Results: We
identified one small change to the performance work statement to add a task
previously omitted; however, the amount was about $92,000 or less than one
percent of total contract costs.

Other Mitigating Factors: Actual contract payments exceeded estimates for
fiscal year 1998 by $1.3 million because the cost of the 2-month transition
period was not included in the cost comparison estimates for either the
contractor or the most efficient organization. The cost was excluded because
Air Force officials said they believed it would be about the same for the
contractor and the most efficient organization. During our review, a
contract change occurred due to a Department of Labor mandated wage increase
of about $440,000, which increased costs by about 1 percent.

Summary Observations Regarding A-76 Savings: This cost study appears to have
produced sizeable savings over previous operating costs, yet the precise
amount of those savings remains unclear given limitations in baseline
costs--these costs were estimated based on average personnel costs. The
government's most efficient organization indicated about
14 percent of its costs were not personnel-related. Precisely how much this
would affect the initial projection of about $14 million in annual savings
is unclear, but the data suggest a majority of the savings would remain if
those costs were fully accounted for. The savings would then be reduced by
about $6 million in study, implementation, and contract administration
costs, but those reductions represent less than a year's projected savings.

Ohio

Overview: In 1998, after completing an A-76 study, the Defense Finance and
Accounting Service retained its vendor payment activity in-house but
implemented the most efficient organization used in the cost comparison
study. The activity processes payments to Defense Commissary Agency vendors.
The agency estimated it would realize $10 million in annual savings, for a
total of $50 million in savings over 5 years. This total suggested an
estimated savings of 79 percent over previous operations. For the costs
identified in the most efficient organization estimate, about
87 percent were personnel-related, whereas about 69 percent of the costs
associated with the contractor's offer were personnel-related.

Basis for Initial Savings Projection: The Defense Finance and Accounting
Service developed its savings projection based on a reduction from 240 to 75
authorized positions. However, only 206 personnel were actually assigned at
the beginning of the A-76 study, about 14 percent less than the authorized
level. The 165 authorized positions eliminated were multiplied by a $61,000
average cost of salary and fringe benefits to arrive at the $10 million
annual savings.

Study and Implementation Costs: Agency officials indicated that they
incurred costs of $828,000 to complete their A-76 study, including in-house
personnel and contracted support. They identified implementation costs in
fiscal year 1998 of about $419,000, which included separation incentives,
retaining staff to assist in the conversion, and the priority placement
program. Additional costs of almost $16,000 were incurred in fiscal year
1999 for remaining priority placement program costs. About $117,000 of the
fiscal year 1998 implementation cost was for implementing
reduction-in-force actions. Officials said they could not provide separate
costs for administration and oversight of the most efficient organization
because these costs are considered overhead.

Changes in Work Requirements Following Implementation of Study Results:
There were no cost increases associated with changes to the performance work
statement. However, changes to the grade structure and a reassessment of
tasks allowed the activity to reduce nine positions in 1999, saving an
estimated $452,000 between 1999 and 2002, or about $158,000 per year.

Other Mitigating Factors: As part of the A-76 study, Defense Finance and
Accounting Service officials decided to consolidate the workload from three
locations into one, and savings from this effort were included in the A-76
savings calculation. This action, however, was unrelated to determining
whether the activity should be performed by the in-house organization or by
a contractor. According to a Defense Finance and Accounting Service
official, the consolidation would likely have taken place regardless of
whether an A-76 study was accomplished. Also, as mentioned before, not all
of the 240 authorized positions were filled. Thus, the A-76 estimated
savings were overstated. Columbus Center officials estimated that the
savings from the A-76 study would have been about
$25 million or about $5 million annually, if the savings from the
consolidation action were not included. However, according to a senior
Defense Finance and Accounting Service official, the consolidation decision
was made as part of the A-76 study and, therefore, the agency's projected
savings estimate will remain at $50 million.

Actual operating costs of $2.5 million roughly equaled estimated costs for
fiscal year 1999. Actual operating costs for fiscal year 1998 were
$2.3 million, just under the $2.6 million estimated cost for the year.

