Space Shuttle: Costs for Hubble Servicing Mission and		 
Implementation of Safety Recommendations Not Yet Definitive	 
(19-NOV-04, GAO-05-34). 					 
                                                                 
Hubble's continued operation has been dependent on manned	 
servicing missions using the National Aeronautics and Space	 
Administration's (NASA) shuttle fleet. The fleet was grounded in 
early 2003 following the loss of the Space Shuttle Columbia, as  
NASA focused its efforts on responding to recommendations made by
the Columbia Accident Investigation Board (CAIB). In January	 
2004, NASA announced its decision to cancel the final planned	 
Hubble servicing mission, primarily because of safety concerns.  
Without some type of servicing mission, NASA anticipates that	 
Hubble will cease to support scientific investigations by the end
of the decade. NASA's decision not to service the Hubble prompted
debate about potential alternatives to prolong Hubble's mission  
and the respective costs of these alternatives. This report	 
addresses the basis of NASA's cost estimates to (1) service	 
Hubble using the shuttle and (2) implement recommendations made  
by the CAIB. GAO is continuing its work on the Congressional	 
request that GAO examine the potential cost of a robotic	 
servicing mission to the Hubble Telescope.			 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-34						        
    ACCNO:   A13852						        
  TITLE:     Space Shuttle: Costs for Hubble Servicing Mission and    
Implementation of Safety Recommendations Not Yet Definitive	 
     DATE:   11/19/2004 
  SUBJECT:   Cost analysis					 
	     Program management 				 
	     Repairs						 
	     Satellites 					 
	     Space exploration					 
	     Space shuttles					 
	     Cost estimates					 
	     NASA Hubble Space Telescope Program		 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO 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.                                         **
**                                                              **
******************************************************************
GAO-05-34

     

     * Results in Brief
     * Background
     * NASA's Cost Estimate for a Hubble S對ervicing Mis
     * Estimate for Implementing CAIB Reco對mmendations Is
     * Agency Comments and Our Evaluation
          * Appendix I: Scope and Methodology
          * Appendix II: Comments from the Nati對onal Aeronauti
          * Appendix III: GAO Contact and Staff Acknowledgments
     * GAO Contact
     * Acknowledgments
          * Order by Mail or Phone

                 United States Government Accountability Office

Report to the Subcommittee on

GAO

VA/HUD-Independent Agencies, Committee on Appropriations,

                                  U.S. Senate

                                 November 2004

SPACE SHUTTLE

Costs for Hubble Servicing Mission and Implementation of Safety Recommendations
                               Not Yet Definitive

GAO-05-34

SPACE SHUTTLE

Costs for Hubble Servicing Mission and Implementation of Safety
Recommendations Not Yet Definitive

  What GAO Found

Although a shuttle servicing mission is one of the options for servicing
the Hubble Space Telescope, to date, NASA does not have a definitive
estimate of the potential cost. At our request, NASA prepared an estimate
of the funding needed for a shuttle servicing mission to the Hubble. NASA
estimates the cost at between $1.7 billion to $2.4 billion. However,
documentary support for portions of the estimate is insufficient. For
example, NASA officials told us that the Hubble project's sustaining
engineering costs run $9 to 10 million per month, but they were unable to
produce a calculation or documents to support the estimate because they do
not track these costs by servicing mission. Additionally, the agency has
acknowledged that many uncertainties, such as the lack of a design
solution for autonomous inspection and repair of the shuttle, could change
the estimate.

At the same time, NASA has yet to develop a definitive cost estimate for
implementing all of the CAIB's recommendations but has developed a budget
estimate for safely returning the shuttle to flight-a subset of activities
recommended by the CAIB as needed to return the shuttle to full
operations. NASA currently estimates return to flight costs will exceed $2
billion, but that estimate will likely be refined as the agency continues
to define technical concepts. NASA provided support for portions of the
estimate, but we found the support to be insufficient-either because key
documents were missing or the estimates lacked sufficient detail. Further,
NASA cautions that return to flight costs will remain uncertain until the
first return to flight shuttle mission, which is scheduled to go to the
International Space Station in spring 2005.

