Missile Defense: Status of the National Missile Defense Program (Letter
Report, 05/31/2000, GAO/NSIAD-00-131).

Pursuant to a congressional request, GAO reviewed the National Missile
Defense program, focusing on: (1) programmatic changes to the program
and whether significant performance and schedule risks remain; and (2)
program officials' efforts to control costs, program estimate, and the
potential for cost increases.

GAO noted that: (1) to reduce risks, the Department of Defense (DOD) has
delayed initial fielding of the National Missile Defense system from
fiscal year 2003 to 2005, delayed program decisions on the production of
radars and interceptors, and increased funding for testing; (2) even
with these programmatic changes, significant performance and schedule
risks remain for several reasons; (3) developing a highly reliable
hit-to-kill capability is a difficult technical challenge; (4) in
various missile defense programs, this capability has been demonstrated
in about 30 percent of the attempted intercepts outside the atmosphere;
(5) only 3 of the 19 planned intercept attempts are scheduled prior to
the July 2000 deployment readiness review; (6) none of these attempts
will expose the interceptor kill vehicle to the higher acceleration and
vibration loads of the much faster, actual system booster; (7) because
the program has a very aggressive schedule, it is vulnerable to delays;
(8) compared to the 15 years estimated to develop and field another
missile defense system called the Theater High Altitude Area Defense
system, the schedule to develop and field the National Missile Defense
system covers only 8 years; (9) DOD told Congress last October that the
development and deployment schedule was compressed by at least 4 years
because of the seriousness of the emerging missile threat from rogue
nations; (10) to control costs, DOD has implemented several measures,
including a process called "cost as an independent variable," which sets
realistic cost objectives and trades off the system's performance and
schedule to control costs; (11) officials estimate that the savings from
this process include $4.3 billion from early trade studies conducted by
the prime contractor on alternative mixes of radars and interceptor
locations; (12) after incorporating the expected cost savings from these
measures, the National Missile Defense program office estimated that the
approved program would cost $36.2 billion over the life of the
Capability I system or about $7.5 billion more than its 1999 estimate;
(13) the cost increased mostly because of the decision to increase the
number of interceptors and add flight tests, ground-test equipment, and
a more capable radar facility; and (14) using current spending rates the
program office estimates that each month of delay would increase program
costs by $124 million.

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

 REPORTNUM:  NSIAD-00-131
     TITLE:  Missile Defense: Status of the National Missile Defense
	     Program
      DATE:  05/31/2000
   SUBJECT:  Ballistic missiles
	     Weapons systems
	     Defense cost control
	     Future budget projections
	     Air defense systems
	     Defense capabilities
	     Schedule slippages
	     Operational testing
IDENTIFIER:  DOD National Missile Defense System
	     Patriot Missile Advanced Capability-Three Upgrade
	     DOD National Missile Defense Program
	     SDI Theater High Altitude Area Defense System

******************************************************************
** 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-00-131

National Security and
International Affairs Division

B-283281

May 31, 2000

The Honorable Daniel K. Akaka
Ranking Member, Subcommittee on International Security,
Proliferation, and Federal Services
Committee on Governmental Affairs
United States Senate

Dear Senator Akaka:

The Department of Defense is developing a National Missile Defense system to
protect the United States against a ballistic missile attack with weapons of
mass destruction from "rogue" nations such as North Korea and Iran.
Following a departmental review by the end of July 2000, the administration
plans to decide on whether to deploy the system. Factors in this decision
are likely to include the severity of the threat, the maturity of the
technology involved, and affordability. The deployment decision is among the
most important defense issues facing the nation this year.1

The National Missile Defense system, when fully deployed, would include (1)
space- and ground-based sensors to provide early warning of attacking
missiles and to initially identify and track them; (2) ground-based radars
to further identify and track the threatening warheads and assess whether
the system destroyed the warheads; (3) ground-based interceptors, each
consisting of a three-stage booster and payload (called a kill vehicle)
capable of guiding itself to collide with and destroy incoming warheads
(a concept called hit-to-kill) outside the atmosphere; and (4) a battle
management, command, control, and communications system. (See fig. 1.)

