Department of Energy: Additional Opportunities Exist for Reducing
Laboratory Contractors' Support Costs (09-SEP-05, GAO-05-897).	 
                                                                 
In fiscal year 2004, about two-thirds of the Department of	 
Energy's (DOE) $26.9 billion in spending went to 28 major	 
facilities--laboratories, production and test facilities, and	 
nuclear waste cleanup and storage facilities. DOE spent about	 
$2.9 billion in fiscal year 2004 to support the mission of its	 
five largest laboratories. GAO was asked to examine (1) recent	 
trends in indirect and functional support cost rates for these	 
five laboratories, noting key differences in how contractors	 
classify costs, and (2) the efforts of DOE and its contractors to
reduce indirect and other support costs and identify additional  
opportunities for savings.					 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-897 					        
    ACCNO:   A36198						        
  TITLE:     Department of Energy: Additional Opportunities Exist for 
Reducing Laboratory Contractors' Support Costs			 
     DATE:   09/09/2005 
  SUBJECT:   Contract administration				 
	     Contractors					 
	     Cost analysis					 
	     Cost control					 
	     Facility maintenance				 
	     Facility management				 
	     Federal facilities 				 
	     Financial analysis 				 
	     Internal controls					 
	     Laboratories					 
	     Management and operating contracts 		 

******************************************************************
** 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-05-897

United States Government Accountability Office

GAO	Report to the Subcommittee on Energy and Water Development, Committee
on

                    Appropriations, House of Representatives

September 2005

DEPARTMENT OF ENERGY

  Additional Opportunities Exist for Reducing Laboratory Contractors' Support
                                     Costs

                                       a

GAO-05-897

[IMG]

September 2005

DEPARTMENT OF ENERGY

Additional Opportunities Exist for Reducing Laboratory Contractors' Support
Costs

                                 What GAO Found

For fiscal years 2000 through 2004, laboratory-reported rates for indirect
costs-those not charged directly to a specific program-increased at two
laboratories and decreased at three. However, indirect cost rates cannot
be compared across laboratories because contractors classify different
portions of support costs as indirect. To facilitate analysis, DOE
requires the laboratories to report what it called "functional support
costs," or costs that support missions, regardless of whether they are
classified as direct or indirect costs. Using this measure, three
laboratories' rates-that is, functional support costs divided by total
costs-increased and two laboratories' rates decreased over the 5-year
period. While functional support cost rates improved comparability,
several DOE and contractor officials said that the definitions for some
categories of support costs, such as "facilities management," are unclear,
leading to confusion and inconsistent reporting.

DOE and its contractors have initiated several steps to reduce indirect
and other support costs but can take additional actions to improve their
implementation. First, DOE's laboratory contracts have increasingly
included incentives to encourage cost reductions. In fiscal year 2004, for
example, the National Nuclear Security Administration began an "awardterm"
pilot program that allows a contractor to earn extra contract years based
on performance and cost-saving achievements. However, DOE is expanding use
of this incentive without evaluating it. Second, DOE requires its
contractors to benchmark employee benefits and to reduce benefits if they
exceed the benchmark, but DOE did not promptly enforce these requirements
at one laboratory and exempted two others. Third, DOE has begun to address
a $1.9 billion backlog of deferred maintenance to reduce long-term costs.
However, without a more rigorous approach, the backlog will persist well
into future decades. Lastly, while some laboratories have used process
improvement programs to streamline business processes and reduce costs,
others do not have such programs, nor are they required to have them.

Functional Support Costs for Five DOE Laboratories Reviewed, Fiscal Year
2004

Dollars in millions

                         National laboratory Contractor

Functional support costs

                             Idaho        Battelle Energy Alliance     $377.5 
                Lawrence Livermore        University of California      573.2 
                        Los Alamos        University of California      889.1 
                         Oak Ridge                UT-Battelle, LLC      292.9 
                            Sandia   Lockheed Martin Corporation        718.0 

Source: DOE.

United States Government Accountability Office

Contents

     Letter                                                                 1 
                                           Results in Brief                 4 
                                              Background                    6 
                             Some Laboratories' Cost Rates Have Increased  
                                          While Others Have                
                             Decreased, but Not All Rates Are Comparable    7 
                            DOE and Its Contractors Have Taken Actions to  
                                         Reduce Indirect and               
                             Other Support Costs, but Opportunities Exist  
                                             for Further                   
                                              Reductions                   13 
                                             Conclusions                   25 
                                 Recommendations for Executive Action      26 
                                           Agency Comments                 26 
                                        Scope and Methodology              27 
Appendixes                                                              
                Appendix I:    Award-Term Provisions at Four DOE Sites     30 
               Appendix II:     Comments from the Department of Energy     32 
              Appendix III:     GAO Contact and Staff Acknowledgments      33 
                              Table 1: Functional Support Cost Rates for   
     Tables                               Five Laboratories,               
                                         Fiscal Years 2000-04              10 
                             Table 2: Results of the DOE-Funded Employee   
                                            Benefit Value                  
                              Studies for Each of the Five Laboratories    16 
                             Table 3: Award-Term Provisions for Four DOE      
                                              Facilities                   30

Abbreviations

CFO chief financial officer
DOE Department of Energy
NNSA National Nuclear Security Administration

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.

A

United States Government Accountability Office Washington, D.C. 20548

September 9, 2005

The Honorable David L. Hobson
Chairman
The Honorable Peter J. Visclosky
Ranking Minority Member
Subcommittee on Energy and Water Development
Committee on Appropriations
House of Representatives

In fiscal year 2004, about two-thirds of the Department of Energy's (DOE)
$26.9 billion in spending went to 28 major facilities, including
laboratories,
nuclear weapons test and production facilities, and nuclear waste cleanup
and storage facilities. DOE primarily uses contractors-industrial firms
and nonprofit organizations, including educational institutions-to manage
and operate these facilities. DOE oversees these contractors' activities
through its headquarters program offices-primarily the National Nuclear
Security Administration (NNSA), the Office of Environmental Management,
and the Office of Science-and site offices located at each facility.

DOE reimburses its contractors for the costs incurred in carrying out the
department's missions. These include costs that can be directly identified
with a specific DOE program (known as direct costs) and costs of
activities
that indirectly support a program (known as indirect costs), such as
administrative activities, utilities, and building maintenance. To ensure
that
DOE programs are appropriately charged for incurred costs, contractors'
accounting systems assign the direct costs associated with each program
and collect similar types of indirect costs into pools and allocate them
proportionately among the programs.

Historically, DOE obtained information on contractors' overall indirect
cost
rates-the ratio of indirect costs to total operating costs-as a basis for
assessing contractors' efficiency in performing their missions. However,
the indirect cost rates of different facilities cannot readily be compared
(1)
because cost accounting standards and federal regulations provide
contractors with flexibility regarding the extent to which they identify
incurred costs directly with a specific program and how they collect
similar
costs into indirect cost pools and allocate them among programs and (2)
because of differences in the facilities' missions, corporate structures,
and

accounting systems.1 As a result, contractors' methods for accumulating
and allocating indirect costs vary-that is, a cost classified as an
indirect cost at one laboratory may be classified as a direct cost at
another. For example, electricity and other utility costs are usually
classified as indirect because they are not associated with a single
program; however, electricity costs could be charged directly if, for
example, a laboratory installs a meter to track the electricity
consumption in a building used solely by one program.

In the mid-1990s, DOE's chief financial officer (CFO) created 22 standard
categories of "functional support costs" to obtain more consistent
information about the support costs at DOE's major contractor-operated
facilities. These categories include, for example, executive direction,
information services, procurement, maintenance, and facilities management.
Each of the 22 categories is defined to cover all related costs,
irrespective of whether contractors classify them as direct or indirect.
Beginning in fiscal year 1997, the CFO and the Financial Management
Systems Improvement Council, composed of DOE and contractor financial
officials, have required the department's primary contractors to annually
report these costs. To oversee the quality of these data, contractors'
financial personnel generally peer review the data for each facility once
every few years. In fiscal year 2004, functional support costs accounted
for $7.2 billion, or nearly 40 percent, of the contractors' $18.1 billion
total costs. The functional support costs for the five largest DOE
laboratories were $2.9 billion.

