-------------------------Indexing Terms------------------------- 
REPORTNUM:   NSIAD-98-151						        

TITLE:     EVOLVED EXPENDABLE LAUNCH VEHICLE
- DOD GUIDANCE NEEDED TO PROTECT
GOVERNMENT'S INTEREST

DATE:   06/11/1998 
				                                                                         
----------------------------------------------------------------- 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


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


Report to the Chairman, Subcommittee on National Security, Committee
on Appropriations, House of Representatives

June 1998

EVOLVED EXPENDABLE LAUNCH VEHICLE
- DOD GUIDANCE NEEDED TO PROTECT
GOVERNMENT'S INTEREST

GAO/NSIAD-98-151

Expendable Launch Vehicle

(707290)


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

  DOD - Department of Defense
  EELV - Evolved Expendable Launch Vehicle
  IR&D - independent research and development
  NASA - National Aeronautics and Space Administration
  NPV - net present value
  SBIRS - Space-Based Infrared System

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


B-277970

June 11, 1998

The Honorable C.W.  Bill Young
Chairman, Subcommittee on National Security
Committee on Appropriations
House of Representatives

Dear Mr.  Chairman: 

The primary purpose of the Department of Defense's (DOD) Evolved
Expendable Launch Vehicle (EELV) program is to develop a family of
vehicles that will (1) reduce the costs of launching satellites into
space and (2) at a minimum, maintain the reliability, operability,
and capability levels of current launch systems.  As you requested,
we reviewed the EELV program, with emphasis on DOD's revised
acquisition approach.  We specifically reviewed whether (1) DOD's
goal of reducing recurring space launch costs could be achieved, (2)
DOD's planned investment would result in commensurate benefits, and
(3) there are risks that could affect the program. 


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

In 1994, by congressional direction, DOD developed a space launch
modernization plan (known as the Moorman study) that led to the EELV
program.  In 1995, the Air Force entered a low-cost concept
validation phase with four competing contractors.  In 1996, the Air
Force proceeded into the current pre-engineering and manufacturing
development phase with two competing contractors--McDonnell Douglas
Aerospace, which later became part of The Boeing Company, and
Lockheed Martin Astronautics.  In June 1998, the Air Force plans to
proceed into the final development phase with the primary purpose of
fabricating launch vehicles and activating the launch sites. 

DOD's initial acquisition strategy was to select one contractor for
final development and production.  For development, the plan was to
issue a cost-plus-award-fee contract, whereby the government would
have paid all of the approximate $1.5 billion in development costs. 
However, in November 1997, DOD approved a revised acquisition
approach designed to maintain the ongoing competition between the two
contractors for final development and production.  The revised
approach was based on forecasts that growth in the commercial space
launch services market would support more than one U.S.  contractor. 
Also, the approach anticipates that DOD and the contractors would
share in the cost of developing the EELV system, which the Air Force
defines as the launch vehicles, infrastructure, support systems, and
interfaces.  DOD's cost share is planned to be fixed at an amount not
to exceed $1 billion--$500 million for each contractor.  The
contractors are expected to contribute their own funds, as necessary,
to complete EELV development. 

To provide the contractors sufficient flexibility in financing their
share of development costs, the Air Force is proposing to use an
acquisition instrument that is referred to as an "other transaction."
Such instruments, which are authorized under 10 U.S.C.  2371, are
agreements other than contracts, cooperative agreements, or grants.\1

Consequently, other transaction instruments are not subject to
federal procurement laws or the regulations that specifically govern
contracts, cooperative agreements, or grants.  They (1) permit a
deregulation of the government research and development system and
allow rules and regulations to be applied by agreement on a selective
basis if deemed to add value and (2) allow significant flexibility in
negotiating terms and conditions with recipients.  They are, however,
subject to certain laws that have general applicability, such as
civil rights and trade secret statutes. 

With the signing of two other transaction instruments (one for each
development contractor), the Air Force intends to concurrently (1)
award one or two firm-fixed-price initial launch service contracts
for 30 or more satellite launches that are to occur during fiscal
years 2002 through 2005 and (2) execute leasing, licensing, and base
support agreements for launch site and facility use.  According to
the Air Force, this approach is intended to establish an
interdependency among the instruments, contracts, and agreements to
better ensure that a full family of vehicles--medium-lift,
intermediate-lift, and heavy-lift--is developed.  The Air Force
believes that the contractors would not develop this family of
vehicles if the contractors were not concurrently obligated to
provide a full range of launch services. 


--------------------
\1 From a federal government perspective, standard business
arrangements use (1) contracts, which are characterized as
acquisition instruments to acquire goods and services for the direct
benefit of the government or (2) cooperative agreements and grants,
which are characterized as assistance instruments to stimulate or
support a public purpose, rather than to acquire goods or services
for the government.  "Other transactions" are referred to as
nonstandard business arrangements and are characterized as either (1)
assistance instruments to carry out research projects or (2)
acquisition instruments to carry out prototype projects directly
relevant to weapons or weapon systems. 


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

DOD's goal in acquiring the Evolved Expendable Launch Vehicle system
is to reduce recurring production and launch costs by at least 25
percent (in fiscal year 1995 dollars) for fiscal years 2002 through
2020 from the costs that would be incurred if the existing Delta,
Atlas, and Titan launch vehicles were used.  Using DOD's methodology,
we estimated that the program would exceed the 25-percent goal. 
However, the number, type, and timing of launches specified in the
vehicle's mission model have continued to fluctuate, making a cost
reduction estimate, based on the model, uncertain.\2

The major reasons for the fluctuations were that (1) satellites were
assigned to the wrong type of launch vehicle, (2) launch requirements
were unverified, and (3) satellite downsizing has changed launch
requirements.  The Air Force is in the process of developing a new
launch cost baseline and cost reduction estimate, based on the most
current EELV mission model, in preparation for the DOD milestone II
review in June 1998. 