Source: Ballistic Missile Defense Organization.

Three levels of system capability are being planned. The Capability I
system, which is approved for development by the Department of Defense, is
designed to address a threat involving a few enemy missiles and simple
countermeasures.2 Capability II and III systems, which have not yet been
approved for development, would evolve over time through the incorporation
of additional interceptors at multiple defense sites, more sensors, and more
advanced technologies to defend against more missiles with increasingly
sophisticated countermeasures.

In June 1998, we reported on the program's funding requirements and schedule
and technical risks. We found that even with increased funding for risk
mitigation, technical and schedule risks were high.3

This report responds to your request for assistance in monitoring changes in
the National Missile Defense program since our June 1998 report.
Specifically, you asked us to (1) identify programmatic changes and
determine whether significant performance and schedule risks remain and (2)
identify program officials' efforts to control costs, the current program
cost estimate, and the potential for cost increases. We also plan to report
within the next month on the status of two other missile defense systems
that are under development--the Navy Theater Wide and Patriot Advanced
Capability-3 systems.4

To reduce risks, the Department of Defense has, since our June 1998 report,
delayed initial fielding of the National Missile Defense system from fiscal
year 2003 to 2005, delayed program decisions on the production of radars and
interceptors, and increased funding for testing. Even with these
programmatic changes, significant performance and schedule risks remain for
several reasons.

ï¿½ Developing a highly reliable hit-to-kill capability is a difficult
technical challenge. In various missile defense programs, this capability
has been demonstrated in about 30 percent of the attempted intercepts
outside the atmosphere, and a panel of experts convened by the Department
noted in 19985 that the difficulty of performing the hit-to-kill capability
very reliably had been underestimated.

ï¿½ Only 3 of the 19 planned intercept attempts are scheduled prior to the
July 2000 deployment review. None of these attempts will expose the
interceptor kill vehicle to the higher acceleration and vibration loads of
the much faster, actual system booster.

ï¿½ Because the program has a very aggressive schedule, it is vulnerable to
delays. Compared to the 15 years currently estimated to develop and field
another missile defense system called the Theater High Altitude Area Defense
system, the schedule to develop and field the National Missile Defense
system covers only 8 years. The Department of Defense told the Congress last
October that the development and deployment schedule was compressed by at
least 4 years because of the seriousness of the emerging missile threat from
rogue nations.

To control costs, the Department of Defense has implemented several
measures, including a process called "cost as an independent variable,"
which sets realistic cost objectives and trades off the system's performance
and schedule to control costs. Departmental officials estimate that the
savings from this process include $4.3 billion from early trade studies
conducted by the prime contractor on alternative mixes of radars and
interceptor locations. After incorporating the expected cost savings from
these measures, the National Missile Defense program office estimated that
the approved program would cost $36.2 billion over the life of the
Capability I system or about $7.5 billion more than its 1999 estimate.6 The
cost increased mostly because of the decision to increase the number of
interceptors and add flight tests, ground-test equipment, and a more capable
radar facility. Estimated costs are likely to increase further because of
the vulnerability to schedule delays. Using current spending rates the
program office estimates that each month of delay would increase program
costs by $124 million. The Department has not prepared official cost
estimates for the Capability II and III systems. However, a decision to
expand the program by funding a Capability II or III system would increase
program costs by billions of dollars because it would involve more
interceptors and more and better ground- and space-based elements.

In commenting on a draft of this report, the Department of Defense concurred
with the information presented. (Written comments are provided in appendix
I.)

Schedule Risks Remain

To reduce program risks, the Department of Defense (DOD) restructured the
National Missile Defense (NMD) program in 1999 by lengthening the time
available for system development and adding independent reviews. However,
performance and schedule risks remain significant because of the technical
challenge, test limitations, and the ambitious schedule.