You asked us to examine (1) recent trends in reported indirect and
functional support cost rates at the largest DOE contractor-operated
laboratories, noting any key differences in how the contractors determine
which costs are indirect and how these rates compare with those of similar
laboratories in other federal agencies, and (2) the efforts of DOE and its
laboratory contractors to reduce indirect and other support costs,
identifying additional opportunities for potential savings. In response,
we reviewed DOE's five laboratories with the greatest total operating
costs- NNSA's Lawrence Livermore National Laboratory, Los Alamos National
Laboratory, and Sandia National Laboratories; the Office of Science's Oak
Ridge National Laboratory; and the Office of Nuclear Energy's Idaho

1See the Cost Accounting Standards (48 C.F.R. Part 9904) for detailed cost
accounting requirements.

National Laboratory.2 The total costs of these five laboratories, $7.5
billion, accounted for more than a quarter of DOE's total fiscal year 2004
budget. For purposes of comparison with other federal laboratories, we
identified two similar large, contractor-operated laboratories-the Jet
Propulsion Laboratory, located in Pasadena, California, and operated by
the California Institute of Technology for the National Aeronautics and
Space Administration; and Lincoln Laboratory, located in Lexington,
Massachusetts, and operated by the Massachusetts Institute of Technology
for the Department of the Air Force. We determined that the selection of
these seven laboratories was appropriate for our design and objectives and
would generate valid and reliable evidence to support our work.

To examine recent trends in the indirect cost rates of the five DOE
laboratories, we obtained indirect cost rates from each laboratory for
fiscal years 2000 through 2004 and identified cost-rate trends, reviewed
differences in what types of costs the laboratories included in their
indirect cost pools and how they allocated these costs, and reviewed the
laboratories' indirect cost rates and those of the Jet Propulsion
Laboratory and Lincoln Laboratory. We did not include indirect cost rates
in this report because some of these data are proprietary. We also
examined the fiscal years 2000 through 2004 data that DOE's CFO published
in its Fiscal Year 2004 Support Cost by Functional Activity Report to
compare these costs for the five DOE laboratories. To examine the efforts
of DOE and its laboratory contractors to reduce indirect and other support
costs, we reviewed contractual provisions, key cost-saving initiatives,
and audits. Specifically, we reviewed the management and operating
contracts for each of the five laboratories, including clauses focused on
reducing functional support costs; analyzed several key initiatives that
DOE and its contractors have undertaken to reduce costs and the
initiatives' applicability to other DOE facilities; and examined reports
of DOE's Office of Inspector General and the contractors' internal audit
teams. We performed our work between January 2005 and August 2005, in
accordance with generally accepted government auditing standards.

2Prior to February 2005, Idaho National Laboratory was known as Idaho
National Engineering and Environmental Laboratory. DOE's Office of Nuclear
Energy is responsible for the laboratory, while DOE's Office of
Environmental Management is responsible for the environmental cleanup of
the site.

Results in Brief	For fiscal years 2000 through 2004, reported indirect
cost rates increased at two laboratories operated by DOE contractors and
decreased at three of them. Los Alamos had the largest reported increase,
10.4 percent, which it attributes to its July 2004 suspension of almost
all of the laboratory's operations in response to safety and security
concerns. Idaho had the largest decrease, 32.1 percent, primarily because
of changes in its accounting for indirect costs. Although a contractor's
overall indirect cost rates are generally comparable over time, some
rates, such as Idaho's fiscal year 2004 rate, are not comparable.
Specifically, in fiscal year 2004, the contractor at Idaho reclassified a
large portion of its indirect costs as direct to prepare for a new
environmental cleanup contract that is separate from its laboratory
operations contract. In addition, indirect costs cannot be compared across
laboratories because one contractor may classify a greater portion of a
particular support cost as an indirect cost than another contractor. For
example, from fiscal years 2000 through 2003, Idaho classified all of its
administrative support as indirect costs, while other laboratories, such
as Oak Ridge, classified administrative support as both direct and
indirect costs. In addition, Lawrence Livermore and Oak Ridge treat
maintenance for roads and grounds as indirect expenses, while Idaho treats
them as direct expenses.

Regarding functional support costs, three laboratories' rates increased
and two laboratories' rates decreased over the 5-year period we examined.
Again, Los Alamos had the largest increase, and Idaho had the largest
decrease. While functional support cost rates facilitate greater
comparability across laboratories, several DOE and contractor officials
told us that the definitions for some categories are unclear, leading to
confusion among categories. Notably, the "facilities management" and
"maintenance" categories are ambiguous and, hence, are not fully
comparable across laboratories. Because of the differences in how DOE
contractors categorize direct and indirect costs, and because only DOE
contractors report functional support costs, the rates at DOE's
laboratories cannot be compared with those of the Jet Propulsion
Laboratory or Lincoln Laboratory. For example, Jet Propulsion Laboratory
officials told us that they categorize all their costs as direct in
accordance with their contract with the National Aeronautics and Space
Administration.

DOE and its contractors have taken several actions to reduce indirect and
other support costs, but improved implementation could also produce
savings-particularly in the following areas of contract incentives,

contractors' employee benefits, deferred maintenance, and process
improvements:

o 	DOE's management and operating contracts have increasingly included
incentives to encourage cost reductions. For example, beginning in fiscal
year 2003, DOE contracts have placed greater emphasis on performance
objectives and fees based on the efficiency of laboratories' business
operations. In addition, in fiscal year 2004, NNSA began a pilot
"award-term" program that allows the contractor at Sandia to earn up to 5
additional years on its 5-year contract if it receives an overall rating
of "outstanding" each year and achieves cost savings sufficient to fund
completion of projects that were approved but did not receive full
funding. In the first year of the pilot program, Sandia's contractor
earned a 1-year extension on its contract. DOE has proposed to expand this
pilot program to the Los Alamos contract-and to allow contract extensions
of up to 13 years beyond the proposed 7-year contract term-even though the
award-term program is less than 2 years old and DOE has not yet evaluated
its effectiveness. For example, DOE has not evaluated whether the cost
savings achieved impaired the quality of work.

o 	DOE requires its contractors to benchmark the value of pension and
other benefit programs with those of industry and to reduce the value of
benefits if they exceed the overall benchmarked average by 5 percent or
more. However, DOE has exempted the University of California, which
manages Lawrence Livermore and Los Alamos, from benchmarking the value of
its benefits because those laboratories use the university's benefit
program. Of the three remaining laboratories, Idaho and Oak Ridge have
benefits whose values fall within the allowable range, while Sandia's
benefits have substantially exceeded the overall benchmarked average since
2002, with current pension benefits being 68 percent higher. DOE did not
request that Sandia's contractor propose corrective actions until May
2005, 3 years after discovering the benefits were too high. While
enforcing the limits on benefit values is a step in the right direction,
DOE has not set any limits on benefit costs. Because the value of the
benefits does not necessarily correlate with their costs, controlling the
benefits' value alone may not be the most effective means to manage costs.
Although DOE proposed in November 2003 to require contractors to evaluate
both the value and cost of their benefits, it has not yet finalized this
requirement, over 2 years later.

o 	After decades of neglect, DOE has begun to address a backlog of
deferred maintenance at its facilities to reduce support costs in the long
term. As required by DOE, the five laboratories have 10-year plans to
reduce their maintenance backlog, which was valued at $1.9 billion in
fiscal year 2004. However, without a more rigorous approach, the backlog
will persist well beyond 10 years. To speed the backlog reduction for NNSA
facilities-including Lawrence Livermore, Los Alamos, and Sandia-the
Congress has funded the first 3 years of a 10-year effort to "buy down"
the backlog. However, only Lawrence Livermore and Sandia have demonstrated
success with long-term approaches to further reduce their maintenance
backlogs and minimize reaccumulation. For example, Lawrence Livermore
charges DOE and other agency programs that use the laboratory about $8 per
square foot for maintenance in an effort to reduce the backlog.

o 	Idaho, Lawrence Livermore, and Sandia have used process improvement
programs to assess their operations and reduce costs. For example, Idaho
used process improvement methods to reduce the average cost of each of 16
safety assessment reports, for a total 2-year savings of $907,000. While
it is generally recognized that using a process improvement program is a
good business practice, neither Los Alamos nor Oak Ridge has one.