                             Hubble Space Telescope

                 United States Government Accountability Office

Contents

Letter        Results in Brief BackgroundNASA's Cost Estimate for a 1 2 3  
                     Hubble Servicing Mission Using the Shuttle Is Not 4 8 13 
                             Definitive Estimate for Implementing CAIB 
                Recommendations Is Not Fully Developed Agency Comments 
                                                    and Our Evaluation 
Appendix I   Scope and Methodology                                      17 
Appendix II  Comments from the National Aeronautics and Space           18 
                Administration                                         
Appendix III         GAO Contact and Staff Acknowledgments              22 
Tables       Table 1: Budget Estimate for Shuttle Servicing of             
                Hubble Table 2: Return to Flight Budget Estimates as   
                of July 2004 Table 3: Return to Flight Budget Estimate 
                Review Status as of July 2004                          6 9 10

                                 Abbreviations

CAIB               Columbia Accident Investigation Board                   
NASA                         National Aeronautics and Space Administration 
PRCB               Program Requirements Control Board                      
UCA                undefinitized contract actions                          

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

United States Government Accountability Office Washington, DC 20548

November 19, 2004

The Honorable Christopher S. Bond Chairman The Honorable Barbara A.
Mikulski Ranking Member Subcommittee on VA/HUD-

Independent Agencies Committee on Appropriations United States Senate

For more than a decade, the National Aeronautics and Space
Administration's (NASA) Hubble Space Telescope has provided unique images
of the universe and has given scientists critical data needed to help
understand a number of space mysteries. Hubble's continued operation has
been dependent on NASA's space shuttle fleet, which carried crewmembers to
the telescope to perform periodic maintenance and upgrades. The grounding
of the shuttle fleet following the tragic loss of Space Shuttle Columbia
in February 2003 put these missions to Hubble on hold, and in January
2004, NASA announced its decision to cancel the fifth and final planned
Hubble servicing mission. With no future servicing missions, NASA
anticipates that Hubble will cease to support scientific investigations by
the end of the decade.

NASA said that its decision to cancel the servicing mission was based
largely on concerns about shuttle safety-specifically, the need to
implement recommendations made by the Columbia Accident Investigation
Board (CAIB) for safely returning the space shuttle to flight. 1 NASA also
recognized that a Hubble servicing mission would necessitate further
changes both to shuttle hardware and operational procedures. However,
NASA's decision has prompted considerable debate. Congressional members,
scientists, and space policy experts have called for an examination of
alternatives to prolong Hubble's mission, such as the possibility of
servicing Hubble robotically. The CAIB Chairman suggested that only a
study of the benefits and risks of a shuttle servicing mission to extend
Hubble's life could determine whether it is worth the risks.

1

The CAIB was created by NASA to investigate the February 1, 2003, loss of
the space shuttle Columbia.

  Page 1 GAO-05-34 Space Shuttle Hubble Servicing

                                Results in Brief

This report addresses the basis for NASA's cost estimates to (1) service
the Hubble Telescope by using the shuttle and (2) implement the CAIB's
recommendations, including those recommendations directly related to
safely returning the shuttle to flight. We are continuing our work on your
request that we examine the potential cost of a robotic servicing mission
to the Hubble Telescope.

To assess the basis for NASA's Hubble servicing mission cost estimate, we
examined NASA's rationale for its decision to cancel the final planned
shuttle servicing mission, analyzed available funding estimates, and
requested analytical and documentary support for selected high-dollar
items to identify the sufficiency of support. In reviewing the basis for
NASA's cost estimate for implementing the CAIB's recommendations, we
analyzed available funding estimates and requested analytical and
documentary support for selected high dollar portions of the estimate to
determine the sufficiency of such support. We performed our review from
March through September 2004 in accordance with generally accepted
government auditing standards. For a complete description of our scope and
methodology, see appendix I.

Definitive cost estimates to facilitate decisions regarding options for
servicing the Hubble are critical. However, NASA has not yet developed
such an estimate for one of the options-a shuttle servicing mission. At
our request, NASA prepared an estimate of the funding needed for a shuttle
servicing mission to Hubble. However, NASA could not provide documented
support for key portions of the estimate, stating that there are many
uncertainties that could change the estimate, such as the lack of a design
solution for two safety-related requirements-autonomous inspection and
repair and crew rescue mission capabilities.

Similarly, NASA does not have a definitive cost estimate for implementing
all of the CAIB recommendations. The agency has been focusing primarily on
those recommendations it considers necessary to return the shuttle fleet
to flight. NASA's current estimate for implementing those recommendations
is more than $2 billion. NASA provided us with documentary support for
portions of the estimate, but we found some of the support to be
insufficient. In NASA's view, the estimate for returning the shuttle fleet
to flight will remain uncertain until the first shuttle mission to the
International Space Station.

In written comments, which are reprinted in appendix II, NASA stated that
the agency believes that both the estimate and the methodology used in

                                   Background

calculating the costs of reinstating the servicing mission are sound and
accurate given the level of definition of the mission at this point in
time. However, the agency agreed that some portions of the servicing
mission activities lacked the design maturity required to estimate the
costs according to NASA accepted and established procedures.