In April 1996, DOD changed the purpose of the NMD program from a
"technology" readiness program intended to develop and mature technologies
for possible use in a NMD system to a "deployment" readiness program. The
deployment readiness program was designed to develop and demonstrate, by
fiscal year 2000, an initial system with 20 interceptors that could be
deployed by fiscal year 2003. In February 1998, a panel of high-level
military and civilian experts issued its report on risks associated with
ballistic missile defense flight-test programs. The expert panel reported
that the strategy of accepting a high level of risk to shorten the
deployment schedule is more likely to cause program slips, higher costs, and
even program failure. It found that the NMD program would benefit from
immediate restructuring to reduce program risks.

In January 1999, the Secretary of Defense restructured the NMD program. The
Secretary established the objective of completing system development and
fielding the 20 interceptors by the end of fiscal year 2005 (or about
2 years later than previously planned), if so directed by the President. In
addition, the Secretary established a DOD-level review--called the
deployment readiness review--to assess the system and to recommend to the
President whether it should be deployed. This review has been rescheduled
from June 2000 to July 2000. Assuming an affirmative decision to deploy, DOD
plans to decide on production of the NMD site radar and upgrading existing
early warning radars in the second quarter of 2001 and production of the
interceptors and associated equipment in the first quarter of 2003. If the
administration decides that the system does not warrant deployment following
the readiness review, DOD plans to continue technology development.

Following the Secretary's actions, the expert panel was reconvened to assess
the impact of the program's restructuring. In November 1999, the panel
reported that while the restructuring reduced risk by providing more time
for system development, program risks remain high.7 To address these risks,
it recommended expansion of the ground-testing capability and provision of
additional kill vehicle components for testing and flight-test backups.

In the President's budget submission for fiscal year 2001, DOD has requested
funding to address the expert panel's concerns. Specifically, it has
requested funding for additional ground and flight testing following the
deployment readiness review. In a separate action unrelated to the expert
panel's report, DOD expanded the Capability I system to meet a larger threat
by planning to deploy 80 additional interceptors by 2007.

Even with DOD's risk reduction actions, the NMD program (1) remains a
technically challenging program, (2) will undergo only a few of the planned
flight tests before the deployment readiness review, (3) has increased
performance risk because of flight-test restrictions, and (4) continues to
have a very aggressive schedule.

Hit-to-kill Capability Presents a Significant Technical Challenge

The challenge presented by the requirement for a highly reliable, very
effective hit-to-kill capability is a difficult one. The NMD interceptor
must hit an incoming target to destroy it. In addition, because the
consequences of an attack can be so catastrophic, the NMD system is required
to defend, with a very high probability of success, all 50 states against
the threat of weapons of mass destruction carried aboard long-range
ballistic missiles. Since 1983, various missile defense programs have
attempted intercepts outside the atmosphere where the NMD system would be
required to engage missiles. These missile defense programs have
successfully demonstrated the hit-to-kill capability in 4 of 14 intercept
attempts--about 30 percent. While the 4 successful intercepts provide
support for the hit-to-kill concept, the 10 failed attempts raise questions
about reliability. In its November 1999 report, the expert panel concluded
that the NMD program office and contractor have continued to underestimate
the challenge of reliably performing hit-to-kill intercepts.

Only a Few of the Planned Flight Tests Are Scheduled Before the Deployment
Readiness Review

As more testing is conducted, programmatic decisions tend to become less
risky. At most, only 3 of the 19 planned intercept attempts will be
completed prior to the July 2000 deployment readiness review. These three
tests are described below.

ï¿½ Ιn October 1999, the first intercept attempt successfully hit the
target. This flight test demonstrated the kill vehicle's ability to locate
the target, perform guidance maneuvers, and collide with the target. The
October test was not designed to rely on actual NMD system elements such as
sensors and a battle management, command, and control system to guide the
interceptor into relatively close proximity to the target before the kill
vehicle located it. Instead, data from test range instruments and the target
directed the kill vehicle until it was close enough for the kill vehicle to
autonomously locate the target and collide with it.