We are making several recommendations to the Secretary of Energy to
improve the comparability of functional support cost data among
laboratories and to reduce these costs by assessing the overall
effectiveness of initiatives and ensuring that other DOE laboratories
adopt important cost-saving initiatives. In commenting on a draft of this
report, DOE generally concurred with all of the recommendations.

Background	The Federal Acquisition Regulation and Cost Accounting
Standards provide overall requirements for allocating incurred costs
either directly to a program or indirectly to pools of similar types of
costs that are allocated proportionately among the programs. For example,
to avoid double counting and ensure that DOE programs and other federal
agencies pay an appropriate share of indirect costs, the standards require
that a contractor use consistent methods for estimating costs for each
project or activity. That is, if the laboratory charges one project $8 per
square foot for maintenance, it must charge other projects in the same
manner. Contractors submit for DOE approval accounting policy statements
describing how they will classify costs as direct or indirect.

DOE improved its ability to compare laboratories' costs in fiscal year
1997, when it began requiring contractors to report all functional support
costs, regardless of how they were classified. Functional support cost
data facilitate comparisons of laboratories' costs because they are
intended to include all costs that support laboratory missions, regardless
of whether a particular laboratory has classified the support costs as
direct or indirect. However, functional support costs have limitations in
that they cannot account for differences in the mission, size, age, or
location of DOE facilities. Facility comparisons need to factor in the
differences, for example, in (1) maintenance costs for a 50-year-old
manufacturing facility as compared with those of a modern research
facility or (2) safety and health costs at a facility that uses nuclear
materials as compared with a facility with no nuclear materials.

  Some Laboratories' Cost Rates Have Increased WhileOthers Have Decreased, but
  Not All Rates Are Comparable

From fiscal years 2000 through 2004, indirect cost rates increased at two
of the five DOE laboratories we examined and decreased at the other three
laboratories. A laboratory's indirect cost rates are generally comparable
over time, but the indirect cost rates of different laboratories are not
comparable because contractors often categorize costs differently.
Regarding functional support costs, three of the five laboratories' rates
increased, while rates of two laboratories decreased during the same
5-year period. While functional support cost data help DOE to compare
rates across laboratories, several DOE and contractor officials told us
that the definitions for some categories are unclear, leading to confusion
among categories. Finally, because of the differences in how DOE
contractors categorize direct and indirect costs, and because other
federal, non-DOE contractors do not collect and report functional support
costs, the rates at DOE's laboratories cannot be compared with those of
the Jet Propulsion Laboratory or Lincoln Laboratory.

    Two DOE Laboratories' Indirect Cost Rates Have Increased and Three Have
    Decreased, but These Rates Are Not Comparable across Laboratories

From fiscal years 2000 through 2004, indirect cost rates increased at two
laboratories operated by DOE contractors and decreased at three.3 Los
Alamos had the largest increase-10.4 percent. Nearly all of this increase
occurred during the 4th quarter of fiscal year 2004 and, according to Los
Alamos officials, was attributable to the "stand down" of activities that
the laboratory director ordered in July 2004 in response to a series of
safety and security incidents. In particular, the Los Alamos officials
told us that the stand down resulted in lower program costs without
similarly lower indirect costs. For example, $8 million in staff costs for
the stand down's first 2 days was treated as indirect costs as laboratory
managers developed plans for assessing and resolving the safety and
security issues. Once risk assessment and mitigation activities began,
stand-down costs were charged directly to benefiting programs. Los Alamos
officials stated that general and administrative costs, especially costs
in the "executive direction" category, were higher than expected, while
direct program costs were lower. The indirect cost rate for Lawrence
Livermore increased as well by 2.9 percent from fiscal years 2000 through
2004 because of additional costs, such as those related to facilities
safety, maintenance, environmental protection, and hazard control.

In contrast, Idaho National Laboratory's indirect cost rate decreased by
32.1 percent during the 5-year period, mainly because the contractor
reclassified a large portion of its costs associated with environmental
cleanup operations from indirect to direct in fiscal year 2004. According
to contractor officials, these costs were reclassified in response to
DOE's decision to split the management and operating contract, which
expired during fiscal year 2005, into two parts by awarding separate
contracts for laboratory operations and the environmental restoration of
the site. Laboratories can make changes in how they account for indirect
costs, and, when they do, they are required to document these changes in
their disclosure statement for DOE approval. The indirect cost rates for
Oak Ridge and Sandia decreased by 7.3 percent and 2.2 percent,
respectively, between fiscal years 2000 and 2004. According to Oak Ridge
contractor officials, the decrease resulted from several factors,
including management's decision to limit the growth of indirect costs
while the laboratory's total spending grew-when total costs increase at a
higher rate

3Because the Congress began providing funding specifically for security in
fiscal year 2001, we asked the laboratories to exclude security from their
prior indirect rates to provide consistent trend data.

than indirect costs, the indirect cost rate will decrease because the rate
equals the indirect costs divided by total costs.

Indirect cost rates cannot be meaningfully compared across laboratories
because one contractor may track costs more closely, allowing the
contractor to classify a higher proportion of the cost of a support
activity, such as administration, as a direct cost than another contractor
does. For example, from fiscal years 2000 through 2003, Oak Ridge and
Sandia classified administrative support costs as both indirect and direct
costs, while Idaho classified all administrative support costs as
indirect. Similarly, Idaho classified road and ground maintenance as
direct costs, while Lawrence Livermore and Oak Ridge treated them as
indirect costs. Two contractor officials provided other examples of costs
that laboratories are likely to classify differently: subcontract
administration, any type of fringe benefit, program management,
organizational management and administration, facility management and
maintenance, and information technology functions. A Lawrence Livermore
official noted, for example, that one contractor may treat desktop
software used by many programs as a direct cost, while another contractor
may bundle these software purchases into a common site license that is
paid from an indirect account. The indirect cost rate may be higher in the
latter case, but the goods may be obtained at a lower cost. Thus, higher
indirect costs do not necessarily equate with less efficiency.

Further, the five DOE contractors use different methods to classify
support activities-that is, they differed in how indirect costs are
collected and distributed. For example, one contractor used 4 major
indirect cost pools while another contractor used 12 indirect cost pools
for distribution among programs. Similarly, the number of major service
centers at the five DOE laboratories ranged from 6 to 14. Service centers
are accounts where costs of specific services are accumulated and charged
on the basis of services rendered, either to a program or to other
indirect cost pools. Common service centers are telecommunications and
computing centers.

Functional Support Cost Table 1 shows that for fiscal years 2000 through
2004, three laboratories' Rates Provide More functional support cost rates
increased and two laboratories' decreased. Comparability than Indirect
These rates are the functional support costs as a percentage of total
costs

without capital construction. Again, Los Alamos had the largest
increase,Costs, Although the 8.1 percent, and Idaho had the largest
decrease, 11.8 percent.4 Definitions for Some

    Categories Are Unclear

Table 1: Functional Support Cost Rates for Five Laboratories, Fiscal Years
                                    2000-04

Laboratory FY 2000 FY 2001 FY 2002 FY 2003 FY 2004

                                            Percentage change for FYs 2000-04

                Idaho  52.5%    52.1%    51.4%    51.7%    46.3%      (11.8)% 
             Lawrence                                              
            Livermore     32.4   34.5     35.1     36.1     34.0   
           Los Alamos     39.7   41.0     40.1     39.5     42.9   
            Oak Ridge     36.6   34.6     34.3     36.2     35.8        (2.2) 
               Sandia     33.5   33.5     32.5     34.7     33.7   

Source: DOE's CFO, Fiscal Year 2004 Support Cost by Functional Activity
Report (Washington, D.C.).

Note: This table excludes the "safeguards and security" category from the
functional support costs because DOE and the laboratories have treated
them as direct costs after they began receiving lineitem funding. We also
excluded "capital construction" from total costs. While these rates are
more comparable than indirect cost rates, they are not entirely
comparable, as discussed in this report.