More importantly, the Air Force's recurring cost methodology does not
adequately measure the economic benefits of the program.  The reason
is that nonrecurring investment costs, which DOD plans to incur to
develop the system in order to achieve a cost savings, are not
included.  The standard criterion for deciding whether a government
program can be justified on economic principles--the primary purpose
of this program--is net present value, which would include both
recurring and nonrecurring costs and the time value of money.\3 DOD
has not yet officially performed a net present value analysis and has
not identified all government costs to do so.  For example, DOD has
not identified the amount of independent research and development
costs related to the system that the contractors could charge,
through an overhead rate, to government contracts.\4 Each contractor
could invest between $800 million and $1.3 billion, a portion of
which could be reimbursed by the government.  Until the government's
total costs are determined, the net program savings will be unknown. 

The use of other transaction instruments for Evolved Expendable
Launch Vehicle development will challenge DOD in determining how best
to protect the government's interests.  The reason is that (1) in
general, other transaction instruments are not subject to federal
procurement laws or regulations and lack prescribed guidance from DOD
and (2) specifically, the use of such instruments for prototype
projects are relatively new.  The two proposed instruments (one to
each contractor), with a government cost share of $500 million each,
could be among the largest other transactions for prototype projects,
in terms of dollar value, that DOD will have negotiated.  In December
1996, the Under Secretary of Defense for Acquisition and Technology
provided some criteria for the use of such instruments for prototype
projects in what he characterized as a relatively unstructured
environment.  More recently, the DOD Inspector General (1) expressed
concern about the lack of controls over the other transaction
process, including the lack of government audit authority, and (2)
emphasized the need for DOD to issue regulations on the use of these
instruments, as required by 10 U.S.C.  2371(g).  The significance of
the proposed cost share and the lack of DOD regulations for other
transactions increase the fiduciary responsibility of DOD officials
who are authorized to negotiate and monitor such instruments. 

Risks are inherent in the program.  Under DOD's revised acquisition
approach, the contractors are not willing to guarantee system
performance because their financial risk would be open ended and
DOD's investment would be limited.  Despite this position, the Air
Force is counting on the contractors to provide launch services to
satisfy the government's requirements, based on their financial
interest in a growing commercial market for launch services.  If this
scenario is not fulfilled, the government could face some risk of not
having its launch requirements satisfied.  In addition, Air Force
planning documentation states that the primary program risk is in
meeting launch site facility preparation schedules.  Other Air Force
planning documentation shows the continued use of certain launch
facilities for several months after the facilities are scheduled to
undergo site preparations for the vehicle. 


--------------------
\2 This mission model extrapolates launches for fiscal years 2011
through 2020 based on the Air Force Space Command's national mission
model, which is a long-range, semiannual, requirements plan that
projects U.S.  space launches through 2010. 

\3 Net present value is computed by assigning monetary values to
benefits and costs, discounting future benefits and costs using an
appropriate discount rate, and subtracting the sum total of
discounted costs from the sum total of discounted benefits.  See
Office of Management and Budget Circular No.  A-94, Guidelines and
Discount Rates for Benefit-Cost Analysis of Federal Programs. 

\4 Independent research and development is defined as a
contractor-initiated, -funded, and -managed technical effort that (1)
consists of projects falling within the areas of basic and applied
research, development, and systems, and other concept formulation
studies and (2) is not sponsored by, or required in performance of, a
government contract or grant. 


   ANALYSIS OF RECURRING COSTS IS
   NOT THE PREFERRED MEASURE OF
   EELV PROGRAM SAVINGS
------------------------------------------------------------ Letter :3

DOD's goal of reducing the cost of launching satellites into space is
measured in terms of recurring production and launch costs.  However,
fluctuations in the contents of the EELV mission model make the
results of analyses, based on the model, uncertain.  More
importantly, the methodology itself is inadequate for measuring
potential program savings because it does not include the investment
costs that DOD plans to incur in EELV system development to achieve
cost savings.  A net present value (NPV) analysis, which would use
total program costs, is preferred. 


      METHODOLOGY USED FOR
      MEASURING RECURRING COST
      REDUCTION
---------------------------------------------------------- Letter :3.1

The Air Force's methodology for measuring recurring cost reduction is
described in the following way:  EELV recurring costs, meaning
production and launch costs, should be a minimum of 25 percent less,
with an objective of 50 percent less, than the recurring costs of
using existing expendable launch vehicles--the Delta, Atlas, and
Titan class systems.  To measure this goal, estimated recurring costs
for the EELV system, which are provided by the competing contractors,
are subtracted from the equivalent recurring costs for existing
vehicles, which is known as the launch cost baseline.  These costs
are based on projected government launch requirements for fiscal
years 2002 through 2020.  The launches from 2011 through 2020 are
extrapolations, therefore less certain, and are done solely for EELV
program purposes. 

To illustrate this methodology, we estimated the launch cost baseline
for existing launch vehicles to be about $15.4 billion (in fiscal
year 1995 dollars) by using a total of 164 launches through fiscal
year 2020.\5 If the minimum 25-percent cost reduction goal were
achieved, the estimated savings would be about $3.9 billion through
fiscal year 2020; if the objective 50-percent cost reduction goal
were achieved, the estimated savings would be about $7.7 billion for
the same period. 


--------------------
\5 See appendix I for how we derived at 164 launches. 


      FLUCTUATIONS IN MISSION
      MODEL CONTENTS MAKE COST
      SAVINGS UNCERTAIN
---------------------------------------------------------- Letter :3.2

Since program inception in 1995, the total number, type, and timing
of launches contained in the Air Force's EELV mission model have
fluctuated considerably, making a cost reduction estimate, based on
the model, uncertain.  The major reasons for the fluctuations were
(1) assignment of satellites to the wrong type of launch vehicle, (2)
inclusion of unverified launch requirements, and (3) reductions in
the number of heavy-lift launches because of satellite downsizing. 
The total number of launches has varied from 169 to 204, with the
current Air Force estimate at 183.  The most significant fluctuations
occurred for fiscal years 2011 through 2020. 