ï¿½ In January 2000, the second attempt did rely on actual ground- and
space-based sensors and a battle management command and control system for
interceptor guidance. Although these system elements reportedly worked well,
this test failed to intercept its target because the kill vehicle device
responsible for locating the target malfunctioned just seconds before the
target intercept should have occurred.

ï¿½ The third intercept attempt is scheduled for early July 2000. This will be
the first intercept attempt with all system elements, except the actual
booster, integrated. Since this test occurs less than 1 month prior to the
July 2000 readiness review, DOD's Director, Operational Test and Evaluation,
is concerned about the limited time available to fully assess test results.
In a February 2000 memorandum to DOD's Under Secretary of Defense for
Acquisition, Technology and Logistics, the director stated that previous NMD
flight-test reports have taken from 10 to 18 weeks to complete. According to
the NMD program office's Deputy Director, System Test and Evaluation, all
that will be known by the deployment readiness review from the early July
flight test is whether the intercept occurred and whether the system
elements worked as expected. Little time will be available to analyze any
causes for a less-than-expected level of system performance.

The two intercept attempts to date did not--and plans for the third
intercept attempt do not--include the system's actual booster. Instead, the
tests used or will use a substitute booster, called the payload launch
vehicle, to carry the kill vehicle outside the atmosphere and into
relatively close proximity to the target. The actual three-stage booster
reaches a much greater velocity than the two-stage payload launch vehicle.
Hence, the actual booster places much higher acceleration and vibration
loads on the kill vehicle. In its November 1999 report, the expert panel
viewed the kill vehicle's unproven ability to withstand loads from the
actual system booster as a high risk to the program. According to the
booster product manager, recent ground testing indicates that the kill
vehicle can withstand the loads of the actual booster, but according to the
manager, the analysis will not be complete until the kill vehicle is flown
on the actual booster. However, the actual booster is not planned for use in
an intercept attempt until early 2001.

The risk in deployment and production decisions tends to decrease as the
amount of testing increases. More flight-test data will be available in the
second quarter of 2001, when DOD plans to decide on producing sensors and
upgrading radars, and considerably more data will be available in the first
quarter of 2003, when it plans to decide on producing the interceptors.
Other types of test data will also increase over time. The test data are
intended to increasingly demonstrate the system's ability to meet
requirements and to validate the predictions of system performance based on
the use of models and simulations. The type and numbers of primary tests
completed over time are shown in figure 2.

Note: Ground tests consist of data processing hardware and software tested
together to replicate the system. Risk reduction tests consist typically of
reliability tests of offensive U.S. Air Force missiles, which are observed
with NMD sensors. Booster tests are launches of the actual system boosters
before their integration with other operating NMD elements. Beginning in
2001, all flight tests include the actual system booster.

Source: GAO, based on NMD program office data.

Performance Risks Are Increased Because of Flight-test Restrictions and
Uncertainties Regarding the Threat

A number of test limitations and uncertainties affect the program office's
ability to test, analyze, and evaluate system performance.

ï¿½ During testing, a target missile cannot be launched from a threat country
such as North Korea and interceptors cannot fly from deployment sites yet to
be constructed. Test targets are being launched westward from Vandenberg Air
Force Base, California, while the ground-based interceptors are being
launched eastward from the Kwajalein Missile Range in the Pacific Ocean. An
actual enemy missile could approach the United States from several
directions, depending on the attacking country. In its November 1999 report,
the expert panel expressed concern over the limitations on test geometry.

ï¿½ The early warning radar used during testing to track the target missile
immediately after launch is located close to the launch point of the target
missile. In an actual attack, the early warning radar would be alerted by
space-based sensors, but would have to track an enemy missile flying toward
the United States from thousands of miles away. A target missile launched
near the early warning radar presents a large, easy target for the radar to
detect.

ï¿½ The flight path of an enemy missile, which could be launched from several
locations, determines the intercept angle between the target and the
interceptor. However, in flight testing, safety restrictions limit the
flight path of both the interceptor and target missile to a predictable
point over a narrow band of the Pacific Ocean.