Summary Observations Regarding A-76 Savings: According to available data,
savings realized from this A-76 study were much less than initially
projected by the Agency because savings from unrelated management efforts
were combined with results of the A-76 study. Also, data indicated that a
certain amount of previously existing operating costs were not identified
since the savings calculation was based only on personnel costs (the most
efficient organization data indicated that about 13 percent of the costs
were nonpersonnel-related.) In addition, actual personnel on board was 14
percent lower than the authorized level used to calculate the savings, which
would further tend to overstate savings. However, available data indicate
that savings would remain, even after accounting for short-term offsets to
account for study and implementation costs.

Objectives, Scope, and Methodology

Based on a request from the Chairman of the Subcommittee on Military
Readiness, House Committee on Armed Services, we conducted this review to
assess the extent of savings being realized from completed A-76 studies. We
followed a case study approach, examining the results of 9 of the 53
A-76 studies completed between October 1995 through March 1998, where
projected savings from the 9 represented over half of projected savings from
the 53 studies ($290 million out of a total of $528 million.) We selected
this time frame with the expectation that documentation of costs and any
changes to workload or mission would be more readily available and that
sufficient time would have elapsed since implementation of the study results
to permit examination of savings. We judgmentally selected cost comparison
studies to provide a mix of activities converted to a government most
efficient organization and those converted to a contractor. We also included
studies that were done by the Army, the Navy, the Air Force (the Marine
Corps had not completed any studies) and the Defense Finance and Accounting
Service. The sample we reviewed was insufficient to identify trends and was
not projectable to the universe of
A-76 studies. However, the sample gives some indication of the challenges
DOD faces in characterizing savings from A-76 studies. See table 2 for the
A-76 studies we reviewed.

 Service/Defense                              Positions
 agency          Location        Function     studied         Winner
                                              (authorized)

 Army            Redstone        Missile      52              Government
                 Arsenal, Ala.   maintenance                  organization

                 Fort Riley,     Storage and  29              Contractor
                 Kans.           warehouse

 Navy            San Diego,      Child care   33              Contractor
                 Calif.          center
                                 Family
                 San Diego,      service      88              Contractor
                 Calif.
                                 centers

 Air Force       Altus Air Force Aircraft     1,444           Government
                 Base,Okla.      maintenance                  organization
                                 Base
                 Laughlin Air    operating    278             Contractor
                 Force Base, Tex.
                                 support
                                 Regional jet
                 Laughlin Air    engine       50              Contractor
                 Force Base, Tex.
                                 maintenance
                 Wright-PattersonBase
                 Air Force Base, operating    499             Contractor
                 Ohio            support
                                 Defense
 Defense Finance                 Commissary
 and Accounting  Columbus, Ohio  Agency       240             Government
 Service                         vendor                       organization
                                 payment

To determine whether the DOD's A-76 studies had achieved savings, we
assessed the validity of DOD's savings projections for individual studies
and compared actual costs with projections for each of the nine A-76 studies
selected. We interviewed installation-level officials and reviewed
supporting documentation to determine whether and what original baseline
cost estimates had been established or how savings were otherwise estimated.
Since most A-76 studies result in personnel reductions, we also sought to
determine the number of personnel authorized and actually on-board when the
cost studies began. We also sought to obtain and compare available
information concerning costs associated with the government's most efficient
organization estimate used in the cost comparisons as well as costs
associated with the private sector proposals. Likewise, we also obtained
information concerning costs associated with conducting the studies, as well
as civilian personnel transition and contract administration and oversight
costs.

In general, officials were able to provide supporting documentation for most
of the cost information we requested. However, a significant portion of the
figures were reconstructed estimates based on available documentation and
cognizant officials' assumptions, such as the cost of the original in-house
organization and the conduct of the A-76 study. In some cases, we estimated
costs based on documentation provided by DOD officials. Further, in a few
instances, cost figures were based on an official's best guess because
documentation could not be found. We did not independently verify the data
provided.

To identify DOD's efforts to improve processes for tracking changes to costs
and savings estimates, we interviewed officials from the Office of the
Secretary of Defense; the Departments of the Army, the Navy, and the Air
Force; and the Defense Finance and Accounting Service. We also reviewed OSD
and service guidance on the competitive sourcing program.

We conducted our review from July 1999 through May 2000 in accordance with
generally accepted government auditing standards.