A credible EELV mission model is fundamental to assessing the
program's principal stated purpose--reducing recurring production and
launch costs.  Because the mission model is also provided to the
development contractors to estimate EELV costs, its accuracy is
essential for an assessment of initial launch service costs.  In
commenting on a draft of this report, DOD stated that the Air Force
is in the process of developing a new launch cost baseline, built
around the most current EELV mission model, in preparation for the
milestone II review. 

On the basis of 164 launches, we estimated that the reduction in
recurring costs through 2020 would be about $5.7 billion (in fiscal
year 1995 dollars), or 37 percent.  Although our estimate exceeds the
minimum EELV program goal of 25 percent, there is still uncertainty
regarding this estimate because of persistently questionable launch
requirements. 

Fluctuations in the number of launches can also have a significant
effect on the launch cost baseline of existing vehicles.  Heavy-lift
vehicle costs are particularly sensitive to quantity changes because
the cost to launch a Titan IV can decrease substantially as the
number of launches decreases, depending on when the launches occur. 
Although such a cost decrease initially appears counter-intuitive, it
is because of the high cost associated with operating and maintaining
Titan IV launch capabilities.  For example, an Air Force analysis
shows that the nine Titan IV launches currently in the mission model
would cost about $473 million each, or $4.3 billion, but seven
launches would cost about $395 million each, or $2.8 billion.  Thus,
two Titan IV launches could change the launch cost baseline by $1.5
billion.  The overall effect would be to lower the savings from 37
percent to 32 percent.  Given this degree of cost sensitivity, a
credible mission model is essential. 

A detailed listing of the composition and fluctuations in the EELV
mission model is shown in appendix I. 


      METHODOLOGY IS INADEQUATE
      FOR MEASURING POTENTIAL
      PROGRAM SAVINGS
---------------------------------------------------------- Letter :3.3

Although measuring a reduction in recurring costs is one method of
assessing potential program savings, this method is inadequate
because it does not include nonrecurring investment costs that DOD
plans to incur to achieve cost savings.  The standard criterion for
deciding whether a government program can be justified on economic
principles is NPV, which would include both recurring and
nonrecurring costs, as well as the time value of money.  Programs
with positive NPVs are generally preferred whereas programs with
negative NPVs should generally be avoided. 


   ALL GOVERNMENT COSTS NOT
   AVAILABLE FOR NPV ANALYSIS;
   THUS, SAVINGS ARE UNKNOWN
------------------------------------------------------------ Letter :4

Our initial NPV analysis showed that DOD would achieve a positive
return on its investment in the EELV program.  However, our analysis
does not include all government costs because the total development
costs are unknown.  DOD does not know the total costs because the
effect of reimbursing the competing contractors for their independent
research and development (IR&D) costs, as a result of using an other
transaction instrument, has not been determined.  Considering that
each contractor could invest between $800 million and $1.3 billion in
an EELV system, a portion of which could be reimbursed by the
government, the potential program savings could be substantially
lower. 


      INITIAL NPV ANALYSIS SHOWS A
      COST SAVINGS
---------------------------------------------------------- Letter :4.1

We performed an NPV analysis based on 164 launches.  We then
determined the program's net savings through 2020 using DOD's total
planned development costs of $1.4 billion, which includes $1 billion
in incremental costs starting in June 1998.\6 We also determined,
separately, the net savings of DOD's planned $1 billion incremental
investment--$500 million per contractor--to determine whether it was
economically prudent to continue with the program. 

We repeated these two approaches, based on launch projections through
2010, to eliminate the period of greater launch uncertainty that
extends from 2011 through 2020.  Both the Air Force Space Command's
national mission model and the Department of Transportation's
Commercial Space Transportation Advisory Committee's commercial
mission model only make launch projections through 2010 because of
uncertainties in making longer range forecasts.  In addition, a
shorter time period would be consistent with what an Air Force
official stated was the contractors' expectations for recouping their
investments. 

Table 1.1 shows that the NPV, using total planned development costs,
would be $1.8 billion through 2020.  Based only on the planned
incremental costs starting in June 1998, the NPV would be $2.3
billion through 2020.  The analysis of incremental costs results in a
larger NPV because, by definition, prior year costs are not included
in the cost calculation, but the benefits remain the same.  The year
in which costs equal benefits (referred to as investment payback) is
2006 and 2004 for total and incremental development costs,
respectively. 



                               Table 1.1
                
                     NPV Based on Planned Total and
                 Incremental Program Development Costs
                    (in billions of fiscal year 1995
                                dollars)


                                           Fiscal           Percent of
                                             year           discounted
                                           period     NPV    savings\a
Development costs                          ------  ------  -----------
Total                                       1995-    $1.8           19
                                             2020
costs of $1.4                               1995-    $0.7           11
                                             2010
Incremental                                 1998-    $2.3           21
                                             2020
costs of $1.0                               1998-    $1.0           14
                                             2010
----------------------------------------------------------------------
\a The percentage of savings represents the change between the
discounted launch cost baseline and the discounted EELV costs.  The
NPV analysis for incremental costs also reflects $200 million in
planned program office costs. 

Also, table 1.1 shows the NPV based on a shorter time period.  If
total planned development costs were considered, the NPV would be
$693 million through 2010.  If incremental costs only were
considered, the NPV would be $984 million through 2010.  The
investment payback for both calculations also would be 2006 and 2004,
respectively. 