Functional support costs were primarily developed to facilitate the
analysis of each facility's costs. While not intended for comparison
purposes, they provide more comparability across laboratories than
indirect costs because they are developed on the basis of standard,
defined cost categories. However, detailed analysis is required to
determine whether rate differences are the result of inefficiencies or
other factors, such as differences in each facility's mission, activities,
location, or size. For example, costs for safety and health, maintenance,
and utilities at Los Alamos are higher than costs at other sites because,
according to DOE officials, the laboratory has 2,224 facilities on 27,800
acres of mesas and canyons. Also, Los Alamos uses plutonium and other
hazardous materials, which require added safety procedures, and
accelerator facilities, which consume large amounts of electricity.

4Functional support costs accounted for about 40 percent of the 28
contractors' fiscal year 2004 operating costs of $17.4 billion.

While functional support cost rates facilitate improved analysis of
laboratory costs, several DOE and contractor officials told us that the
definitions for some categories are somewhat unclear, leading to confusion
in how to categorize certain costs. Notably, the "facilities management"
and "maintenance" categories are somewhat ambiguous and, hence, are not
fully comparable across laboratories. Peer reviewers checking for accuracy
of classification of costs have found that several laboratories have
misclassified costs between the two categories. In July 2003, peer
reviewers found that Sandia had classified $1.1 million of facilities
management costs as maintenance and $8.8 million in maintenance costs as
facilities management. In June 2004, peer reviewers found that Los Alamos
had classified $550,000 in maintenance costs as facilities management. In
addition, in fiscal year 2003, a Lawrence Livermore internal review found
that the laboratory had categorized plant facility engineering costs as
maintenance, while other laboratories had categorized these costs as
facilities management. In 2004, after discussing the categories with DOE
and contractor officials involved in reviewing functional support cost
data, Lawrence Livermore moved $15 million from the fiscal years 2002 and
2003 maintenance category to the facilities management category. Most of
the peer reviews for the DOE laboratories and other facilities found
difficulties with which cost elements were placed in or omitted from
"facilities management" and "maintenance," according to our analysis. In
addition, peer reviews at Los Alamos and Oak Ridge found over $2 million
of legal or information services costs that was misclassified in the
"executive direction" category, another example of a category whose
definition may be unclear. Idaho officials reported that discrepancies in
executive direction cost data between Idaho and other sites resulted from
uncertainty about how many levels of management or what type of site
development and strategic planning costs are to be included in the
executive direction category.

Differences in interpretation result from insufficiently detailed guidance
for developing functional support costs. The guidance primarily consists
of 10 pages of 22 support category definitions. DOE's Web site does not
have more detailed instructions that contractors can turn to when they are
uncertain whether a cost should be classified under one category or
another. Contractor officials developing cost data often turn to different
DOE or contractor officials with responsibility for these data for help,
increasing the likelihood of getting different advice, despite the fact
that consistency is key to data quality. Several DOE and contractor
officials with responsibility for these data agreed that more specific
guidance would cost little to develop and would increase consistency in
reporting. For

example, some officials said they could post a list of common laboratory
errors on the Financial Management Systems Improvement Council's Web site,
based on peer review findings of the past few years.

    DOE Laboratories' Cost Rates Cannot Be Compared with Those of Other Federal
    Laboratories

Because comparisons of indirect cost rates are not very meaningful and
non-DOE contractors do not report functional support costs, the cost rates
for DOE's laboratories cannot be compared with those of the Jet Propulsion
Laboratory or Lincoln Laboratory. The Jet Propulsion Laboratory, operated
by the California Institute of Technology, is the lead center for robotic
exploration of space. Virtually all of the work it performs is under a
single National Aeronautics and Space Administration agreement, which
states that all of the laboratory's costs are direct, according to
laboratory officials.5 Lincoln Laboratory, located on Hanscom Air Force
Base, conducts applied research to develop advanced technology in remote
sensing, space surveillance, missile defense, battlefield surveillance and
identification, communications, air traffic control, and biological and
chemical defense for the Department of Defense and other federal agencies.
The laboratory's indirect cost rate cannot be compared with those of DOE
laboratories without a detailed understanding of differences in (1)
contract provisions and other requirements; (2) how contractors classify
costs as direct or indirect; and (3) research missions and activities,
such as the added costs at the DOE laboratories associated with safety
requirements for handling radioactive and other hazardous materials.
Lincoln Laboratory's indirect cost rate increased by 13.9 percent between
fiscal years 2000 and 2004 because of a new enterprisewide accounting and
management reporting system and infrastructure improvements for laboratory
test facilities, according to laboratory officials.

5The Jet Propulsion Laboratory tracks "distributive costs," which are
similar to, but not comparable with, the DOE laboratories' indirect costs.
These distributive costs increased by 9.5 percent between fiscal years
2000 and 2004.

  DOE and Its Contractors Have Taken Actions to Reduce Indirect and Other
  Support Costs, but Opportunities Exist for Further Reductions

DOE and its contractors have numerous efforts under way to reduce indirect
and other support costs; however, we identified several efforts that could
be strengthened to further reduce costs. First, DOE is including
incentives in its contracts to encourage indirect cost reductions. DOE
officials stated that one of these incentives, a pilot to award additional
contract years for performance, had produced cost savings. DOE is
expanding this incentive to additional laboratories, although it has not
evaluated its effectiveness. Second, DOE generally requires contractors to
offer employee benefits that are similar in value to those of comparable
organizations, but the department has done little to enforce this
requirement. Third, DOE has begun requiring contractors to address a
backlog of maintenance projects while they also manage current maintenance
needs. Although this effort will involve costs in the near term, it could
reduce support costs in the long term. However, only Lawrence Livermore
and Sandia have programs that have been shown to be sustainable over
several years and appear to be promising models. Finally, while DOE and
some contractors have reduced costs through process improvement programs,
consolidated procurement actions, and audits by DOE's Inspector General
and DOE contractor audit groups, opportunities exist for further cost
savings through these activities.

    DOE Is Increasingly Using Contractual Incentives to Encourage Cost Savings
    but Is Expanding a Key Program without Evaluating Its Effectiveness

Recently, NNSA has taken several actions to improve business operations
and achieve support cost savings through contractual incentives. For
example, Sandia's management and operating contract that NNSA extended to
the Sandia Corporation, a subsidiary of Lockheed Martin Corporation, in
October 2003 gives higher priority to improved performance and greater
efficiency in business operations. Specifically, 40 percent of the
contract's annual award fee in fiscal years 2004 and 2005 is based on
Sandia's performance in areas such as information technology, procurement,
human resources, and maintenance. Similarly, NNSA's management and
operating contracts for Lawrence Livermore and Los Alamos have given
greater emphasis to improved performance and greater efficiency in
business operations. For example, the Los Alamos contract's performance
measures that focused on business operations increased from

about 28 percent in fiscal year 2002 to 40 percent in fiscal year 2005.
NNSA also has added provisions to some of its contracts to allow the
laboratories to reinvest cost savings in other activities considered to be
indirect costs.6

In addition, the Sandia contract initiated a pilot award-term program
under which an additional year may be awarded to the life of the contract
for each year the contractor achieves an overall outstanding performance
rating. A key performance target is finding sufficient cost savings to be
applied to unfunded projects.7 In fiscal year 2004, the first year of the
pilot program, Lockheed Martin earned a 1-year extension on its contract
and documented $38 million in cost savings, which it spent on agreed-upon
projects, such as the following:

o 	$14 million for reprogramming security and safeguards to meet the new
design basis threat,

o  $9.8 million for investing in computer clusters for defense projects,

o 	$3 million for purchasing equipment to refurbish the pulsed power
accelerator,

o  $2 million for enhancing the classified network,

o  $2 million for cleaning up beryllium contamination, and

o 	$0.3 million for negotiating an agreement with Russia on polymer
research.

6A January 2005 NNSA policy letter expanded the use of reinvested savings
to other approved areas. The final request for proposals for the contract
at Los Alamos allows a percentage of savings in indirect costs to be
applied to one of these approved areas- contractor-directed research and
development at universities and small technology companies in northern New
Mexico. Until recently, DOE and NNSA used a contract provision that
rewarded contractors for cost savings by allowing them to keep up to 25
percent of documented cost savings that DOE or NNSA officials had reviewed
and approved. However, NNSA officials stated that this provision was not
effective because the claimed cost savings could not be verified.