A-76 Study Process

In general, the A-76 process consists of six key activities--(1) developing
a performance work statement and quality assurance surveillance plan;
(2) conducting a management study to determine the government's most
efficient organization; (3) developing an in-house government cost estimate
for the most efficient organization; (4) issuing a Request for Proposals or
Invitation for Bids; (5) evaluating the proposals or bids and comparing the
in-house estimate with a private sector offer or interservice support
agreement and selecting the winner of the cost comparison; and
(6) addressing any appeals submitted under the administrative appeals
process, which is designed to ensure that all costs are fair, accurate, and
calculated in the manner prescribed by the A-76 handbook.

Figure 1 shows an overview of the process. The solid lines indicate the
process used when the government issues an Invitation for Bids or Request
for Proposals requesting bids or proposals on the cost of performing a
commercial activity. This type of process is normally used for more routine
commercial activities, such as grass-cutting or cafeteria operations, where
the work process and requirements are well defined. The dotted lines
indicate the additional steps that take place when the government wants to
pursue a negotiated, "best value" procurement. This type of process is often
used when the commercial activity involves high levels of complexity,
expertise, and risk.

Source: Air Force Air Education and Training Command documents.

The Circular A-76 requires the government to develop a performance work
statement. This statement, which is incorporated into either the Invitation
for Bids or Request for Proposals, serves as the basis for both government
estimates and private sector offers. If the Invitation for Bids process is
used, each private sector company develops and submits a bid, giving its
price for performing the commercial activity. While this process is taking
place, the government activity performs a management study to determine the
most efficient and effective way of performing the activity with
in-house staff. Based on this "most efficient organization," the government
develops a cost estimate and submits it to the selecting authority. The
selecting authority concurrently opens the government's estimate along with
the bids of all private sector firms.

If the Request for Proposals--best value process--is used, the Federal
Acquisition Regulation and the A-76 supplemental handbook require several
additional steps. The private sector offerors submit proposals that often
include a technical performance proposal and a price. The government
prepares an in-house management plan and a cost estimate based strictly on
the performance work statement. On the other hand, private sector proposals
can offer a higher level of performance or service.

The government's selection authority reviews the private sector proposals to
determine which one represents the best overall value to the government
based on such considerations as (1) performance levels, (2) proposal risk,
(3) past performance, and (4) price. After the completion of this analysis,
the selection authority prepares a written justification supporting its
decision. This includes the basis for selecting a contractor other than the
one that offered the lowest price to the government. Next, the authority
evaluates the government's most efficient organization and determines
whether it can achieve the same level of performance and quality as the
selected private sector proposal. If not, the government must then make
changes to its most efficient organization to meet the performance standards
accepted by the authority. This is intended to ensure that the
in-house cost estimate is based upon the same scope of work and performance
levels as the best value private sector offer. After the authority
determines that the most efficient organization and the private sector
proposal are based on the same level of performance, the cost estimates are
compared.

According to Office of Management and Budget's A-76 guidance, the activity
will be converted to performance by the private sector if the private sector
offer is either lower by an amount equal to 10 percent of the direct
personnel costs of the in-house estimate or is $10 million less over the
performance period than the in-house estimate, whichever is less. The Office
of Management and Budget established this minimum cost differential to
ensure that the government would not convert performance for marginal
estimated savings.

Participants in the process--for either the Invitation for Bids or Request
for Proposals process--may appeal the selection authority's decision if they
believe the costs submitted by one or more of the participants were not
fair, accurate, or were not calculated in the manner prescribed by the A-76
handbook. Appeals must be submitted in writing and within 20 days after the
date that all supporting documentation is made publicly available. The
appeal period may be extended to 30 days if the cost comparison is
particularly complex. Appeals are supposed to be adjudicated within
30 days after they are received.

Comments From the Department of Defense

GAO Contacts and Staff Acknowledgments

Barry Holman (202) 512-5581
Marilyn Wasleski (202) 512-8436

In addition to those named above, Christine Frye, Thad Rytel, and
Jonathan Silverman made key contributions to this report.

Related GAO Products

Competitive Contracting: The Understandability of FAIR Act Inventories Was
Limited (GAO/GGD-00-68 , Apr. 14, 2000).

DOD Competitive Sourcing: Potential Impact on Emergency Response Operations
at Chemical Storage Facilities Is Minimal (GAO/NSIAD-00-88 , Mar. 28, 2000).