Since it was launched in 1990, the Hubble Space Telescope has sent back
images of space that have made a significant contribution to our
understanding of the universe. The telescope uses pointing precision,
powerful optics, and state-of-the-art instruments to explore the visible,
ultraviolet, and near-infrared regions of the electromagnetic spectrum. To
keep it at the forefront of astronomical research and extend its
operational life, Hubble's instruments have been upgraded through a series
of shuttle servicing missions. The fifth and final planned servicing
mission was intended to install new science instruments, replace the
telescope's insulation, and replace the batteries and gyroscopes.
According to NASA, the lifetime of the observatory on orbit is ultimately
limited by battery life, which may extend into the 2007-2008 time frame,
but scientific operations are limited by the gyroscopes that stabilize the
telescope-whose lifetimes are more difficult to predict. NASA forecasts
that the Hubble will likely have fewer than three operating gyroscopes by
mid-2006, and fewer than two by mid-2007.

In response to congressional concerns about NASA's decision to cancel the
servicing mission, NASA requested that the National Research Council 2
conduct an independent assessment of options for extending the life of the
Hubble Space Telescope. In May 2004, the Council established a committee
to assess the viability of a shuttle servicing mission, evaluate robotic
and ground operations to extend the life of the telescope as a valuable
scientific tool, assess telescope component failures and their impact, and
provide an overall risk-benefit assessment of servicing options. In an
interim report issued in July 2004, the committee urged NASA to commit to
a Hubble servicing mission that accomplishes the

2

The National Research Council is part of the National Academies, which
also comprise the National Academy of Sciences, National Academy of
Engineering, and Institute of Medicine. They are private, nonprofit
institutions that provide science, technology, and health policy advice
under a congressional charter. The Research Council was organized by the
National Academy of Sciences in 1916 to associate the broad community of
science and technology with the Academy's purposes of further knowledge
and advising the federal government.

  NASA's Cost Estimate for a Hubble Servicing Mission Using the Shuttle Is Not
  Definitive

objectives of the canceled servicing mission and to take no actions that
would preclude using a space shuttle to carry out this mission. According
to a NASA official, the agency is not actively pursuing the shuttle
servicing option but is not precluding it.

NASA is currently evaluating the feasibility of performing robotic
servicing of the Hubble Telescope. To facilitate the evaluation, the
agency has formulated a robotic mission concept, which includes a vehicle
comprised of a robotic servicing module and another module that can be
used to eventually de-orbit the telescope. The potential task list of
activities for robotic servicing includes replacing the gyroscopes and
batteries, installing new science instruments, and de-orbiting the
observatory at the end of its life. According to a NASA official,
contracts to facilitate the robotic mission were recently awarded for work
to begin on October 1, 2004.

The CAIB concluded that the Columbia accident was caused by both physical
and organizational failures. The Board's 15 return to flight
recommendations necessary to implement before the shuttle fleet can return
to flight primarily address the physical causes of the accident and
include eliminating external tank debris shedding and developing a
capability to inspect and make emergency repairs to the orbiter's thermal
protection system. NASA publishes periodic updates to its plan for
returning the shuttle to flight to demonstrate the agency's progress in
implementing the CAIB recommendations. The most recent update is dated
August 27, 2004. This update identifies the first shuttle flight as
occurring in spring 2005.

NASA does not currently have a definitive cost estimate for servicing the
Hubble Telescope using the shuttle. The agency focused on safety concerns
related to a servicing mission by the space shuttle in deciding not to
proceed, and did not develop a cost estimate. At our request NASA prepared
an estimate of the funding needed for a Hubble servicing mission by the
space shuttle. NASA could not provide documented support for its estimate.
The agency recognizes that there are many uncertainties that could change
the estimate. NASA has now begun to explore the costs and benefits of
various servicing alternatives, including robotic servicing, which should
enable NASA to make a more informed decision regarding Hubble's future.

At our request NASA began development of an estimate of the funding needed
for a shuttle servicing mission to the Hubble. The estimate provided
captures additional funds over and above NASA's fiscal year 2005 budget
request that would be required to reinsert the mission in the shuttle
flight manifest for launch in March 2007. The estimate does not include
funding already expended to support the canceled servicing mission and
develop the science instruments. NASA has determined that the additional
funds needed to perform a shuttle servicing mission for Hubble would be in
the range of $1.7 billion to $2.4 billion. According to NASA, this
estimate is based on what it might cost, but it does not take into account
the technical, safety, and schedule risks that could increase the cost
and/or undermine the viability of the mission. For example, NASA cites
uncertainties related to two safety-related requirements: inspection and
repair and crew rescue mission capabilities that would be autonomous of
the International Space Station and for which NASA currently has not
formulated a design solution. 3 In addition, NASA cautions that it did not
examine whether design solutions could be accomplished in time to service
Hubble before it ceases operations. Table 1 shows NASA's budget estimate
phased by fiscal year (FY) for shuttle servicing of the Hubble Space
Telescope, including ranges for some of the estimates.