ï¿½ The NMD system could be required to engage more than one missile and fire
more than one interceptor at each missile to destroy it. But the flight-test
plans only include testing one interceptor against a single target missile.
In his February 2000 Annual Report, the Director, Operational Test and
Evaluation, identified this shortfall in flight testing as negatively
impacting the NMD program's ability to evaluate system performance.

ï¿½ The intercontinental ballistic missile threat posed by rogue nations is
uncertain, which increases performance risk. The threat from Russia and
China is better understood because both countries have deployed
intercontinental ballistic missiles. North Korea tested a three-stage
missile for the first time in August 1998, and according to the September
1999 national intelligence estimate, could test a more capable version at
any time.8 The 1999 estimate also shows that Iran could test an
intercontinental ballistic missile in the latter half of this decade.
Similarly, the intelligence community is uncertain about what
countermeasures a rogue nation would employ in attempting to defeat a
missile defense system. In February 2000, DOD's Director, Operational Test
and Evaluation, expressed concern regarding the lack of information on the
real threat. He stated that the NMD test program risks building test targets
not representative of enemy missiles or countermeasures.

To address the shortfalls in actual flight testing, the program office plans
to use computer-based simulations. For example, the launch of enemy missiles
and interceptors from other locations will be simulated, as will multiple
interceptors engaging multiple targets. It is uncertain, however, whether
the contractor's primary simulation tool, called LIDS,9 will be fully
available for predicting performance in time for the deployment readiness
review in July 2000. The prime contractor noted this concern in February
2000, reporting a high risk of the LIDS tool not being available by the
readiness review to support analysis of system effectiveness. If so, the
contractor plans to use two other, older simulation tools. However, the
agencies responsible for operational testing of the NMD system expressed
concern that these older simulation tools produce overly optimistic results
and would not accurately represent the system. In addition, according to the
program office's October 1999 status report, the LIDS simulation tool is the
only model that can represent the integrated NMD system.

The NMD system's June 1999 Test and Evaluation Master Plan, which was
approved by test agencies responsible for NMD operational assessments,
attempts to address the threat uncertainties, including countermeasures, in
future rogue nation ballistic missile deployments. The test plan calls for
the use of flight-test targets designed to exhibit a variety of individual
characteristics relating to specific threats, including countermeasures.
Also, the NMD program office is exploring alternatives for development and
validation of future targets.

Defense Acquisition Schedules

The NMD acquisition schedule requires completion of a large number and
complex set of development and deployment activities in a relatively short
amount of time. In restructuring the program in 1999, DOD added 2 years,
allowing about 8 years for the development and deployment of the initial
capability in 2005. However, the schedule is still much shorter than the
schedules for most missile defense acquisitions. For example, the schedule
is only about two-thirds as long as the Safeguard's--the only ballistic
missile defense system that has been fielded in the United States.10 The
Theater High Altitude Area Defense system--currently under development--is
projected to require 15 years before the system's initial fielding. While
the Patriot Advanced Capability-3 system is only a modification to the
existing Patriot air defense system, this acquisition is projected to take
over 7 years from the beginning of engineering and manufacturing development
to initial fielding.11 In addition, technical problems are already eroding
some of the additional time provided for NMD development. For example, the
first attempt at intercepting a target was delayed about 4 months--from June
to October 1999--because of multiple problems with the kill vehicle.

DOD acknowledges that even with the 2-year extension in the time available
for development and deployment activities, the program's schedule is risky.
In October 1999, the Under Secretary of Defense for Policy testified before
the House Armed Services Committee that the "[NMD] program remains risky but
we accept this risk." He stated that the development and deployment schedule
was compressed by at least 4 years because of the seriousness of the
emerging missile threat from rogue nations.

To accelerate the schedule, DOD abandoned the conventional four-phase
acquisition process and implemented a tailored two-phase program structure.
The conventional acquisition process is contrasted to the NMD program
structure in figure 3.

Source: GAO, based on NMD program office chart.