DOD Competitive Sourcing: Plan Needed to Mitigate Risks in Army Logistics
Modernization Program (GAO/NSIAD-00-19 , Oct. 04, 1999).

DOD Competitive Sourcing: Air Force Reserve Command A-76 Competitions
(GAO/NSIAD-99-235R , Sept. 13, 1999).

DOD Competitive Sourcing: Lessons Learned System Could Enhance A-76 Study
Process (GAO/NSIAD-99-152 , July 21, 1999).

Defense Reform Initiative: Organization, Status, and Challenges
(GAO/NSIAD-99-87 , Apr. 21, 1999).

Quadrennial Defense Review: Status of Efforts to Implement Personnel
Reductions in the Army Materiel Command (GAO/NSIAD-99-123 , Mar. 31, 1999).

Defense Reform Initiative: Progress, Opportunities, and Challenges
(GAO/T-NSIAD-99-95 , Mar. 2, 1999).

Force Structure: A-76 Not Applicable to Air Force 38th Engineering
Installation Wing Plan (GAO/NSIAD-99-73 , Feb. 26, 1999).

Future Years Defense Program: How Savings From Reform Initiatives Affect
DOD's 1999-2003 Program (GAO/NSIAD-99-66 , Feb. 25, 1999).

DOD Competitive Sourcing: Results of Recent Competitions
(GAO/NSIAD-99-44 , Feb. 23, 1999).

DOD Competitive Sourcing: Questions About Goals, Pace, and Risks of Key
Reform Initiative (GAO/NSIAD-99-46 , Feb. 22, 1999).

OMB Circular A-76: Oversight and Implementation Issues (GAO/T-GGD-98-146 ,
June 4, 1998).

Quadrennial Defense Review: Some Personnel Cuts and Associated Savings May
Not Be Achieved (GAO/NSIAD-98-100 , Apr. 30, 1998).

Competitive Contracting: Information Related to the Redrafts of the Freedom
From Government Competition Act (GAO/GGD/NSIAD-98-167R , Apr. 27, 1998).

Defense Outsourcing: Impact on Navy Sea-Shore Rotations (GAO/NSIAD-98-107 ,
Apr. 21, 1998).

Defense Infrastructure: Challenges Facing DOD in Implementing Defense Reform
Initiatives (GAO/T-NSIAD-98-115 , Mar. 18, 1998).

Defense Management: Challenges Facing DOD in Implementing Defense Reform
Initiatives (GAO/T-NSIAD/AIMD-98-122 , Mar. 13, 1998).

Base Operations: DOD's Use of Single Contracts for Multiple Support Services
(GAO/NSIAD-98-82 , Feb. 27, 1998).

Defense Outsourcing: Better Data Needed to Support Overhead Rates for A-76
Studies (GAO/NSIAD-98-62 , Feb. 27, 1998).

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

Financial Management: Outsourcing of Finance and Accounting Functions
(GAO/AIMD/NSIAD-98-43 , Oct. 17, 1997).

Base Operations: Contracting for Firefighters and Security Guards
(GAO/NSIAD-97-200BR , Sept. 12, 1997).

Terms Related to Privatization Activities and Processes (GAO/GGD-97-121 ,
July 1997).

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

Base Operations: Challenges Confronting DOD as It Renews Emphasis on
Outsourcing (GAO/NSIAD-97-86 , Mar. 11, 1997).

Public-Private Mix: Effectiveness and Performance of GSA's In-House and
Contracted Services (GAO/GGD-95-204 , Sept. 29, 1995).

Government Contractors: Are Service Contractors Performing Inherently
Governmental Functions? (GAO/GGD-92-11 , Nov. 18, 1991).

OMB Circular A-76: Legislation Has Curbed Many Cost Studies in Military
Services (GAO/GGD-91-100 , July 30, 1991).

OMB Circular A-76: DOD's Reported Savings Figures Are Incomplete and
Inaccurate (GAO/GGD-90-58 , Mar. 15, 1990).

(709452)

Table 1: Competition Outcomes and Our Assessment of Whether
Savings Were Achieved in Nine Case Studies 8

Table 2: A-76 Cost Comparison Studies Reviewed 38

Figure 1: Overview of the A-76 Process 41
  

1. This figure does not include approximately 2,500 positions directly
converted to or from contractor performance without a cost comparison study.