Although the autonomous inspection and repair capability of the shuttle's
thermal protection system is a CAIB requirement for all shuttle missions,
the currently planned method depends on International Space Station
assets, such as the International Space Station's robotic arm to stabilize
a spacewalk crew making repairs to the shuttle. A Hubble mission would not
have this asset available; therefore, NASA would have to develop an
alternate method for stabilizing a crewmember making repairs. The second
requirement- NASA's requirement for a crew rescue shuttle-would apply to
all shuttle flights, but according to NASA, the agency would need to
dedicate two shuttles to a Hubble servicing mission because of the shorter
amount of time to react to an emergency because the crew would not have
the benefit of the International Space Station as a safe haven. The agency
contends that the second (rescue) shuttle for the Hubble servicing mission
would need to be on the adjacent launch pad in countdown mode at the same
time the first shuttle is launched. The rescue shuttle capability would
require the development of a second generation boom to be used to transfer
the crew from the stricken orbiter and an enhanced camera to process
imagery in time to support a go-no go rescue mission decision. In
addition, a crew return kit would be needed to provide seats on the rescue
shuttle for the rescued crew, and a special crew would have to be trained
for the rescue mission.

Table 1: Budget Estimate for Shuttle Servicing of Hubble
Real year dollars in millions a
                                       Fiscal year
                                                                                Budget to
Description          2004  2005   2006   2007       2008   2009     Complete       Totala 
Sustaining engineering for                                                   
Hubble                      $115  $141   $105        $26   $30           $40         $457 
project                                                                      
Delay de-orbit mission                                                       
until 2012                               (69)      (136)       79        166           40 
Extend Hubble operations                                                     
to 2012                                                                  117          117 
Shuttle services unique to                                                   
servicing                                                                                 
mission and                                                                  
extravehicular                                                               
activityb                            18    15                                          33
Payload                          7    7    13         10        7                      43 
processing                                                                   
Autonomous inspection and                                                    
repair        $116-134  74-86   52-60   18-21                                     260-300 
capability                                                                   
Autonomous rescue mission                                                    
capability     85-98  116-134   61-71   31-36                                     293-338 
Rescue mission unique                                                        
requirements and ground                                                      
operations                               1 22                                          22 
International Space Station                                                  
Program impact (3-month                                                      
delay)                               (5) (15)       (13)      (4)         50           13 
Delay shuttle phase-out                                                      
3 months                                  9-0      17-72  26-249     350-757    401-1,078 
Totala    $200-231 $312-341 $275-293 $128-127 $(95)-(41) $137-361 $723-1,130 $1,679-2,441 

Source: NASA.

Note: Estimates are in full cost. The concept of full cost ties all agency
costs (including civil service personnel costs) to major activities. This
includes costs that are directly related to a specific project, such as
contractor-supplied hardware, and indirect costs such as administrative
costs, rents, utilities, and travel costs that cannot be directly
identified with a specific project but which can be allocated based on
direct labor hours, square footage, or other methods.

a

In some cases, totals do not add because of rounding.

bExtravehicular activity is a space walk.

While we did not independently verify each component of NASA's estimate,
we requested that NASA provide the analytical basis and documentary
support for selected portions of the estimate, primarily those with large
dollar values. NASA could not provide the requested information. For
example, NASA officials told us that the Hubble project's sustaining
engineering costs run $9 to10 million per month, but they were unable to
produce a calculation or documents to support the estimate because they do
not track these costs by servicing mission. We also requested the basis of
estimate for the costs to delay shuttle phase-out and for tools
development for vehicle inspection and repair without the International
Space Station (a component of extravehicular activity above). In response,
NASA provided the assumptions upon which the estimates were based and
stated that the estimates were based on information provided by Johnson
Space Center and Kennedy Space Center subject matter experts. NASA also
added that rigorous cost estimating techniques could not be applied to the
tools development estimate because a rescue mission currently is only a
concept. No analytical or documentary support was provided. In estimating
the cost for the autonomous inspection and repair and rescue mission
capabilities, NASA used a 30 to 50 percent uncertainty factor because of
the very high uncertainty in the cost of developing and conducting a
mission that is not adequately defined-i.e., NASA's estimate of $425
million plus 50 percent equals the $638 million upper range shown in the
table above for these two items added together. As with the other
estimates for which we requested analytical and documentary support, NASA
was not able to provide it because the agency could not do a risk analysis
without a design solution, according to a NASA official. The lack of
documented support for portions of NASA's estimate increases the risk of
variation to the estimate. Further, NASA recognizes that there are many
uncertainties that could change the current estimate.