Regarding DOD's $1 billion incremental investment cost, Air Force
officials informed us that they determined this amount in two ways. 
First, they estimated that government launches will represent about
one-third of the U.S.  commercial launch market and that the
investment amount should be proportionate to this market.  Therefore,
about $500 million a contractor, or one-third of about $1.5 billion
estimated per contractor to develop its version of the EELV system,
was considered reasonable.  Second, the officials stated that the
contractors advised the Air Force that about $500 million each was
needed to ensure a competitive corporate rate of return on
investment.  The officials stated that without the DOD investment,
the contractors would not develop an EELV system to meet the full
range of DOD's launch requirements or within the planned time period
to transition from existing vehicles to an EELV. 

Using NPV analysis, the net program benefits are positive when these
planned incremental costs are considered.  Such an analysis for an
EELV system should be positive, given that DOD's primary program
objective is to actively reduce costs and not simply break even on
its investment.  However, DOD does not know what its total costs will
be because the effect of reimbursing the competing contractors for
their IR&D costs, as discussed in the following section, has not been
determined.  Until the total costs are determined, the net program
savings will be unknown. 


--------------------
\6 For the EELV program, the net savings would be the difference in
the discounted costs to launch the satellites listed in the mission
model compared with the discounted costs to launch the same
satellites using existing launch vehicles. 


      CONTRACTORS' IR&D COULD
      AFFECT DOD'S EXPECTED
      SAVINGS
---------------------------------------------------------- Letter :4.2

As a matter of policy, DOD recognizes contractor costs incurred for
IR&D projects as a necessary cost of doing business and considers the
projects as a valuable contributor to DOD's overall research and
development effort.  Generally, when a contractor charges an
allowable cost to IR&D, the cost is accumulated as overhead and later
applied as an overhead rate to government contracts.  According to an
Air Force document, IR&D costs could include, under Federal
Acquisition Regulation 31.205-18(e), the costs contributed by the
contractors for work under the EELV other transactions instrument. 

Air Force and EELV contractor representatives have discussed charging
all or a portion of the contractors' share of planned EELV
development costs to IR&D.  In addition, Air Force documentation
indicated that the value of prior research and development funded
projects could be part of the contractors' IR&D efforts.  As stated
in our March 1996 report on acquiring DOD research by nontraditional
means, accepting the value of prior research in lieu of concurrent
financial or in-kind contributions may not accurately depict the
relative financial contributions of the parties.  (The scope of the
report did not include prototype projects.)\7

In May 1996, the Senate Committee on Armed Services provided
clarifying comments on the use of other transactions authority for
research (not prototype projects) under 10 U.S.C.  2371.  It stated
that

     ".  .  .  the committee intended that the sunk cost of prior
     research efforts not count as cost-share on the part of the
     private sector firms.  Only the additional resources provided by
     the private sector needed to carry out the specific project
     should be counted."\8

The amount of IR&D costs associated with the EELV program has yet to
be resolved within DOD.  According to a DOD representative, the
amount could be quite high, considering that each contractor could
invest between $800 million and $1.3 billion.  To the extent that
IR&D costs would be reimbursed by the government, the result would be
to decrease the EELV contractors' investment and reduce the
government's savings.  An Air Force document indicates that it is
important to determine the IR&D amount in order to reduce the risk of
a dispute regarding the allowance of such costs.  The usual means of
doing this under a contract is with an advance agreement. 
Determining the amount also would assist DOD in performing an NPV
analysis to estimate EELV program savings. 


--------------------
\7 DOD Research:  Acquiring Research by Nontraditional Means
(GAO/NSIAD-96-11, Mar.  29, 1996). 

\8 National Defense Authorization Act for Fiscal Year 1997, Committee
on Armed Services, U.S.  Senate Report 104-267, May 13, 1996, pp. 
313-314. 


   REVISED ACQUISITION APPROACH
   CONTAINS CHALLENGES, RISKS, AND
   BENEFITS
------------------------------------------------------------ Letter :5

The use of a relatively new acquisition method, called other
transactions, will challenge DOD in determining how best to protect
the government's interests.  Also, risks are inherent in the program
because of (1) DOD's plan to limit its investment and the
contractors' resulting unwillingness to guarantee a system to meet
the government's launch requirements and (2) a chance that certain
launch facilities may not be available as currently scheduled. 
However, to the extent that the risks can be mitigated, the primary
program benefit is expected to be reduced costs to the government. 


      OTHER TRANSACTION
      INSTRUMENTS WILL CHALLENGE
      DOD TO PROTECT GOVERNMENT'S
      INTERESTS
---------------------------------------------------------- Letter :5.1

Initially, under DOD's revised acquisition approach, the Air Force
planned to award firm-fixed-price contracts to both EELV contractors
for the development effort.  However, after the Air Force released a
draft request for proposal in late November 1997, EELV program
officials stated that both contractors were unwilling to accept
firm-fixed-price contracts.\9 According to these officials, the
contractors' unwillingness was because of the resulting risk to
corporate financing--a situation whereby the contractors' long-term
contractual liability would require committing their share of EELV
development costs in advance. 

As a result, the Air Force is proposing to use other transaction
instruments, instead of standard government contracts, to develop the
EELV system.  The specific other transactions authority cited by the
Air Force is section 845 of the National Defense Authorization Act
for Fiscal Year 1994 (P.L.  103-160, Nov.  30, 1993), as modified by
section 804 of the National Defense Authorization Act for Fiscal Year
1997 (P.L.  104-201,
Sept.  23, 1996).  These sections provide DOD with authority, under
10 U.S.C.  2371, to carry out prototype projects that are directly
relevant to weapons or weapon systems proposed to be acquired or
developed by DOD.  The authority, however, is very broad because it
includes not only prototype systems but also lesser projects such as
subsystems, components, and technologies.  Also, the authority is
temporary, expiring on September 30, 1999. 