7Under the contract, Lockheed Martin shall apply cost efficiencies
achieved during a given fiscal year only to unfunded priority direct
mission work that NNSA and Lockheed Martin had agreed upon at the
beginning of the fiscal year, provided that this work is within the same
appropriation and budget and reporting category, unless NNSA approves a
formal reprogramming action.

Although Sandia and NNSA officials stated that they believe the awardterm
program emphasizes improved performance and cost savings better than
provisions in prior contracts, NNSA has not evaluated the nearly
2-year-old pilot. Such an evaluation could compare the benefits of
redirecting funds for better mission uses with the costs of forgoing
recompetition of the contract and examine whether the cost savings
resulted in any negative effects on reduced work quality. The evaluation
also could determine whether award-term incentives need to be revised in
other contracts to improve their effectiveness and sustainability,
particularly since the mission and level of performance among contractors
vary. By expanding the incentive without evaluating it, DOE does not know
if it is receiving benefits commensurate with awarding extra years to the
contract term. Despite the lack of evaluation, DOE's Office of Science
extended the award-term incentive to Lawrence Berkeley National
Laboratory, and NNSA plans to extend it to Los Alamos and the Nevada Test
Site later this year when it awards new contracts. As a result, Lawrence
Berkeley's contractor, the University of California, can potentially earn
up to 15 additional years on its recently awarded 5-year contract; the
request for proposals for the Nevada Test Site states that the contractor
can potentially earn up to 5 additional years on its 5-year contract; and
the request for proposals for Los Alamos states that the contractor can
potentially earn up to 13 additional years. (See app. I for more
information on this topic.)

    DOE Has Not Always Enforced Its Requirement That Contractors' Employee
    Benefits Be Comparable with Those of Similar Organizations

To ensure that the value of each contractor's employee benefits are
comparable with its competitors and that costs are reasonable, DOE Order
350.1 requires its management and operating contractors to periodically
benchmark the value of their employee benefit packages-including
retirement pensions, health care, death, and disability-with those of
organizations with whom the contractors compete in hiring employees.8 The
DOE order requires that if the value of a contractor's benefits exceeds
the average benchmarked value by more than 5 percent, the contractor will
provide DOE with a plan to adjust the benefits so that they fall within 5
percent of the benchmarked value. DOE must ensure that the contractor's
proposed adjustments are acceptable and reasonable. More specifically, the
DOE order requires that the contractors use a professionally

8The value of the benefit packages varies by laboratory, and the potential
liability for DOE may be substantial. For example, DOE estimates that, for
the five laboratories, its long-term liabilities for postretirement
medical and pension benefits are at least $2.9 billion.

recognized measure to compare the value of their benefits with those of
other organizations. The contractors can use a nationally recognized
consulting firm with expertise in benefit value studies to perform such a
study every 3 years or perform an annual employee benefit comparison
survey through the U.S. Chamber of Commerce. A benefit value study
determines the average of each benefit for 15 organizations with similar
workforces. The average value of the benefits becomes the benchmark
against which the contractor's benefits are assessed.

The benefit value studies conducted for Lawrence Livermore, Los Alamos,
and Sandia in 2004 show that the value of employee benefits for these
laboratories exceeded the benchmark by more than 5 percent in several of
the four primary categories of benefits. More importantly, the studies
showed that the overall benefits for those three laboratories exceeded the
allowable 5 percent variance for the overall benefits (see table 2).
Lawrence Livermore and Los Alamos both had benefit values that far
exceeded the benchmark and, in many categories, both laboratories exceeded
all comparators. For example, pension benefits for both laboratories
exceeded those of all 15 comparators and were nearly twice those of the
benchmarked value. Lawrence Livermore, Los Alamos, and Sandia were highest
or second highest in most benefit categories. Sandia's defined benefit
pension was second highest of all 15 comparators and exceeded the
benchmarked value by 68 percent. In contrast, the value of benefits for
Idaho and Oak Ridge did not exceed the 5 percent allowable range.

Table 2: Results of the DOE-Funded Employee Benefit Value Studies for Each
of the Five Laboratories

Benefits exceeded benchmark by more than 5 percent

        Laboratory Retirement Health care Death Disability All benefits

Idaho

Lawrence
Livermore X X X X

                               Los Alamos X X X X

Oak Ridge

Sandia X XXX X

Source: DOE.

Note: The health care and retirement values represent the benefits with
the highest average values, while the death and disability values
represent the benefits with the lowest values. All health care

values reflect both pre- and post-retirement health care values. We did
not independently verify the data used for the comparison studies.

DOE did not require prompt action to adjust benefits at the three
laboratories that exceeded the benchmarked value. Initially, DOE and,
later, NNSA exempted Lawrence Livermore and Los Alamos from the DOE
benchmark because the contract with the University of California, which
manages the laboratories, allowed the university to extend its own
benefits package to laboratory employees. When the DOE order was issued in
1996, the existing contract with the university took precedence, according
to a university document. DOE did not request a benefit value study until
2004, after it was determined that the Los Alamos and Lawrence Livermore
contracts were scheduled to be competed and DOE became concerned about
long-term liabilities for employees' postretirement costs. DOE has no
short-term liability for the pension benefits because the pension plan is
fully funded and is projected to remain fully funded for at least the near
term, according to university officials. However, DOE's liability for
longterm pension benefits for these laboratories remains undetermined. In
an effort to reduce this liability, NNSA is requiring a stand-alone
pension plan for the winning bidder of the Los Alamos contract, which DOE
plans to award at the end of this year.

Similarly, after a benefit value study in 2001 showed that the value of
Sandia's benefits exceeded the benchmarked value, NNSA did not require
Sandia to adjust its benefits. In this case, an actuarial study showed
that NNSA had minimal risk that it would have to contribute to Sandia's
pension plan for at least 5 years. However, the amount of long-term
liability is again undetermined because the study could not reliably
determine the risk of NNSA having to contribute beyond 5 years. When a new
benefit value study was completed 3 years later, in May 2004, NNSA
required Sandia to submit a corrective action plan to adjust the benefits.
NNSA received Sandia's plan for making adjustments in June 2005, but
officials are requiring modifications before approving the plan.

Finally, while benchmarking the value of benefits is a step in the right
direction, DOE does not require contractors to benchmark the costs of
their benefits. NNSA officials stated that cost studies are needed because
the value of benefits may not be directly proportional to their costs. For
example, while value and cost are generally highly correlated, it is
possible that a contractor may negotiate high-value benefits that have a
low cost, or low-value benefits that have a high cost. DOE has not yet
finalized the revisions to DOE Order 350.1, which includes a draft
provision to require benchmarking of costs, in addition to benefits. In
commenting on this

report, DOE officials stated that, since late 2004, DOE solicitations and
awards for management and operating contracts required contractors to
conduct benefit value and cost studies. In addition, our April 2004 report
(1) noted that DOE should review postretirement costs because they have a
continuous and compounding effect as they are paid out for each year of
retirement and (2) recommended that DOE strengthen its oversight of
postretirement benefits by focusing more attention on long-term costs.9

    DOE Has Begun to Focus on Deferred Maintenance, but Efforts at Some of the
    Laboratories Are Not Sustainable

For more than 2 decades, DOE orders and policies have required that
contractors' maintenance programs ensure the safe, reliable, and efficient
operation of buildings and equipment. They have also required that these
programs be adequately funded to ensure that the design requirements of
the buildings and equipment are met or exceeded for their operating lives,
and that industry standards for maintenance are applied.10 Industry
standards, for example, require that maintenance budgets be developed
using historical and other information on the resources required to
maintain the structure or equipment in good repair. Industry standards and
the National Academy of Sciences have recommended that day-to-day
maintenance requires continuous annual funding of about 2 percent to 4
percent of the replacement plant value.11

Despite requirements, for many decades, DOE and its contractors have
neglected the routine maintenance of buildings and equipment-including
inspection of fire alarms, upgrades to electrical systems, and testing of
equipment critical to the nuclear weapons program. DOE and contractor
officials stated that the mission always took priority for resource
allocations. This practice has resulted in a maintenance backlog that will
cost an estimated $1.9 billion for the five laboratories, excluding
deferred

9GAO, Department of Energy: Certain Postretirement Benefits for Contractor
Employees Are Unfunded and Program Oversight Could Be Improved, GAO-04-539
(Washington, D.C.: Apr. 15, 2004).