The 2004 NASA Cost Estimating Handbook states that cost analysts should
document the results of cost estimates during the entire cost estimating
process and that the documentation should provide sufficient information
on how the estimate was developed so that independent cost analysts could
reproduce the estimate. According to the handbook, the value of the
documentation and analysis is in providing an understanding of the cost
elements so that decision-makers can make informed decisions.

  Estimate for Implementing CAIB Recommendations Is Not Fully Developed

Recently, we also reported that dependable cost estimates are essential
for establishing priorities and making informed investment decisions in
the face of limited budgets. 4 Without this knowledge, a program's
estimated cost could be understated and thereby subject to underfunding
and cost overruns, putting programs at risk of being reduced in scope or
requiring additional funding to meet their objectives.

Since we began our review, attention has focused on alternatives to a
shuttle mission, such as robotic servicing of Hubble. NASA has formed a
team to evaluate Hubble servicing alternatives, including cost
information. This analysis should enable NASA to make a more informed
decision about Hubble's future and facilitate NASA's evaluation of the
feasibility of robotic servicing options.

Currently, NASA has developed budget estimates for implementing the CAIB
recommendations required to return the space shuttle to flight but not for
all of the CAIB recommendations. NASA provided us with documentary support
for portions of the return to flight estimate, but we found it to be
insufficient. According to NASA, the agency's cost for returning the
shuttle to flight, which is slightly over $2 billion, will remain
uncertain until the completion of the first shuttle missions to the
International Space Station in fiscal year 2005.

NASA's return to flight activities involve enhancing the shuttle's
external tank, thermal protection system, solid rocket boosters, and
imagery system to address the physical cause of the Columbia accident-a
piece of insulating foam that separated from the external tank and struck
a reinforced carbon-carbon panel on the leading edge of the orbiter's left
wing. To address this cause, NASA is working to eliminate all external
tank debris shedding. Efforts are also in place to improve the orbiter's
thermal protection system, which includes heat resistant tiles, blankets,
and reinforced carbon-carbon panels on the leading edge of the wing and
nose cap of the shuttle, to increase the orbiter's ability to sustain
minor debris damage. NASA is also redesigning the method for catching
bolts that break apart when the external tank and solid rocket boosters
separate as well as providing the capability to obtain and downlink images
after the separation. NASA and the United States Air Force are working to
improve

4

GAO, NASA: Lack of Disciplined Cost-Estimating Processes Hinders Effective
Program Management , GAO-04-642 (Washington, D.C.: May 28, 2004).

Page 8 GAO-05-34 Space Shuttle Hubble Servicing

the use of ground cameras for viewing launch activities. Table 2 shows
NASA's budget estimates for return-to-flight activities.

Table 2: Return to Flight Budget Estimates as of July 2004
Real year dollars in millions a
                                  Fiscal year
Activity                                  2003 2004 2005 2006-2009a Totalb 
Orbiter reinforced carbon-carbon           $2  $38    $7            
inspections                                                         
On-orbit thermal protection system inspection                       
and extravehicular activity tile repair     20  68  130             
Orbiter workforce                           0   5     37            
Orbiter thermal protection system hardening 0   28    34            
Orbiter certification/verification          0   47    26            
Orbiter other                               0   15    16            
External tank items (camera, bipod ramp,    11 114    94            
etc.)                                                               
Solid rocket booster items (bolt catcher,                           
camera, etc.)                                1  8     26            
Ground camera ascent imagery upgrade         8  40    58            
Kennedy Space Center ground operations                              
workforce                                    0  32    36            
Other (system integration, hardware processing                      
and operations systems verification, and space                      
shuttle main engine technical                0  67  177             
assessment)                                                         
Stafford-Covey team                             3      1            
Totalb                                   $42   $465 $643   $1,067   $2,217 

Source: NASA.

Note: According to NASA, not all elements of full cost have been
distributed to return to flight activities.

aFiscal years 2006-2009 are NASA preliminary estimates, and a cost
breakout by activity is not available.

b

In some cases, totals do not add because of rounding.

However, the majority of NASA's budget estimates for returning the shuttle
to flight are not fully developed-including those for fiscal year 2005-as
indicated by the agency's internal approval process. The Program
Requirements Control Board (PRCB) is responsible for directing studies of
identified problems, formulating alternative solutions, selecting the best
solution, and developing overall estimates. 5 According to NASA, actions
approved with PRCB directives have mature estimates, while those with
control board actions in process-that is currently under review but with
no issued directives yet-are less mature. Both the content and estimates
for return to flight work that have not yet been reviewed by the control
board are very preliminary and subject to considerable variation. Table 3
shows the status of control board review of NASA return to flight budget
estimates and the percent of the total estimate at each level of review.