In December 1996, the Under Secretary of Defense for Acquisition and
Technology notified the secretaries of the military departments and
the directors of defense agencies about the use of other transaction
instruments for prototype projects.  He mentioned the flexibility
associated with using such instruments as alternatives to contracts,
listing 19 statutes that apply to contracts, but which are not
necessarily applicable to other transactions.  He emphasized that the
use of such instruments should incorporate good business sense and
appropriate safeguards to protect the government's interest,
including assurances that the cost to the government is reasonable,
the schedule and other requirements are enforceable, and the payment
arrangements promote on-time performance.  He also emphasized that
DOD officials who are delegated the authority to use such instruments
should have the level of responsibility, business acumen, and
judgment to enable them to operate in this relatively unstructured
environment. 

In a March 1997 report, the DOD Inspector General's office identified
problem areas in awarding and administering other transactions.  The
office reviewed 28 randomly selected other transactions valued at
$1.2 billion that were issued by the Defense Advanced Research
Projects Agency--4 were section 845/804 prototype projects and 24
were for research.  In general, the report stated that no guidance
existed for (1) evaluating proposed contributions, (2) monitoring
actual research costs, or (3) including an interest provision in
other transaction instruments.\10

In March 1998, the Inspector General testified about a continuing
concern regarding the lack of controls over the other transaction
process since normal rules and procedures generally do not apply.\11
The Inspector General emphasized that although 10 U.S.C.  2371
requires the Secretary of Defense to issue regulations on other
transactions, none have been published.  On the basis of the 1997
report, the Inspector General stated that there is a need to (1)
ensure that cost-sharing arrangements are honored, (2) monitor the
actual cost of work against the funds paid, (3) place funds advanced
to recipients into an interest bearing account until used, and (4)
standardize the audit clause.  She also testified that a more current
review of 78 other transactions had found problems similar to those
in the 1997 report. 

With regard to an audit clause, the Under Secretary of Defense for
Acquisition and Technology identified 10 U.S.C.  2313 in his December
1996 memorandum as being an inapplicable statute for other
transactions.  This statute provides audit authority to a defense
agency awarding certain types of contracts and to the Comptroller
General for defense contracts awarded other than through sealed bid
procedures.  Safeguards, such as government audit authority, that are
common to government contracting would not be available under other
transaction instruments unless such authority was negotiated as part
of the instrument.  An official of the Inspector General's Office of
General Counsel emphasized the importance for the government to be
able to verify and audit certain aspects of other transactions.  He
stated that a prudent business practice would provide for audits to
verify contribution valuation, cost share, performance milestones,
and final costs. 

In commenting on our draft report, DOD stated that because (1) the
government is providing funding to private contractors to develop a
commercial item and (2) the government's funding is significantly
less than the contractors' funding, the contractors do not intend to
provide, and the government does not expect to get, visibility into
corporate investment and financing.  DOD stated that this unique
situation is not reasonably subject to audit requirements that
generally apply to contracts.  DOD, instead, emphasized the
importance of government insight into the contractors' development
efforts, stating that a methodology will be established to audit the
accomplishment of milestones prior to disbursing funds. 

The amount of government funds planned to be used to develop the EELV
system through other transaction instruments raises a question of
materiality.  There are indications that most of DOD's other
transactions for prototype projects, historically, have been
relatively small in dollar value.  For example, the Inspector General
testified that for fiscal years 1990 through 1997, she believed that
59 other transaction agreements for prototype projects were valued at
$837 million.  Although no cost-sharing breakout between the
government and the recipient was provided, the average value per
agreement was about $14 million.  In a DOD report on cooperative
agreements and other transactions entered into during fiscal year
1997 that was submitted to the congressional defense authorizing
committees, we noted that of 50 other transactions for prototype
projects, the government's contribution on the largest 1 was $60
million.  These data contrast sharply with DOD's intentions to
negotiate two EELV other transaction instruments with a government
contribution of $500 million each.  The significance of these
proposed amounts and the lack of DOD regulations for other
transactions not only increase the fiduciary responsibility of DOD
officials who are authorized to negotiate such instruments but also
may necessitate that some degree of government audit authority be
established. 


--------------------
\9 Firm-fixed-price contracts place maximum risk and full
responsibility on the contractor for all costs and resulting profit
or loss. 

\10 Award and Administration of Contracts, Grants, and Other
Transactions Issued by the Defense Advanced Research Projects Agency
(DOD Inspector General Report No.  97-114, Mar.  28, 1997). 

\11 Statement of Eleanor Hill, DOD Inspector General, before the
Subcommittee on Acquisition and Technology, Committee on Armed
Services, U.S.  Senate, Mar.  18, 1998. 


      PROGRAM RISKS ARE INHERENT
---------------------------------------------------------- Letter :5.2

According to Air Force documents, the two contractors are not willing
to guarantee system performance under a firm- fixed-price contract or
an other transaction instrument for EELV development.  This
unwillingness is because DOD's financial risk is to be capped at $500
million per contractor, while the contractors' financial risk would
be an open-ended commitment.  As a result, the contractors would only
agree to provide a "best effort" in developing the EELV system,
meaning that they would not guarantee a launch vehicle capability to
meet the government's requirements. 

One DOD representative indicated a possible inconsistency between
such a system development agreement and the expectation that the
contractors would subsequently deliver fully functional launch
services.  Such an inconsistency could create a risk to the
government of not satisfying its launch requirements.  However, the
Air Force is relying on the contractors being motivated by a
compelling financial interest in an expected lucrative international
commercial launch services market.  Also, the Air Force intends to
negotiate performance-based milestones that represent significant
activities under the development effort and to pay the contractors
based on completing each milestone.  In the case of nonperformance,
the Air Force should withhold payment because no payment would be
earned. 