10For example, DOE Order 4330.4B, effective in 1994, specified that
sufficient resources should be budgeted to ensure the reliability, safety,
and operability of structures, systems, and components, and that
maintenance programs should meet equivalent industry guidelines.

11Building Research Board, Committing to the Cost of Ownership:
Maintenance and Repair of Public Buildings (Washington, D.C.: June 1990).
NNSA determines replacement plant value by a formula that estimates the
value of replacing structures and equipment at each site.

maintenance for unused buildings. Maintenance that continues to be
deferred, particularly on older structures and equipment, will contribute
to an increasing growth in deferred maintenance costs, including
replacement costs for certain equipment parts. For example, Lawrence
Livermore reports that its deferred maintenance costs continue to escalate
each year because of the higher probability of failure in the operability
of aging structures and equipment. In fiscal year 2004, Los Alamos
reported that it had the oldest structures of the three weapons
laboratories, with an average structure age of 33 years. Los Alamos also
reported that it had the highest level of deferred maintenance of the five
laboratories, accounting for about one-third, or $547 million, of the
backlog for the five laboratories.

The backlog could also jeopardize the safe, reliable, and efficient
operation of the buildings and equipment. At Los Alamos, for example, the
backlog put at risk the safety of some workers. Officials inspecting the
fire protection system in early fiscal year 2004 reported that the fire
sprinklers were not properly winterized to protect them from freezing, as
required by fire codes and standards. The inspectors reported that at
least three fire sprinkler pipes had frozen since November 2003, creating
a safety concern in the event of a fire. At Idaho, the backlog has placed
reliable operation of the Advanced Test Reactor at risk. Specifically, the
deferral of maintenance and recapitalization for several key
systems-including the waste control system and the digital monitoring
system-has resulted in the reliance on an outdated computer system for
which technical support is no longer available and replacement parts can
only be found in used parts markets. Loss of any these key systems could
require Idaho to temporarily shut down the Advanced Test Reactor,
hindering test plans and DOE's nuclear energy mission. Idaho plans to
continue replacing failing parts in the computer system until it can fund
a full system replacement.

DOE has begun to focus on deferred maintenance in an effort to reduce the
backlog and long-term costs. DOE's Defense Programs, the predecessor to
NNSA, first began to address the maintenance backlog in fiscal year 2000.
Defense Programs required the nuclear weapons facilities to develop 10year
plans to evaluate short-and long-term maintenance needs and develop
long-term efforts to better manage the maintenance backlog. At about the
same time, the Congress began providing funding for certain maintenance
programs, such as NNSA's Readiness in Technical Base and Facilities
Program. Some DOE laboratories also are using different types of indirect
costs, such as space charges, to help address maintenance. Additionally,
in fiscal year 2002, the Congress began funding the Facilities and
Infrastructure Recapitalization Program to reduce long-term deferred

maintenance for NNSA facilities. The program can also be used to demolish
certain structures. Demolishing or consolidating structures can help
reduce maintenance and other operating costs. Between fiscal years 2002
and 2005, the Congress provided about $1 billion for this program and, in
fiscal year 2005, NNSA requested an additional $1.7 billion for fiscal
years 2006 through 2009. When established, the program had two key goals.
First, contractors are to stabilize deferred maintenance by the end of
fiscal year 2005 so that there is no further growth in backlog. NNSA
informed the Congress that contractors have met this goal of stabilizing
deferred maintenance. The second goal is for contractors to reduce the
maintenance backlog so that the backlog for mission-critical and
nonmission-critical structures is less than 5 percent and less than 10
percent, respectively, of replacement plant value by fiscal year 2009.
However, an NNSA official stated that most of NNSA's contractors will not
be able to meet this goal because the $1.7 billion budget request for
fiscal years 2006 through 2009 was reduced by $574 million.

Furthermore, in fiscal year 2003, DOE adopted NNSA's lead by addressing
maintenance in 10-year plans to better manage maintenance overall,
including deferred maintenance. As with NNSA's long-term planning
requirements, DOE Order 430.1B requires that all DOE facilities develop
10-year plans to assess, among other things, their short-term and
long-term maintenance needs. Although the order does not have specific
time lines associated with reducing deferred maintenance, DOE's
headquarters program offices are responsible for approving the 10-year
plans and for tracking the performance of the facilities against the
plans. DOE's Office of Engineering and Construction Management has general
oversight over DOE Order 430.1B, including the review of the 10-year plans
and the tracking of program office performance.

Despite all of these efforts, DOE contractors report that deferred
maintenance will continue to be a significant problem in the short term
and long term. Of the five laboratories, Lawrence Livermore has
demonstrated the most sophisticated and sustainable approach that fully
funds maintenance needs and reduces deferred maintenance over the long
term. Lawrence Livermore's plan, first implemented in 1998, relies on a
combination of funding, including two directly funded NNSA programs- the
Readiness in Technical Base and Facilities Program and the Facilities and
Infrastructure Recapitalization Program-and an indirect cost pool
established for its Maintenance Management Program. The Maintenance
Management Program collects funds ($8 per square foot) from all users of
its buildings-including NNSA and other DOE and non-DOE programs.

About 20 percent of the collected funds are spent on deferred maintenance
at the laboratory, in accordance with the program. The program identifies
the most critical needs for maintenance on the basis of a matrix balancing
"mission-essential" and "probability of failure" criteria. In addition,
the charge itself encourages more efficient use of space and the return of
unneeded space for use for other purposes or demolition. Also, other
Lawrence Livermore cost pools, such as the Institutional General Plant
Project, help fund capital improvements that can help address maintenance
by upgrading structures and equipment. By managing the various elements of
its maintenance effort, Lawrence Livermore stopped the growth of its
maintenance backlog by fiscal year 2002. Despite the anticipated
reductions in the Facilities and Infrastructure Recapitalization Program,
Lawrence Livermore projects that it can reduce its deferred maintenance to
within NNSA's industry standards by 2011.

Sandia has an approach similar to Lawrence Livermore's, relying on a
combination of funds for addressing its maintenance and deferred
maintenance costs. Using an established methodology, a committee sets
maintenance priorities and determines how to spend funds collected from an
internal cost recovery pool. Sandia charges building users $12 per square
foot and expects to initiate increases to space charges over the next 4
years to compensate for expected cuts in maintenance funding. As a result,
despite the anticipated reductions in the Facilities and Infrastructure
Recapitalization Program, Sandia projected in March 2005 that it would
reduce its deferred maintenance to within NNSA's industry standards by
2011.

Los Alamos officials acknowledge that they cannot achieve NNSA's goals for
reductions in deferred maintenance given the expected cuts in the
Facilities and Infrastructure Recapitalization Program in the foreseeable
future. Contractor and NNSA officials acknowledge that Los Alamos'
maintenance program is not sustainable, particularly given its current
level of funding for maintenance. Any reductions in backlog as a result of
the Facilities and Infrastructure Recapitalization Program would
reaccumulate when that funding ended. Similarly, even though Oak Ridge
provides additional support for maintenance by charging for space used,
officials report that without additional infrastructure renewal or
recapitalization funds to address aging facilities, the charge is
insufficient to fully fund its maintenance needs. Without this additional
funding, Oak Ridge reports that its maintenance backlog will continue to
grow. Finally, Idaho has passed many of its deferred maintenance
liabilities on to another contractor responsible for cleaning up part of
the laboratory site; nevertheless, Idaho

reports that without additional funding, its maintenance backlog will
continue to grow.

    DOE and Some Contractors Have Reduced Support Costs through Process
    Improvement Programs, Audits, and Procurement Consolidation

Three Laboratories Have Reduced Costs through Process Improvement
Programs, While Two Laboratories Do Not Have Such Programs

Idaho, Lawrence Livermore, and Sandia have reduced support costs through
programs to improve business and other processes, while Los Alamos and Oak
Ridge do not have such programs. In addition, DOE and some of the
laboratories we reviewed have some efforts under way to consolidate
procurement actions. Finally, DOE's Inspector General and contractors'
internal audit groups have contributed to some cost savings by auditing
support functions.