2. Activities involving an additional 42,000 positions are expected to be
reviewed under a business process reengineering initiative, referred to as
strategic sourcing, which is expected to help DOD achieve a total of $11.7
billion in savings for these combined efforts by 2005 and increase the
amount of recurring savings to $3.5 billion annually thereafter.

3. DOD Competitive Sourcing: Results of Recent Competitions (GAO/NSIAD-99-44
, Feb. 23, 1999). Also, a list of related products is included at the end of
this report.

4. DOD Competitive Sourcing (GAO/NSIAD-99-44 , Feb. 23, 1999).

5. Section 385, P.L. 105-85, amending 10 U.S.C. 2463.

6. See Base Operations: Challenges Confronting DOD as It Renews Emphasis on
Outsourcing (GAO/NSIAD-97-86 , Mar. 11 1997), DOD Force Mix Issues:
Converting Some Support Officer Positions to Civilian Status Could Save
Money (GAO/NSIAD-97-15 , Oct. 23, 1996), and DOD Force Mix Issues: Greater
Reliance on Civilians in Support Roles Could Provide Significant Benefits
(GAO/NSIAD-95-5 , Oct. 19, 1994).

7. As discussed later, in many of the case studies examined in this report,
including Altus, the actual number of persons on board at the time of the
cost comparison studies was often less than authorized, which inflated the
projected level of savings.

8. OMB Circular A-76, as revised, does not require organizations to
calculate the cost of performing the original in-house activity to complete
an A-76 study. The circular requires calculating the estimated cost for the
government's proposed most efficient organization to fulfill requirements of
the new performance work statement.

9. Available data indicated that all agencies are now using contractor
personnel to help facilitate the completion of the studies.

10. 10 U.S.C. 2463.

11. DOD has been attempting to improve its financial data for many years and
in October 1998 issued to Congress a Biennial Financial Management
Improvement Plan. The plan provided a first-ever vision of DOD's future
financial management environment and identified an array of improvement
initiatives. See Financial Management: Analysis of DOD's First Biennial
Financial Management Improvement Plan (GAO/AIMD-99-44 , Jan. 29, 1999) for
our assessment of the plan. Also see Department of Defense: Progress in
Financial Management Reform (GAO/T-AIMD/NSIAD-00-163 , May 9, 2000).

12. The Federal Accounting Standards Advisory Board was created in October
1990 by the Secretary of the Treasury, the Director of OMB, and the
Comptroller General to consider and recommend accounting principles for the
federal government.

13. Statement of Federal Financial Accounting Standards No. 4, Managerial
Cost Accounting Standards (July 31, 1995).

14. DOD Competitive Sourcing: Lessons Learned System Could Enhance A-76
Study Process (GAO/NSIAD-99-152 , July 21,1999).

15. This figure differs from the $100,000 in estimated multiyear savings we
reported in an earlier report. An Army headquarters official said the
estimates differ because installation officials originally calculated
savings based on an average salary of $50,000 per position. Army
headquarters revised the estimated savings figure in its management
information system to reflect fringe benefits in the average salary used in
the most efficient organization estimate. The later reflects a more accurate
estimate, the official said.

16. As discussed, the savings figure in the Army's Commercial Activities
Management Information System database is higher from what we reported
previously. The Army's revised estimated savings figure is based on the
average salary and fringe benefits in the most efficient organization
estimate and, according to an Army headquarters official, the $1.2 million
figure is more accurate.

17. OMB Circular A-76 Revised Supplemental Handbook provides simplified cost
comparison procedures that may be used for activities involving 65 full-time
equivalent positions or less.

18. The Navy's Commercial Activities Management Information System reports a
smaller 3-year savings figure ($700,000). According to a Navy headquarters
official, the estimate provided by installation officials was more accurate.
Once data are entered into the Navy's system, the headquarters official
said, entries such as estimated savings are not changed or revised. Further,
the installation officials' figure was more reliable because the
installation has current information and, since this contract is no longer
in effect, reconstructing the basis for the savings figure in the system
would be difficult, the official said.

19. The Navy's Commercial Activities Management Information System includes
a higher estimated savings figure ($10.3 million). However, a Navy
headquarters official said the
$9 million revised figure provided by installation officials was more
reliable because the installation has access to current information.
*** End of document. ***