Table 3: Return to Flight Budget Estimate Review Status as of July 2004
Real year dollars in millions
                                  Fiscal year
                                                                      Percent
Review status         2003 2004  2005  2006 2007 2008 2009 Totala       of 
                                                                       totala 
Control board review                                              
complete; directive    $31 $319   $117  $53 $47  $49  $39    $655       29 
issued                                                            
Been to control board;                                            
directive not yet       11     146  217 117 125  134  109     859       39 
issued                                                            
In review process                   309 162  84   79   70     704       32 
Total Return to Flight                                            
activities a           $42 $465  $643  $331 $257 $261 $218 $2,217      100 

Source: NASA.

Note: According to NASA, not all elements of full cost have been
distributed to return to flight activities.

a

In some cases, totals do not add because of rounding.

The membership of the PRCB includes the Space Shuttle Program Manager,
Deputy Manager, all Project and Element Managers, Safety and Mission
Assurance personnel, and the Team Leader of the return to flight Planning
Team.

Page 10 GAO-05-34 Space Shuttle Hubble Servicing

NASA provided us with the PRCB directives and in some cases, attachments
which the agency believes support the estimate. 6 However, we did not find
this support to be sufficient. According to NASA's cost estimating
handbook, estimates should be documented with sufficient detail to be
reproducible by an independent analyst. Nevertheless, in many cases, there
were no documents attached to the directive, and in cases where documents
were attached to the directives, the documents generally provided
high-level estimates with little detail and no documentation to show how
NASA arrived at the estimate. For example, a request for $1.8 million to
fund the network to support external tank camera transmissions indicated
that $1.516 million of the amount would be needed for Goddard Space Flight
Center to provide the necessary equipment at receiving stations, labor,
subcontractor costs, and travel and that the remaining $290,000 would be
needed for improvements to the receiving antennas ($104,000) and recurring
costs ($62,000 per flight) for three trucks and the associated transponder
time. However, the documents did not show how the requester for the $1.8
million arrived at the estimates. NASA officials told us that the reason
for this was that the managers approving the directives trusted their
employees to accurately calculate the estimate and maintain the support.
In addition, our review of the documents indicated and NASA confirmed that
quite a few of the estimates were based on undefinitized contract actions
(UCA)-that is, unnegotiated contract changes. Under these actions, NASA
officials can authorize work to begin before NASA and the contractor agree
on a final estimated cost and fee. As we have stated in our high-risk
series, relying on unnegotiated changes is risky because it increases the
potential for

We attempted to obtain analytical and documentary support for the return
to flight estimates on two occasions. First, in May 2004, we requested
support for three high-dollar items for which we were told the estimates
ranged from strong to weak to a mixture. In response to this request, NASA
told us that the detailed bases of estimate for the activities were being
developed and that NASA would report on return to flight expenses to
Congress and GAO in a few weeks. Then, several weeks later when NASA
released the July 28 return to flight update, we requested support for the
three largest estimates, two of which were the same items for which we
requested support in May. In response, NASA provided the PRCB directives
and backup documents underlying the estimates that the agency considers to
be mature. After we reviewed these documents and concluded that the
support was not adequate, NASA offered to select some additional examples
to show that the estimates were credible.

unanticipated cost growth. 7 This, in turn, may force the agency to divert
scarce budget resources intended for other important programs. As of July
31, 2004, NASA records showed 17 UCAs related to return to flight with
not-to-exceed amounts totaling $147.5 million. NASA's estimate for the
entire effort under these UCAs totals about $325 million, or 15 percent of
NASA's current $2.2 billion return to flight estimate.

In June 2004, NASA established additional requirements for funding
requests submitted to the PRCB. Under the new policy, an independent cost
estimate must be developed for requests greater than $25 million, and a
program-level cost evaluation must be completed for requests over $1
million. The program-level evaluation consists of a set of standard
questions to document the rationale and background for cost-related
questions. The responses to the questions are initially assessed by a cost
analyst but are reviewed by the Space Shuttle Program Business Manager
before submission to the PRCB.