Process improvement is the practice of taking an analytical look at
different steps that go into developing a product or delivering a service
and assessing how those processes could be conducted better. In a process
improvement initiative, participants may define the process and metrics,
measure performance, analyze root causes of problems, improve areas of low
performance, and develop controls for the process. Typically, the process
improvement initiative involves iterative cycles of identification and
improvement, fueled by employee participation in identifying problem areas
and recommending steps to improve them. It is generally recognized that
process improvement is a good business practice. Process improvement
programs can increase product or service quality, while decreasing costs.
Public and private organizations have reported significant returns on
investment through process improvement programs.

Several DOE laboratories and production facilities, including Idaho,
Sandia, and Lawrence Livermore, have used process improvement methods,
such as Six Sigma and Lean Six Sigma, to reduce costs.12 Six Sigma and
Lean Six Sigma are rigorous and disciplined methodologies that use data
and statistical analyses to measure and improve the performance of a
company's operations by identifying and eliminating defects in
manufacturing and service-related processes. Idaho used Six Sigma to
reduce the average cost of 16 Safety Assessment Reports for a total
savings of $907,000 from fiscal years 2002 through 2003. Sandia used Lean
Six Sigma to streamline its accounts payable purchase order process,
potentially leading to savings of more than $1 million over a 5-year
period.

12Other commonly used process improvement methods include Total Quality
Management and ISO 9000.

Finally, Lawrence Livermore used Six Sigma and other methods to improve
processes in the areas of safety, security, resource management, and
property management. For example, the laboratory reduced security staff
overtime by increasing the workday to 12 hours and reported a $2.3 million
annual savings. The laboratory also standardized the hours of the
perimeter security gates and closure of redundant gates in an effort to
save $300,000 annually. Lawrence Livermore plans to train 400 managers by
the fall of 2006 in a new process improvement method based on a
combination of Lean Six Sigma and a method used at the United Kingdom's
Atomic Weapons Establishment, which provides warheads for nuclear
deterrence.

Neither Oak Ridge nor Los Alamos have process improvement programs,
although some DOE and laboratory officials agreed that such programs can
help reduce costs. When laboratories do not have some type of process
improvement program, DOE has little assurance that managers are
identifying and addressing inefficient or ineffective processes.

Many of the DOE Inspector Many of the DOE Inspector General's reports have
examined support

General's Reports Have activities at DOE laboratories and other
facilities, including security,

Reviewed Support Activities	procurement, property management, information
resources, financial management, and financial controls. The Inspector
General has found opportunities for reducing indirect and other support
costs. For example, in 2003, the Inspector General's audit of central
office expenses for the Thomas Jefferson National Accelerator Facility
questioned $4.6 million in costs claimed by and paid to the contractor for
central office expenses from November 1999 to September 2002. These
questioned expenses included costs that were not allowable, such as
alcoholic beverages, and costs that were not adequately supported or
documented. The audit resulted in $3.5 million in savings to DOE. In 2003,
an audit of Los Alamos reported support and other costs of $14.6 million
that were potentially unallowable, including meals, excessive travel
costs, and an internal audit function that did not meet DOE requirements.
DOE officials are in the process of determining allowability of the
questioned amount. In addition to the Inspector General audits,
contractors' internal auditors annually perform an audit of the
allowability of costs as claimed by contractors.13 This audit addresses
both indirect and direct costs, identifying whether or not costs are
allowable under the terms of the contract. For example, one such audit,
conducted at Sandia in fiscal year 2003, projected $112,000 in

13The Inspector General, the cognizant auditor for DOE's management and
operating contractors, supplements its audit program with each
contractor's internal audit activities.

unallowable costs-DOE is in the process of determining the precise amount.

DOE Has Consolidated Some DOE and its management and operating contractors
have reduced support

Procurement Actions	and other costs by consolidating their procurements of
computer equipment, office supplies, and other bulk items into single
contracts that result in volume discounts and the need for fewer
purchasing resources. In particular, DOE established an Integrated
Contractor Procurement Team that pursues buying opportunities and has
negotiated over 40 purchasing agreements with vendors to reduce
procurement staff time and obtain favorable prices, according to a DOE
official. The team, composed of contractor purchasing agents, surveys
sites to determine what contractors are paying for a product and whether
they can negotiate a better price. A contractor that wants to purchase
goods (e.g., paper) can review a list of agreements and supplier contact
information on the Integrated Contractor Procurement Team's Web site and
use its Basic Order Agreement. This saves contractors and suppliers time
because they do not need to renegotiate all the terms of the contract.
Contractors do not track savings resulting from the use of Integrated
Contractor Procurement Team, but they have collected some examples of
savings that total about $12 million to $15 million, annually.

Other efforts to save time and money through consolidation of procurement
actions include the following:

o 	NNSA has required Lawrence Livermore, Los Alamos, and Sandia to begin
analyzing their spending using similar software so that opportunities for
consolidated purchases among the three laboratories can be identified.

o 	Lawrence Livermore has hired a specialist who is focused full-time on
analyzing the laboratories' spending patterns and identifying
opportunities for consolidation. As a result of opportunities being
identified, the procurement office expects to reduce the number of
subcontracts by 10 percent, saving procurement staff time. The laboratory
also has developed 13 Electronic Ordering System agreements through which
suppliers provide electronic catalogs with over 2 million commercial items
(e.g., computers, electrical and plumbing supplies, and chemicals) that
laboratory employees can purchase online. The purchases are automatically
routed through an approval system.

o 	Oak Ridge consolidated its new radio system, which is used by security
and maintenance staff, with other DOE facilities in the area, reportedly
saving $900,000 on purchase prices plus $475,000 in annual costs.

Laboratory officials told us that opportunities exist for further cost
savings through reduced staff time and better prices by consolidating
procurements with each contractor's parent organization, other DOE
facilities, or other federal agencies' facilities in the same region.
Idaho officials cited the potential for reducing costs for using
consolidated procurements with nearby military bases and other facilities.
An NNSA procurement official noted that laboratories could save through
increased use of Integrated Contractor Procurement Team agreements, noting
that some procurement officials may not be aware of available
opportunities.

Conclusions	In an era of federal budget constraints, it is crucial to
efficiently manage support costs at DOE laboratories, thereby maximizing
funds available for laboratory missions. DOE and its contractors have
taken steps to reduce support costs, but additional opportunities exist.
To help decision makers analyze support costs across the laboratories,
several years ago DOE began to require laboratories to report functional
support costs. However, peer reviews have revealed some problems with
these data and challenges remain in comparing costs, in part because of
ambiguity in the definitions for some categories of support costs. To
encourage contractors to reduce support costs, DOE also piloted a program
to award contract years for performance improvement and cost savings.
However, by expanding the pilot incentive program without evaluating it,
DOE does not know if it is receiving benefits commensurate with awarding
extra years to the contract term. In another effort to save money, DOE
developed requirements to ensure that contractors' employee benefits are
comparable with those of similar organizations with whom they compete for
critically skilled staff. However, DOE has not always required its
contractors to reduce employee benefits that substantially exceed the
value of their competitors' benefits and has not required contractors to
benchmark the costs of their benefits, potentially adding billions of
dollars in long-term costs. Furthermore, while DOE has begun to address
the $1.9 billion backlog of deferred maintenance to reduce long-term costs
and improve the safe, efficient, and reliable operation of equipment and
buildings, only Lawrence Livermore and Sandia have demonstrated
sustainable approaches that successfully reduced their backlogs. Lastly,
process improvement programs are generally considered a good business
practice, and three laboratories have reduced costs through their own
programs. DOE, however, does not

require laboratories to have such programs, and some laboratories have not
instituted one. Overall, while DOE has made progress, without additional
attention to these initiatives, the department may miss the opportunity to
produce significant savings.

  Recommendations for Executive Action

To improve the quality and comparability of DOE facilities' support cost
data, we recommend that the Secretary of Energy direct the CFO to work
with the Financial Management Systems Improvement Council to clarify
definitions of functional support cost categories.