In our June 1997 report on the EELV program, we identified three
factors that could create a risk in achieving a smooth launch
facility transition at the Cape Canaveral and Vandenberg launch
ranges in Florida and California, respectively.\12 They were (1)
conflicts associated with existing facilities that the contractors
expected to use or that would be affected by an EELV system, (2)
completion of environmental regulatory requirements before funds can
be committed to engineering and manufacturing development, and (3)
the amount of time needed for facility modification and new
construction.  We did not reassess these factors for this report;
however, current Air Force planning documentation identifies meeting
launch site facility preparation schedules as the primary program
risk.  The reason is that construction must begin shortly after the
milestone II decision in June 1998 to support the first EELV launch
in fiscal year 2002.  Other Air Force planning documents show the
continued use of certain launch facilities for several months after
they are scheduled to undergo site preparations for EELV.  In
commenting on our draft report, DOD cited a Titan IV launch complex
as an example. 

We also reported on vehicle propulsion, systems integration, and
software as technical risk factors that could adversely affect
program cost and schedule goals.  Current Air Force documentation
also identifies these three factors as risks common to both
contractors and indicates that mitigation efforts are underway. 


--------------------
\12 Access to Space:  Issues Associated With DOD's Evolved Expendable
Launch Vehicle Program (GAO/NSIAD-97-130, June 24, 1997). 


      PROGRAM BENEFITS ARE
      EXPECTED
---------------------------------------------------------- Letter :5.3

The primary benefits to the EELV program are expected to culminate in
lower costs, whether they are measured in terms of recurring
production and launch costs or NPV.  Before revising its acquisition
approach, DOD was planning on a natural synergy between the federal
government and the commercial space industry because of a common
requirement for space launch.  In our June 1997 EELV report, we
discussed DOD's interest in seeing the EELV used for commercial
purposes in order to expand the customer base and help lower costs. 
At that time, DOD was planning to pay for all development
costs--about $1.5 billion--but the contractors indicated a
willingness to invest in EELV development.  We recommended that the
Secretary of Defense devise a cost-sharing mechanism for EELV
development to help reduce the government's investment, particularly
in view of the expected compensating benefits to the winning
contractor to enhance its competitive position in the international
commercial launch services market. 

In July 1997, the House Committee on Appropriations noted that while
partners share benefits, they also share costs, and it suggested that
the Air Force aggressively pursue commercial cost sharing.\13 In
August 1997, DOD responded to our report by agreeing with the
recommendation and stating that the cost-sharing issue would be
reviewed as the acquisition strategy was developed over the next 12
months.  In September 1997, the Conference Committee on the fiscal
year 1998 DOD appropriations bill suggested that the Air Force
require a successful bidder to share in the EELV development cost.\14
In November 1997, when DOD approved the Air Force's proposal to
revise the acquisition strategy, contractor cost sharing was one of
the requirements. 

With the Air Force's proposal for the government's share to not
exceed
$1 billion for the two contractors, about $500 million in DOD
development costs were expected to be avoided, based on the original
$1.5 billion estimate.  However, this cost avoidance will be reduced
by the need to acquire two additional launches with procurement funds
under the initial launch services contracts.  The Air Force had
originally planned to acquire these two launches for test purposes
using development funds.  Thus, the net cost avoidance is expected to
be about $295 million, with the remaining $205 million to be shifted
to a procurement account. 

In our March 1996 report on DOD research by nontraditional means, we
discussed the importance of leveraging the private sector's financial
investment by using other transactions and cooperative agreements. 
In doing so, DOD can first stretch its research and development funds
by having commercial firms contribute to the cost of developing
technologies with both military and commercial applications.  Second,
cost sharing is appropriate and a matter of fairness when commercial
firms expect to benefit financially from sales of the technology. 
Third, a cost-sharing arrangement demonstrates a commitment to the
project, enabling less rigid government oversight requirements. 
These three elements appear to exist in the case of the revised EELV
acquisition approach. 

Air Force officials emphasized that more recent information regarding
the projected growth in the commercial launch services market,
primarily based on the expected growth in commercial communication
satellites, was a key factor in revising the EELV acquisition
approach.  The recent projection contrasts sharply from DOD's 1994
space launch modernization plan whereby the commercial market was not
considered to be nearly as promising.  As a result, the Air Force
concluded that this growing market was sufficient to support two EELV
contractors, instead of one.  Two contractors would ensure more
effective competition for future government launch requirements and
would result in a change from cost-based contracting to price-based
contracting, using the commercial market for launch services. 

In its November 1994 implementation plan for national space
transportation policy, DOD envisioned that the EELV system would (1)
maximize common systems and components to reduce procurement costs
and enhance production rates and (2) decrease the number of launch
complexes, launch crews, and support requirements to reduce
operations costs.  Although the gains envisioned may not be as large
because two contractors are to be supported, the Air Force is still
expecting standardization--launch pads configured for all EELV sizes
(medium-lift, intermediate-lift, and heavy-lift) and standard
payload-to-vehicle interfaces--that should help reduce overall costs
and achieve more efficient launch operations than with existing
vehicles.  In addition, the availability of two launch vehicle
manufacturers that use standard payload interfaces would better
ensure that government satellites are launched if one contractor's
fleet of vehicles were grounded. 


--------------------
\13 DOD Appropriations Bill, 1998, Report of the Committee on
Appropriations, House of Representatives Report 105-206, July 25,
1997, p.  202. 

\14 Making Appropriations for the Department of Defense for the
Fiscal Year Ending September 30, 1998, and for Other Purposes,
Committee of Conference, House of Representatives Report 105-265,
Sept.  23, 1997, pp.  128-129. 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

DOD's revised EELV acquisition approach represents a significant
departure from the standard government procurement approach.  The
revision was brought about primarily because commercial interests are
expected to dominate the worldwide space launch service market.  When
making its investment decision in the EELV system, DOD should apply a
market-oriented approach, using NPV analysis, to ensure that expected
savings are suitable, including consideration for unforeseen future
costs.  This approach would help protect the government's interests
and be consistent with the EELV program goal of reducing the cost of
launching satellites into space. 