Flight tests scheduled over the next several months could be critical to the
NMD schedule. The Under Secretary of Defense for Acquisition, Technology and
Logistics, established criteria for the system's passing the deployment
readiness review scheduled for July 2000. One criterion is having two
successful intercepts--one of which relies on the actual elements of the
integrated NMD system. One of two intercept attempts, to date, has been
successful. The third attempt is scheduled for early July 2000. Based on the
criteria, the award of a site construction contract can be made if one
attempt was successful, but actual site construction--an event critical to
the schedule--cannot begin until a second successful intercept. With the
exception of the actual system booster, the third intercept attempt in July
2000 is expected to rely on elements of the integrated system.12

and Further Increases Are Likely

As discussed below, DOD has implemented over the course of the program
several measures directed at controlling NMD program costs.

ï¿½ In April 1998, DOD awarded a 36-month, cost-plus-award-fee contract to
Boeing North American to be the prime contractor, or lead system integrator.
The prime contractor is accountable for completing development on each NMD
element and integrating it into the overall system. The cost-plus-award-fee
contract rewards superior performance in the following areas: system
performance, schedule, cost, and program management. Award fee contracts
provide the government with control over the contractor's fee--unlike
fixed-fee type contracts--and, therefore, more direct influence over the
contractor's efforts.

ï¿½ The NMD program office established a "cost as an independent variable"
process with its prime contractor. The process has the goal of setting
realistic cost objectives and following through by making trade-offs between
the system's performance and schedule to achieve a balanced set of goals.
This process involves management of risks to achieve cost, performance, and
schedule objectives; devising appropriate measures for tracking progress in
achieving cost objectives; and providing incentives for the reduction of
operating and support costs of the fielded system. In December 1999, the
prime contractor reported that development and implementation of the "cost
as an independent variable" process has been slow. However, according to the
program office, notable savings have resulted from "cost as an independent
variable" trade-off evaluations. For example, early trade-off studies
conducted by the prime contractor on alternative mixes of radars and
interceptor locations reduced estimated costs over the life of the program
by about $4.3 billion.

ï¿½ The "family of radars" acquisition strategy is being used to reduce costs
related to the NMD X-band radar. The strategy emphasizes commonality of
hardware and software components to satisfy radar requirements for both
theater and national missile defense. The strategy was used to procure the
NMD prototype X-band radar. Advances in technology that were not available
for the prototype radar are planned for the deployed radar. In every case,
contractor-suggested changes in radar design to incorporate advances in
technology stem from the Theater High Altitude Area Defense system. The NMD
program office maintains that reuse of Theater High Altitude Area Defense
hardware and software saved months of design and production time and $80
million to $100 million in engineering and production costs.

Since expanding the program in 2000, the program office estimates that the
total NMD program cost is $36.2 billion compared to the $28.7 billion
estimated in 1999 before the program expansion. The decision to field
100 rather than 20 interceptors accounts for almost $2 billion of the cost
increase. Other factors include the added costs for more flight tests,
ground-test equipment, kill-vehicle hardware, and upgrades to the
early-warning radar facilities. Table 1 shows the total program cost for the
Capability I system by appropriation type, selected fiscal years, and cost
to complete.

 Dollars in millions
                     Fiscal Fiscal FiscalFiscal Fiscal Fiscal
              Prior  year   year   year  year   year   year   Cost to  Total
              years                                           complete
                     2000   2001   2002  2003   2004   2005
 Research,
 development,
 test, and    $4,717 $950   $1,740 $850  $792   $689   $681   $1,543   $11,962
 evaluation
 Procurement  0      0      75     1,537 1,222  1,238  1,079  2,643    7,792
 Military
 construction 10     15     102    192   127    38     15     0        498
 Acquisition
 total        4,726  965    1,916  2,578 2,140  1,965  1,775  4,186    20,252
 Operation and
 support      0      0      0      0     0      0      0      15,993   15,993
 Total        $4,726 $965   $1,916 $2,578$2,140 $1,965 $1,775 $20,179  $36,245

Note: Costs are expressed in then-year dollars. Totals may not add due to
rounding.