To determine whether the department receives benefits commensurate with
awarding 1 or more extra years to the contract term, we recommend that the
Secretary of Energy direct NNSA to evaluate the effectiveness of its pilot
award-term program at Sandia National Laboratories, particularly the
nature and extent of work quality improvements, prior to extending the
program to other laboratories.

To provide competitive but economical employee benefits, we recommend that
the Secretary of Energy complete the revision of DOE Order 350.1 and
ensure that the order (1) extends the requirement to benchmark the value
of employee benefits to all contractors; (2) requires prompt corrective
action if the value of benefits exceeds the allowable range; and (3)
extends the benchmarking requirements to include the costs, as well as the
values, of the benefits.

To reduce long-term maintenance costs at contractor-operated facilities,
we recommend that the Secretary of Energy develop a long-term sustainable
approach that meets day-to-day maintenance requirements, reduces the
maintenance backlog, and minimizes its reaccumulation.

To facilitate the further reduction of support costs, we recommend that
the Secretary of Energy require that each DOE management and operating
contractor implement a process improvement program that routinely assesses
the efficiency and effectiveness of business practices and other
operations.

Agency Comments	We provided DOE with a draft of this report for its review
and comment. In written comments, DOE generally concurred with our
recommendations.

(See app. II.) DOE also provided a number of technical comments, which we
incorporated in this report as appropriate.

  Scope and Methodology

To examine indirect cost-rate trends at each of the five largest DOE
laboratories, we analyzed each laboratory's indirect cost-rate data for
fiscal years 2000 through 2004. Specifically, we interviewed laboratory
financial officials and analyzed documents to determine how each
laboratory calculated its overall indirect costs rate by (1) classifying
costs as direct or indirect and (2) collecting indirect costs into pools
and distributing them among other cost pools or directly to program
sponsors. Furthermore, because DOE and its management and operating
contractors use functional support cost data to help assess certain
activities, we analyzed these rates for the five laboratories for fiscal
years 2000 through 2004, along with the results of peer reviews performed
at each laboratory.

We surveyed laboratory financial officials on the reliability of the
indirect and functional support cost data, covering issues such as data
entry access, quality control procedures, and the accuracy and
completeness of these data. Follow-up questions were added whenever
necessary. In addition, we reviewed all data provided by the laboratories,
investigated instances where we had questions regarding issues such as
categories or amounts, and made corrections as needed. On the basis of
this work, we determined that the financial data provided were
sufficiently reliable for the purposes of our report. We presented
percentage changes in the overall indirect cost rates for fiscal years
2000 through 2004 that the five laboratories reported. However, because of
limitations discussed in this report, analyses of increases or decreases
in these rates mean little without a careful analysis of how a contractor
classifies costs, and one contractor's rates cannot be compared with those
of others. Moreover, we analyzed changes in classification of costs or
changes in mission over time to determine if these data were comparable
over several years at a single laboratory. We noted in our report when we
found changes in classification that affected the comparability of data at
a single location over time.

To assess the efforts of DOE and its laboratory contractors to reduce
support costs and identify additional opportunities for savings, we
visited each of the five laboratories to interview senior managers and
obtain supporting documentation, interviewed DOE officials, and examined
prior GAO and DOE Inspector General reports. In the course of this work,
we identified opportunities for further potential cost savings. We then
interviewed cognizant laboratory and DOE officials about actions taken to

address each opportunity and reviewed supporting documentation.
Specifically, to review DOE's contract provisions, we interviewed DOE
contracting officers and obtained information about recent contractual
provisions that emphasized potential cost savings or improved efficiency.
To address DOE's liabilities related to employee benefits, we analyzed
benefit value studies for each of the laboratories and compared the
results of our analysis with the department's requirements. We also
interviewed DOE, NNSA, and contractor officials to clarify the results of
the studies and to provide us with documentation on proposed resolutions,
as appropriate. It was not our intent to verify, nor would we have been
able to independently verify, the accuracy of actuarial calculations,
assumptions, or data used in the comparison studies due to the proprietary
nature of benefits that consulting firm databases used to conduct the
studies.

To address the maintenance backlog, we analyzed data from DOE's Office of
Engineering and Construction Management, which collects information on
deferred maintenance. We also analyzed each laboratory's 10-year plan and
related documents and compared the results of our analysis with NNSA's and
the department's requirements. We spoke with cognizant officials at DOE
headquarters and its site offices and with laboratory managers to verify
the results of our analysis and to determine the actions being taken to
address the backlog. To examine the laboratories' use of process
improvement programs, we reviewed examples that the three laboratories
provided of improved effectiveness and of reduced costs for their business
operations and interviewed senior managers at the other two laboratories
regarding what they had done to improve business operations.

As agreed with your offices, unless you publicly announce the contents of
this report, we plan no further distribution of it until 30 days from the
date of this letter. At that time, we will send copies to the Secretary of
Energy; the Director, Office of Management and Budget; and other
interested parties. We will also make copies available to others on
request. In addition, the report will be available at no charge on the GAO
Web site at http://www.gao.gov.

If you or your staffs have any questions about the report, please contact
me at (202) 512-3841. Contact points for our Offices of Congressional
Relations and Public Affairs may be found on the last page of this report.
GAO staff who made key contributions to this report are listed in appendix
III.

Jim Wells Director, Natural Resources and Environment

Appendix I

                    Award-Term Provisions at Four DOE Sites

The Department of Energy (DOE) introduced award-term incentives as a pilot
program to Sandia in fiscal year 2004, and then expanded the use of the
incentives to the Lawrence Berkeley contract and to the final request for
proposals for Los Alamos and the Nevada Test Site, as shown in table 3.
The language in the final request for proposals for Los Alamos, if placed
in the contract, would award additional years to the contract term if the
contractor (1) achieves a certain level of performance to be determined by
the DOE contracting officer and (2) meets other conditions, such as
finding cost savings and using these savings to adequately perform
approved work. As shown in table 3, Los Alamos can earn 13 additional
years on its 7-year contract.

    Table 3: Award-Term Provisions for Four DOE Facilities Final request for
               proposals Existing contracts Provisions Los Alamos

Nevada Test Site

Lawrence
Berkeley Sandia

                      Duration of contract (years) 7 5 5 5

Total number of years that 13 5 15 5 can be added to contract

Possible total number of years of contract with award term

      20 10 2010 Level of performance Annual Based on First 3 Outstanding

needed to be assessed at to earn award term

determination by NNSA

incentive fee earned

years: satisfactory

Subsequent years: Outstanding

                                   each year

Requires cost savings to Yes No No Yes earn award term

Source: DOE.

Key differences between the Los Alamos final request for proposals and the
Sandia contract are as follows:

o 	The level of performance required to receive additional contract years
can be determined by the contracting officer for Los Alamos, but Sandia's
contract requires outstanding performance. NNSA officials stated that the
recent history of performance at Los Alamos may take several years to
reverse, and the contractor may not be able to achieve

Appendix I Award-Term Provisions at Four DOE Sites

overall "outstanding" ratings in the initial years of the contract. The
officials also noted that the flexibility may allow the contracting
officer to focus on certain areas for improvement, such as business
operations.

o 	The contract term for Los Alamos could be 20 years, rather than the 10
years allotted to Sandia. The 20-year term for Los Alamos requires a
deviation from DOE Acquisition Regulation 970.1706-1, which limits DOE
contracts to 10 years.

                                  Appendix II

                     Comments from the Department of Energy

Appendix III

                     GAO Contact and Staff Acknowledgments

                      GAO Contact Jim Wells (202) 512-3841

  Staff Acknowledgments

(360536)

In addition to the contact named above, Richard Cheston, Beverly Peterson,
Robert Sanchez, James Espinoza, and Cynthia Norris made key contributions
to this report. Also contributing to this report were Chuck Bausell,
Virginia Chanley, Nancy Crothers, Alison O'Neill, and Omari Norman.

  GAO's Mission

Obtaining Copies of GAO Reports and Testimony

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 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

  To Report Fraud, Contact:

Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm

E-mail: [email protected] Programs Automated answering system: (800)
424-5454 or (202) 512-7470

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

Public Affairs 	Paul Anderson, Managing Director, [email protected] (202)
512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

                           PRINTED ON RECYCLED PAPER
*** End of document. ***