NASA provided us with two examples of requests falling under the new
requirements. Both of the examples had better support than those with PRCB
directives, but documentary support was still not apparent. For example,
the funding request for a debris radar indicated that the estimate was
based on a partnering agreement with the Navy and the Navy's use of the
technology. However, the program-level evaluation pointed out that no
detailed cost backup was provided. The other example, which was a funding
request to change the processes currently in place for the Space Shuttle
Program's problem reporting and corrective action system, was very well
supported in terms of analysis, as the requester prepared detailed
spreadsheets calculating the funding requirements according to a breakdown
of the work to be performed, cited sources for labor rates, and provided
assumptions underlying the calculations. However, as pointed out in the
program evaluation of the request, there was no support provided for the
estimate other than the initiator's knowledge of the change. We believe
that future compliance with NASA's new policy establishing additional
requirements for funding requests and the inclusion

GAO, NASA: Major Management Challenges and Program Risks, GAO-03-849T
(Washington, D.C.: June 12, 2003); GAO, High-Risk Series: An Update,
GAO-03-119 (Washington, D.C.: January 2003); GAO, Major Management
Challenges and Program Risks: National Aeronautics and Space
Administration, GAO-03-114 (Washington, D.C.: January 2003); GAO,
High-Risk Series: An Update, GAO-01-263 (Washington, D.C.: January 2001);
and GAO, Major Management Challenges and Program Risks: National
Aeronautics and Space Administration, GAO-01-258 (Washington, D.C.:
January 2001).

                                Agency Comments
                               and Our Evaluation

of documentary support could potentially result in more credible return to
flight budget estimates.

According to NASA, estimates for fiscal year 2005 and beyond will be
refined as the Space Shuttle Program comes to closure on return to flight
technical solutions and the return to flight plan is finalized. NASA
expects that by late fall of 2004, a better understanding of the fiscal
year 2005 financial situation will be developed. However, NASA cautions
that the total cost of returning the shuttle to flight will remain
uncertain until completion of the first shuttle missions to the space
station, scheduled to begin in spring 2005.

In written comments on a draft of this report, the NASA Deputy
Administrator stated that the agency believes that both the estimate and
methodology used in the calculation of costs for reinstating the Hubble
Space Telescope servicing mission are sound and accurate, given the level
of definition at this point in time. Notwithstanding that belief, the
agency agreed that portions of the servicing mission activities lacked the
design maturity required to estimate the costs according to accepted and
established NASA procedures.

Specifically, NASA agrees that the Hubble Space Telescope work breakdown
structure was not constructed to collect program costs. At the same time,
NASA believes it is erroneous to suggest that NASA has no valid basis for
the numbers provided, citing the "Servicing Mission 4 Resources Management
Plan," which describes the effort required for completion of a servicing
mission. According to NASA, although the program's accounting system does
not capture sustaining engineering costs in GAO's preferred format, the
Servicing Mission 4 Resources Management Plan details mission schedules
and staffing, and applying contractor and civil service rates to that
staffing level can accurately reflect the effort required to execute a
servicing mission. We requested this type of analysis and documentary
support, but NASA representatives did not offer such a calculation.
Rather, the officials stated that the sustaining engineering costs were
based on management's assessment of contractor financial data and in-house
service pool charges and that these activities could not be traced back to
source documentation. Without adequate supporting data, we cannot assess
the accuracy and reliability of such information.

NASA acknowledged that the agency does not have a technical design from
which to derive the cost for the on-orbit inspection and repair of the
shuttle independent of support from the International Space Station. In
the case of the unsupported cost estimate for delaying the phase out of
the space shuttle in order to complete a manned Hubble servicing mission,
NASA stated that it used approved budget projections for the operating
years affected by the insertion of the Hubble servicing mission and
prorated the extension of the service life. According to NASA, a range was
added to the estimate to account for uncertainties and retention of
critical skills. The estimates were presented as a rough order of
magnitude. NASA stated that it provided its assumptions to demonstrate the
reasonableness of the estimates. Nevertheless, in spite of the
uncertainties in the estimate, which we recognized in our report, NASA
guidance states that cost estimates should be documented during the entire
cost estimating process and that the documentation should provide
sufficient information on how an estimate was developed to be reproducible
by independent cost analysts. NASA did not provide us with this type of
documentation. Without adequate supporting data, we cannot assess the
accuracy and reliability of such information. We do not agree that the use
of approved budget projections is a reliable cost estimating methodology,
particularly given the long-term budget implications of the extension of
the space shuttle's service life.

NASA believes that the examples it provided of the actions to implement
several of the CAIB recommendations attest to the rigor of the process and
approved procedures NASA utilized to validate the costs. According to
NASA, the estimates will mature as the technical solutions mature, but the
estimates were not refined at the time of our review. The agency believes
the outstanding technical issues necessary to return to flight are
beginning to be resolved. However, the examples that NASA provided were in
support of estimates that the agency considers mature. We requested
support for high dollar portions of NASA's estimate, which the agency did
not provide. However, NASA selectively provided examples of what it
considered to be mature estimates. We reviewed the examples but found that
most of them contained insufficient documentation to assess the
reliability of the estimates. In many cases, there were no documents in
the approval packages to support the estimates, and in cases where there
were documents, they generally provided high-level estimates with little
detail and no documentation to show how NASA arrived at the estimates. We
believe that because of its difficulty providing reliable cost estimates,
NASA cannot provide the Congress assurance that its budget request for the
shuttle program for fiscal year 2006 will be sufficient and that
shortfalls would not need to be met through reductions in other NASA
programs.