Source: Department of Defense.

Estimated costs are likely to increase further because the previously
discussed risk factors could cause schedule delays. The costs of delay would
depend on the reasons for and extent of the delay. However, using current
spending rates, the program office estimates that each month of delay would
increase NMD program costs by $124 million.

A decision to enhance the program by evolving to the Capability II and III
systems would substantially increase the overall cost of the program. For
example, the Capability II or III systems would require from four to nine
X-band radars. Excluding all research and development costs, DOD estimates
that one X-band radar will cost $1.2 billion over the life of the Capability
I system. In addition, Capability III calls for at least
250 interceptors at multiple sites. DOD's cost estimate to produce, operate,
and maintain 100 interceptors at a single site for the Capability I system
is $3.9 billion. Table 2 shows differences in the three levels of NMD
capability designed to counter larger and more sophisticated threats.

                Capability I (approved)  Capability II    Capability III
                                         (planned)        (planned)
                                         Tens of missiles More missiles and
 Threat         A few missiles and       and              more
                simple countermeasures   sophisticated    sophisticated
                                         countermeasures  countermeasures
 Fielding       2005 through 2007        Undecided        Undecided
 Site elements

 Interceptors   100 (single site)        100 (single      250 plus
                                         site)            (multiple sites)
 Ground-based                            4 (1 each, at 4  9 (1 each, at 9
 X-band radars  1 (single site)          sites)           sites)
 Ground-based
 early warning  5 (1 each, at 5 sites)   5 (1 each, at 5  6 (1 each, at 6
 radars                                  sites)           sites)
                Defense Support Program  Spaced-Based     Space-Based
 Space-based    Satellites/Space-Based   Infrared         Infrared
 sensors        Infrared System-High     System-High/Low  System-High/Low
                Earth Orbit              Earth Orbit      Earth Orbit
 Battle
 management,                             To link the      To link the
 command,       To link the elements     elements with    elements with
 control, &     with existing sensors    existing sensors existing sensors
 communication

 Cost           $36.2 billion            Not released by  Not released by
                                         DOD              DOD

Source: GAO, based on NMD program office data.

In April 2000, the Congressional Budget Office (CBO) reported DOD's cost
estimate for a Capability I system, given certain CBO adjustments, as
$25.6 billion.13 Among the adjustments, CBO assumed a deployment period
ending in 2015 rather than at the end of the NMD system's projected
deployment in 2026. Therefore, its estimate did not include procurement
costs of $0.91 billion and operation and support costs of $8.99 billion for
fiscal years 2016 through 2026. CBO also excluded $0.70 billion in
prior-year costs. Accordingly, CBO cited a DOD estimate that is
significantly less than the $36.2 billion estimate used in this report.

In commenting on a draft of this report, DOD concurred with our findings. It
also suggested technical changes, which we incorporated as appropriate.
DOD's comments are reprinted in appendix I.

To identify changes in the program since our June 1998 report and determine
whether significant performance and schedule risks remain, we interviewed
program officials and other relevant sources within and outside DOD,
including a representative from the Office of the Director, Operational Test
and Evaluation. We obtained and analyzed documentation such as system
requirements documents, intelligence summaries of the threat, test plans and
results, and the results of program reviews conducted by outside agencies.
To determine what program officials are doing to control cost risks, what
the program currently costs, and whether cost increases are likely, we
interviewed program cost officials and contractor representatives from
Boeing North American. We also discussed the program cost estimate with
other knowledgeable officials within DOD. We obtained and analyzed copies of
budget documents such as budget decisions, the contract, contractor voucher
costs, and an independent cost estimate. During the course of our review, we
visited the Ballistic Missile Defense Office in Washington, D.C.; the
National Missile Defense Joint Program Office in Arlington, Virginia; the U.
S. Army National Missile Defense Ground Based Elements Office in Huntsville,
Alabama; the National Air Intelligence Center at Wright-Patterson Air Force
Base, Ohio; the Central Intelligence Agency, Langley, Virginia; the Defense
Intelligence Agency, Washington, D.C.; the Space and Missile Defense Command
in Huntsville, Alabama; the Naval Surface Warfare Center in Dalghren,
Virginia; the Office of Science and Technology Policy in Washington, D.C.;
the U.S. Army Test and Evaluation Command in Arlington, Virginia; the Cost
Analysis Improvement Group, Strategic Systems in Washington, D.C.; the
Institute for Defense Analyses in Alexandria, Virginia; and the Office of
the Assistant Secretary of Defense (Acquisition, Technology, and Logistics),
Strategic and Tactical Systems in Washington, D.C.