The means by which DOD intends to negotiate an agreement with the
competing contractors--other transaction instruments--calls for
specific guidance to govern the EELV development effort.  Such
guidance is particularly important considering the general lack of
DOD regulations on the use of such instruments.  It is also important
considering the high-dollar EELV development program that is to be
executed in what is characterized as a relatively unstructured
environment. 

Assuming that the challenge in using other transaction instruments
can be met and program risks can be overcome, the primary benefits
associated with the EELV system should be reduced costs to the
government.  Reduced costs would include lower (1) short-term
nonrecurring costs by forming a cost-sharing partnership with space
industry contractors to develop a product that has mutual benefits
for the government and commercial space launch sectors and (2)
long-term recurring costs by designing a family of common launch
vehicles, standardizing launch facilities and payload interfaces, and
establishing price-based competition between two contractors for
future launch services. 


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

To protect the government's interest, and to be consistent with
entering a business partnership with launch industry contractors for
EELV development, we recommend that the Secretary of Defense take
steps to ensure that an NPV analysis of the program is performed
before making a milestone II decision.  The analysis should include
(1) DOD's total planned incremental investment costs for development,
(2) the most current EELV costs from the contractors' proposals and
DOD's estimate for launch services, and (3) a time period for which
launch requirements can be verified and reasonably forecasted.  The
Secretary should (1) establish criteria for judging the results of
the analysis that would provide a suitable margin for discounted
savings and unforeseen future costs and (2) determine the amount of
IR&D costs that need to be factored into the analysis.  If the
results of the NPV analysis do not meet the criteria, we recommend
that the Secretary review the program to either (1) reduce the amount
of the government's planned incremental investment or (2) rejustify
the program on a basis other than cost reduction. 

Because DOD has not prescribed regulations for other transactions, as
required under 10 U.S.C.  2371(g), we recommend that the Secretary
review the Air Force's planned use of other transaction instruments
for EELV development to ensure that the government's interest is
protected.  Consideration should be given to (1) the criteria
expressed by the former Under Secretary of Defense for Acquisition
and Technology and (2) the DOD Inspector General's concerns regarding
the other transactions process, including some degree of government
audit authority. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :8

DOD agreed with our recommendation to perform a NPV analysis.  DOD
stated that such an analysis (1) was a more appropriate affordability
measure for determining EELV program viability than the financial
analysis performed to date and (2) would be presented during the
milestone II decision process.  DOD did not specify how the analysis
would be used to support the decision.  Our intent was to emphasize
the importance of using such an analysis as a rigorous means of
measuring economic benefits to the government, considering the unique
business arrangement DOD is planning with launch industry
contractors. 

DOD also agreed with our recommendation concerning protection of the
government's interest in the use of other transaction instruments for
EELV development.  DOD stated that adequate visibility into the
contractors' progress would be obtained by a clause in the
development agreements to provide insight into technical and schedule
performance--for example, to verify the accomplishment of milestones
prior to payment.  Regarding the issue of government audit authority,
or oversight, DOD differentiated between (1) other transactions for
research projects that have a statutory requirement for cost sharing
by the recipients to the extent the Secretary of Defense determines
practicable and (2) other transactions for prototype projects, that
have no such statutory requirement, thus leaving the determination of
a fair and reasonable amount of government development funding for
the EELV program up to the contracting officer. 

Collectively, these statements imply that some degree of government
audit authority may not be needed for the EELV program.  Given that
such matters are negotiable, our intent was to stress the importance
of the Secretary of Defense giving due consideration to some degree
of government audit authority because of the (1) significant amount
of government development funds planned to be used for EELV and (2)
lack of DOD regulations on the use of other transactions for either
prototype projects or research. 

DOD's comments on a draft of this report are reprinted in their
entirety in appendix II.  DOD also provided clarifying comments,
which we have incorporated, as appropriate. 


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

To evaluate the Air Force's plans and progress in developing the EELV
system, we examined acquisition planning documents, budget
information, cost assessment methodologies, launch requirements, and
information related to other transaction authority and guidelines. 

We performed our work primarily at the Air Force Space and Missile
Systems Center in El Segundo, California.  We held discussions with
representatives of the Office of the Secretary of Defense, the
Department of the Air Force, the National Aeronautics and Space
Administration (NASA), the Federal Aviation Administration,
Washington, D.C., and the Air Force Space Command, Colorado Springs,
Colorado.  We acquired limited launch requirement information from
the National Reconnaissance Office, Chantilly, Virginia.  In
addition, we held discussions with private industry representatives
from Lockheed Martin Telecommunications, Sunnyvale, California, and
Space Systems/Loral, Palo Alto, California; The Boeing Company,
Huntington Beach, California; Hughes Space and Communications
International, Inc., Los Angeles, California; and TRW Space and
Electronic Group, Redondo Beach, California. 

Because we noted considerable fluctuations in the contents of the Air
Force's EELV mission model during the past 2 years, we adjusted the
latest mission model data based on discussions with Air Force
satellite program office representatives and NASA representatives and
a review of satellite program documentation.  Specifically, we
excluded 19 NASA and classified launches because they were not fully
justified.  We used the adjusted mission model data to analyze
recurring costs and to perform an NPV analysis. 

In performing our recurring cost analysis, we obtained current
production and launch costs for Delta, Atlas, and Titan launch
vehicles from the respective launch program offices.  We obtained
EELV production and launch costs from the EELV program office, which
were based on contractors' proposals and the Air Force's evaluation
during selection of the two contractors in 1996.  (The Air Force is
currently revising EELV cost estimates in preparation for the
milestone II decision in June 1998.)