NASA stated that it believes the use of UCAs is both reasonable and
necessary for return to flight activities. We agree that UCAs may be
justified to facilitate work outside the scope of existing contracts to
expedite the return to flight activities. However, the use of UCAs appears
to be a growing trend and is a risky contract management practice because
it increases the potential for unanticipated cost growth. In the past, we
cited the agency's use of UCAs as one of the reasons we retained contract
management as a high-risk designation for NASA to focus management
attention on problem areas that involve substantial resources. 8

Finally, NASA agrees that cost estimates for significant development
activities should be appropriately documented. According to NASA,
additional requirements for cost estimates and internal controls recently
established by the program represent a step in ensuring the appropriate
documentation is developed as solutions are identified. As stated in our
report, we believe that future compliance with this new policy could
potentially result in more credible budget estimates.

In a broader context, reliable and supportable cost estimating processes
are important tools for managing programs. Without this knowledge, a
program's estimated cost could be understated and thereby subject to
underfunding and cost overruns, putting programs at risk of being reduced
in scope or requiring additional funding to meet their objectives.
Further, without adequate financial and nonfinancial data, programs cannot
easily track an acquisition's progress and assess actions to be taken
before it incurs significant cost increases and schedule delays.

As agreed with your offices, unless you announce its contents earlier, we
will not distribute this report further until 30 days from its date. At
that time, we will send copies to the NASA Administrator and interested
congressional committees. We will make copies available to others upon
request. In addition, the report will be available at no charge on the GAO
Web site at http://www.gao.gov.

High-Risk Series: An Update (GAO-01-263, January 2001).

If you or your staff have any questions concerning this report, please
contact me at (202) 512-4841 or [email protected]. Key contributors to this
report are acknowledged in appendix III.

Allen Li Director Acquisition and Sourcing Management

                       Appendix I: Scope and Methodology

To assess the basis for NASA's Hubble servicing mission cost estimate, we
analyzed NASA's estimate of the funding needed for a shuttle servicing
mission and supporting documentation, and we reviewed NASA documents
explaining the rationale for the decision and identifying alternatives to
shuttle servicing. We interviewed program and project officials to clarify
our understanding of the available cost information and NASA's rationale
for the decision. To test the sufficiency of the support for the estimates
provided by NASA, we requested the analytical basis and documentary
support for selected portions of the estimates, primarily those with large
dollar values. In addition, we compared NASA's decisionmaking process with
relevant Office of Management and Budget and NASA guidance on information
and analyses recommended to enable decisionmakers to select the best
alternative.

To determine the basis for NASA's cost estimate for implementing all of
the CAIB recommendations, we reviewed the CAIB report (volume 1), NASA's
return to flight implementation plan and budget estimates, and agency
documentation discussing the return to flight budget estimate. We
interviewed program officials to obtain a better understanding of NASA's
plans for returning the space shuttle to flight, the status of that
effort, and the estimated cost. To test the sufficiency of the support for
NASA's return to flight estimate, we requested the analytical basis and
documentary support for selected high dollar portions of the estimate.

To accomplish our work, we visited NASA Headquarters, Washington, D.C.;
and Goddard Space Flight Center, Maryland.

We performed our review from March through September 2004 in accordance
with generally accepted government auditing standards.

Appendix II: Comments from the National Aeronautics and Space Administration

  Appendix II: Comments from the National Aeronautics and Space Administration
  Appendix II: Comments from the National Aeronautics and Space Administration
  Appendix II: Comments from the National Aeronautics and Space Administration

Appendix III: GAO Contact and Staff Acknowledgments

Allen Li (202) 512-4841

  GAO Contact

Staff making key contributions to this report were Jerry Herley, Erin
Schoening, Karen Sloan, and Jonathan Watkins.

(120335)

    GAO's Mission

The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO's
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site ( www.gao.gov ). Each weekday, GAO posts GAO
Reports and newly released reports, testimony, and correspondence on its
Web site. To

have GAO e-mail you a list of newly posted products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."

                             Order by Mail or Phone

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

Contact:

To Report Fraud, Web site: www.gao.gov/fraudnet/fraudnet.htm

    E-mail: [email protected]

Federal Programs Automated answering system: (800) 424-5454 or (202)
512-7470

Gloria Jarmon, Managing Director, [email protected] (202) 512-4400 U.S.
Government Accountability Office, 441 G Street NW, Room 7125 Relations
Washington, D.C. 20548

Susan Becker, Acting Manager, [email protected] (202) 512-4800

    Public Affairs

U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
*** End of document. ***