We did not assess the impact of the Anti-Ballistic Missile Treaty on the NMD
program.14

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

As arranged with your staff, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days from
its issue date. At that time, we plan to provide copies of this report to
the Honorable William Cohen, Secretary of Defense; the Honorable Lewis
Caldera, Secretary of the Army; the Honorable Jacob Lew, Director, Office of
Management and Budget; and key committees of the Congress. We will make
copies available to others upon request.

If you or your staff have questions concerning this report, please contact
me at (202) 512-4841. The major contributors to this report were Bob Levin,
Stan Lipscomb, and Dayna Foster.

Sincerely yours,
Allen Li
Associate Director
Defense Acquisitions Issues

Comments From the Secretary of Defense

(707437)

  

1. The President's budget submission for fiscal year 2001 contains a
$1.9-billion request for the National Missile Defense system.

2. Countermeasures refer to the enemy's use of devices and techniques
intended to impair the National Missile Defense system's operational
effectiveness.

3. National Missile Defense: Even With Increased Funding Technical and
Schedule Risks Are High (GAO/NSIAD-98-153 , June 1998).

4. The ship-based Navy Theater Wide system is designed to intercept enemy
ballistic missiles while they are still above the Earth's atmosphere. The
ground-based Patriot Advanced Capability-3 system is designed to intercept
such missiles within the atmosphere.

5. Report of Panel on Reducing Risk in Ballistic Missile Defense Flight Test
Programs, Institute for Defense Analyses, February 27, 1998

6. All costs in this report are in then-year (adjusted for expected
inflation) dollars.

7. National Missile Defense Review, Institute for Defense Analyses, November
1999.

8. Foreign Missile Developments and the Ballistic Missile Threat to the
United States Through 2015, National Intelligence Council, September 1999.

9. The NMD prime contractor is referred to as the lead system integrator and
the simulation tool is called the Lead System Integrator Integrated
Distributed Simulation, or LIDS.

10. Development of the Safeguard system components began in 1963 and the
system's single site at Grand Forks, North Dakota, achieved full operational
capability in 1975. The system was decommissioned in 1976.

11. The Theater High Altitude Area Defense and Patriot Advanced Capability-3
systems are being developed to protect U.S. assets and forces deployed
overseas and U.S. allies and friends from theater ballistic missile attacks.
Theater ballistic missiles have shorter ranges than intercontinental
ballistic missiles and are designed for use in major regional conflicts,
such as Operation Desert Storm.

12. The actual system booster is not scheduled for flight testing as part of
the entire system until 2001. It is not included as an actual element of the
system for purposes of the deployment readiness review intercept criteria.

13. The CBO also provided its independent estimates for Capability I, II,
and III systems. See Budgetary and Technical Implications of the
Administration's Plan for National Missile Defense, Congressional Budget
Office, April 2000.

14. The Anti-Ballistic Missile Treaty between the United States and the
former Soviet Union governs the conditions under which anti-ballistic
missile systems and components can be developed and deployed. The Treaty as
currently formulated would limit deployment to a single site near Grand
Forks, North Dakota. The Secretary of Defense has directed that NMD
development comply with the terms of the treaty, but according to the
Secretary's current annual report to the President and the Congress, actual
deployment would require modifications to the treaty.
*** End of document. ***