In performing our NPV analysis, we used our adjusted mission model
data and the data we obtained for our recurring cost analysis.  In
addition, we obtained DOD's planned investment costs based on a
combination of congressional appropriations and funds programmed by
the Air Force for EELV development.  We used the real discount rate
of 3.7 percent, adjusted for forecasted inflation, based on
marketable Treasury debt with maturity comparable to that of the EELV
program. 

We performed our review between August 1997 and April 1998 in
accordance with generally accepted government auditing standards. 


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

We are sending copies of this report to the Ranking Minority Member,
Subcommittee on National Security, House Committee on Appropriations;
the Chairmen and Ranking Minority Members of the House Committee on
National Security; and the Chairmen of the Senate Committee on Armed
Services and the Subcommittee on Defense, Senate Committee on
Appropriations.  We are also sending copies to the Secretary of
Defense and the Director, Office of Management and Budget.  We will
make copies available to others upon request. 

If you or your staff have any questions concerning this report,
please call me on (202) 512-4841.  Major contributors to this report
are listed in appendix III. 

Sincerely yours,

Louis J.  Rodrigues
Director, Defense Acquisitions Issues


COMPOSITION AND FLUCTUATIONS IN
EELV MISSION MODEL
=========================================================== Appendix I

Since program inception in 1995, the total number of launches
contained in the Air Force's Evolved Expendable Launch Vehicle (EELV)
mission model has fluctuated from 171 to 194 to 204 to 169 to 183. 
The types of launch vehicles--medium-lift, intermediate-lift, and
heavy-lift--and the timing of launches have also varied.  The
composition of, and fluctuations within, the model, including our
adjusted model are shown in table I.1. 



                                    Table I.1
                     
                       Air Force EELV Mission Model and Our
                                  Adjusted Model

                                                                             Our
                                      Sept                                adjust
Type of launch  Satellite       Dec.     .  July     Mar. 9,    Mar. 24,      ed
vehicle         system          1995  1996  1997        1998        1998   model
--------------  --------------  ----  ----  ----  ----------  ----------  ------
Medium          Space test         7     6     6           5           5       5
                 satellites
                GPS               55    53    53          40          56      56
                DMSP               4     4     5           4           4       4
                NPOESS             9     9     8           6           5       5
                NASA               6     7     3           2           2       0
                 satellites
                NASA-              9     9     5           5           5       0
                 Discovery
                SBIRS-LEO          0    28     0           9           9       9
================================================================================
Subtotal                          90   116    80          71          86      79
Intermediate    DSCS               2     2     2           2           2       2
                SBIRS-GEO         16    15    14          10          10      10
                SBIRS-LEO          0     0    29           0           0       0
                Advanced          21    21    20          21          20      20
                 MILSATCOM
                Mission A/B/E     25    25    50          56          56      44
================================================================================
Subtotal                          64    63   115          89          88      76
Heavy           Mission C/D       17    15     8           8           8       8
                DSP                0     0     1           1           1       1
================================================================================
Subtotal                          17    15     9           9           9       9
================================================================================
Total                            171   194   204         169         183     164
--------------------------------------------------------------------------------
LEGEND

DMSP = Defense Meteorological Satellite Program
DSCS = Defense Satellite Communications System
DSP = Defense Support Program
GPS = Global Positioning System
MILSATCOM = Military Satellite Communications
Mission A/B/C/D/E = Classified programs
NPOESS = National Polar-orbiting Operational Environmental Satellite
System
SBIRS-GEO = Space-Based Infrared System-Geosynchronous Earth Orbit
SBIRS-LEO = Space-Based Infrared System-Low Earth Orbit


   TYPES OF VEHICLES
--------------------------------------------------------- Appendix I:1

The number of medium-lift vehicles has fluctuated from 90 to 116 to
80 to 71 to 86.  The major reasons were (1) incorrect assignment of
29 Space-Based Infrared System (SBIRS)-Low satellites for launch on
intermediate-lift vehicles in the July 1997 model, rather than
medium-lift vehicles; (2) a decision that after 2010, SBIRS-Low
satellites would be launched on an existing commercial launch vehicle
system, called Athena, which is smaller than a medium-lift EELV, and
(3) the omission of 16 Global Positioning System satellites from the
March 9, 1998, model. 

The number of intermediate-lift vehicles has also fluctuated, from 63
to 115 to 89.  The major reasons were (1) the incorrect assignment of
29 SBIRS-Low satellites for launch on intermediate-lift vehicles
rather than medium-lift vehicles and (2) adding 31 classified
satellites, of which 12 were not included in a launch summary
document and were considered unverified requirements, according to
Air Force Space Command representatives. 

The number of heavy-lift vehicles has decreased almost 50 percent,
from 17 to 9.  The major reason was because of downsizing the number
of satellites.  This downsizing was stimulated by the high cost of
launching heavy payloads on the Titan IV launch vehicle. 


   OUR ADJUSTED MISSION MODEL
--------------------------------------------------------- Appendix I:2

On the basis of our analysis, we identified 164 satellite launches
from 2002 through 2020.  We determined these launches through
discussions with Air Force satellite and launch vehicle program
office representatives and National Aeronautics and Space
Administration (NASA) representatives and from satellite program
documentation. 

Compared with the Air Force's March 24, 1998, EELV mission model, our
adjusted model excluded seven NASA launches because NASA plans to
downsize the satellites associated with these seven launches and use
vehicles that are smaller than the EELV system.  Our adjusted model
also excluded 12 classified launches because they were considered to
be optional; were not listed as launch requirements in a February
1998 launch summary; and according to Air Force Space Command
representatives, were not based on validated requirements. 




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



(See figure in printed edition.)



(See figure in printed edition.)


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix III

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

Homer H.  Thomson
James A.  Elgas

LOS ANGELES OFFICE

Larry J.  Bridges
Steve Martinez
James D.  Moses
Allan Roberts
Allen D.  Westheimer


*** End of document. ***